## Svalgaard #6

1. Posted May 1, 2008 at 9:11 AM | Permalink

606 (Sam): Because you do not follow our convention of supplying a reference number to an earlier post, I have no idea what you talking about. If the 5 W/m2 is with reference to the uncertainty of the absolute accuracy of TSI, then so be it. The people that measure this can’t get their story straight to any better than that. for the models, for our understanding of this, for ANY practical purpose I can think of, the 5 W/m2 make no difference, no matter how many square meters are involved.

2. Posted May 1, 2008 at 10:14 AM | Permalink

Erl: what confuses me are these two quotes:

599: My top graph simply points out that there is indeed a negative and a positive phase and it corresponds to which is dominant, El Nino or La Nina, or as I would prefer to describe it, generalised warming in the tropics or generalised cooling.

601: A cool Pacific (and global tropical waters) is described as negative PDO

In the first you say negative [I presume cold] PDO is warming, in the second it is cooling.

3. Sam Urbinto
Posted May 1, 2008 at 10:14 AM | Permalink

Leif #1 Sorry, I quoted Nasif # 594

Dennis, Leif, Sam,
Just try to get the energy caught by a square meter of grassland considering 1360 W/m^2 and 1360.5 W/m^2 at the top of the atmosphere. 5 W/m^2 against -0.1523 W/m^2 makes an important difference (a priori), true?.

I think it was a comparison between calculations involving 1360 versus 1366 in the real world.

4. Sam Urbinto
Posted May 1, 2008 at 10:16 AM | Permalink

Me #2 His next post was #595

Thus, I prefer to do my evaluations on best terrains for growing veggies with 1360.5 to 1361.5 W/m^2 than using the awful “normal” 1366.5 W/m^2. I need positive numbers hitting on my crops. :)

5. Posted May 1, 2008 at 10:36 AM | Permalink

4 (Sam): I think it was [as yours] just a botched attempt at levity. The problem is that on these blogs [present company excepted, of course] some posts are just so ridiculous that it is hard to tell if that is voluntary or not.

6. Sam Urbinto
Posted May 1, 2008 at 10:58 AM | Permalink

Leif #5; I think he’s just saying it messes the calculations up. Of course, in terms of growing things, if it’s 1500 or 1000 is not important as long as your stuff grows. :) But it can mess up the calculations, not that we have any control of what it really is.

On the other hand, if you just start at zero and compare +5 with -.15 there’s a difference. If you put that on a 1360 floor it becomes a bit less important.

Two different subjects it seems. :D

7. Dennis Wingo
Posted May 1, 2008 at 11:17 AM | Permalink

(3) Sam

Just try to get the energy caught by a square meter of grassland considering 1360 W/m^2 and 1360.5 W/m^2 at the top of the atmosphere. 5 W/m^2 against -0.1523 W/m^2 makes an important difference (a priori), true?.

This is not how it works on the ground. Solar irradiance at the surface is dependent upon the amount of water vapor in the air, pollution, and the altitude of the site. For example, I can point you to the Department of Energy’s site where they have radiometers that measure the total flux (TSI) at sites all over the country. The “standard” Air Mass 1 (AM1) Irradiance is 1000 W/m2 but in the solar industry there is another standard that puts the number at 800 W/m2 based upon measurements that have been taken. It is my experience that the irradiance values vary from about 875 to 1000 W/m2. This varies on an hourly, daily, seasonal, and yearly basis. In that variance, 5 W/m2 does not matter to the ground. Now I would posit that it does, at least somewhat over the entire sunward face of the atmosphere.

8. Posted May 1, 2008 at 11:25 AM | Permalink

6,7 (Sam,Dennis): Hey, hey, hey! the Sun did not change 5 W/m2, just our instruments, you guys are arguing about what is hotter 50F or 40C.

9. Sam Urbinto
Posted May 1, 2008 at 11:28 AM | Permalink

Dennis #7 Acutally that’s Nasif #594 :)

Sure, 5 out of 900 +/- 100 isn’t much. And if you are at 900, that is only about 2% Unless, like you said if it’s over the entire surface….

Sort of like increasing something in the atmosphere by .0001 in 150 years. :D

10. Bill
Posted May 1, 2008 at 11:30 AM | Permalink

Re: Amt of downwelling solar

As someone who makes surface measurements of the energy balance, Summertime, noon, clear-sky, downwelling shortwave irradiance is roughly 800 to 100 W/m2 depending on where you are. However, this number could easily be less than half that value due to clouds

11. See - owe to Rich
Posted May 1, 2008 at 1:18 PM | Permalink

Leif,

You may be interested in my new article on the Hale cycle. Here is the summary: a statistically significant difference is demonstrated between mean-adjusted temperatures in even and odd solar cycles, with the variation during an odd cycle being the greater.

Steve Mc forgot to provide a link at the top of this thread to Svalgaard #5 so here it is better late than never.

Rich.

12. Erl Happ
Posted May 1, 2008 at 2:02 PM | Permalink

605 (Bender )

the illusion of lead results from differences in magnitude of heating.

Hey, where is the heating happening?

Which leads- which lags? What is the cause? Does the tail wag the dog? This is the confounding puzzle in climate science. Unfortunately the question is frequently ignored or even wilfully dismissed. E.g. the episodic heating and cooling of the tropics is seen as ‘an internal oscillation in the climate system’. If it is so, the result should be temperature neutral. Plainly, it is not.

If one insists on this proposition the question must be posed. By what mechanism does the climate system suddenly determine that more or less heat shall be taken into the Pacific? Or, by what mechanism does the Pacific determine that for the next thirty years there will be a colder than normal current running down the west coast of North America? Does the Arctic autonomously cool? Does East Asia such the energy out of the northwards trending waters?

13. Erl Happ
Posted May 1, 2008 at 2:20 PM | Permalink

2 (Leif)

599: My top graph simply points out that there is indeed a negative and a positive phase and it corresponds to which is dominant, El Nino or La Nina, or as I would prefer to describe it, generalised warming in the tropics or generalised cooling.

Sorry, I see your point. The sequence is important and I had it out of order. It shoud be ‘Negative PDO is promoted by generalised cooling in the tropics’.

We are being told that negative PDO will have consequences for weather in North America, even the entire Northern Hemisphere. The important question to address is how we get this generalised cooling in the tropics that is in turn, in my view at least, (not Benders apparently) responsible for the negative PDO. At this time the middle troposphere is about 4°C cooler than it has been for most of cycle 23. The lower troposphere is about 0.5°C cooler. This promotes more cloud and enhances albedo. My contention is that the difference relates directly to sunspot activity and the middle troposphere leads. In other words, that is where the change is mediated. Yours?

14. bender
Posted May 1, 2008 at 5:40 PM | Permalink

#12 Erl
1. Where is the heating happening? The ocean surface largely, I would assume. Why?
2. lead or lag (ENSO vs PDO): what do you prefer, chicken or egg? My point is that ENSO is strong and PDO is weak. So lead-lag is a red herring.

15. Posted May 1, 2008 at 6:16 PM | Permalink

I hope my last posts are not what Leif has addressed like ridiculous. Perhaps they are.

From Leif’s work, the Solar Irradiance has not increased so much in the last two decades… That’s true. Coincidentaly, the tropospheric temperature has not increased in the last nine years. Perhaps the Sun is the real driver of climate?

16. Posted May 1, 2008 at 6:24 PM | Permalink

15 (Nasif):

I hope my last posts are not what …

which ones? :-)

17. Posted May 1, 2008 at 6:25 PM | Permalink

Can’t they see that every minimum change is “masked” immediately by “mother nature”? Well, the solar activity has decreased, and so the climate change… ;)

18. Posted May 1, 2008 at 6:37 PM | Permalink

# 16

Leif,

My last posts about the convenience of managing a range of solar irradiance between 1360.5 to 1365.5 W/m^2 like normal or standard instead 1366.5 W/m^2. I find it is difficult to work with negative numbers when modeling the possibilities of a land for crops. Of course, this is not the only factor included. You, Earl, Dennis and Sam have indicated other environmental factors that we are considering also, but for our work it is easier to work with positive numbers than with negative numbers, and when we try with the standard “1366.5 W/m^2″ the output is always negative. For example, in July, the solar irradiance in our latitude is around 21892 kJ/m^2 per day. If I make my calculations considering 1360.5 to 1365.5 W/m^2 my numbers are positive… almost always. Perhaps that’s trivial for other applications, but it is very important for the evaluation of insolation above tomatoes and potatoes.

19. Posted May 1, 2008 at 6:47 PM | Permalink

# 18

Leif,

Yes, I know it applies for the last four hundred years, but I collated your work with Lockwood’s paper. Sorry… :)

Regarding insolation, if I use 1366.5 W/m^2 the output is -0.3 W/m^2*s. It would be easier working with 0.3 W/m^2*s. I hope I have explained it correctly. ;)

20. Posted May 1, 2008 at 7:01 PM | Permalink

18 (Nasif): “negative numbers”. I’m a bit lost. Explain again where they come from. I must have had a ‘senior’ moment.

21. Posted May 1, 2008 at 8:30 PM | Permalink

# 20

Leif,

I’m doing a graph that could explain what I’m referring to. Please, give me some time. Thanks for your interest on helping to me.

22. Posted May 1, 2008 at 10:05 PM | Permalink

Dennis, Erl, Sam, Leif, cbs, etc.,

Here the graph is:

This is a very coarse work because I was pressed by time. Leif, I used the data from the second database of Judith Lean, for no special reasons. I would have used your database, but you have not answered my message. I only hope that this graph says more than thousand words. Thanks to all you for your patience.

23. Posted May 1, 2008 at 10:11 PM | Permalink

# 22

Oops! Perhaps, you’ll find something interesting in that graph. Dinner time… My next program of radio will turn on virial algorithms…. Hehehe… ;)

24. Posted May 1, 2008 at 10:15 PM | Permalink

# 21

Me, again… I apologize, I didn’t convert W/m^2 to cal per minute per square cm; if you do the conversion you’ll find it worst.

25. cba
Posted May 2, 2008 at 6:58 AM | Permalink

22 (Nasif):

Your total kJ/m^2 apparently looks fairly reasonable for the energy coming in during a day. However, your chart is indicating something a long way from what it should be indicating, value wise. It looks like an anomoly value of insolation chart rather than of insolation.

quick averages for a general order of magnitude reasonableness check:

1366 = 684 w/m^2 averaged over the surface of the earth for 12 hrs per day lit hemisphere.

684/1.7 = 400 w/m^2 = a rough (average) value during a 12 hr day for a moderate latitude

this gives 17280 kJ/m^2 per day versus the 21892 kJ/m^2 you mentioned – which is probably somewhat better – but i’m rushed for time now and we want an order of magnitude.

the value also needs a reasonable albedo fraction removed.

however, your numbers say 0-10 w/m^2 when they should be 100- 400 w/m^2

you need to explain why your graph y-axis is so vastly different from what I (we) are understanding it should be.

26. Posted May 2, 2008 at 8:09 AM | Permalink

# 25

cba,

It looks like an anomaly value of insolation chart rather than of insolation.

Yes, it is an anomaly of insolation chart.

however, your numbers say 0-10 w/m^2 when they should be 100- 400 w/m^2 you need to explain why your graph y-axis is so vastly different from what I (we) are understanding it should be.

We use another theorem and consider albedo when calculating the annual total income of energy; then, the incident solar radiation on ground acquires the values you are asking about. For example, in 1987 the total fixed by tomatoes was around 83000 Kcal/m^2/year.

27. Posted May 2, 2008 at 8:41 AM | Permalink

26 (Nasif): fascinating: tomatoes

28. Posted May 2, 2008 at 8:50 AM | Permalink

# 27

Leif,

(OT) Tomatoes grow faster than potatoes… ;)

29. Posted May 2, 2008 at 9:42 AM | Permalink

# 28

Leif,

Now you see the reason by which your work is very important for us, even for humble biologists like me. We have a 9 hc terrain where we grow greenbeans, corn, tomatoes, chili, etc. The spaceweather forecasts are useful also. ;)

30. cba
Posted May 2, 2008 at 10:04 AM | Permalink

Nasif

It sounds like yet another factor of interest might be spectrum. Not all radiant energy incoming is of interest to plants (as I understand it).

What’s more, I would expect massive variations due to moisture content (rainfall) and fertilization of soil along with the distribution of that insolation over the time frame – not to forget cloud fraction effects – how much cloudy vs clear skies are there.

In California, harvest season for tomatoes ends around late aug. early sept. (with the first rains as that generates some sort of blight or fungus with the remaining tomatoes (water mold?). (memories from a mis spent youth)

31. Posted May 2, 2008 at 10:41 AM | Permalink

# 30

cba,

It sounds like yet another factor of interest might be spectrum. Not all radiant energy incoming is of interest to plants (as I understand it).

What’s more, I would expect massive variations due to moisture content (rainfall) and fertilization of soil along with the distribution of that insolation over the time frame – not to forget cloud fraction effects – how much cloudy vs clear skies are there.

You are right… all those factors your talking about are taking into account in models, besides the important insolation anomalies that can be deduced more appropiately from Leif’s database. Moisture and cloudiness are considered in the same moment they occur (we have not good forecasts for humidity and cloudiness yet).

In California, harvest season for tomatoes ends around late aug. early sept. (with the first rains as that generates some sort of blight or fungus with the remaining tomatoes (water mold?). (memories from a mis spent youth)

In my location… well, not my location but that of the terrain, harvest season for tomatoes lasts about 20 weeks, from April-May to August-September. Rain of excessive humidity can cause the lost of the whole crop. We have caterpillars, flies, snails, fungus and virus affecting our crops; however we have had good control of them almost always. We found a good region at the Anahuac Plateau (coordinates 24° North Lat. /103° 01′ Long.); unfortunately the warm season ends too early and the harvest season is limited to only 10 weeks (May to July-August); once a year because most of our veggies doesn’t survive through winter.

Thus, the “standard” Solar Irradiance value is useful not only for climatology and solar physics, but for basic and applied biology also.

32. Posted May 2, 2008 at 10:49 AM | Permalink

# 30

cba,

and fertilization of soil along with the distribution of that insolation

We take samples of soils each year for fertilizing only when it is necessary. We prefer organic fertilizers over chemical mixtures, so don’t worry about some genotoxic products in your tomatoes. ;)

33. Posted May 2, 2008 at 12:12 PM | Permalink

13 (Erl): Deep convection in the tropics will result in a saturated adiabatic rate having a dominant influence on the vertical temperature profile, and hence result in the upper tropical troposphere warming faster than the surface. The convection is still controlled from below.

34. CWells
Posted May 2, 2008 at 3:09 PM | Permalink

From Ice Age #2;
DennisW:
Thanks for the article- didn’t have that one.
But to my earlier point- going to Fawcett/Jones; after smoothing/refining out ENSO, doesn’t the remaining temp curve follow PDO?? And NOT the continuous CO2 increasing curve/line from Mauna Loa?? (Sorry, maybe someone better at graphics than I can get the plots and move them here. SteveMc??) And do not give me the particulate arguement for the flat period ‘46-’77; there was particulate before and after, and there wasn’t continuous volcanic eruptions during the period either.
I’m excited that I think I see something and want someone to explain it if possible. Thanks.
C
OK Leif, ‘splain it to me here….

35. CWells
Posted May 2, 2008 at 3:17 PM | Permalink

Bender:
RE: Your ENSO vs PDO strength arguement in 14. Have you seen the article I ref above?
C

36. DeWitt Payne
Posted May 2, 2008 at 3:45 PM | Permalink

Leif,

…hence result in the upper tropical troposphere warming faster than the surface.

That is true only if the specific humidity at the surface increases with temperature. There is some controversy about whether that is actually happening on average over the whole of the tropics. Constant relative humidity as a function of temperature is a plausible assumption, but the observed trends imply that it isn’t true. Not that the models specifically assume constant RH, but the parameterizations used seem to result in constant RH for the most part.

37. bender
Posted May 2, 2008 at 4:04 PM | Permalink

#35 CWells.
What article? Give me a full citation and I will read it.

38. Erl Happ
Posted May 2, 2008 at 8:27 PM | Permalink

As a grape grower interested in what happens in other parts of the world I am familiar with the effect of very low temperatures on vineyards in Washington State. Complete wiped out is possible if the trunks of the vines freeze in mid winter and crop loss occurs if the fresh shoots are frozen in spring. At http://thedaily.washington.edu/2008/4/23/uw-scientists-explain-mid-april-snow/ we see a report of April snow that is regarded as unusual.

As the climatologists quoted in the Washington newspaper say, this cold snap reflects changes in Pacific sea surface temperatures and air pressures that relate to the Pacific Decadal Oscillation but this is ultimately driven by the state of energy intake in the tropics.

This morning I was asked if the Arctic ice cap is going to shrink this northern summer like it did last year. My reply may be of interest to those interested in sun/atmosphere relations. The linkages are many but it is really quite a simple story.
Data shows that both poles have an inordinately large ice capture currently. There is a lot of cold water in the North Pacific and some commentators, including NASA are saying that the Pacific Decadal Oscillation has swung into its cool phase. But the PDO simply reflects the rate of heating being experienced in the tropics and the latter is directly related to sunspot activity. Long periods without sunspots cut the amount of short wave radiation that enters the atmosphere and it seems that the response to short wave radiation can be seen as low as the middle troposphere where temperature change of as much as 4°C can be induced. No sunspots bring marked cooling in all layers up to 50Km in elevation. Check this for yourself at http://www.cpc.ncep.noaa.gov/products/stratosphere/strat-trop/ where you can download a pictorial representation of the progress of temperatures in the atmosphere for the current year. Look at both poles and the tropics. You will see a strong solar signature that can be related to daily sunspot activity. Latest month at http://sidc.oma.be/products/ri/ Other recent months at http://www.ngdc.noaa.gov/stp/SOLAR/ftpsunspotnumber.html#international

In the middle troposphere (4-10 km, temperatures less than 0°C and mainly ice crystal cloud known as cirrus) relative humidity increases, dew point is reached, more clouds form, incident radiation is reflected. Down the line, the Earth gives off less heat and the lower troposphere (much more cloud here and temperatures > °C) cools, increasing cloud cover further. So, we will cool until sunspot cycle 24 gets going properly. That could be 18 months off. It may not get going properly at all inducing Maunder minimum type conditions.

Temperatures at US base Mawson Scott have been falling since 1957 but will now rise with the relative collapse of tropical convection. The air returns to the Earth in the Brewer Dobson circulation (http://www.atmosphere.mpg.de/enid/1__Dynamics___Aviation/-_Dynamics_n8.html), up at Equator and down at poles. Antarctica is the area of Australia or the US, and the area doubles in winter. Ice is grounded on cold terra firma. This is not the case in the Arctic where the ice melts from below. However, I would be surprised if we see a less than average retention of Arctic Ice in this coming NH summer. This is because the underlying waters are much colder.

39. Erl Happ
Posted May 2, 2008 at 9:02 PM | Permalink

33 (Leif)

Deep convection in the tropics will result in a saturated adiabatic rate having a dominant influence on the vertical temperature profile, and hence result in the upper tropical troposphere warming faster than the surface. The convection is still controlled from below.

Here are some things to consider in relation to that statement:
1. The thunderstorm convection zone tends to be localised and shifts with the seasons.
2. The latitude band of the pictorial is quite narrow at 10N to 10S and shows little seasonal influence other than heating of the upper layers at the equinox.
3. The solar signature is very strong from the top of the atmosphere at 50km right down to the surface. To see it for yourself download the pictorials for the NH and the SH and plot against daily sunspot activity. My post #38 provides the necessary links. You will see that marked cooling occurs in the stratosphere as the period of zero sunspot activity lengthens. Outside the tropical zone the solar signal is more marked than in the tropics, particularly so in the winter (northern) hemisphere at latitudes between 60°N and 90°N.
4. Most of the cloud in the thunderstorm zone appears in the afternoon as you would know from living in Singapore and I imagine that release of latent heat occurs mostly in the lower stratosphere at temperatures above zero.
5. If the air is saturated all the way I can not see why the lapse rate should change with elevation except to diminish with height, especially in the middle and upper troposphere as UV absorbtion comes into play.

40. bender
Posted May 3, 2008 at 8:28 AM | Permalink

#39
In your theorizing it would be helpful if you could outline where your thinking is consistent and where it is inconsistent with the GCMs. Keep at it!

41. Erl Happ
Posted May 3, 2008 at 10:45 AM | Permalink

40 (Bender)
Thanks for the encouragement. I have no familiarity with the GCM’s . As a person trained in Economics I am well aware of the importance of the assumptions that you make in setting up a model. I can not conceive that CO2 is a forcing of any significance. The mechanisms that operate to remove heat from the surface and transfer it into space are very strong and the atmosphere exceedingly thin. Temperature change has a strong spatial and seasonal element. The lack of any attempt to find a mechanism to explain the temperature change that is due to heating and cooling cycles in the tropics, the most influential and dramatic temperature change that we see, and a change that can force the seasonal and spatial warming that is observed, represents a failure at the earliest stage of diagnosis in the problem solving exercise. I believe the treatment of water vapour and cloud in the GCM’s is patently wrong, as Roy Spencer has been at pains to point out.

Bender, if you can download the pictorials of temperature change in the NH, SH, and tropics for the first four months of 2008 at http://www.cpc.ncep.noaa.gov/products/stratosphere/strat-trop/ and compare with daily sunspot activity you will have no difficulty recognising a strong solar connection with temperature and inferred cloud cover in the middle troposphere. Then, develop the theory to explain what you see.

My theory: Cloud cover is primarily driven by the sun but within limits according to the evaporative response driven by nature of surface (ocean, soil or leaf) surface temperature and speed of air movement across the wet surface. The time at which the heating event (increase in UV intensity due to sunspot activity) occurs is critical because it determines how much is absorbed and how much is immediately returned to space (extent of ocean versus land). However, that said increases in UV irradiation run into diminishing returns very quickly as surface warming forces atmospheric humidity upwards and creates more cloud to replace that which has evaporated. The strongest and most sustained increase of UV intessity occurs at the upswing stage of a new solar cycle. However, change is great also in the downswing stage.

The evidence is that we have grossly underestimated the impact of UV radiation on the atmosphere. It looks as if slightly wet air, or clouds made of ice strongly absorb UV light. The air warms considerably and in doing so, the clouds evaporate. In the middle troposphere there is not much water vapour, temperatures are less than 0°C but that which is present is very important in altering the Earths albedo. Very little extra UV can have a big impact here because the amount of water that can be held by air at temperatures lower than °C is slight. According to the Max Planck institute ozone is created in the troposphere (www.atmosphere.mpg.de/enid/1__Dynamics___Aviation/-_Dynamics_n8.html). So, the energy is available.

Nasif
Where are you growing your vegetables? Looks like you have strong seasonal limitations. I have an index that simulates the generation of carbohydrate production in grape vines that is based on temperature, usually the most limiting factor in plant growth. For grape vines carbohydrate production is maximal at 25°C, suffers declining returns for temperature increase above 25°C and is minimal at temperatures less than 12°C. Exposure to light at temperatures below 10°C can inhibit productivity when temperatures actually rise into the favourable zone. Generally, as long as light intensity is above about one third full sunlight, light is sufficient for grape vines. However, the arrangement of leaves in space is critical.

42. Posted May 3, 2008 at 12:01 PM | Permalink

# 41

Erl,

We are growing veggies in Zacatecas, MX, and we are looking for new fields in Nuevo Leon, MX. Here, in Nuevo Leon our biggest problem are the scorching temperatures (for veggies) and the excess of humidity. In Zacatecas is the cold season, which begins in September and ends in April. I think the next graph will be of interest for you, Earl.

Leif,

I think I’ve found a correlation between TSI and the variability of Tropospheric Temperature:

43. John West
Posted May 3, 2008 at 12:07 PM | Permalink

#38, #39, & #41

Hi, Erl . . . Thanks very much for these interesting links. Could you briefly explain what the Wave 1, Wave 2, and Wave 3 columns represent in the table at the Stratosphere-Troposphere Monitoring site? Thanks.

44. Posted May 3, 2008 at 12:08 PM | Permalink

# 42

Me,

When we compare pears with apples (or apples with pears) the things go wrong, but if we compare apples with apples, the things are clearer. No doubt… There is a close correlation between sunspots and temperature anomalies. ;)

45. Posted May 3, 2008 at 12:33 PM | Permalink

38,39 (Erl): We have been down this road before and there is simply no evidence for your statement: “The solar signature is very strong from the top of the atmosphere at 50km right down to the surface. To see it for yourself download the pictorials for the NH and the SH and plot against daily sunspot activity.”
Here is such a plot for 2007:

We cannot make any progress as long as your basic premise does not square with reality. I can assure you that if you were correct and “The solar signature is very strong from the top of the atmosphere at 50km right down to the surface” that this would be one of most important discoveries in this field in the last 150 years and would be Nobel Prize stuff. Unfortunately I do not see a trip to Stockholm in the cards for you, and that is not because ‘conventional wisdom’ is ‘head in the sand stuff’, but rather because there really is no substance to your claim. This is a pity, because I, for one, would dearly welcome such a discovery [would make my work all that more valuable and relevant]

46. Posted May 3, 2008 at 12:45 PM | Permalink

42 (Nasif): If you had used a correctly calibrated TSI I could make some meaningful comment. Use mine :-) [the red curve in:
]

47. Posted May 3, 2008 at 1:12 PM | Permalink

Leif,

Mmmh… Ok, I’ll use your database, thanks! :)

48. Posted May 3, 2008 at 1:42 PM | Permalink

42 (Nasif): If I use my TSI and fot polynomials to the interval 1975-2008 I get this:

If you include 1645-1725, where we had the Little Ice Age you would, of course, find that both Temperature and Solar activity went down then, but that in itself does not prove anything.

49. Posted May 3, 2008 at 2:16 PM | Permalink

# 48

Leif, except if I use a three periods polynomial:

If I use a decadal period I get inversed polynomial trends.

50. Posted May 3, 2008 at 2:21 PM | Permalink

# 49

Leif,

Something in Earth’s system is modifying the effects of SI variability, but… What it could be? Perhaps, gravitational oscillations?

51. Posted May 3, 2008 at 2:23 PM | Permalink

49 (Nasif): try a 3rd degree [you already did], then a 4th and a 5th, and 2nd for good measure. The degree you like the most will reveal what you think beforehand the relation should be.

52. Posted May 3, 2008 at 2:25 PM | Permalink

# 49

Me and to whom it concerns,

For the last plot, I used 1360.5 as standard.

53. Posted May 3, 2008 at 2:30 PM | Permalink

# 51

Leif,

I’ve tried and found opposed results. The second degree gives a lineal trend on temperature anomalies.

54. Posted May 3, 2008 at 2:42 PM | Permalink

53 (Nasif): so what you find depends on what degree you choose and that was my point.

55. Posted May 3, 2008 at 2:45 PM | Permalink

# 53

Leif,

Wait, wait, wait… The differences between the # 42 graph deduced from the second Lean’s database and the # 49 graph calculated from your database is enormous! The plot of the anomalies calculated from your database is smoother than the plot from the second work of Judith Lean. You’re not on the mainstream, Leif… That’s pretty nice! :) Ok! If I’ve found that physically the CO2 is not related to the change of temperature in the last two decades, and there is not enough correspondence between TSI and Ttrop… What a hell (I apologize) is happening in the solar system?

56. Posted May 3, 2008 at 2:54 PM | Permalink

55 (Nasif):

You’re not on the mainstream, Leif…

Thank goodness for that! Mainstream science includes AGW [‘the science is settled’], Strong cycle 24 [‘stunning breakthrough’ – NASA], solar and geomagnetic activity has more than doubled in the last 100 years [‘highest in 11,000 years’], etc, etc.

57. Posted May 3, 2008 at 4:23 PM | Permalink

Leif,

My article was published here. Please, read it so you can let me know about any errors. I have made clear that the errors are mine alone in the last paragraph of the article. ¨lease make me knoe if you don’t wish your data are included in my article; I will make the changes immediately.

58. Posted May 3, 2008 at 4:36 PM | Permalink

57 9Nasif): the article is fine; I don’t know which language it will be in, but you might fix “El Niño y La Niña evolve”.

59. Posted May 3, 2008 at 4:50 PM | Permalink

Thanks, Leif. I’ll correct that y.

60. John West
Posted May 3, 2008 at 7:54 PM | Permalink

re my #43: Never mind, Erl. I figured it out. They’re Rossby waves.

61. Erl Happ
Posted May 3, 2008 at 8:13 PM | Permalink

43 (John West)

Could you briefly explain what the Wave 1, Wave 2, and Wave 3 columns represent in the table at the Stratosphere-Troposphere Monitoring site?

I have no idea. An email on a question other than the one you asked have so far not met with a response. Please try direct to NOAA via the website. It would also be nice to have data for the latitudes 10° to 60°.

62. Erl Happ
Posted May 3, 2008 at 8:29 PM | Permalink

Leif,
I am puzzled to know the relationship between sunspot activity and UV radiation? Can UV radiation pulse before sunspots appear or at times when sunspots are not present? Are there other solar phenomena that are related to the intensity of UV radiation recieved at the edge of the Earth’s atmosphere? Is there an index that can be used than better represents UV radiation than sunspot activity? A daily time resolution is necessary. Perhaps the spectrum that is thought to be responsible for ozone formation.

Re #45. I will look at the data for 2007 and come back to you. I need a larger magnification and also the anomalies.

Have you looked at the table for the tropospheric ozone budget by Dr. Elmar Uherek – Max Planck Institute at http://www.atmosphere.mpg.de/enid/1__Dynamics___Aviation/-_Dynamics_n8.html

63. Erl Happ
Posted May 3, 2008 at 8:31 PM | Permalink

63 (John)
Good to have you look at this in detail. Looking forward to some input.

64. cba
Posted May 3, 2008 at 9:00 PM | Permalink

57 (Nasif):

65. Posted May 3, 2008 at 9:12 PM | Permalink

62 (Erl):

the relationship between sunspot activity and UV radiation?

All the solar parameters (sunspots, magnetic field, TSI, F10.7 radio flux, MgII UV, CaI flux, etc) are so well correlated that you can deduce one from the other. In a sense they are all equally good (or bad). They are also averages over the whole disk, so 1-day resolution is kind of an overkill. You might as well ask for 1-minute resolution, doesn’t make much difference. This is, of course, not strictly true: there are short-lived phenomena related to strong solar flares, but these are rare and short-lived [seconds to hours] so have little impact on climate. If you are particularly interested in UV, then the MgII index is probably the best as it is an excellent proxy for the UV that creates the ozone layer. You can find a brief explanation at http://www.swpc.noaa.gov/AboutUs/Review2000/Solar_Irrad_Poster.ppt
You can find the data at http://www.sec.noaa.gov/ftpdir/sbuv/NOAAMgII.dat
There are some calibration issues [which for your purpose doesn’t matter] as you can see here: http://www.leif.org/research/MgII%20Calibration.pdf

I have not looked before at the ozone-link you mentioned. Going there now, I see that the site is a nice non-technical exposition of what I have been saying ad nauseam in this blog.

66. Dennis Wingo
Posted May 3, 2008 at 10:52 PM | Permalink

Leif

We do know, beyond a shadow of a doubt that the solar cycle does have an effect on the atmosphere. The drag on the International space station and all low earth orbiting spacecraft are directly effected by the intensity of the solar cycle. During solar minimums the drag is as much as several times less than at solar maximum. There is a well known density variation of the atmosphere at orbital altitudes. I cannot but think that this effect extends downward at least some distance into the atmosphere. So it is a bit disengenious to imply that there is little or no effect on the atmosphere due to the solar cycle. If it is not TSI related, then the data is absolutely indicating some effect. Therefore it would be interesting to further investigate these differences.

There is a great gap in our knowledge of the atmosphere. From the altitude above where planes and balloons can travel and below orbital altitude there is a region of the atmosphere known as the ignorosphere jokingly by scientists because it is only through very infrequent sounding rocket flights that we have any good information on what is happening there.

We do know that there are definite terawatt class energy transfers between the ionosphere and the upper atmosphere clouds that was finally conclusively discovered in the 1990’s by scientists that were later confirmed by the Gamma Ray Observatory and the Space Shuttle.

I guess they are calling them transient luminous events now. These are real and are a path for the conduction of energy between the clouds and the ionosphere. Since the strength of the ionosphere is absolutely effected by the solar cycle, there is at least one concrete link between the ionosphere and thunderstorms.

Here is an explanatory picture.

67. Erl Happ
Posted May 3, 2008 at 11:09 PM | Permalink

65 (Leif)
Re the ozone link: My point is that the Max Planck institute posits ozone creation in the troposphere (3500 Tg/yr) and show it in their budget for ozone in the troposphere, only a small part of that present 600 Tg/yr) being due to ozone drifting down from the stratosphere. What is creating it?

Thanks for the links for MGII. Sorry to be a pain but is there a file in text format that will open in Excel. The file you cite is a .dat file that my gear does not recognise.

Thanks for the link to the issue of MGII calibration. The interchange is illuminating.

Why does 10.7 radio flux vary more than MGII. I imagined that MG II would fairly capture a large proportion of the variation in short wave energy, that energy component being responsible for a large part of the the variation in irradiance. Perhaps not?

68. Posted May 3, 2008 at 11:50 PM | Permalink

# 64

cba,

I’m sorry… The webmaster does it when he’s changing something, like the “y” from my bad Spanish. The password has been erased.

69. cba
Posted May 4, 2008 at 5:12 AM | Permalink

68 (Nasif):

The scientific method is a method of progress where by in the long run, great strides in learning can be made. However, I describe it as 2 steps forward, 1 step backwards, and 3 steps to the side. At the moment, there is evidently some serious turbulence in solar physics and the trend has been among most all of them to be following Leif’s lead of less variation based upon the more recent data available and upon reinterpretations, hopefully improvements upon previous data. However, Leif’s approach calls for a minimum flatline of no variation at the low level while the others have simply reduced the overall amount of variation they had assumed. As I understand, it remains to be seen as to whether more data will support Leif’s hypothesis and whether others will jump aboard this view or whether it will put it out of the running for what goes on.

In my view, which is not that of a solar physicist, it seems a bit early to jump on the Leif’s train or place any dependence on who is right. Rather, a sit and wait attitude to see where this is going would be more prudent. This leaves one with having to do contigency plans as to what if Leif is right and what if Leif isn’t right on this. For me personally, I suspect Leif may well be right or mostly right, but I don’t know enough of this area to say one way or the other.

Actually, this whole thing with solar variability is good in that simple modeling of the system has brought about the realization that a more sensitive earth undergoing less solar change leads to better results that accomodate time lags due to specific heat capacities than a less sensitive one with larger solar variation. Also, it has brought up the question (or helped) as to the nature of what might be affecting the earth from this solar variability as TSI is but one measure of the sun’s variations and it would seem to be the one that has about the variation.

I did note in your article a sensitivity for the earth of 0.1 K per 0.1 W/m^2 based upon the last 20 years. I don’t know if that was intentional or a typo but I would suggest that is off by about a factor of 10 for the sensitivity. There is an average power from the sun hitting the earth of 340-342w/m^2 (before albedo considerations that reduce it slightly)and a total effect of 33 deg K. A crude average suggests about 0.1 deg K per W/m^2 of incoming. Also, one would expect that since radiative output is a T^4 proposition and that is related to balance, that each W/m^2 additional would have a little less than the earlier ones with T rise, suggesting that each new W/m^2 change in total incoming would have less effect than the crude average of contributions would suggest.

On the flip side, the reconstruction of albedo as shown on one of these threads at CA (by Leif) suggest a variation over this same 20 years of about 8-10%. Such an albedo change would be responsible for a total of 10 W/m^2 difference in the total power absorbed in the earth system, 100 times that of the sensitivity implied if you assume that the 0.1 K change in earth’s temperature over the last 20 years is all due to 0.1w/m^2 variation in solar flux. If this were the case and the earth were that sensitive, it should have changed temperature by 10 degrees average, well beyond what has been measured.

A change of 0.1 K for 10 W/m^2 suggests 0.01K/W/m^2 – which would be plausible for the decreasing sensitivity idea presented above.

70. cba
Posted May 4, 2008 at 7:31 AM | Permalink

69 (me):

ugh – too early – too much rush – the numbers are screwed totally screwed up
the 33 K has to be in regards to the 150 -170 w/m^2 blocking going out. That puts it at 0.2K/W/m^2

71. Bob B
Posted May 4, 2008 at 7:42 AM | Permalink

Leif, is there indeed a new cycle 24 set of spots in the SH?

72. John West
Posted May 4, 2008 at 9:22 AM | Permalink

#63:

Erl, there’s a short, non-technical intro to Rossby waves (atmospheric and oceanic) at:
and some more on oceanic Rossby waves at:
http://www.noc.soton.ac.uk/JRD/SAT/Rossby/Rossbyintro.html
Much more technical material can be found by googling ‘Rossby waves’ or ‘planetary waves.’ I don’t know what (if any) bearing they have on the present discussion, except perhaps to further demonstrate the vast complexity of the atmosphere.

73. Posted May 4, 2008 at 9:31 AM | Permalink

71 (Bob): yes, there is a cycle 24 spot. It has been seen by the Sunspot Center in Brussels too and will likely get a NOAA region number. We’ll see how long it lasts, it already looks a bit weaker, so may not last until tomorrow. We shall see.

74. Posted May 4, 2008 at 10:02 AM | Permalink

# 69

cba,

And you are right, cba (?). However, it is an article, not a peer reviewed paper, and I think it will be submitted to scrutiny so we could find the errors. You may have noticed also that I’m referring only to sunspots number, not to the Solar Irradiance and that I opened the gate for other alternatives, like the solar magnetic field and the resonance between the solar magnetic field and the geomagnetic field. As you noticed, I have also noticed that the plot from calibrated Leif’s data is flat, or smooth. Don’t you agree with me that it would be convenient to compare the Leif’s numbers with other numbers based in both, sunspots and isotopes? Perhaps the iron stained grains could be a good starting point. OTOH, Thanks a lot for noticing the error on the sensitivity for the Earth’s surface; it was a typo, sorry… The real factor is 0.85 K per W/m^2, which many have rounded out to 1 K per W/m^2. I’ll make the correction. BTW, I’m not either a solar physicist, so I have to be cautious with this solar stuff. Comparing Lean’s work with Leif’s work, If I consider the database of Lean, the TSI anomaly in 400 years has been 1.68 W/m^2, which, a grosso modo, could mean a temperature anomaly of 1.428 K in four centuries. What’s your opinion? I think it is congruent with reality, isn’t it? :)

75. Posted May 4, 2008 at 10:10 AM | Permalink

# 69

cba,

From Leif’s database, the change of temperature would be 1.0455 K in only 258 years.

76. Posted May 4, 2008 at 10:15 AM | Permalink

# 69

cba,

On the other hand, from the first database of Judith Lean, the change of temperature in 400 years would be 2.7535 K. Throw your cards, ladies and gentlemen.

77. cba
Posted May 4, 2008 at 12:38 PM | Permalink

74 (Nasif):

That sensitivity of 0.85 K per W/m^2 doesn’t match the basics. There’s too little actual T contribution and too many W/m^2 variations going on for it to be correct. If it were true and the climate were that sensitive, then our global averaged temperature would be varying by over 10 degrees in the last 20 years.

78. Posted May 4, 2008 at 2:59 PM | Permalink

# 77

O.85 K per W/m^2 of increase/decrease of SI. It’s a factor to calculate variability, not for calculating the absolute temperature produced by 360 W/m^2 of radiation. We have formulas to calculate delta T of the surface by the incident solar radiation on ground.

79. Posted May 4, 2008 at 4:05 PM | Permalink

67 (Erl):

Re the ozone link: My point is that the Max Planck institute posits ozone creation in the troposphere (3500 Tg/yr) and show it in their budget for ozone in the troposphere, only a small part of that present 600 Tg/yr) being due to ozone drifting down from the stratosphere. What is creating it?

After a thunderstorm you can smell the ozone…

Thanks for the links for MGII. Sorry to be a pain but is there a file in text format that will open in Excel. The file you cite is a .dat file that my gear does not recognise.

Here is a file for data since 2005: http://www.leif.org/research/MgII-and-UV.xls

Why does 10.7 radio flux vary more than MGII. I imagined that MG II would fairly capture a large proportion of the variation in short wave energy, that energy component being responsible for a large part of the variation in irradiance. Perhaps not?

The UV and its variations are not a large part of the variation in irradiance. Rather it is a very small part of the TSI variation. I have lost count of how many times I have said that…

To put things in perspective, here are some graphs from the XLS file. The first panel shows (red) the MGII index, (blue) the f10.7 flux scaled to match MgII. You can see that they follow each other or so, the MgII is really crummy [as you could see from the calibration paper, NOAA doesn’t really give a damn about maintaining quality], so the scaled values from F10.7 are better. The very short-wave stuff that doesn’t get past thermosphere [30.4 nm] can be derived form the MgII index:

From the SORCE data you can get the integrated UV flux from 180nm to 310nm, which is the flux that creates the ozone-layer. It is shown in the next panel. Note that the short-term variations match the MgII-index quite well and that there is a clear solar cycle dependence as the UV-flux is going down; also note the the flux is tiny, only 1.5 W/m2, with the variations being only 0.01-0.02 W/m2. The next panel shows the integrated flux that gets by the ozone cutoff at 242nm, namely between 242 and 310 nm. The flux here is a bit higher [a little more that 1% of TSO]. The short-term variations are of the order 0.05 W/m2, but, perhaps more important, the solar cycle variation is reversed: with decreasing solar activity, the UV flux above 242 nm goes up. Just so that you should not suspect a programming error on my part, here is the flux as calculated by the SORCE website:

80. cba
Posted May 4, 2008 at 5:13 PM | Permalink

78 (Nasif):

I wouldn’t mind seeing some more detail on that. I find it rather incredulous regardless of whether it’s some sort of localized instantaneous value or whether it’s a global average. In either case, the variations are tremendous compared to the delta T involved.

Here, today it was overcast. The T was perhaps 5-10 degrees C less than what it would have been with mostly clear skies. A rough guess on the surface insolation, even at mid day would be probably half of normal – suggesting again 10 deg for 100 W/m^2 averaged insolation over the day as a rough guess. Solar power people talk about total overcast reducing surface insolation by typically as much as 90%.

I assume something else is involved here with your efforts which is not obvious. Otherwise, I suspect the use of such numbers will likely result in a great deal of problems trying to get them to agree with the reality of your application.

81. DeWitt Payne
Posted May 4, 2008 at 5:25 PM | Permalink

Erl,

In addition to direct formation due to electrical discharge as mentioned by Leif above, lightning e.g., there’s the photochemical smog reactions that produce ozone when you have sunlight, branched hydrocarbons, terpenes from pine trees work here as well as anthropogenic emissions (see Smoky Mountains), and nitrogen oxides other than N2O. Lightning makes nitrogen oxides as well.

82. Erl Happ
Posted May 4, 2008 at 6:04 PM | Permalink

79 (Leif)
Thanks very much. That throws a lot more light on things. What is happening to at 310-400nm towards solar minimum? Is that rising even more than 282-310nm?

With less very low frequency UV (presumably more energetic in ozone creation)and more UVA and UVB (with less ozone to trap it) one would expect a greater penetration of the atmosphere by the latter towards solar minimum (and more interaction with water vapour in the middle troposphere) Is that interpretation fair enough?

What are the implications, if any, for the height and depth of the ozone layer in the stratosphere of this change in the relative strength of these frequencies in the UV? My gut feeling is that it would descend.

83. Posted May 4, 2008 at 6:46 PM | Permalink

82 (Erl):

With less very low frequency UV (presumably more energetic in ozone creation)and more UVA and UVB (with less ozone to trap it) one would expect a greater penetration of the atmosphere by the latter towards solar minimum (and more interaction with water vapour in the middle troposphere) Is that interpretation fair enough?

I think wishful thinking got the better of you. You have this backwards. Your “very low frequency” is actually “very high frequency” or UV-C. And, as I have said a zillion times, none of the ‘ozone-creating’ UV penetrates to the troposphere; the ozone in the troposphere is created from below or brought down by the very slow (years) Brewer-Dobson circulation. And none of the solar radiation below 290nm reaches the lower stratosphere and the troposphere.
UV above 210nm does not interact with water vapor, so what penetrates or not is moot as far as H2O is concerned.

84. Posted May 4, 2008 at 8:23 PM | Permalink

# 80

I wouldn’t mind seeing some more detail on that. I find it rather incredulous regardless of whether it’s some sort of localized instantaneous value or whether it’s a global average. In either case, the variations are tremendous compared to the delta T involved.

Well, try this algorithm.

I’m not sure I have undestood your objection.

I assume something else is involved here with your efforts which is not obvious. Otherwise, I suspect the use of such numbers will likely result in a great deal of problems trying to get them to agree with the reality of your application.

85. Posted May 4, 2008 at 8:27 PM | Permalink

# 84

As you can see it’s not so simple as to multiply the factor 0.85K/(W/m^2) by the anomaly of Solar Irradiance expressed in W/m^2.

86. cba
Posted May 4, 2008 at 9:04 PM | Permalink

(Nasif):

Some of your numbers are not making sense there. You have 50% of incoming solar insolation listed as being absorbed. That is not the average of incoming over a full day or over the surface of the earth. Also, it would seem you are trying to relate real world values to conduction only – a non real world situation. Both conduction and radiation are occuring in the real world. Basically, it’s not making sense to me as it is being described.

87. Posted May 4, 2008 at 9:12 PM | Permalink

# 86

cba,

My purpose on showing you that article was only to let you know that the things are not so simple. I have another article more elaborated about the issue. Perhaps you will find some sense there.

88. Erl Happ
Posted May 4, 2008 at 9:44 PM | Permalink

83 (Leif)
Yes, I expressed the thing at 180° to my intent. No, its not wishful thinking. Must be more than slightly dyslexic. I do understand that UVC is short wave high frequency radiation and most active in ozone creation. My question remains. What happens to the stratosphere when the most energetic frequencies diminish and the proportion of less but still energetic radiation at lower frequencies UVA, and UVB increases as we approach solar minimum? My guess: 1. Ozone concentration diminishes and the zone of highest concentration moves upwards. 2. There is less ozone to absorb UVB which appears at greater intensity at lower elevation. 3. The lower stratosphere will dry slightly enabling greater ozone persistence there. 4. Temperature increase in the lower stratosphere due to greater UVB absortion there. Result is a dispersion of the zone of maximum concentration at solar minimum. This should be testable.

UV above 210nm does not interact with water vapor, so what penetrates or not is moot as far as H2O is concerned.

Others disagree with this statement.

I disagree based on the observed change in atmospheric temperature when radiation changes.

We do know that UVB and UVA is attenuated as it travels through the atmosphere.

At http://physicsworld.com/cws/article/print/17402 you will see that others disagree on the basis of their understanding of physics. The authors are:
Ahilleas Maurellis is in the Earth-Oriented Sciences Division of the SRON National Institute for Space Research, Sorbonnelaan 2, 3584 CA Utrecht, the Netherlands, e-mail a.n.maurellis@sron.nl. Jonathan Tennyson is head of the Atomic, Molecular, Optical and Positron Physics group at University College London, Gower Street, London WC1E 6BT, UK, e-mail j.tennyson@ucl.ac.uk

89. Posted May 4, 2008 at 10:02 PM | Permalink

Leif,

A quick question, if I may. If solar variation has been so small, do you have any opinion on what kind of factors may have led to the temp increase of the early 20th century?

(Without solar variation, the famous, mainstream Meehl et al attribution study would basically leave it unexplained).

Thanks,

Mikel

90. Posted May 4, 2008 at 10:20 PM | Permalink

Leif,

Could you be so kind as to reccomend me a good source for solar magnetic field databases since 1880? Thanks a los in advance.

91. Posted May 4, 2008 at 10:23 PM | Permalink

Oops! My Spanish is terrible:

It must say: “Thanks a lot in advance”. :)

92. Posted May 4, 2008 at 11:09 PM | Permalink

88 (Erl):

What happens to the stratosphere when the most energetic frequencies diminish and the proportion of less but still energetic radiation at lower frequencies UVA, and UVB increases as we approach solar minimum?

This paper concludes that “ozone solar cycle variation in the low-latitude middle stratosphere (5-10 hPa) is very small or negligible” :
The 11-year solar cycle variation of stratospheric ozone as obtained from the SBUV and HALOE ozone profile measurements
B. Soukharev and L. Hood, IAGA 2005:
The observed solar cycle variation of ozone is a key constraint on climate models that include solar UV / ozone / dynamical coupling as a sun-climate forcing mechanism. Previous empirical analyses indicate that the largest percentage ozone increase from solar minimum to maximum is in the upper (1-3 hPa) and lower (30-100 hPa) tropical and subtropical stratosphere while a minimum percentage increase occurs in the tropical middle stratosphere (5-10 hPa). This observationally derived altitude dependence in the middle and lower stratosphere contrasts with the results of radiative photochemical models which predict a maximum ozone increase in the middle stratosphere and a small increase in the lower stratosphere. Here, we apply standard multiple regression statistical model to estimate the solar cycle component of stratospheric ozone variability using two different satellite ozone profile data sets (Version 8 SBUV (/2) over the 1979 to 2003 time period and HALOE over the 1992 to 2004 time period). Despite large difference in the length of the time periods considered (25 years for SBUV and only 13 years for HALOE) the results for both data sets are in qualitative agreement. They show that the ozone solar cycle variation in the low-latitude middle stratosphere (5-10 hPa) is very small or negligible.

From the article you cited:

Water vapour is responsible for 70% of the known absorption of incoming sunlight, particularly in the infrared region

Erl: We do know that UVB and UVA is attenuated as it travels through the atmosphere.

but not by water. http://www.geology.wmich.edu/haas/envs2150/12s.pdf has this nice figure:

Temperature increase in the lower stratosphere due to greater UVB absorption there

The temperature in the stratosphere doesn’t determine the temperature at the surface or in the lower troposphere.

93. Posted May 4, 2008 at 11:31 PM | Permalink

90 (Nasif): http://www.leif.org/research/IDV,%20B%20analysis%20-%20rotations.xls
gives the strength of the interplanetary magnetic field near Earth since ~1882.
The ‘solar magnetic field’ is sort of undefined – what do you mean?

94. Posted May 4, 2008 at 11:32 PM | Permalink

93 (me): should have been 90 (Nasif): http://www.leif.org/research/analysis/IDV,%20B%20analysis%20-%20rotations.xls

95. Posted May 4, 2008 at 11:38 PM | Permalink

89 (Mikel): the short answer is “no”. I don’t know. they could be many factors. No matter what I say, it will clash with somebody’s pet theory. But, anyway, I can speculate: there are longer term variations [of unknown causes] of ~1500 years, on which are superposed PDO-type variations with periods ~60 years, on which is superposed some [small?] CO2 effect.

96. Erl Happ
Posted May 5, 2008 at 1:38 AM | Permalink

92 (Leif)
Re the article I cited and the question of UV absorption by water in one or other of its various states:

but not by water

You missed the critical issues.

Water absorption bands are also present in the optical region and extend all the way to the ultraviolet, although they are less strong at shorter wavelengths.

Radiation that is absorbed by the atmosphere in the ultraviolet (UV) and visible (VIS) regions of the spectrum contributes to the emission in the infrared region (IR). The incoming solar radiation is either reflected directly back into space, absorbed by the atmosphere or absorbed by the Earth’s surface. The 67 W m-2 ultraviolet-visible absorption that is due to atmospheric trace gases, such as water, also translates into infrared emission, which combines with the infrared surface heat to emerge as some 195 W m-2. However, the measured atmospheric absorption is up to 30 W m-2 higher than models predict, and this is known as the absorption anomaly.

We don’t know for sure what is going on but the study strongly suggests that part of that anomaly is due to water. Your figure contradicts this but is superseded by more recent work. The work is ongoing.

In this situation observations of temperature change in the atmosphere are vital. You do not invalidate the observation simply by saying that this or that can not happen because we know….. We don’t know.

97. Posted May 5, 2008 at 7:18 AM | Permalink

96 (Erl):
You are over-interpreting this statement:

ultraviolet-visible absorption that is due to atmospheric trace gases, such as water,

All they mean is the ‘non-infrared’ absorption. They absorption properties of water has been known for decades. This is not something for ‘ongoing research’. As you can see from Segelstein’s graph [and everybody agrees with him – this is not contentious, see http://omlc.ogi.edu/spectra/water/abs/index.html%5D:

Water is most transparent in visible and near-UV. In the far-UV, H2O will absorb, but all that UV is already absorbed by Oxygen.

However, the measured atmospheric absorption is up to 30 W m-2 higher than models predict, and this is known as the absorption anomaly

The anomaly is a deficiency of the models, not a property of the atmosphere, specifically with how clouds are modelled. In the discussion about absorption by clouds, the misleading term ‘shortwave’ radiation is often used. This does not mean UVB, but simply ‘non-infrared’. This is because water absorbs best in the infrared. You can find a good discussion of the ‘absorption anomaly’ here: http://ams.allenpress.com/archive/1520-0442/14/6/pdf/i1520-0442-14-6-1147.pdf

The UVB flux at the top of the atmosphere is 13 W/m2. Only 10% of that gets through the stratosphere, that is 1.3 W/m2. The variation from day to day of the TOA UVB-flux is 0.1 W/m2, 10% of that is 0.01 W/m2. The solar cycle variation [in the ‘wrong’ direction] is 0.5 W/m2, 10% of that is 0.05 W/m2. We are talking about so minute variations that it boggles the mind to ascribe variations in climate and weather to direct heating by UV [which is what I think your theory actually is at the bottom]

So, even if we allow that absorption of all non-infrared radiation is by water vapor, what bearing would that have on the solar influence on climate, because, as we have seen, there is very little solar cycle variation of UVB and it goes the wrong way, so when you said [back in #38,#39] that “The solar signature is very strong from the top of the atmosphere at 50km right down to the surface. To see it for yourself download the pictorials for the NH and the SH and plot against daily sunspot activity.” We seem to have a problem with that.
My point here is not about minutia about anomalous absorption caused by the models not including clouds correctly, but about your claim of strong solar signature right down to the surface. Now, I realize that your statement is worded in such an ambiguous way that it can be misinterpreted, like “what do you mean the sun does not influence weather and climate at the surface?, of course it does, just take away the sun and you’ll understand”. So in more precise terms you should have said ‘solar activity’, which I shall assume that you meant, otherwise my comments are void.

98. kim
Posted May 5, 2008 at 7:25 AM | Permalink

97 (Leif) and Erl. How about an amplifier effect from UV on convection and CR on clouds, not to neglect chemical and biological modifiers?
===================================

99. Posted May 5, 2008 at 7:39 AM | Permalink

98 (kim):

How about an amplifier effect from UV

And what would that amplifier be? There are two ways to claim [or suppose] amplification:
1) there is very strong observational support for an effect
2) there are good theoretical reasons and mechanisms that specify how amplification is achieved.
Since we don’t have any of these, there is no basis for worrying about an ‘amplifier’.

100. Posted May 5, 2008 at 7:43 AM | Permalink

97 (me): I should have added that the ‘absorption anomaly’ really is about calculation of the albedo; we should perhaps properly call it the ‘reflection anomaly’.

101. cba
Posted May 5, 2008 at 8:31 AM | Permalink

97 (Leif):

The right side seems to indicate optical path length (1/e) and the label is misleading – suggesting microns). What’s more, if that is correct, it’s saying that the path length (1/e attenuation distance) for visible light is 100m and that for 100nm uv, it’s 1/10th wavelength.

Oops – just looked at your ref (getting rid of the : at the end.

That’s not atmosphere, that’s ocean absorption distances!!!

102. Posted May 5, 2008 at 8:35 AM | Permalink

101 (cba): yes the distances are for liquid water, but the important thing is the left Y-axis. Showing the absorption. That shouldn’t change.

103. Erl Happ
Posted May 5, 2008 at 9:07 AM | Permalink

Leif,
I thank you for your attention again.

The interaction of the atmosphere and solar radiation is made complex by the number of interacting variables.

1. The amount of water vapour in the atmosphere changes strongly with temperature and therefore elevation.
2. Solar radiation is an ever changing mix of frequencies some of which interact with the atmosphere and its water and particulate matter very directly causing atmospheric heating. Heating of the troposphere (from whatever source) causes changes in albedo which strongly amplifies the heating tendency.
3. Water vapour absorbs across a wide spectrum and especially in the infrared. Infrared and short wave radiation will work in tandem but they may not vary together.
4. The atmosphere does not conserve energy. It dissipates it. The entire atmosphere is moving about in response to temperature and pressure differences so as to resolve and dissipate the forces being applied to it.
5. Position of the Earth in its elliptical orbit affects strength of irradiance.
6. The Earth is an orb and the strength of irradiance is greatest when it is incident at the vertical. Seasonal factors involving the tilt of the Earths axis changes the location of the zone of greatest heating.
7. The asymmetrical arrangement of the land and the sea at the surface affects the rate of return of energy to the atmosphere and on the flipside, the extent of storage in the oceans.
8. UV and or infrared is not proportional to sunspot activity.
9. The interaction of a variable amount of ozone with a variable amount of UVB, the latter increasing towards solar minimum and the former declining with the flux in UVC.
10. Brewer Dobson circulation produces enhanced downdraft in the polar vortex in the winter hemisphere as a result of extra uplift in the tropics. So, the polar signal is opposite to that in the tropics. Although the rate of uplift in the stratosphere over the tropics is slow, the air descends at a much faster rate at the poles due to the difference in surface area involved and the consequent funnelling effect. Winter winds at the poles are gale force and as cool as minus 80°C. This markedly increases pole to Equator temperature gradients when the equator heats.
11. The extent of the tropical ocean that is warm or cool affects evaporation. So does rainfall and soil moisture content on land.
12. Enhanced UVC at solar maximum should increase ozone concentration.
13. Feedback from the surface influencing temperature and humidity in the troposphere through to the lower stratosphere is different at different times of the year.
14. Both the solar and the atmospheric parameters change over a short period of time. Nothing is static.

So, we can forget about the sort of linear and proportional relations that are frequently seen with solar indices that enables one to be calibrated according to the variations in others. This is not a ‘space environment’ where the medium has no affect on the transfer of energy. Once irradiance gets into the atmosphere a lot of different things are happening. The upshot is that we must look a little harder at the data. The response to extra energy will nowhere be linear. You can’t afford to be a hard nosed statistician or you will miss the signal.

The progression towards extremes of cooling in sea surface temperatures and very low sunspot activity that has occurred in 2008 has created a picture that is relatively easy to read. That was not the case in 2007 when the tropics slowly transitioned from mild warming to mild cooling over nine months, but the signal is still there, if less readily apparent. We have to be more diligent to locate it. The critical response so far as albedo is concerned is that in the middle troposphere and here it is apparent that temperature change of up to 4°C in response to solar radiation is common. This represents a very strong response to small changes in irradiance but it is magnified again by what happens in the lower troposphere. It is in the lower troposphere that albedo can really change most markedly because this is where the great bulk of cloud is located. In this zone temperatures are above the freezing point of water. Strangely, temperatures here vary much less than in the mid troposphere. I believe they do so because of firstly the cooling power of water and secondly the heat transport capacity of the release of the latent heat of condensation in alliance with convection.

Whereas changes in temperature in the stratosphere tend to contemporaneous with changes in irradiance there is a lag in the troposphere associated with the time it takes for land masses to begin to heat up and push energy back into the atmosphere. Typically the hottest month of the year is one month after the sun reaches its zenith. Accordingly we would expect temperatures in the middle troposphere to show cooling a month after sunspot activity ceases. That seems to be the pattern.

104. Erl Happ
Posted May 5, 2008 at 10:02 AM | Permalink

Leif,
To assist with the interpretation of this imagery each successive period that is free of sunspots is outlined in a different colour to the last. So, between lines of the same colour there is no sunspot activity. Between lines of different colour there is sunspot activity.

Periods where sunspots disappear for a relatively long time are marked with heavier lines.

Cooling of the mid troposphere is evident soon after sunspot activity ceases. At higher levels of the atmosphere the pattern is much less coherent.

A cooling response can be generated at any level of sunspot activity (any stage of the solar cycle) because the Earth and its atmosphere will adjust its albedo to the prevailing background level of irradiance.

It follows that changes in the level of irradiance over long periods of time are less important than the impact of a particular warming episode on the temperature of the oceans. This very much depends upon when it occurs, close to January or July.

That we can not coherently explain why these changes occur does not invalidate the observation of a coherent pattern of change that is obviously influential in determining surface temperature.

Furthermore it is established that generalised warming and cooling in the tropics is related to solar activity. ENSO is co-incident, but the strength of the response in the Pacific owes a lot to factors that are unique to the Pacific Ocean.

105. Posted May 5, 2008 at 10:45 AM | Permalink

# 94

Leif, Thank you very much for the two tips… :) Now my question is… Is the series G (from your databases) the most appropiate for being collated with sunspots data? (scratch on my head).

106. cba
Posted May 5, 2008 at 10:45 AM | Permalink

102 (Leif):

Off the cuff, it’s hard to say just how accurate that left side is going to be for something other than liquid ocean water. First off, it’s 90%+ h2o with very little gases involved and there’s a tremendous amount of it. Second, it’s liquid so not quite the same as a gas.

I haven’t figured out why you put in the chart. It seems indicative of uV penetration into the ocean below the surface ‘skin’ that exists for IR which permits the injection of power further down but it doesn’t succeed in showing what reaches the surface in terms of medium to long uV.

Besides, it would seem that this el nino type stuff is churning the ocean quite substantially in a way.

107. Posted May 5, 2008 at 11:02 AM | Permalink

105 (Nasif): Column G is a good choice. I can already tell you that you will find B = 4.5 + 0.28 sqrt(SSN), as we did in some of the papers on my website.

108. Posted May 5, 2008 at 11:10 AM | Permalink

102 (cba): The graph is there to show that H2O is most transparent in visible and near-UV. I think what matters is simply the number of water molecules in the optical path. For the atmosphere that is equivalent to a water depth of 1-5 cm [depending on humidity]. It was simply to show that water is not a significant absorber of visible or near-UV. clouds do not absorb sunlight, they reflect it. right now a thunderstorm with a very dark cloud is passing overhead and it is very dark [I have to turn on the lights]. This is not because the cloud has absorbed all that sunlight.

109. Posted May 5, 2008 at 11:30 AM | Permalink

104 (Erl):

Furthermore it is established that generalised warming and cooling in the tropics is related to solar activity

That you keep saying it does not make it so. As usual, your diagrams make little sense. Maybe we can have some of the lurkers or other interested parties, explain them to me. As much as I hate to dilute the blog with these diagrams, I’ll just show one of yours where you have marked areas in the middle troposphere as Warm or Cold with little boxes [red and green]. I have tried, in vain, to see how the boxes relate to the temperature anomalies. For illustration I have circled a W/C pair with a light blue circle and another pair with a pink circle. To the best of my ability I have tried to see what the difference in the contours are between the W and the C within each pair. the background contours look identical or nearly so for the area where the box is:

The same could be said of most of the other boxes elsewhere in the diagram.
Your long list of complexities in #103 does make your missive look well-researched, but is just the opposite of making a good scientific point, namely to cut to the core issue.

110. Posted May 5, 2008 at 12:58 PM | Permalink

300 years cycle variation of Solar Irradiance: 0.110528628 W/m^2
∆T = 0.184 K

∆T = 0.11 (J/s) / 0.60099 (J/s*K) = 0.184 K

The asterisk in J/s*K doesn’t denote multiplication.

Does it mean that the change of temperature in the last 300 years has been only 0.184 K by the variation of SI?

111. Posted May 5, 2008 at 1:27 PM | Permalink

110 (Nasif): the 300-year TSI increase is too small, apart from the excessive number of decimals. A more reasonable value would be ∆I = 0.2 W/m2. Since T varies about 0.05K per 1 W/m2, the result would seem to be ∆T = 0.01 K.

112. cba
Posted May 5, 2008 at 2:22 PM | Permalink

111 (Leif):

I guess I can’t keep up with it all. Not that I disagree with the value so much for dT = 0.05K per w/m^2 but where does that one come from? It’s less than the simple calc’s I’ve seen by almost an order of a magnitude and it’s 2 orders of magnitude below what the IPCC and video game artists claim. :)

113. Posted May 5, 2008 at 2:32 PM | Permalink

108 (cba): I have been thinking a bit about the difference between gas and liquid. The water molecule with its two ‘arms’ or ‘Mickey Mouse Ears’ can twist, vibrate, rotate, stretch, bend, etc in many different ways giving rise to thousands and thousands of absorption lines. In the liquid some of these vibrations are restricted by the bond or proximity to other molecules, so there is a difference between the two phases. But this is mostly up in the infrared.
The large number of possible vibrational and rotational transitions is responsible for a large number of absorption lines, grouped in absorption bands, often referred to as polyads. An absorption band or polyad typically consists of several hundred lines that are grouped in certain [at times overlapping] wavelength regions. In the near-infrared and the visible there are bands at: 900-1030nm, 780-860nm, 690-740nm,630-660nm,585-605nm, etc.
In general, the intensity of absorption bands decreases with decreasing wavelength. The atmospheric absorption in weak absorption bands with polyad numbers up to the ones listed above are relevant for atmospheric absorption of solar light due to the high water vapor abundance in the Earth’s atmosphere. Absorption bands with higher polyad numbers [i.e. of wavelengths shorter than 600nm] are known but are so weak that they are considered irrelevant for radiative transfer in the Earth’s atmosphere.

This is clearly seen in the decreasing absorption peaks of for water vapor in this graph:

114. Posted May 5, 2008 at 2:36 PM | Permalink

# 111

Leif,

Thank you… you’re right. My mistake was to take 11yr cycles for the calculations of ∆I. If I convert ∆I to ∆q the result is 0.185 J/s. :(

∆T = 0.185 (J/s) / 0.60099 (J/s*K) = 0.3 K

No, it doesn’t coincide with your cipher. Wait, wait, wait… I’m wrong because I’m considering the Cp of CO2 alone. Let’s try with the mix of air in the troposphere at 1.5 m above surface:

∆T = 0.185(J/s) / 0.0481 J/s*K) = 3.8 K

Mmh… What’s wrong?

115. cba
Posted May 5, 2008 at 2:50 PM | Permalink

113 (Leif):
There’s that stuff and then there’s the broadening factors. I also expect there to perhaps be somewhat of a bit of a planck continuum – but that’s supposition rather than knowledge on my part. One of the things I’ve left til later so far has been the nature of h2o in the nonvapor form in the atmosphere.

High thermal energy gas create a continuum while solids do so at low energy. Water liquid should fit in between more towards the solid than the gas I would expect. Another thing delayed in my studies is the nature of radiative concerning small particulates and water droplets. General concepts suggest that items small compared to radiative wavelengths are not good at radiating and FIR is getting down into the realm of size vs wavelength. I don’t have a feel for how this plays out – bouncy springy molecules versus classical EM radiating dipole antenna stuff.

Clouds are going to have superb visible (solar spectra) reflectivity topside – maybe not to 90% but never the less – way up there with some absorption and some translucence. On the bottom side, they should have extremely low reflectivity and no transmission, only absorption – or so I would expect and then perhaps primarily convective transfer inside and topside, probably a continuum radiation as it’s likely to be ice crystals – although again – I’ve got no feel for any effects from the size of the particles other than EM suggests problems.

116. Posted May 5, 2008 at 3:02 PM | Permalink

112 (cba): I don’t know how the IPCC circus does it, but here is the standard way of getting those numbers:
Incoming solar energy at the surface of the Earth [actually TOA] is Fs = (1-A)S/4 where S is TSI = 1361 W/m2 and A is the albedo = 0.3 [say]. For a black-body of temperature T, the outgoing longwave flux averaged over the globe is Fl = sT^4, where s = 5.671×10^-8 W/m2/K4 is the Stefan-Boltzmann constant. Equating Fs and Fl gives an expression for T =((1-A)S/(4s))^(1/4). Inserting S = 1361.0 gives T = 254.5711 K. Inserting S = 1361.2 gives T = 254.5805 for a difference of ∆T = 254.5805 – 254.5711 = 0.0094 K ~ 0.01K as quoted. If I have made an error in the fourth decimal place, I’m sure that many would delight in correcting me :-)

117. Posted May 5, 2008 at 3:09 PM | Permalink

114 (Nasif): see #116.
115 (cba): Nothing of this matters, because the only question was Erl’s insistence that water vapor would suck up UVB and thus heat the middle troposphere, and that that would be the mechanism for the purported ‘strong solar influence’. Clearly we can agree that that doesn’t happen, bouncy, springy molecules notwithstanding.

118. DeWitt Payne
Posted May 5, 2008 at 3:12 PM | Permalink

cba,

Slightly OT, but did you ever manage to include the water vapor continuum absorption/emission, thought by some to be caused by a water dimer, in your atmospheric model?

119. cba
Posted May 5, 2008 at 3:22 PM | Permalink

116 (Leif):

that of course gives value for a hypothetical ball with no absorbing atmosphere yet with similar albedo. It doesn’t seem applicable as even a simple rendition of what we’ve got – which is a mean surface of 288K and enough stuff in the air to absorb around 160W/m^2 of outgoing. The calc is very little different from your’s and permits one to use a delta due to all atmospheric absorption (288-255 K) which then relates all absorption or change in w/m^2 to T complete with all existing active feedback mechanisms as the 33K is what is – even though averaged in effects from 0. Seems like it’s about 0.2 K/w/m^2 and it’s real, it’s what has happened. :)

120. Posted May 5, 2008 at 3:32 PM | Permalink

118 (cba): no, I’m not that sophisticated. I did not in #116 include greenhouse gases [mostly water vapor] which would add another 33K to the effective temperature. And then there is this: The net radiative flux, ∆F = Fs – Fl, is zero at equilibrium. If you had some drastic changes in Fs and Fl, the change in net energy would be ∆F = ∆Fs – ∆Fl. To get back into equilibrium a temperature change ∆T is required, which we parameterize by this sleight of hand: ∆T = L∆F, where L is called the climate sensitivity factor. If we neglect feedbacks in the system, L can be estimated as 1/(∆F/∆T), hence: L = 1/(4sT^3) = T/(4Fl), which would yield L = 0.3 K/(W/m2). If, say, the effect of a doubling of CO2 would be 3 W/m2, then with L = 0.3, we would get ∆T = 1K. With feedbacks, the change would be larger, depending on your model and assumptions, etc. This is how far my model sophistication goes. Remember, I’m just a solar physicist.

121. Posted May 5, 2008 at 3:44 PM | Permalink

119 (cba): You might present the ‘calc’ here to the same level of simple understanding as my model in 116 for the benefit of all. For me, 0.01K is not substantially different from 0.2K, both say that the difference between 1700 (the end of the Maunder Minimum) and, say, 1975 , or from 1850 to 1975, is not mainly due to the Sun [and clearly not to AGW], the conclusion must be [speculation:] that there are longer ‘cycles’ [of unknown, but natural, causes] that control most of the climate on a centennial scale [/speculation]]

122. Posted May 5, 2008 at 4:01 PM | Permalink

121 (me): if I add 155 W/m2 to Fl in my expression for T, I do get T = 288.5570 for S=1361.0 and T = 288.5634 for S=1361.2, for an even smaller difference ∆T = 0.006K. So, show me how to fix the expression for T.

123. cba
Posted May 5, 2008 at 4:39 PM | Permalink

118 (DeWitt):

Things have pretty much been stagnant since the first of the year. I need a computer that isn’t going to fry when it becomes processor bound for a few hours. My little laptop started developing a simmering thermal grease smell to it that lasted for weeks after the last run. That and my teaching load officially hit 3/4 full time for this semester (and unofficially full load) and I’m not an experienced prof. This summer promises to be worse although I think a desktop quad core may be in my immediate future. However, it still has to take a bit of a back seat to the official unoffical requirements – like the need to rewrite a couple of lab manuals and revise the lab into the 21st century.

I had been hoping to get more done this time around but it’s been fairly incredible. I was putting off trying to learn about the continuum until other things were in better shape with the model and perhaps some comparisons could be made to verify it. And now, it’s finals week and I’m starting off the summer with a class I’ve not taught before to an even more naive crew.

So consequently, aerosols, particulates and continuum emissions are probably going to have to wait for a bit longer.

124. cba
Posted May 5, 2008 at 5:29 PM | Permalink

122 (Leif):

I’ll give it a shot here with a couple of approaches and some additional info to try to make it fit together better. It’ll be off the cuff again so hopefully I won’t scramble it.

Assuming 1361 W/m^2 is coming in that’s 1361/4 =340.25 to convert power / m^2 from the disk cut out by the Earth at the nominal 1 AU distance, into the average power over the surface of the Earth as a sphere. This becomes the averaged power received per area assuming uniform illumination of all parts of the Earth at all times.

The estimated 1976 mean Earth surface temperature is 288.2 degrees kelvin. Using Stefan’s Law Power = epsilon * sigma * T^4 , where sigma is Stefan’s constant = 5.67E-8 (W/(m^2 * k^4) and epsilon is just under 1.0 for this temperature and for the Earth’s surface and can be very accurately approximated as 1.0. epsilon is the emissivity or radiative efficiency and is itself an approximation.

For Earth to maintain a stable temperature, the total power incoming must equal the total power outgoing on the moderate to fairly short term (multiyear). The power in (using average) is the Solar radiation minus the albedo or energy reflected from Earth. The nominal value for albedo – a number 0.00 to 1.00 – is 0.30. Measurements of this albedo have been rather low priority and sporadic – even in the modern era but it turns out it is more variable and more important than just about anything that has been measured. It varies by several percent during the year and by even more as reconstructed over the last two decades, apparently varying as much as 8-10% over a multiyear time frame.

As Leif’s post already show, the total energy in is incoming solar minus albedo, Fs*(1-a) since a is given as a fraction. Note the higher the albedo, the lower the incoming power and the nominal value is 340.25 * (1 – 0.30) = 238.18 W/m^2. The Earth must shed on average, this amount of power. Note that 1 W /m^2 is 1 Joule/second/m^2 or 1/4 calorie /second m^2 – the energy needed to raise 1 gram of h2o by 1/4 degree K (roughly speaking). Any imbalance will heat or cool the Earth and it will not take all that much time as there are 88,400 seconds in a day – and we’re using daily averages – the amount of power that would be incoming 24 hours per day, not just during daylight.

Applying Stefan’s law to the surface T=288.2K, we get power/m^2 outgoing = 391.16. This output power depends upon the T of the surface and its emissivity – assumed to be 1.0 for now. Doing this for clear skies to keep it simple, 391.16 – 238.18 = 152.98 ~= 153 W/m^2. This means that there must be something headed to the surface to balance our equation to 0 to keep the T from having to change as thermal energy is depleted.

Considering that as Leif has calculated, the Earth should be at 255 K assuming the same amount of incoming power from the sun yet the average T is 288, this is a difference of 33 degrees Kelvin.

This means all contributions from water vapor, clouds, aerosols, particulates, and other GHG components in the atmosphere contribute a warming of 33 degrees total from this 153 W/m^2 back radiation. One can then get an average of 33/153 = 0.21 degrees K per W/m^2.

This is the real effect on the current atmosphere although it is an average T per W/m^2 not the instantaneous value required to add a little more to it. It also includes all feedbacks in current operation.

125. cba
Posted May 5, 2008 at 7:03 PM | Permalink

124 (continued):

Another way to view this is that the surface is radiating 391 W/m^2 yet only 238 W/m^2 can be allowed to escape the Earth or else we will lose energy on an ongoing basis and so must cool. Again note this is highly simplified for understanding and there’s a lot of the radiated energy coming from the atmosphere going out rather than just the surface.

We can assume that this 153 W/m^2 is the total absorbed by the atmosphere (if we assume that all power out to space originates at the surface). there is then 153/391 = 0.39 as the fraction of radiated power absorbed on the way out.

Now, let’s assume there is an extra 1 W/m^2 coming in from the Sun or an extra 1 W/m^2 being absorbed in the atmosphere. For 1 W/m^2 to be radiated from the surface to reach space, one can assume a fraction, 0.39, is absorbed or (1.00 – 0.39 = 0.61 is passed). This means that 1.64 W/m^2 must be radiated additionally. 288.2 radiates 391.164 w/m^2 so adding 1.64 to this gives 392.8 W/m^2. Running Stefan’s law backwards yields 288.5, a 0.3 deg K increase for balance. This is a worst case scenario that assumes no radiation from the atmosphere. Asumming no additional absorption for the increase in output yields a result of 288.38 K for an increase of 0.18 K. This provides more of a best case scenario and suggests that the real value is between 0.18 and 0.3K.

Referring back to the previous post, we see that the average value is about 0.21 – 0.22 K per W/m^2 which is within the range of 0.18 and 0.3K per m^2.

Going back to the situation where there is virtually no atmospheric absorption, the T is 255 (and radiative output is 239.74) and the increment of 1W/m^2 raises the T by 0.27 K / W/m^2. Considering that the earlier T rise is greater than the average, the latter T rise must be less than the averaged value. If one takes a simple average between the first direct value and second direct calculated value, (0.18+0.27)/2 = 0.22 K/W/m^2, a right at the overall averaged value. Taking the midpoint of 16.5 K temperature rise (271.5k, 308 W/m^2) gives a final T of 271.72K with a 1 W/m^2 increase or 0.22K rise in T for 1 W/m^2. This suggests a best value of 0.18K/W/m^2 for an increase of 1 W/m^2 power and also suggests that there is not an increase at the surface required to handle an additional power to compensate for absorption by the atmosphere as assumed above.

Net result 0.18K/W/m^2 for a calculation of the most recent added 1 W/m^2 assuming no added power output necessary for more absorption and it also agrees with the calculation of what would be expected by using the real average and adjusting for the calculated difference between the first ghg W/m^2 and the last.

126. Posted May 5, 2008 at 7:04 PM | Permalink

124 (cba): So far we agree on the numbers and the theory, but there is one little step that needs attention: You said “One can then get an average of 33/153 = 0.21 degrees K per W/m^2“. Now, imagine that the original TSI increases from 1361.0 to 1361.2, i.e. by 0.2 W/m2. Are you then suggesting that one should multiply that 0.2 W/m2 by your 0.21 K/W/m2 to get a ∆T of 0.2*0.21 = 0.042K effect? Or, as I am certain you will agree to that we should divide the 0.2 W/m2 by 4, so that the total effect becomes 0.2 * 0.21 / 4 = 0.01K? which is in agreement with my number.

127. cba
Posted May 5, 2008 at 7:32 PM | Permalink

125 (continued):

The generally accepted W/m^2 for co2 changes are something like 1.6 W/m^2 for changes from 1750 to now and an expected ~3.6 W/m^2 for a doubling from 1750 to (whenever).

The results of 0.18K/W/m^2 turn out to be 0.29 deg K for the rise since 1750 and 0.65 K for the ultimate expected doubling. Remember, this is consistant with the real atmospheric response based upon both the REAL LIFE average response and the closest calculated response to the real one, including all actual feedbacks currently operating in the atmosphere.

Now, for another comparison. Leif posted an albedo reconstruction around here for the Earth over the last 20 some odd years (from Palle & ???) that indicates multiyear variations as much as 8-10% in albedo. Since albedo is nominally 0.30, 10% is about 0.03 and this affects incoming average radiation of 340.25 and provides a variation of 0.03 * 340.25 = 10.2 W/m^2 which is over 6 times the variation presumed to be caused by the current increased co2 levels from 1750 and roughly 10 times the variation in solar TSI as estimated by Leif.

As such, it would seem that this is the smoking gun in what is causing the T variations. The fact that not nearly 1.8K T variation has occurred in the last 20 years would suggest that the maximum variations of the albedo were not sufficiently long to overcome the inertia of the thermal system. It would also suggest that there is a very strong negative feedback that uses cloud albedo to regulate Earth’s T.

An example is that 1998 was roughly the hottest year in a long time but it was also the year that albedo hit a relative minimum. The fact that 1998 was a relative peak T which has since plummeted should falsify any claims to positive forcings (clouds & probably anything else) at work as co2 has continued to rise slightly while T has fallen (and albedo has risen, despite having been at a somewhat higher T).

As to what affects the cloud albedo, this becomes a suitable effort to concentrate on. It is clouds as surface albedo is only a small overall fraction of the total albedo. The Cosmic Ray / magnetic field / cloud fraction continues to remain ellusive. It seems to be there and it seems there is some sort of imprint of the Sun upon Earth’s weather and climate and possibly cosmoclimatological long term factors based upon the solar system’s location in the galaxy arms. Magnetic fields, solar wind, variations in upper atmospheric chemistry due to high energy particles and uV all may have some sort of influence as well. Also, it seems that random variations can occaisionally combine to create even more phenomena.

128. cba
Posted May 5, 2008 at 7:42 PM | Permalink

126 (Leif):

Of course that delta P has to be dealt with as apples and apples so it does have to be divided by 4. The radiative output is the averaged value as it transpires regardless of incoming. You’ll note that for my Tome here that I’m using an averaged value, requiring solar insolation changes to be divided by 4 to relate to the averaged value (insolation of 340.25) as you assumed and anything related to outgoing such as a ghg change must be dealt with on the average but these don’t have the /4 factor involved as they tend to be already averaged.

In other words I’m in agreement except that it should be 0.18 rather than 0.21 or 0.22 as this is actually a better number for the next or last W/m^2. 0.18*0.2/4 = 0.009 ~= 0.01 K.

129. Posted May 5, 2008 at 7:53 PM | Permalink

128 (cba): just to rub it in, we agree that your problem in #112:

where does that one come from? It’s less than the simple calc’s I’ve seen by almost an order of a magnitude and it’s 2 orders of magnitude below what the IPCC and video game artists claim.

is now resolved and you no longer are puzzled or disagree with my number?

130. Posted May 5, 2008 at 7:58 PM | Permalink

127 (cba): It is obviously the albedo that is the key. The [unresolved] problem is that this is a chicken and egg problem: is the albedo changes causing the temperature changes or is it the other way around? or is it some intricate zig-zag process with positive/negative feedbacks coming and going where both A and T change together in a coupled manner? IMHO it does not further progress to be too dogmatic about this.

131. Posted May 5, 2008 at 8:02 PM | Permalink

# 126

Leif,

Your number (0.01 K) is correct:

∆I = 0.184 W/m^2

∆T = ∆I (Cp [d])

∆T = 0.184 (W/m^2)(0.0518929249 K/(W/m^2) = 0.0095482981816 K; rounding out, 0.01 K

132. Erl Happ
Posted May 5, 2008 at 8:14 PM | Permalink

109 (Leif)
The data on 2008 is unreadable because the depth required for the graphic is resulting in a compression of its width. I will email it to your gmail address. This is the graphic to discuss because the contrast in solar activity is much more obvious and the temperature response in the atmosphere very plain. This is not the case for 2007.

Re your query on the 2007 data: If the vertical extent of the warm anomaly increases I interpret that as warming. If the depth of the warm anomaly diminishes or disappears entirely I interpret that as cooling. If I have made an error anywhere (more than possible) I am human.

Let’s cut to the core issue as you suggest:
Joe D’Aleo has an article on the great Pacific climate shift.
Does this represent an amazing degree of ‘internal variability’ OR something else entirely? Joe links the phenomena with ENSO. The cool phase shift of the PDO depends upon a reduction in the heat gained in the tropics.

The extent of the warmth gained by the oceans in the tropics determines the air temperature at high latitudes where solar radiation is weak or absent. I have shown that during La Nina conditions (generalised cooling across the tropics) the temperature in the middle troposphere in the tropics close to the equator falls by up to 4°C. The tropic, close to the Equator, is the only part of the globe that experiences very little seasonal variation. That clarifies the signal. I can demonstrate this association between temperatures in the tropical troposphere, the tropical oceans and the temperature in the mid troposphere over the entire period since 1979 when the satellite data begins. The data array necessary to display the course of temperature variation over years is too large to be displayed here. But I can email them to anyone interested. I can be contacted via my website.

I see a thread on this blog devoted to the extent of ice in Antarctica. That is the source of next year’s cold up-welling current off Peru. If you look at current sea surface temperatures in the Pacific adjacent to Peru (http://www.eldersweather.com.au/climimage.jsp?i=sstag) you will see La Nina refreshing itself for another assault on the Pacific. So, we had La Nina in SH summer 2006. The El Nino of SH summer 2007 was weak. Much cloud cover in the SH summer of 2008 has meant that the heat gain in the Pacific this summer has been weak. The positive temperature anomalies in the oceans of the southern hemisphere are currently evaporating by the day. The big cooling to be experienced in the Northern Hemisphere is still to come.
Mechanisms: If the mid troposphere warms by 4°C cloud cover and reflection of radiation falls away. The surface heats and the ocean gains energy. If the mid troposphere cools (my post 103), reflection increases and the oceans lose energy. We can’t explain it…too bad. Does it make it any less real? No. Is this the temperature fluctuation that is commonly observed on Earth? Yes. How big is it? The biggest we see. What causes it? The temperature fluctuations in the middle troposphere are clearly tied in with those in the upper atmosphere. QED solar.

Is long term change in the suns irradiance important in determining surface temperatures? Unlikely. It is the dynamic that drives tropical temperature fluctuation that determines that.

133. Posted May 5, 2008 at 8:26 PM | Permalink

131 (Nasif): I’m glad that we are all converging to the ‘truth’. As Napoleon once said: “truth is a bunch of lies agreed upon” :-)

134. cba
Posted May 5, 2008 at 9:00 PM | Permalink

129 (Leif):

Evidently, my problem was not realizing the 0.05 value was related to the disk value rather than the averaged sphere power/area tends to be somewhat customary. As I mentioned in 112, I was more curious as to where the number came from than was in disagreement with it and was mostly curious. Usually, I just use the averaged 33/153 relation that is slightly higher than the most recent W/m^2 delta T value.

As a quick asise, the accuracy of this extremely crude single valued method (0 dimensional versus 3 d modeling) is probably far better than most would imagine. After all, It’s a T^4 relationship to power, not a linear one.

The only very minor disagreement is merely that the 288k T should be used in your approach rather than 255 – which is basically the airless ball approach. It’s still fairly accurate and it’s probably within the overall error margin but it conveys a confusing situation as 255 K isn’t the current mean T. It is also a surprising result that there appears to be no additional absorption effect by the increased surface radiating despite the fact that it would seem that there should be. There again, the whole reality of the simplification would seem to be a bit unreal yet the results still come out great.

135. cba
Posted May 5, 2008 at 9:12 PM | Permalink

130 (Leif):

that 1998 was a peak T with low albedo suggests causality. The system apparently ‘recovered’ with no other apparent assistance of any consequence. If albedo were driven down by rising T, creating a positive feedback, the system would be horribly unstable and couldn’t have recovered without the presence of a yet greater than 10W/m^2 type of forcing and/or a forcing of the albedo. It’s clear that albedo has to have an effect on T, an extremely strong negative effect relative to just how much it can vary. It’s possible in some areas that there could be positive effects but the overall must be negative. However, in the overall situation, I would think to claim it has both negative and positive feedbacks would be like claiming a number (not zero) is both positive and negative. It can only be one or the other.

136. Erl Happ
Posted May 5, 2008 at 9:13 PM | Permalink

130 (Leif)

is the albedo changes causing the temperature changes or is it the other way around?

At http://www.bom.gov.au/climate/enso/ You will see that the BOM has decided that the La Nina is finished. The meteorological signals now point towards warming. The Sun may have something else in mind. Your question is about to be answered.

137. Posted May 5, 2008 at 9:38 PM | Permalink

# 136

From your link, which is a trustable source I deduce that La Niña is declining, but it has not finished.

From myself: Perhaps La Niña will finish in July.

138. Posted May 5, 2008 at 9:45 PM | Permalink

135 (cba): what I meant was that the sign of the feedback could vary from place to place and from time to time and the overall could then zig and zag, but, of course, not at the same time. I have also previously [I think over in the ‘ice age’ thread] suggested that the most likely feedback is negative, because of the ‘blowup’ problem, but this is something that needs careful consideration [and I haven’t done it justice been busy with sunspots and things solar]. Erl’s suggestion in #136 that my question is about to be answered, is, of course, silly as it must have been answered innumerable times in the past. There is nothing special about what is just about to come to pass. in #132 “you will see La Nina refreshing itself for another assault on the Pacific” and now we hear that “the BOM has decided that the La Nina is finished. The meteorological signals now point towards warming.” Evidently the BOM sees things differently, maybe it was this disagreement that Erl is calling attention to. In general, it is better to just state your view, rather than pack it in in layers of sarcasm, suggestive hints, and such. Anything to say? : say it!

139. Posted May 5, 2008 at 9:47 PM | Permalink

137 (Nasif): La Nina finished? I have no idea. Ask some Peruvians.

140. Posted May 5, 2008 at 10:37 PM | Permalink

# 139

Leif,

I didn’t say that La Niña finished, but Earl Harper (# 136). La Niña didn’t finish yet. La Niña is still there, although it is declining. Perhaps La Niña will persist until July.

141. Posted May 5, 2008 at 10:47 PM | Permalink

Oops! Erl, I apologize… I wrote your name from my memory.

Someone has sent to me a message asking for the meaning of d… Well, d is not for distance, but for action, and it is expressed in Kg/m^2*s. I know some authors use A, but it could be taken like area.

I’m using this medium for answering your question because all the e-mails I have sent to you have been rebounded. No name, no address… Wow! How did you do that?

142. Erl Happ
Posted May 6, 2008 at 12:08 AM | Permalink

138 (Leif)
Perhaps I have been too cryptic. My opinion is that, if the sun continues in the current vein of low sunspot activity the current episode of cooling ocean temperatures and cooling tropical troposphere will continue. It will do so until we see a significant warming of the middle troposphere. That warming will be accompanied by a warming of the upper atmosphere as sunspot activity picks up.

The ENSO models presumably rely on atmospheric variables and past experience to predict the course of the warming cycle. They are predicting an end to La Nina cooling. None of them predicted the timing or the strength of the current La Nina. There are about 20 of them. Their predictions do not agree with each other. They are not likely to have improved over the last few months. The BOM is wedded to the viewpoint that CO2 is forcing more frequent El Nino events.

The change in the speed of the trade winds and the pressure zones that are seen to be responsible for ENSO may in fact be a response to ENSO. I suggest that the current warming signal in the meteorological variables is likely to be short lived.

I do not observe any marked upswing in solar activity. I observe a new up-welling of very cool waters in the Eastern Pacific. I suggest that the Southern oceans have not gained the energy that is usual in summer (because of extensive cloud cover) and these waters will soon become dominant in determining sea surface temperatures at the equator. Even if solar activity picks up, it will not make much difference due to the fact that we are entering NH summer when global cloud cover normally falls away by 3% as the land masses of the NH shed heat into the atmosphere as fast as it arrives.

Next NH winter will be cold and people will conclude that the 30 year flip flop in the PDO has indeed occurred (sarcastic). This will happen whatever changes the sun brings to the middle troposphere and above.

143. Posted May 6, 2008 at 7:54 AM | Permalink

April is the worst time of year for the ENSO models – this is the period of the well known ‘predictability barrier’ where the models are are all over the place and there is no clear idea where things will actually go.

During this time of year the BoM tends to wing it by guessing where they think it might go, and the poor track record of their guesses reflects the true uncertainties of the situation.

Things usually slowly improve as May progresses, with the models getting a far better handle on what’s happening by around June, with more accurate ENSO forecasts in the later part of the year.

I guess what I’m saying is it is still a bit early to take much notice of the ENSO models and agency forecasts, so a decent pinch of salt is still required.

For what it’s worth, my take is that the cool SST anomaly along coastal South America may be the seed of a return to a deeper La Nina later in the year, but I could be wrong.

144. Erl Happ
Posted May 6, 2008 at 10:31 AM | Permalink

Leif,
What is the UV absorption characteristic of minute crystals of dirty ice of the sort that might be found in cirrus clouds of the middle troposphere?

145. Posted May 6, 2008 at 11:14 AM | Permalink

144 (Erl): Water [and ice] is basically transparent to near-UV. Now any dirt [e.g. black soot] will absorb all UV and visible [unless you have transparent dirt :-) ]. But the issue is really not about UV as the dirt will also absorb the much larger energy of the visible light. The fixation you have to get out from under is that the minute amount of near-UV will control any heating or absorption. Also, there is not much dirt on the ice crystals in cirrus clouds and cirrus clouds usually form above the middle troposphere or at the very upper edge, so are not really a middle troposphere thing, and finally, I think that reflection by the crystals is the primary way of blocking UV and visible, rather than absorption.

146. Posted May 6, 2008 at 11:27 AM | Permalink

145 (me): The absorption of radiation by ice has become important for a completely different reason because of the construction of the AMANDA neutrino detector in Antarctic ice: http://www.atmos.washington.edu/~sgw/PAPERS/2006_icemcx.pdf

From the article:
Between 300 and 600 nmthe absorption by ice is so weak that for some geophysical purposes it may as well be set to zero, for example, when computing absorption of solar radiation by ice clouds, because path lengths of photons through atmospheric ice crystals are very small compared to the absorption length. Clean fine-grained snow reflects 97–99% of the incident sunlight in this spectral region,2 and the error bars sometimes encompass 100% reflectance, again consistent with a near-zero absorption coefficient.

There is also some discussion of the influence of dust. All in all, fascinating reading.

147. DeWitt Payne
Posted May 6, 2008 at 2:09 PM | Permalink

Leif,

If the absorption coefficient for visible light by water and ice weren’t small and nearly constant over the bandwidth, then clouds and snow wouldn’t be white. Deep water (or deep clear ice for that matter) is blue because there is a small increase in absorption in the visible as the wavelength increases so red and green are absorbed and blue is scattered back.

148. DeWitt Payne
Posted May 6, 2008 at 2:13 PM | Permalink

cba,

I have a quad core desktop that doesn’t do much unless I’m processing video files, playing Assasin’s Creed or a Blu-ray disk. Even with a Blu-ray, it’s running at less than 20% capacity. The logistics are probably impossible, but otherwise I don’t have a problem.

149. Posted May 6, 2008 at 2:54 PM | Permalink

147 (DeWitt): glad to see that you agree and even give a non-technical explanation for the benefit of Erl [although he was more interested in UV]. Thanks.

148 (DeWitt): OT, please refrain from this.

150. Sam Urbinto
Posted May 6, 2008 at 5:16 PM | Permalink

Over at the BB there’s a post on the idea that odd solar cycles affect the anomaly more than even ones do. Something like 23 and 25 will have higher anomalies than 22 and 24. I have no idea, but it sounds just as easy to explain this as anything else. http://www.climateaudit.org/phpBB3/viewtopic.php?f=6&t=271

As far as the rest, isn’t it just easier to say what the sun does at .25 to 2.5 microns when it comes in versus when the surface lets go that much larger wattage in the 2.5 to 70 range, other things happen?

Or that insolation is surface constrained and outsolation is atmosphere constrained? (or whatever the wording of it)

151. Sam Urbinto
Posted May 6, 2008 at 5:25 PM | Permalink

Leif #149 I think (pretty sure) DeWitt was talking about what cba said in #123 about modeling things with computer power. Not that it’s still not off topic (I don’t know), just wanted to mention the why of the answer.

152. Andrew
Posted May 6, 2008 at 5:31 PM | Permalink

Leif, is there any particular reason why the data here only go back to 1700? I’m interested in getting the data extending through the Maunder Minimum, do you know where I could find it?

153. Posted May 6, 2008 at 6:05 PM | Permalink

152 (Andrew): The reason is that we don’t know what it is before 1700. If you don’t mind that caveat, you can find Lean’s data back to 1610 here:
Use the column marked ’11yrCYCLE’, the other one is just fantasy.

Lean’s reconstruction is based on what I think is a faulty sunspot series. If you believe my corrected series, then you can find my fantasy here: http://www.leif.org/research/TSI-LEIF.xls
Use column L in bold typeface. It is a running 12-month mean of reconstructed TSI for each month [except that reconstruction on a monthly basis is very noisy, so don’t use the individual monthly values]

154. Andrew
Posted May 6, 2008 at 6:12 PM | Permalink

153 (Leif): Thanks. I’ll keep the caveats in mind. :)

155. Posted May 6, 2008 at 6:43 PM | Permalink

150 (Sam): I would not put much into the even-odd thing. Here are the numbers from your BB cite:

The text says:

So, “even” cycles are fairly flat but ending with an upward flourish, and “odd” cycles start with low temperatures followed by a peak in temperatures 2-4 years after maximum.

I have plotted the cycles back-to-back [and then repeated the pattern]. The upward flourish must be followed by a much larger precipitous drop to get down to the next [odd] cycle. As the change from cycle to cycle takes place in the depth of a solar minimum where nothing is happening, it seems unlikely that that would cause such a large drop in the temperature anomaly. I think the whole thing is just noise.
From a physical point of view, you might argue [as some do – not me] that Galactic Cosmic Rays have some effect. There is a slight difference in the behavior of GCRs from cycle to cycle, but that is from maximum to maximum, halfway through an even cycle, say. I also don’t see any pattern there, so I don’t feel any need to speculate about cause and effect.

156. Posted May 6, 2008 at 7:13 PM | Permalink

Have you read this new? Another superflare, like in 1998… scaring.

157. Posted May 6, 2008 at 7:29 PM | Permalink

156 (Nasif): maybe you mean 1859? there was also a big one in 1989. Ed Cliver and myself have studied the 1859 flare in detail. You can learn more here: http://stinet.dtic.mil/oai/oai?verb=getRecord&metadataPrefix=html&identifier=ADA434906
then click on the Handle / proxy Url : http://handle.dtic.mil/100.2/ADA434906

158. Erl Happ
Posted May 6, 2008 at 10:21 PM | Permalink

145 (Leif)
Its the below zero through to -70°C portion of the troposphere (what I call the middle troposphere) that is showing the large temperature swings between episodes of tropical surface and atmospheric cooling and warming (4°C swing in the middle troposphere), much larger swings occur in the middle troposphere than are seen in the portion of the troposphere where temperatures are above freezing(1°C or less).

The question is: What has caused the below zero portion of the troposphere to cool so much? Why is there progressively more OLR in the tropics over the last 15 months as the surface cools. It has to be cloud. The Earths albedo has increased.

I have shown in previous posts that positive anomalies in the troposphere are associated with surges in irradiance? I have also shown that negative anomalies are associated with slumps in irradiance and a slackness in sunspot activity. You are the expert on solar matters. What could be responsible?

159. Erl Happ
Posted May 6, 2008 at 10:42 PM | Permalink

145 (Leif)
Perhaps this is the sort of cloud created in the middle troposphere that sets off a cycle of cooling. The Sahara is one place where the heat returning to the atmosphere on a daily basis would be expected to maintain a cloud free environment. Photo is 5th March 2008.

160. Posted May 6, 2008 at 11:00 PM | Permalink

Leif,

From your book, if you and you co-author only compiled flares = or > than X11 class, than the 1997 flare was not so important? BTW, the table VIII is very interesting.

161. stargazer
Posted May 7, 2008 at 1:29 AM | Permalink

159(ERL)
I dont know if this means anything, but I look at the sky on a daily basis,

and have noticed that (at least where I live on the south west coast of England) that the sky (even when clear of clouds) looks ‘hazy’ that is it looks blue with a thin ‘white’cast over it. I have not seen a ‘truly blue’ sky in months. I remember skys so blue that at the apex you could ‘see’ the sky as dark blue. but as I said I have not seen this for a very long time.

162. Erl Happ
Posted May 7, 2008 at 3:00 AM | Permalink

The hallmark of climate change in California since the 1970’s has been warmer winter and spring with less frost. Maximum temperatures have not risen. Sea surface temperatures drive the local climate.

Spring 2008 has been different, as this news report reveals:

“It was cold in lots of places,” said Karen Ross, president of the California Association of Winegrape Growers. “There’s lots of misery to be shared.”

The frosts hit all over Northern California, including Mendocino and Lake counties in Northern California as well as the Napa Valley and Sonoma County regions.

The cold also was felt in Central Coast vineyards.

Damage was spotty, a hallmark of frost, with some vineyards singed in only in a few corners.

Sinking temperatures dragged growers out of bed as frost alarms on vineyard thermometers went off.

Although a little frost isn’t unusual, the cold snap that lasted 20 or more nights in some places was a first for many vintners.

“The last one we had that was anywhere near this brutal was back in the ’70s,” said Napa County Agriculture Commissioner Dave Whitmer.

163. Posted May 7, 2008 at 5:48 AM | Permalink

(cba):
email from Enric Palle: “reading the sentence it seem obvious that we were not very fortunate in the choosing of words, or made a typo. Normally “on the time scales of solar evolution” means millions to billions of years, and clearly the Sun has been dimmer in the past (the faint Sun paradox for example). The sentence refers to the irradiance variability over the past few centuries to millennia.

164. See - owe to Rich
Posted May 7, 2008 at 2:45 PM | Permalink

155 (Leif)

Thanks for doing a plot of my data – but you’ve missed the point a bit in your “The upward flourish must be followed by a much larger precipitous drop to get down to the next [odd] cycle”. Since each cycle has had its mean subtracted, the mean level of the next cycle might be higher (and on average has been over the 150 years) so the drop to the new cycle isn’t as great as you suggest.

The point is that at about 3% significance, the data for even and odd cycles (after subtracting the mean) are different. It could just be chance, or it could be a very weak signal from the changing polarity of the odd and even cycles, and some unknown interaction with Earth’s magnetic field.

Rich.

165. Posted May 7, 2008 at 3:11 PM | Permalink

164 (Rich): so tell us what the two means were, then we can substitute the ‘might’ with an ‘is’, perhaps.

166. Erl Happ
Posted May 7, 2008 at 4:12 PM | Permalink

Leif,
How is the task of deriving the actual temperatures for the tropical troposphere going? Is it possible to get data for the different levels, especially for that above and below the freezing point of water?

Water temperatures along the Equator reach a maximum in April each year reflecting the high rate of absorbtion by the more extensive ocean of the Southern Hemisphere between September and March. It is a little unsettling to know that the waters along the Equator are not going to get any warmer than they are right now, for another 12 months.

Looking at the flux in ocean temperatures since 1982 and going simply on the morphology of the current cooling event by comparison with those of the past, and timing of heating events in relation to the solar cycle minima, I estimate solar minimum July 2009. Next El Nino 2011.

167. Posted May 7, 2008 at 4:34 PM | Permalink

166 (Erl): I’m using the standard definition of LT [lower troposphere] and MT [middle troposphere].

Is it possible to get data for the different levels, especially for that above and below the freezing point of water?

Those levels will move up and down from day to day and from place to place so it is not clear what you are asking for here. Off the cuff, I guess the answer will be “no”.

168. Erl Happ
Posted May 7, 2008 at 4:39 PM | Permalink

167 (Leif)
What is the standard definition? Hopefully the two will not be that different.

169. Posted May 7, 2008 at 5:12 PM | Permalink

168 (Erl): The microwave/satellite measurements apply ‘weighting functions’ to each channel and the result can best be described by this Figure:

Where the nomenclature is:
TLS – lower Stratosphere
TTS – Troposphere/Stratosphere boundary
TMT – Middle Troposphere
TLT – Lower Troposphere

170. Posted May 7, 2008 at 5:16 PM | Permalink

169 (me): For reference, the TMT [Mmiddle Troposphere] is roughly the 4-7 km layer.

171. Sam Urbinto
Posted May 7, 2008 at 5:41 PM | Permalink

Leif #167

That’s the key: “levels will move up and down from day to day and from place to place”

Not to mention it varies in level just from position on the Earth even if everything else stays the same.

172. DeWitt Payne
Posted May 7, 2008 at 6:28 PM | Permalink

The y axis in the figure should really be expressed as pressure, not altitude. TMT in the Arctic or Antarctic will be considerably lower altitude than in the Tropics. Also, the weighting functions aren’t applied to the channels, they are a property of the channel and the result of things like the finite bandwidth of the channel and the slope of the wing of the oxygen absorption curve at 60 GHz at the channel wavelength. So emission from any altitude where the weighting function is non-zero contributes to the measured intensity for the channel. Satellite atmospheric temperature profile measurement is nowhere near as straightforward as using a thermometer.

173. John Lang
Posted May 7, 2008 at 6:42 PM | Permalink

Leif, what are your latest thoughts on the start of Cycle 24 now that we have seen the third 24 spot? Cycle 23 is coming up on its 12th anniversary this month.

174. Posted May 7, 2008 at 6:45 PM | Permalink

171 (Sam):

That’s the key: “levels will move up and down from day to day and from place to place”. Not to mention it varies in level just from position on the Earth even if everything else stays the same.

Explain what is the difference between place and position

172 (DeWitt):

The y axis in the figure should really be expressed as pressure, not altitude.

I know, but apparently the folks at RSS don’t [ :-) ]. This is what they say about the weighting functions:

The weighting functions are given as instantaneous values at the given height. To calculate the total weight over a layer, one needs to calculate the average over that layer and multiply by the layer thickness.
total_wt_between_level_minus_one_and_level =
0.5*(wt_function(level) + wt_function(level-1))*(h(level)-h(level-1))
where h is in kilometers.

For the illustrative purpose I made of the figure [theirs, BTW] I think we are close enough.

175. Posted May 7, 2008 at 6:56 PM | Permalink

173 (JohnL):

what are your latest thoughts on the start of Cycle 24 now that we have seen the third 24 spot? Cycle 23 is coming up on its 12th anniversary this month

Perhaps the following Figure says it all, it shows a count of ‘active regions with spots’ for the past few cycles. The count is really a count of days the region was visible [and no more than 70 degrees from central meridian] and then summed for every region. Yearly smoothed values are also shown as the ‘smoother’ curves. Different cycles are coded with a different color. The detailed figures show the transitions between cycles. Note that cycle 24 has just barely begun:

I think we still a way to go, before we can speak about being in a new cycle. In case you missed it, cycle 24 is the purple ‘fly-dirt’ at the bottom of the figure.

176. John Lang
Posted May 7, 2008 at 7:03 PM | Permalink

Thanks Leif, great charts I have never seen before. Several months to as much as 1 year to go then.

177. Posted May 7, 2008 at 7:17 PM | Permalink

176 (JohnL): Thanks for the kind words about the chart [which is my invention]. I find it so useful that I have urged both the Sunspot Center in Brussels [contact: hochedez@sidc.be ]and the Space Weather Prediction Center in Boulder [contact: SWPC.CustomerSupport@noaa.gov ] to provide as a public service a similar chart, possibly going back further in time. But to no avail. I have seen such a large interest in this chart [apparently people are really interested in this] that I think these agencies should fulfill their public duty and provide the chart. You [and others that may think so too] are urged to contact the agencies and ask that they provide this service. Maybe people-power can work its magic with this.

178. Pat Keating
Posted May 8, 2008 at 10:06 AM | Permalink

177 Leif

I would also like to thank for an excellent chart. Did you ever plot the T anomaly on it?

179. Posted May 8, 2008 at 10:18 AM | Permalink

178

Pat… Leif gave me the chance and I did a comparison here; however, I plotted anomalies of TSI versus anomalies of T. I’m doing it now including the anomalies from 2000 to date.

180. Posted May 8, 2008 at 10:38 AM | Permalink

178 (Pat): No, I don’t know how to tell which part of T anomaly belongs to old cycles and which part to new cycles. If you tell me, I’ll compile the chart.

181. Posted May 8, 2008 at 10:44 AM | Permalink

Folks, yes it is a great chart. Even the fearless leader of our sunspot prediction panel thought so. He used it on slide 10 of his latest report:
http://helios.swpc.noaa.gov/sww/index.html
click on Doug B.

182. pochas
Posted May 8, 2008 at 10:56 AM | Permalink

180 (Leif):

No, I don’t know how to tell which part of T anomaly belongs to old cycles and which part to new cycles.

I think you could figure it out. But if you did, you’d never believe yourself. :-)

183. Sam Urbinto
Posted May 8, 2008 at 10:59 AM | Permalink

Leif #174 “Explain what is the difference between place and position…”

I think I meant a place (lat/long) and time X (over period Z?) give us a position (say 2.8 KM) we define what the LT/MT are (as you said, roughly 0-4 and 4-7 KM). And that two places at the same or different times where the position is the same at 3.8 KM, there are variable temperatures and water vapor contents at that specific position and compared to the other specific position.

Something like position X (LT/MT altitude) at place Y at time Z has temperature A and humidity B. There’s other combinations that can give us the same AB, the same or different YZ can give us the same or different X, another XYZ that are the same might have a different AB or the same A different B, and so on.

So your answer in #167 (to #166 Q: Is it possible to get data for the different levels, especially for that above and below the freezing point of water?) isn’t really no per se, just that you could only model it. Which is the same I suppose. :D

Or maybe I meant position in the level determines the data to be gotten, which also varies at any given place+time=altitude depending on all the factors in play at that time.

184. Posted May 8, 2008 at 11:21 AM | Permalink

182,183 (pochas,Sam): I think I may have to pass on these two. Too complicated for me.

185. Posted May 8, 2008 at 11:46 AM | Permalink

# 182

Pochas,

I think you could figure it out. But if you did, you’d never believe yourself.

I think we cannot figure it for now. However, I think we will have the answer in 300 years, or perhaps more?

186. Posted May 8, 2008 at 11:54 AM | Permalink

Steve, Leif,

I’m sorry… Definitely, this will be out of topic. Please, give me a chance:

Dear Auditors,

May I count on you for a review of the article I posted in # 179? The committee did not approve it for its publication…

187. Posted May 8, 2008 at 12:08 PM | Permalink

186 (Nasif): This is better done using email. So, I suggest you use that channel from now on for this. Start by sending the comments the ‘Committee’ had as cause for rejection.

188. Sam Urbinto
Posted May 8, 2008 at 12:14 PM | Permalink

184 Leif: “Too complicated for me.” Sarcasm coming from you!?!? :)

So, what is the temperature and water vapor content at the boundry of the lower and middle troposphere? I need to measure it at the position of the boundry at a specific place at a specific time. (Ditto for any level (place) inside LT or MT of course)

We have 5 variables here; at any combination of the first three, time and place (time/space coordinates?) defining a position (a certain altitude) of the LT/MT boundry. At that time/space/position, the temperature and relative humidity can be basically anything, although certainly are affected by both time and place, and the resulting position of the boundry at some altitude (besides the sun, clouds, surface, wind, etc)

I’m just making these numbers up. Start with only the boundry position the same, at altitude 4.3 KM:

V1: 60N50E 3PM 4.3KM 30C 70%
V2: 50S60W 5PM 4.3KM 40C 20%

Now compare V1 and V2 but at the same time:

V1: 60N50E 5PM 4.1KM 40C 30%
V2: 50S60W 5PM 4.3KM 40C 20%

The boundry is at a different position, and we’re at the same place, different time. Now the temp is the same and the RH a little closer.

Now both at 10 PM

V1: 60N50E 10PM 3.8KM 20C 50%
V2: 50S60W 10PM 4.1KM 20C 50%

Time and boundry position have changed, everything else is the same.

How do you measure or calculate or model that? And why? :D

189. Posted May 8, 2008 at 12:19 PM | Permalink

# 187

Yes, Leif… I knew that. The problem is that I have only three E-mail addresses of the auditors. I’m sorry.

190. Posted May 8, 2008 at 12:24 PM | Permalink

188 (Sam): No sarcasm at all. Perhaps I should have added [the always underlying assumption] “for what I’m trying to do”. I’m trying to devise two time series, one for the lower troposphere and one for the middle troposphere of temperature derived from MSU Satellite data. As I understand it, the weighting factors depend on height [only, as per the link already given] so none of these changing complicated conditions apply and I don’t need to deal with them [nor do I want to: “too complicated for the aim of the investigation”].

191. Posted May 8, 2008 at 12:35 PM | Permalink

189 (Nasif): you have three email addresses, how many do you need. Who are the auditors? All of us? I don’t think you can get several thousand people to respond.

192. Posted May 8, 2008 at 12:43 PM | Permalink

190 (me): Maybe I should have added the dictionary definition of sarcasm:
A cutting, often ironic remark intended to wound.
A form of wit that is marked by the use of sarcastic language and is intended to make its victim the butt of contempt or ridicule.
The important thing is the intent. It has to be bad in order for the utterance to be sarcasm. But, let’s agree: no more on sarcasm. I think everything is clear on that.

193. See - owe to Rich
Posted May 8, 2008 at 12:49 PM | Permalink

165 (Leif)

There are 14 means, one for each cycle, as follows.

10 -0.411
11 -0.336
12 -0.295
13 -0.371
14 -0.504
15 -0.361
16 -0.235
17 -0.071
18 -0.119
19 -0.112
20 -0.130
21 -0.008
22 0.139
23 0.391

Hope that helps. The yearly mean minus the cycle mean is what I analyzed, finding that overall the distribution for maximum+i years (i=-4,-3,…,+6) was different between even and odd cycles.

Rich.

194. Sam Urbinto
Posted May 8, 2008 at 1:25 PM | Permalink

Leif #192

And I agree with you, no more of it. (I wasn’t really accusing you of the S-word, just a little joke in asking for clarification of what you meant about “complicated”. :))

Leif #190

Ahh, too complicated for the current goal. Sure. Okay I have it now. I wasn’t actually saying you should do any of those (I think it’s meaningless and/or impossible anyway) I was just explaining what I meant by “position”, as you asked. I probably should have said “resulting altitude level” at a given spot at a given time separating lower and middle troposphere. Obviously there’s too many points, so you have to estimate and average areas as a function of time. Something like averaging over zones into monthly anomalies. :)

You mean the files in the weighting_functions folder I take it. Interesting… Looks complicated enough as is. They have 6 weights from nadir and limb and they go by 300 meter increments starting at the ground. Then you get to start with the data. :D

It’s a bit more than that though, for the weighting; the assumptions:
US standard atmosphere
Surface relative humidity of 70%
PV scale height of 1.5 km
.9 surface emissivity
Wind of 8 m/sec for the sea surface model

Then there’s also the nadir and near nadir differences between water and land that get factored in (with the greatest variation at nadir).

It seems to me you could just average the temperature given for levels 0-13 for the LT (275.48 K) and 14-23 for the GT (252.108 K) and be done with it. :D

Just kidding, I know you said time-series from the MSU data separate for LT and MT. Nobody’s done that before? No online tools where you can break some time series or another into the two ranges? I suppose not, or you’d already have them.

195. Posted May 8, 2008 at 3:56 PM | Permalink

http://www.physorg.com/news129483836.html :
“Over the past century, Earth’s average temperature has increased by approximately 0.6 degrees Celsius. Solar heating accounts for about 0.15 C, or 25 percent, of this change, according to computer modeling results published by NASA Goddard Institute for Space Studies researcher David Rind in 2004.”

Although the article describes modern and upcoming measurements [SORCE] it cites the obsolete paper by Rind, based on Lean’s 2000 reconstruction. The change in TSI using her old numbers is five times what it is believed today, so the 0.15C should rightly be 0.03C. It is amazing how old, obsolete conclusions simply will not lie down and die as long as it fits in somebody’s scheme to use them.

196. Posted May 8, 2008 at 4:02 PM | Permalink

194 (Sam):

MSU data separate for LT and MT. Nobody’s done that before? No online tools where you can break some time series or another into the two ranges?

Of course people have done that and the data is readily available. But they are anomalies, not temperatures. I’m trying to get, find, steal, compute, whatever, the daily means from which the anomalies are computed: Tano = Tobs – Tmean. This turn out to be a lot harder. If somebody knows where, let me know.

197. Posted May 8, 2008 at 4:37 PM | Permalink

# 196

Leif, Sam,

And that is the reason I compared anomalies with anomalies, that is, apples with apples. Comparing absolutes with anomalies gives ambiguous results.

198. Sam Urbinto
Posted May 8, 2008 at 5:07 PM | Permalink

Leif 196

Ah, temps. But climate science says the anomaly is the way to go! :)

It’s my understanding they average the month and compare it to the same month for the average of the base period and figure out the offset (or some such). Certainly somebody has that data? I wish I could help you with whom that is, I have no idea.

you have to pick one of the available climatologies {averages of model runs} and add the anomalies (with respect to the proper base period) to it. For the global mean, the most trusted models produce a value of roughly 14 Celsius, i.e. 57.2 F, but it may easily be anywhere between 56 and 58 F and regionally, let alone locally, the situation is even worse.

And from their data, it appears their base period is decided the same way:

Best estimate for absolute global mean for 1951-1980 is 14C = 57.2F, so add that to the temperature change if you want to use an absolute scale

So are all (everyone’s) base periods modeled and the monthly compared to that? Hmmm.

I wish you luck in locating the raw temperature data, whatever period that’s kept in and whomever has it.

199. Posted May 8, 2008 at 5:21 PM | Permalink

197 (Sam):

It’s my understanding they average the month and compare it to the same month for the average of the base period and figure out the offset (or some such)

They quote anomalies for each day, so there must exist [for their schema to make sense at all] a daily mean to subtract.

200. Posted May 8, 2008 at 5:33 PM | Permalink

197 (Sam):

Ah, temps. But climate science says the anomaly is the way to go!
I wish you luck in locating the raw temperature data, whatever period that’s kept in and whomever has it.

Climate Audit is only possible if the raw data [and the ‘meta-data’] is available, otherwise a lot of ‘adjustment’ and ‘massaging’ could [and maybe does] go on ‘behind the scenes’.

201. steven mosher
Posted May 8, 2008 at 5:34 PM | Permalink

re 196. Leif if you are talking about GISS and HADcru, giss post the average of the
base period (1951-80) on their web site. Hadcru also post a file of absolute
temps. Somewhere, I can go find it again. As for UAH and RSS, I dont know. Christy could

I dont know why these guys dont post simple raw data and let every do their own anomaly dance
or filtering.

202. DeWitt Payne
Posted May 8, 2008 at 5:39 PM | Permalink

Leif,

Extracting a temperature profile just from the MSU channel intensities is ill-posed because the weighting functions of the channels overlap and noise. There are an infinite number of solutions. It appears that RSS and UAH constrain the solutions for calculating LT, MT and LS by assuming the constant lapse rate, pressure and density profile of the 1976 standard atmosphere. I have seen comments indicating that if you have an actual radiosonde sounding somewhere in your vicinity (i.e. a best guess), then temperature profiles can be extracted from the satellite measurements that more accurately reflect the true local temperature profiles unconstrained by the fixed quantities of the standard atmosphere. These can then be inserted into numerical weather prediction models to improve their accuracy. If the actual atmospheric lapse rate has changed over time then it is likely the calculated trends in MT especially are incorrect. Enhanced greenhouse theory with water vapor feedback requires, I think, that the lapse rate decreases over time. Now that I think about it, maybe this comment would go better in the Douglass thread. I’ll leave it here for the moment. I certainly won’t object if the boss moves it.

203. Sam Urbinto
Posted May 8, 2008 at 5:48 PM | Permalink

Leif 198: Okay, so they average the daily anomaly from the mean for the day (which is calculated, from tmin and tmax or some other system appropriate for satellites, whatever). So yes, somebody must have some number that for that day that can be subtracted (an absolute mean). For the schema to make sense (at least as far as we know from what we know, and assuming it actually does make sense). Then another question might be if the daily mean is compared to the number for the base period in total, or something else.

One would imagine somebody at RSS or UAH would know where the data is. If you’ve contacted them all and can’t get it, then who else in the world would have it? The people running the MSU and AMSU sounders? :rolleyes:

Leif 199: I totally agree. It’s taken me quite a while to figure out what GISTEMP actually was, and the adjustment methods still make no sense. We don’t know if they won’t tell us. :D

mosh 200: I don’t know why they don’t post raw, why would there be a reason not to? Might we hurt ourselves? But as I said, the GISTEMP data file just says for the global anomaly “Best estimate for absolute global mean for 1951-1980 is 14C” DId they just model it and start using 14C or is there a real number someplace. Very perplexing and confounding.

204. Andrew
Posted May 8, 2008 at 5:52 PM | Permalink

201 (DeWitt Payne): Isn’t UAH validated against radiosondes? Maybe you should ask Christy.

205. DeWitt Payne
Posted May 8, 2008 at 6:08 PM | Permalink

The answer to your question is yes, AFAIK. But there’s still a lot of potential slop in the trend. The problem also is that nobody (well, a lot of people. see this thread for example) quite believes radiosonde trends either and to make it worse, tropical soundings are few and far between.

206. steven mosher
Posted May 8, 2008 at 6:08 PM | Permalink

RE 202 SAM,

for a long time GISS temp did not post the mean ( 14C) of the base period ( 1951-1980)
Just recently they have. So If you look at the GISSTEMP code, they read in data convert F to
C where required. Then when they finish the whole mess they calculate a mean for 1951-1980
and subtract that mean( 14C) from every value. So its not magic. Hadcru, as far as I know doesnt post the average for 1961-1980 ( it would be about 14.092) but they do post somewhere
hidden away the gridded absolute values in a big nasty file.

This adversion to position raw absolute numbers is an annoyance. I dont get it.

207. Sam Urbinto
Posted May 8, 2008 at 6:19 PM | Permalink

#206 mosh Yeah. But the the absolute temperature page says the “most trusted models” have it at 14C. So who knows if the number is any real quantity for 1951-1980 or not. Does the trend line equal 0 for those years? I think it’s off some. I dunno.

On the other hand, I want to get to a base period of 1981-2010 Cuz all the jump in the monthly anomaly starting in the late 70s is going to be interesting to see.

Here’s how it should work: “You want my data? Here’s the FTP site, in this folder. Here’s the download website. Oh, and you can get the offsets, calculations, algorithms and the like there too. Or I’ll email it to you. Have fun with it!”

Hmmmm.

208. Sam Urbinto
Posted May 8, 2008 at 6:21 PM | Permalink

Andrew #204

Maybe Dr. Christy has the raw data or knows who does?

209. Posted May 8, 2008 at 7:53 PM | Permalink

202 (DeWitt): I recognize all of this, but it does not help me one single bit, so I can safely ignore it. Fact is that anomalies are available for something the data producers call TLT and TMT. I am just a user of that data. I know from my own field that the steps from absolutely raw [e.g. telemetry data] to calibrated ‘science’ data is long, tortuous, varied, almost infinitely complex, etc. But I choose to trust the providers assuming they do what they know best to do, just as I expect users of my data to trust that I do the best job possible. Therefore I don’t want to know or worry about the low-level details [at this stage – there may be exceptions to that if major problems crop up]. To be frank, it is a bit annoying to be lectured about all the intricacies, without getting any actual help in reducing them.

Other posts: you talk about subtracting the mean. Even give a number to several decimal places, but I need 365 numbers, one for each day. That is all. There should be no need for me to spend time in understanding and fighting the gridded, zipped, binary, weighted, etc data. Anomalies are given for every day since the 1970s, relative to a set of 365 numbers. I just want those 365 numbers. If I compute them myself from the ‘raw’, gridded, zipped, binary, etc data, how do I know if I did it correctly, or if they did it correctly?

210. DeWitt Payne
Posted May 8, 2008 at 8:45 PM | Permalink

Leif,

This probably won’t help either, but I think it’s a lot more than 365 numbers. Best case would be 365 numbers for each grid point, and it could be a lot more depending on when the raw temperature is converted to an anomaly or whether the gridded data is an average of a yet finer mesh. IIRC, You only have a few readings each day for any given grid point which could easily add another layer of averaging. Maybe if you ask nicely (co-authorship?) someone at RSS or someone at UAH could produce a program to give you what you want. It sounds like a whole lot of work, though.

211. Posted May 8, 2008 at 9:26 PM | Permalink

210 (DeWitt): Averaging is a linear process, so for each grid point and day we have an anomaly: ∆T = T – <T> where the <> denotes an average. Then the global average over all grid points for the day would be <∆T> = <T – <T>> and because of linearity: <∆T> = <T> – <<T>> hence there would exist the global number <<T>> for the day. No? By the same argument, there would exist a global <<<T>>> over the reference period for the day, no? This is the number I want.

212. Jon
Posted May 8, 2008 at 10:07 PM | Permalink

Sorry to jump in unawares, but while the L-1 norm is linear, temperature should be averaged using the L-4 norm–particularly if your goal is to measure correlation against solar output.

213. DeWitt Payne
Posted May 8, 2008 at 10:12 PM | Permalink

Leif,

That all seems logical. However, what seems logical may not be true if we don’t include an important factor. Ask Judith Curry (and a lot of other people as well) about how radiative forcing by CO2 is enhanced at the poles because there is more CO2 and less water vapor. That’s very logical too, but isn’t true, according to Gavin Schmidt and GISS, because that effect is overwhelmed by the effect of enhanced CO2 emission at 15 micrometers in the high latitudes caused by a lower, warmer tropopause compared to the tropics, not to mention variability with longitude at the same latitude.

If you’re averaging over the globe, I’m not sure why there would be much, if any, variation from day to day in the mean. So use the 1976 standard atmosphere to calculate a temperature at the appropriate altitude, 2 km and 5 km, and add that to the global daily anomaly. Seems to me you’re throwing away a lot of information, though, considering the very wide range of temperature with geography. This also get back to the issue of whether a linear average is the appropriate measure.

214. DeWitt Payne
Posted May 8, 2008 at 10:23 PM | Permalink

Jon,

If you’re looking at emission from the atmosphere, and you really have to, then even L-4 isn’t correct because the atmosphere isn’t gray, isotropic or isothermal. And have I mentioned line broadening? At that point, it starts to get really ugly and you quickly run out of computer power, are forced to use lots of questionable parameterizations, etc., etc.

215. Posted May 8, 2008 at 11:25 PM | Permalink

Leif,

216. Posted May 9, 2008 at 4:29 AM | Permalink

213 (DeWitt):

If you’re averaging over the globe, I’m not sure why there would be much, if any, variation from day to day in the mean.

One reason is that in January TSI is 90 W/m2 larger than in July.

217. cba
Posted May 9, 2008 at 5:46 AM | Permalink

(Leif):

The above is one of the links I stumbled across looking for something else. It has a 2 meter altitude T measure grid. I think they pull the data from NOAA or NASA archive sites and format it into a grid format. It’s available as a map or as a text file array. It’s T values in deg K by grid square. Using this avoids having to try to decode the CDF (???) integer format with offsets etc. There are other similar places.
:)

218. cba
Posted May 9, 2008 at 6:00 AM | Permalink

216 (Leif):

Ah ha! I’m not the only one getting caught with TSI versus TOA insolation faux pas :)

DeWitt – it’s the orbital eccentricity variation aphelion vs perihelion and the difference is staggering even when averaged a bit. An average over 1-3 months in jan is about 9 W/m^2 more than the same centered in july after doing the divide by 4 to go from the lambertian disk capture area to the spherical Earth surface insolation distribution.

There are annual variations in this and in the albedo of several % which is of similar magnitude. There’s also a variation in OLR on an annual basis of several W/m^2 – and I don’t think it tracks the incoming (but then I seem to want to get that factor backwards for some reason).

Basically, the whole enchilada changes on the short and long term. Incoming solar, albedo, OLR vary during the year and over multiple years even with monthly averages being used and reduced down from the grid chart data to a global average. Note that one grid square (mine) has tremendous variations relative to the global while a latitude band has a moderate amount. There is something to be learned from these variations.

219. Posted May 9, 2008 at 3:59 PM | Permalink

GEOPHYSICAL RESEARCH LETTERS, VOL. 35, L09808, doi:10.1029/2007GL032911, 2008
Impact of terrestrial weather on the upper atmosphere
T. J. Fuller-Rowell et al.
CIRES/University of Colorado and NOAA Space Weather Prediction Center, Boulder, Colorado, USA

Abstract
A whole atmosphere model has been developed to demonstrate the impact of terrestrial weather on the upper atmosphere. The dynamical core is based on the NWS Global Forecast System model, which has been extended to cover altitudes from the ground to 600 km. The model includes the physical processes responsible for the stochastic nature of the lower atmosphere, which is a source of variability for the upper atmosphere. The upper levels include diffusive separation, wind induced transport of major species, and uses specific enthalpy as the dependent variable, to accommodate composition dependent gas constants and specific heats. A one-year model simulation reveals planetary waves explicitly up to 100 km altitude. At higher altitude, multi-day periodicities in the dynamics appear as a modulation of tidal amplitudes, particularly the migrating semi-diurnal tide in the lower thermosphere dynamo region. The penetration of planetary wave periodicities from tropospheric weather into the upper atmosphere can explain terrestrial weather sources of variability in the thermospheric and ionospheric.

In discussing solar influence on the weather it often assumed that the influence is first seen in the upper atmosphere, but here is [has been said before – this is not exactly new, just ‘better’] evidence that some of the changes seen in the upper atmosphere are actually caused by weather at the surface propagating upwards.

220. Sam Urbinto
Posted May 9, 2008 at 4:48 PM | Permalink

Leif 219: But isn’t the entire system one where it’s redundant and circular, where it’s almost impossible to separate out the discrete components? Weather upwards seems rather obvious. Like my post here on what warms when. Interesting.

Daily data: I don’t know if this will be helpful, but you can get the text data for the average of 3.3 KM of actual temperature for some time periods. Maybe contacting the people on the page would assist if you haven’t yet already. magellan posted it on the BB (bulletin board, not black body! :)) after I mentioned it.

http://discover.itsc.uah.edu/amsutemps/

221. Posted May 9, 2008 at 5:13 PM | Permalink

220 (Sam): thanks for the link. Now, if one could get the values in digital form rather than the graphs….

222. Sam Urbinto
Posted May 9, 2008 at 5:56 PM | Permalink

Leif 221: You can at least do the “show data as text” But I found this as the back end:

http://discover.itsc.uah.edu/amsutemps/data/

Looks like just the daily for each channel though. :(

223. Posted May 9, 2008 at 6:39 PM | Permalink

222 (Sam): thanks, why the frown? This is great, no?

224. cba
Posted May 10, 2008 at 4:54 AM | Permalink

(Leif):

I’ll try again with the link. It is a site that pulls data from archived data sets and offers both maps and text file formats for all sorts of data.

Hopefully, this link will be intact

225. cba
Posted May 10, 2008 at 4:57 AM | Permalink

224 (me):

I seem to be having trouble getting a link inserted here.

226. Erl Happ
Posted May 10, 2008 at 6:50 AM | Permalink

219 (Lief)

In discussing solar influence on the weather it often assumed that the influence is first seen in the upper atmosphere, but here is [has been said before – this is not exactly new, just ‘better’] evidence that some of the changes seen in the upper atmosphere are actually caused by weather at the surface propagating upwards.

Its very much a two way process:

Data below implies that the moisture vapour content of the atmosphere above the tropopause has a lot to do with its temperature:
This data is recoverable at: http://discover.itsc.uah.edu/amsutemps/AAT_Browse.php?chan=01&satnum=15&aord=d

Notice that the whorls of wet tropical air (marked with white ovals on surface map) that are entrapped in the waves of very cold air of Antarctic and Arctic origin are elevated to about 9km at about latitude 60°N and also at 60°S where they are above the local tropopause (marked with the same white ovals and copied to these maps as a group to maintain the same spatial relationship as at the surface).

At the 250 hPa pressure level (about 9k) these localised air masses, the result of strong uplift in low pressure cells) exhibit a higher temperature than the surrounding air. This warming would seem to be associated with stratospheric/thermospheric processes perhaps involving short wave radiation because the transfer of warmth upwards through the troposphere can not be seen at the 400 hPa (4K) level directly beneath. At the 150hPa (12k) level (not shown) the heated area is more widespread becoming apparent also over the Arctic where it may be a direct reflection of auroral processes.

I should imagine that thermospheric processes involving direct heating by short wave solar radiation are taking place closer to the surface of the Earth at solar minimum than they would be at solar maximum.

This illustrates the strong variability in atmospheric process at different levels of the atmosphere depending upon latitude. It is essential to define the latitude when one is talking about atmospheric processes.

Polewards of 60° latitude temperature change is a mirror image of what is happening at the equator. Heating at the equator can produce cooling, particularly in the Antarctic.

The value of investigating what happens in the 10°N to 10°S latitude band is the low rate of seasonal variability. One can observe the effects of orbital and irradiance variability with little interference from seasonal influences. One also avoids those confounding effects that derive from the difference in the area of land and sea between the hemispheres.

At every step of the way it is a two way interaction with atmosphere as mediator of solar processes. However, in the final analysis it is a self evident truth that the energy to drive the whole thing comes from without.

href=”http://s249.photobucket.com/albums/gg220/erlandlong/?action=view&current=Heatingat9km.jpg” target=”_blank”>

227. Chris Knight
Posted May 10, 2008 at 4:24 PM | Permalink

The third image in #226 (Erl) shows the AMSU channel 1 surface temperature global distribution. It is interesting that in their published data, #222 (Sam), they do not produce a daily channel 1 surface temperature text file of the data. It is obviously possible to process this data as they do for the other channels.

Perhaps channel 1 data bears little relationship to NASA GISS, HadCrut3, or any other established global temperature record, and would draw doubt on other AMSU data, if compared to the force of authority these long-established records command.

228. Posted May 10, 2008 at 6:29 PM | Permalink

227 (Chris): About surface temperature: what would the satellite measure in channel 1 over the Himalayas and the Andes? Do you know how surface elevation is handled?

229. Posted May 10, 2008 at 6:41 PM | Permalink

# 228

Leif,227

Chris): About surface temperature: what would the satellite measure in channel 1 over the Himalayas and the Andes? Do you know how surface elevation is handled?

No matter the decrease of temperature is mostly adiabatic.

230. Sam Urbinto
Posted May 10, 2008 at 7:02 PM | Permalink

223 Leif

222 (Sam): thanks, why the frown? This is great, no?

Oh, yes, it is.

Just that there’s only (seemingly) the day’s daily’s per altitude and only covers 3300 meters (as far as I can see) But at least, it seems, the people with the data are identified. So at the least, it’s a start!

231. Posted May 11, 2008 at 12:13 AM | Permalink

Leif,

Leif, I found this. It seems there is a correlation, after all… :)

232. Chris Knight
Posted May 11, 2008 at 10:29 AM | Permalink

228 (Leif):

I get your point :) Surface elevation, in terms of everything over 1 km high has “already” been accounted for by channel 2,3,4 etc. I guess the same goes for Antarctica with about 80% between 1,000 and 4,000metres asl, and Greenland.

The elevation seems to be handled in the following way:

Apparently,the channel 1 data is not yet supplied as a product, therefore weighting functions are not supplied.

Also “2. The surface contribution and water vapor crosstalk are so large
that the “standard” weighting function concept becomes fatally
flawed.”

And “For channel 2, the surface makes a significant contribution, and
must be treated differently for land and ocean scenes.”

So it all seems to be done with weighting functions, available from the above ftp site.

233. Chris Knight
Posted May 11, 2008 at 10:39 AM | Permalink

I should include the image:

Figure 3. Color coded map of decadal trends in MSU channel TLT (1979 – 2007). Data poleward of 82.5° North and 70° South, as well as areas with land or ice elevations above 3000 meters, are not available and are shown in white.

234. Posted May 11, 2008 at 10:50 AM | Permalink

233 (Chris):

So it all seems to be done with weighting functions.

Thanks, Chris. Now, of course, a lot of ‘auditing’ could be done with the weighting functions :-)
But, that’s OK. This is where must trust the raw data producers.
I’m sure it has been noticed before, but it just looks to me that the NH is has been warming since 1979 while the SH has been cooling. Interesting in view of the ‘Global Warming’ debate.

235. Posted May 11, 2008 at 11:25 AM | Permalink

# 232

Chris,

I’d like to get the raw data for graphing the average deviations of the NH plus SH deviations. Does the NSSSCT-UAH database works well for my purpose?

236. Posted May 11, 2008 at 11:46 AM | Permalink

231 (Nasif):

It seems there is a correlation, after all…

Beauty must be in the eye of the beholder. I don’t see any correlation at all.

237. Posted May 11, 2008 at 2:49 PM | Permalink

# 236

Leif… Ok! OTOH, no “global” warming from NSSTC data:

238. Posted May 11, 2008 at 3:13 PM | Permalink

237 (Nasif): only 0.4C last 100 years.

239. Posted May 11, 2008 at 3:27 PM | Permalink

# 239

Leif,

Yes, it’s correct. And only 0.065 K in the last 30 years. It’s not global.

240. Posted May 11, 2008 at 3:38 PM | Permalink

# 239

And it’s a normal deviation for the Holocene Epoch because the expected deviations for the present period are from -1.5K to 1.5K.

241. Posted May 11, 2008 at 3:46 PM | Permalink

# 238

Leif,

Two links to records during the Holocene:

Steve:
Bond’s data has recently been added to NCDC archive.

242. Chris Knight
Posted May 11, 2008 at 3:56 PM | Permalink

235 (Nasif)

Links to the binary data (and lots of other stuff) from RSS are found here:

http://www.remss.com/msu/msu_data_description.html

243. Posted May 11, 2008 at 4:16 PM | Permalink

# 242

Wow!!! Thank you very much for the link, Chris! :)

244. Posted May 11, 2008 at 5:44 PM | Permalink

Leif,

I’ve combined Möeberg data of the deviations of temperature and NSSTC data and I’ve obtained 0.06 K for the NH deviations in the last 134 years, and 0.13 K for the period 1908 to April 2008. No doubt… Beauty must be in the eye of the beholder. ;)

245. Posted May 11, 2008 at 5:50 PM | Permalink

OT… Happy mother’s day for all climate auditors’ moms! :)

246. Posted May 12, 2008 at 10:39 AM | Permalink

Leif,

Continuing playing with the TSI database after calibration:

247. Dennis Wingo
Posted May 12, 2008 at 10:39 AM | Permalink

Leif

Have you seen this book?

248. Posted May 12, 2008 at 10:45 AM | Permalink

# 246

Leaf,

In the plot of averaged TSI anomalies the hexic trend is not too different from the hexic trend calculated from other TSI reconstructions.

Steve McIntyre,

249. Dennis Wingo
Posted May 12, 2008 at 11:52 AM | Permalink

Leif

Here is an interesting paper from a study of Omani stalagmites that indicate not only Milankovitch variations but shorter term solar variability as an influence on Monsoons.

http://www.sciencemag.org/cgi/content/abstract/300/5626/1737?ck=nck

250. Posted May 12, 2008 at 11:59 AM | Permalink

247 (Dennis): Yes I know about the book. There is a lot of good stuff in it, and some that is just plain wrong. [wrong being defined as disagreeing with me :-) ]

251. Posted May 12, 2008 at 12:06 PM | Permalink

# 249

Denniss, Leif,

Perhaps, we should consider the proxies seriously on reconstructions of TSIs?

252. Posted May 12, 2008 at 12:10 PM | Permalink

246 (Nasif): You can use the width/height tags. The following:

was generated using:
[img src=http://www.biocab.org/Average_11_years_Hexic.jpg width=600 height=400]
where you just have to replace the “[” by “<” and “]” by “>”

253. Posted May 12, 2008 at 12:39 PM | Permalink

251 (Nasif,Dennis):

Perhaps, we should consider the proxies seriously on reconstructions of TSIs?

This topic has three parts:

1) what is the contribution, TSIA, to TSI by solar activity [as given by the combined sunspot darkening and facular brightening]. This is not in doubt, everybody agrees with that and formulae can be found at many places for that.
2) what is the contribution, TSIN, to TSI of the solar ‘network’ [that is the remaining magnetic field when all sunspots have died away]. There is some agreement that this number is 1.7 W/m2 [based on Simon Foster’s PhD theses; for the recent time since 1978].
3) To reconstruct TSI one proceeds as follows
TSI = TSI0 + TSIA*f(SSN) + TSIN*h(SSN?)

where TSI0 is the solar constant [if the sun didn’t have a magnetic field] and ‘f’ is a function of the sunspot number (SSN) and ‘g’ is either constant [=1.7 W/m2] or varying in an unknown way as a function of solar activity.

If we assume [as I do] that g = 1.7 [constant], we still have to know the SSN in order to apply function ‘f’. So it is important that we know what the sunspot number was in the past, and we do not know, and that is the real problem.

I’m preparing a paper on the correction of the SSN [have already presented at AGU and elsewhere on that]. It will take a decade before the impact of this will be felt [there are still people quoting Hoyt&Schatten’s TSI, and Lean’s 2000] and there will be a lot of resistance, just as there was when I claimed the aa-index was wrong [well, I eventually won that battle]. The sunspot number is one of the most used and misused indices in this field, and any correction [whichever way] will step on somebody’s toes.

254. Posted May 12, 2008 at 12:46 PM | Permalink

253 (me) TSI = TSI0 + TSIA*f(SSN) + TSIN*h(SSN?) should have been
TSI = TSI0 + TSIA*f(SSN) + TSIN*g(SSN?)

255. Posted May 12, 2008 at 1:02 PM | Permalink

253 (me): to elaborate a bit more on this.
One could replace SSN by SA, where SA [Solar Activity] is another activity measure. Many exists:
1) Wolf, Zurich, International sunspot number: Rw, Rz, Ri [Rw and Rz have calibration problems]
2) Group sunspot number: Rg [by Hoyt and Schatten; is up to a factor of 2 lower than Rw and Rz at times]
3) American sunspot number: Ra [Boulder]
4) Sunspot Area: SSA; several exists, not the same calibration
5) f10.7 radio flux: S, since 1947
6) MgII: magnesium II index [back to 1976] has calibration problems
7) Latitudinal spread of active regions: s [back to 1876; Simon Foster]
8) Open Flux: Fs; Lockwood et al., IMF strength by satellites; controversial
9) Cosmic ray intensity from ice cores and tree rings; calibration issues
and more…

256. Posted May 12, 2008 at 2:03 PM | Permalink

# 255

Leif,

Thanks for the illustration.

Incise 8) refers as “controversial” to Lockwood et al’s paper or to the IMF strength?

257. Posted May 12, 2008 at 3:12 PM | Permalink

8: Open Flux: Fs; Lockwood et al., IMF strength by satellites; controversial
I didn’t want the silly smiley. The IMF strength after 1963 is not controversial, before 1963 it is.

258. Posted May 12, 2008 at 4:07 PM | Permalink

# 257

Leif,

I didn’t want the silly smiley.

Me either… Heh! :)

The IMF strength after 1963 is not controversial, before 1963 it is.

Doubt cleared. Leif, It seems Lockwood coincides with you on reconstruction of the period 1980-2007… Am I right?

259. Posted May 12, 2008 at 4:20 PM | Permalink

A “jaw-dropping” prominence…

260. Posted May 12, 2008 at 4:23 PM | Permalink

Please, examine closely the image from my # 259… No sunspots there.

261. Posted May 12, 2008 at 4:39 PM | Permalink

258-260 (Nasif): No problem with 1980-2007 as the data is not a reconstruction but actual measured values.
The prominence does not need a sunspot. In fact, most of them are not connected to any spots.

262. Posted May 12, 2008 at 4:57 PM | Permalink

# 261

Leif,

258-260 (Nasif): No problem with 1980-2007 as the data is not a reconstruction but actual measured values.

And that’s the reason all the reconstructions coincide in the last 30 years, right? The problems are in recons. of periods before 1963. Am I interpreting you correctly?

The prominence does not need a sunspot. In fact, most of them are not connected to any spots.

Yes, I know that… However, it’s surplus of energy, isn’t it? And the energy is radiated in all trajectories, correct? Why not to consider prominences and flares for calculating/calibrating the total solar irradiance?

263. Posted May 12, 2008 at 5:23 PM | Permalink

262 (Nasif): Before 1963, yes, that’s is the problem. Because the first reliable and calibrated spacecraft data is from 1963.
The energy in prominences etc is so minute that it doesn’t matter. The same with flares, with the possible exception of a few super-flares of which we have one every twenty years so or.

264. Posted May 12, 2008 at 6:01 PM | Permalink

# 263

Leif,

Thanks for your patience. Each time, I’m approaching more to your hypothesis, that is, I understand it better. Not easy to swim against the stream:

What’s the approximated energy load of a prominence like that (post # 259)?

265. Posted May 12, 2008 at 7:11 PM | Permalink

264 (Nasif): A prominence is basically a very cool [100 times cooler than the surrounding solar atmosphere] parcel of matter hanging in a magnetic ‘hammock’ and can last weeks or months. If the magnetic field in which it hangs is disturbed the prominence can erupt and as such has a lot of kinetic energy derived from the ‘slingshot’ effect of the magnetic field. Such an erupting prominence is often called a coronal mass ejection and a large one has about 10^26 Joule or about a quarter of the luminosity of the Sun for one second. At solar maximum there are about 5 of those per day, so the total kinetic [not radiant] energy is about 1/50,000 of the solar luminosity but does not show up as an increase in TSI. So, the energetic impact on the Earth is minuscule.

266. Posted May 12, 2008 at 7:34 PM | Permalink

# 265

Leif,

Got it! No more questions (well, I think). Thank you, Leif. In one of his/her posts, cba asked what I was doing here (# 80)… I hope cba got the answer.

267. Erl Happ
Posted May 13, 2008 at 6:37 AM | Permalink

Leif, cba, anyone.
Have you any idea what albedo is due to light cirrus cloud in the middle troposphere?

To what extent could a fluctuation in irradiance (of the extent commonly experienced over a year or so) affect the local temperature of the atmosphere in the 250kPa pressure zone at about 9 km in elevation where the absorbtive material comprises water vapour and ice crystals in the form of cirrus cloud. The temperature gain observed is from 229k to 234k.

Do you agree that the infrared component of irradiance could produce a temperature change of this magnitude from observed changes in irradiance intensity at this level?

Perhaps a learned person could work out the mathematics involved and compute the change in relative humidity assuming a saturated atmosphere at 229k warming by 5°k.

268. Posted May 13, 2008 at 6:40 AM | Permalink

All my questions answered, keep up the good work
cheers

269. Posted May 13, 2008 at 6:44 AM | Permalink

267 (Erl):

cirrus cloud in the middle troposphere

Erl, chek your terminology. Cirrus clouds are usually above the ‘middle troposphere’ (4-7) km.

270. Erl Happ
Posted May 13, 2008 at 8:53 AM | Permalink

269 (Leif)
Is there a physical basis for differentiating lower, middle and upper troposphere at any latitude?

In my admittedly creative usage the ‘middle troposphere’ would be hard to find outside the tropics. In the tropics there are three zones according to temperature with lower above zero, middle up to about 9-10km and ending where the temperature that characterises the tropospause in mid latitudes is reached, that is -50°C. Above that point temperatures are cooler than exist at mid latitudes.

UAH refers to middle troposphere with a mid point between 3 and 7 km. Its hard to know where the limits might be considered to be, especially at the pole in the winter hemisphere where the tropopause reaches the surface.

There is no real agreement where the tropopause is. It is dfferent if one is looking at ozone rather than temperature inversion.

But, I am happy to be educated.

For the purposes of my question let’s just take the height reference of 9km or the pressure reference at 250mb and locate that in the tropics. The temperature is as specified.

What I really want to know is whether infrared is capable of that degree of warming with the calculated water content given the sort of fluxes in radiation experienced and the energy available within that irradiance.

The change in temperature that varies with irradiance and generalised tropical warming is centred about 9km in the zone 10°N to 10°S and does not manifest well below 7km or above 12 km except in circumstances like 1997-1998 when temperatures blew out at all altitudes.

271. cba
Posted May 13, 2008 at 9:15 AM | Permalink

267 (Erl):

cirrus clouds well up in the atm are complex and controversial. There are arguments that they provide net increases in IR trapping over their albedo reflection. Do a google search on cirrus albedo and you will find tons of info on the subject including contradicting info.

At 230k (about -40C) concentrations are in a few ppmv rather than in a %, such as typical surface T values provide. Since sublimation exists and there is virtually no concentration possible at that high altitude T, I’d think that it is totally saturated (or 100% humidity) up there with just a few ppmv of h2o present – equating to ultra dry air, even warmed to just 0 deg C.

272. Posted May 13, 2008 at 9:58 AM | Permalink

270 (Erl): The amount of water up there is so small that not much heat can result from absorption [and cirrus clouds reflect rather than absorb]. As said many, many, many times, the troposphere is heated from below. It looks that you are on a ‘fishing expedition': if it is not TSI, then perhaps UV; if it is not UV, maybe it’s IR; if not IR, then maybe something else will be next [cosmic rays?]; if not, then [as Jim Arndt in a recent Chez Tammy post claimed: CME’s are known to have effects on the troposphere that can last up to 10 years, this is known] it may be Coronal Mass Ejections. The list goes on.

273. kim
Posted May 13, 2008 at 10:11 AM | Permalink

272 (Leif) Well, it’s clouds, Leif, but just how isn’t quite fer sher; I’m going with electrical, this week. Ions seem to be a common denominator.
==================================================

274. Jim Arndt
Posted May 13, 2008 at 10:46 AM | Permalink

272, Leif

I corrected myself on Tammy but hasn’t posted yet. Meant to say CME effects on the troposphere are known. That was just a drive by blogging..LOL. 10 year thing is the magnetic field changes, could be wrong. Big CME’s = warming? Look at the 1989 event start to really warm after that.

275. Erl Happ
Posted May 13, 2008 at 10:59 AM | Permalink

272 (Leif)
Definitely fishing. It can’t be warming from below if there is no evidence of warming immediately below the point of warming without the suddenly available presence of an absorber of either long wave emissions (ozone) or some part of the inwards irradiance (water or ice). Which would you prefer at a height of 9km?

One thing is for sure, It ain’t no genie! Got any other possibilities?

276. Jim Arndt
Posted May 13, 2008 at 11:04 AM | Permalink

272, Leif it has to be something and we are just starting to understand the sun. We are at solar minimum (sort of), PDO shift, AMO shifting soon and we are cooling. Just can’t be CO2 since its effect should be almost linear according to the models…. Of course they will “adjust the models” and presto models fit again. I remember hearing that some have said that we will only see El Nino’s and maybe some very weak La Nina’s, but now they say we are cooling and the model say that. Warmers can’t have it both ways. So if not CO2 then what? That is what we are trying to find out, so we pop out some odd stuff but it gets it out in the open. Lucky for us that we have someone like you to bounce some of off of. I do respect your opinion highly and if you say it ain’t so then I will re-examine.

277. Sam Urbinto
Posted May 13, 2008 at 2:09 PM | Permalink

Why does it have to be anything other that what it is on its own?
:)

278. Erl Happ
Posted May 13, 2008 at 7:47 PM | Permalink

272 (Leif)

As said many, many, many times, the troposphere is heated from below.

Agree entirely. And the stratosphere is also heated strongly from below, especially in July to November when the Earth / Northern Hemisphere is pumping out long wave radiation in excess of the energy input.

But in the Southern Hemisphere summer when the Maritime continent and the Indian Ocean heats strongly and there is a swing towards less OLR and more heat removal by convection then the tropical atmospheric column is lifted strongly and even the stratosphere cools by decompression.

But to your assertion that the solar irradiation does not heat the atmosphere below the tropospause I say ‘nuts’, and urge you to look harder by checking the spatial and seasonal flux of temperatures at all elevations, not simply as an average at a particular level but with the complexities that geography introduces. And you can do that at http://www.cdc.noaa.gov/map/time_plot/

Some cautions. It will only handle 5 years at a time. Check ‘operational’, ‘time longitude’, ‘mean’ rather than anomaly and specify 10 and -10 for the latitude band.

If you can not find a solar signal at 300 mb, 250 mb and 200 mb get some assistance from someone who can read the Hovmoller diagram, or a street map.

Any heat that accrues in the atmosphere has only a temporary residence permit. In the main it is for daytime hours only. It is not the maximum temperature that has produced the warming but the minimum. This is the real silliness of the CO2 argument.

The problem with averages is that one loses the details that the careful observer needs to see cause and effect working. If we continue to treat the globe as a homogeneous unit that can be described in terms of an average temperature we will get nowhere. If we continue to average the flux in irradiance and look at the trend over 11 years we will miss the time interval where the action is occurring.

279. Posted May 13, 2008 at 8:00 PM | Permalink

274 (JimA):

Meant to say CME effects on the troposphere are known.

Yes they are: there ain’t any.

280. Posted May 13, 2008 at 9:31 PM | Permalink

# 278

Erl,

The problem with averages is that one loses the details that the careful observer needs to see cause and effect working. If we continue to treat the globe as a homogeneous unit that can be described in terms of an average temperature we will get nowhere.

I agree. However it is not easy to check the local responses second by second, It would be harder globally.

If we continue to average the flux in irradiance and look at the trend over 11 years we will miss the time interval where the action is occurring.

Perhaps if we consider a lag of three years, or 100 years… Remember that the oceans do not show the effects immediately, but many many years after the event took place. Perhaps the current phenomena in climate is the consequence of something that happened in the oceans (or in Sun) hundreds of years ago.

281. Erl Happ
Posted May 13, 2008 at 10:49 PM | Permalink

280 (Nasif)
We can talk more when you have done the exercise. Not second by second but season by season for five years across 16 elevations and the Hovmoller gives you the spatial variations. And five years at a grab. Two grabs gives you the solar cycle.

Re the oceans: It takes just six months for the warm surface waters of the Gulf of Mexico to reach Europe. The turnaround in the Pacific is a little longer. One of the first things I did when I started to get interested in this subject was to download temperature data for an ocean buoy situated near Alaska. The cooling cycle in the North Pacific began a couple of years ago.

282. Posted May 13, 2008 at 11:59 PM | Permalink

# 281

Erl,

We can talk more when you have done the exercise.

I wouldn’t push an aspirine on the floor with my nose… Heh! ;) Seriously, I’ll try it season by season, assuming that each season comprehends four months. But wait, why not a whole SC?

cooling cycle in the North Pacific began a couple of years ago.

Sorry… I do not want to be insidious, could you explain how did you know that the North Pacific modified its temperature two years ago? I thought it had changed eight or nine months ago… What you emphasize is extremely important because the effect of the North Pacific on global temperatures could be prolonged per decades ahead.

283. Jim Arndt
Posted May 14, 2008 at 9:19 AM | Permalink

279 Leif,

CME’s do effect the troposphere. This is an interesting piece.
http://sunspot.lund.irf.se/SMMF_article1.pdf

284. Erl Happ
Posted May 14, 2008 at 9:49 AM | Permalink

282 (Nasif)
Yes, the solar cycle is the way to go. When you have done one you will want to do another because you will see precipitable water falling away on the upswing and rising on the downswing, the zones of heat, radiation, precipitation and surface pressure swinging to the right and then to the left as the solar cycle produces the same rhythmic behaviour one sees in an El Nino/ La Nina cycle that may take only two or three years. It is changes in irradiance that drive the large and the small cycles and you will also be able to work that out for yourself when you see the temperature anomalies at 250mb. Looks like those little cirrus clouds are really important.

Re the North Pacific: We had a La Nina in the first half of 1996. The tropical waters that failed to warm in that year are now in the North Pacific. The North Pacific does nothing by itself for global temperatures. To suggest otherwise is just a bit of North American ethnocentrism. At the 49th parallel in the Pacific you have waters at about 8°C at the moment. It is in the tropics that the waters warmer than 15°C reside. There is no hope of the waters of the North Pacific warming up of their own accord. If you look at a map of sea surface temperature anomalies today you will see the start of a new Pacific ocean cooling. That is normal for this time of the year except that the anomaly looks deep. Here is an Australian version http://www.eldersweather.com.au/climimage.jsp?i=sstag
Australia is in the middle and North America is on the extraneous right. Hope you can handle that.

And below is something to show you what a Hovmoller diagram can do to facilitiate analysis. The blue rectangles and ovals show you the La Nina events. The current one looks pretty important.

This one is just for you Nasif. None of the rest of you guys should look at it.

285. Jim Arndt
Posted May 14, 2008 at 9:53 AM | Permalink

279 Leif,

286. Erl Happ
Posted May 14, 2008 at 11:11 AM | Permalink

272 (Leif)

The amount of water up there is so small that not much heat can result from absorption [and cirrus clouds reflect rather than absorb

And if that is the case a small change in irradiance can make a very big difference.

The change in temperature is just a few degrees but the expansion in the area affected is the noticeable thing. There are parts of the Earth that cool largely by convection and the release of latent heat of condensation (heating occurs low down and no reaction in the stratosphere) and there are other parts that cool mainly by radiation (cool east Pacific) because they have no cloud cover (big reaction in the stratosphere). Both classes are affected. There are times of the year where there is a lot of long wave radiation (big reaction in stratosphere) and other times when there is much less (no reaction in the stratosphere). These times are equally affected by this change in temperature at 250mb (9km).

In January there is another 90 watts per square metre available and in July not so. In the January to March period the entire stratosphere cools due to uplift and decompression. The January solar signal nevertheless shows up at 250mb.

Hypothesis: Infrared absorbtion consequent upon an increase in solar irradiance causes a reduction in area of cirrus cloud and knock on effects follow as the surface returns heat to the lower troposphere. A loss of albedo generates warming across the tropics. The atmosphere recovers cloud as the oceans warm.

287. Posted May 14, 2008 at 2:52 PM | Permalink

283 (JimA):
From the paper you quote:

improved hourly space weather predictions.
Another phenomenon caused by CMEs is their modulation of cosmic radiation a cting on the Earth’s troposphere[Cane et al., 1996] suggesting that coronal mass ejections might have an influence on the Earth’s climate[Feynman and Ruzmaikin,
1999].

1: the space weather refers to conditions in space, not in the earth’s atmosphere.
2: brings up the un-confirmed speculation that cosmic rays might have an influence on climate.
There is no evidence in the paper you cite for any of these assertions.

288. Jim Arndt
Posted May 14, 2008 at 3:38 PM | Permalink

287 Leif,

Did you read the T. Baranyi and A Ludmány paper I posted in 285?
http://fenyi.sci.klte.hu/publ/Praga2002.pdf
From the conclusion;
“The above reported features may help to reveal the
factors contributing to the results which show that the
CMEs and the fast wind streams can affect the lower
atmosphere in different ways and in different latitudinal
distributions. The specific mechanisms of the
atmospheric response need further investigation.”

But I said big CME’s can take out big power grids (known) so they do effect the troposphere and this may also effect cloud forming nuclei, some high energy particals do reach the upper troposphere.

289. Posted May 14, 2008 at 3:42 PM | Permalink

286 (Erl):

272 (Leif): The amount of water up there is so small that not much heat can result from absorption [and cirrus clouds reflect rather than absorb
Erl: And if that is the case a small change in irradiance can make a very big difference.

You sound almost like a believer in homeopathy, where the smaller the concentration of the medicine is, the stronger is the effect, until a maximum effect is reached when there is not a single molecule left of the original medicine. This is implausible and so is your comment.

290. Jim Arndt
Posted May 14, 2008 at 3:47 PM | Permalink

288 me

But big CME’s not But I said

291. Posted May 14, 2008 at 4:20 PM | Permalink

288 (JimA); Papers that say that this or that may help or can reveal or might suggest, etc., curry little favor with me. The paper you cite presents no evidence of CME-effects on the troposphere. It is one of a class of several hundreds I have come across over the last 40 years. Collectively, these papers contradict each other, have been later shown not to hold up, etc. The fact that they still appear may suggest [see I know the lingo] the possibility that one cannot exclude the results reported on the basis of preliminary data that requires further research to substantiate. In short: they are not credible.

292. Erl Happ
Posted May 14, 2008 at 5:16 PM | Permalink

Leif,
Not much I can do if you continue to turn a blind eye to the data.

293. pochas
Posted May 14, 2008 at 5:41 PM | Permalink

Leif,

Here’s an exercise I did for my own amusement. I took the “Lumpy” model that lucia has been talking about and supplied it with “forcings” which were simply sunspot numbers. The output (blue) was scaled and plotted with the Armagh temperature anomalies (red), calculated by averaging the mean monthly max and min temps and appplying a 12 month smooth, then standardizing. I think the result is interesting. The time constant in the “Lumpy” model was 78 months. The result is hardly perfect, but I do think it is suggestive.

294. Posted May 14, 2008 at 6:25 PM | Permalink

292 (Erl): The onus is always on the one claiming an effect. And you have to convince the disbelievers, not by asking them to look at the data and ‘do the exercise’. They have little motivation to so. You have to stick it to them in little pieces, one at a time. This is always possible. It is no argument that things are so complicated that one has to grasp the whole picture before being able to understand a minute corner of it. I tried earlier [a couple of times] to pin down the essentials of what you were saying but always faltered and we got stuck in the morass of peripheral detail that was served up. An incorrect way of doing this is to ask: ‘do you agree to X’ without specifying why this particular question was raised. It should be more like: ‘In order to appreciate Z, we need to agree on X’. Then questions and issues raised should not be just dismissed or glossed over. Back in #109, I went to the trouble [after honest effort] of identifying four Ws and Cs that could not understand. correct procedure is for you to patiently [just as I do, if may be so immodest] explain for each of why points where lies the error of my ways or my lack of understanding. If you just barrel on, or [as you did] answer with a generality [“I’m only human”] you are neglecting reasonable and well-established rules of scientific discourse [even some scientists do that now and then if they are on thin ice]. The ‘proper’ way is to be specific and for each of my specific cases point out where my lack of understanding prevents me from seeing what is plain [as I’m human too]. When that has been done, my response would either be “oh, I see” and we have made progress, and I would present you with yet another a small number of similar cases where things are not plain to me, or I would keep concentrating on the same four cases until it is clear that no agreement can be made. And so we would continue until there are no more issues on that graph or Figure. This may sound tedious but is the only way. Google: “There is no royal road to understanding”.
To say that I “continue to turn a blind eye” to the data belittles my effort and the hundreds of comments and arguments I have engaged in.

295. Posted May 14, 2008 at 6:33 PM | Permalink

293 (pochas): As the sunspot number goes to [almost] zero every minimum it should not be massaged such as to lose that property, all [blue] minima must be at [almost] the same level. Hence I cannot accept that this is a valid approach.

296. pochas
Posted May 14, 2008 at 8:05 PM | Permalink

295 (Leif)

It’s not the sunspot number that is plotted. It is the numeric integral of the “Schwartz” model as lucia calls it:

dT/dt = -T/C + S*F/C

where: T = temp; C = time constant 84 months; S = sensitivity (adjust to fit); F = forcing here sunspot numbers from

ftp://ftp.ngdc.noaa.gov/STP/SOLAR_DATA/SUNSPOT_NUMBERS/MONTHLY

A series of strong cycles force the integral upward, weak cycles pull it downward. I just ran a quick plot
of the sunspot numbers, and indeed between cycles they are all near zero.

297. Posted May 14, 2008 at 9:13 PM | Permalink

296 (pochars): sorry, I should have looked more carefully and not let my knee-jerk reaction get the better of me :-)
1: I see no or little hint of a correlation on the 11-year scale.
2: the longer term increase comes about naturally because both T and S have an increase over the last 120 years. You might get the same result [sans the 11-year wobble that is not in T anyway), by using the US population or the Dow Jones index or any other quantity that has gone up over the last century or so. The number of degrees of freedom is here very small.

298. cba
Posted May 15, 2008 at 3:16 AM | Permalink

295 (Leif):

I think the blue is no longer sunspot numbers but a processed output (ostensibly a temperature) to a simple (I think the term is perhaps a phenominological thermodynamic) model, Lumpy, consisting of something like a gain and a time delay factor or response time factor. The model is probably unrelated to any physical reality of the system it’s intended to model. There was a paper referenced here a few months back on one of these that tried to use Wang’s and Lean’s values – Scafetta & West 07 ????.

299. Basil
Posted May 15, 2008 at 9:18 AM | Permalink

#297,

Leif,

Isn’t “S” a constant?

I agree, though, with the overall sentiment.

What would you say, though, to evidence showing a correlation between temperature and the Hale cycle, rather than the Schwabe cycle? What could it be? Not irradiance, because that doesn’t vary any more over 22 years than over 11 years. At least T. Baranyi and A Ludmány have a theory about how there could be a 22 year variation in solar energy that shows up in earth temperature variations. It is not a very simple theory, but as I’ve seen you say more than once, the Sun is messy.

There is something taking place during the 22 year Hale cycle, as opposed to the 11 year Schwabe cycle, that influences Earth’s temperature. If the previous plot didn’t show enough correlation between temperature and the 11 year variation in sunspots, try this one:

The blue line is actual variation in the rate of change in a smoothed series of HadCRUT3, monthly, 1850:01 through 2008:04. The red line is a simple sinusoidal fit with two signals, one having a frequency of 258 months (or 21.5 years, i.e. the Hale cycle), and the other having a frequency of 111.6 months (or 9.3 years, i.e. half of the 18.6 year lunar nodal cycle). A least squares fit produces periods of 248 months and 110 months. The longer cycle is too long to be the lunar nodal cycle. But it could be the Hale cycle modulated by the lunar nodal cycle to produce a beat cycle (a theory first advanced by P.R. Bell in 1981). And that could well account for the lack of fit in the 1920’s and 1930’s shown in the diagram above, because a beat cycle composed of the Hale cycle and the lunar nodal cycle will go through a beat cycle approximately every 120 years in which frequency and amplitude are distorted somewhat like we see in the above diagram.

I know you are a skeptic of cyclomania. But there is just too much evidence of bidecadal variations in climate to write it off as coincidence. Incidentally, there is more evidence for 22 year variations in climate, than for 11 year variations. I think that’s a case where absence of evidence is evidence of absence. I.e., there is something like a Gnevyshev-Ohl rule at work here in which the even numbered solar cycles have less impact on Earth’s temperatures, allowing the lunar nodal cycle to dominate in determining cycle frequency. With the odd numbered cycles, though, the solar influence becomes noticeable.

FWIW.

300. Posted May 15, 2008 at 10:19 AM | Permalink

299 (Basil): about ‘S’, I meant ‘F’ [clearly – although some people argue that ‘a Watt is not always a Watt, when it comes to sensitivity’].
Why do you bring in the lunar stuff? The more parameters the better is the chance that a fit will show up. Try to add the Dow Jones index. [Not silly as even Hershel thought that the price of wheat was correlated with the solar cycle]. The even-odd cycle is just coincidence [has been broken at times, most recently between cycles 22 and 23 in 1996]. The sun has two magnetic cycles, one from minimum to minimum [controls polarity of sunspots] and one from maximum to maximum [reversal of polar fields – influencing the modulation of cosmic rays], and there are no mechanisms that connect these to any terrestrial weather/climate changes. The notion that sunspots are related in some way to weather/climate goes back 400 years and thousands of ‘papers’ have been written on this. That we are still trashing around is a clear sign that this is wishful thinking. This may never stop, even when we get several more centuries worth of good data, as then the ‘papers’ will be about why the effect is sometimes there and sometimes not; what physical mechanism [apart from Global Warming :-) which is blamed for everything nowadays] could account for such a subtle [but ‘robust’] behavior? etc, etc. You don’t see papers anymore about whether the Sun goes around the Earth or the Earth around the Sun. That has been settled [mostly – Josh.10:13 notwithstanding].

301. hemst101
Posted May 15, 2008 at 4:43 PM | Permalink

Leif,

Just noticed your lovely charts at #175. Could you indicate on them the known past starts of the other cycles – 22 and 23? Is it possible to determine the start of a cycle by what you see on the charts?

I believe that the start of 23 was May 1996. Just looking at the transition of 23/24 and the “fly dirt” and comparing to 21/23, 22/23 it seems to me that the official start of 24 is just a few months away.

I’ll send a sweetly worded email to the addresses posted.

302. Posted May 15, 2008 at 4:51 PM | Permalink

(Erl): as a follow up on #294 and #109, I’ll show one of your plots where you have marked areas in the middle troposphere as Warm or Cold with little boxes [red and green]. I have tried, in vain, to see how the boxes relate plainly to the temperature anomalies. For illustration I have circled a W/C pair with a light blue circle and another pair with a pink circle. To the best of my ability I have tried to see what the difference in the contours are between the W and the C within each pair. The background contours look identical or nearly so for the area where the box is:

So, please explain in detail why those four are where they are, and what makes two of them W and the other two C.

303. Posted May 15, 2008 at 5:19 PM | Permalink

301 (hemst): Thanks for sending ‘sweet words’ to where they are needed. The start of the cycles is in the eyes of the beholder, so I won’t mark them. I am working on extending the graph backwards in time to 1874. But the day only has 24 hours, …

304. David Archibald
Posted May 15, 2008 at 5:41 PM | Permalink

Re 300, Dr Svalgaard you are right. There a plethora of papers showing a Sun – climate connection. A lot of these are tree ring studies. The tree rings are hard physical evidence. Try as you might, you cannot wish them away.

305. Posted May 15, 2008 at 5:45 PM | Permalink

304 (DavidA):

A lot of these are tree ring studies

post three of the best and three of the worst and I’ll discuss them.

306. David Archibald
Posted May 15, 2008 at 6:16 PM | Permalink

Further to 299 and 304, a number of tree ring studies show a bigger Hale cycle effect that Schwabe cycle, including the Japanese cedar one.

307. Posted May 15, 2008 at 9:18 PM | Permalink

I’ve been absent cause I was working on the next chart. Tell me… What do you infer from the hexic trends?

It’s logical that Svalgaard’s reconstruction (only sunspots) coincides with the hexic of the annual sunspots number. Nonetheless, the hexic of the deviations of temperature is also coincident. As Judith Lean based her reconstruction (2001) on sunspots and background (proxies) the hexic of her database doesn’t coincide with the hexic of the sunspots number or the hexic of deviations of temperature. Beholder, perhaps something has been missing there…

308. Posted May 16, 2008 at 7:14 AM | Permalink

307 (Nasif): too many curves on one graph… Extract what you think you see.

309. Erl Happ
Posted May 16, 2008 at 7:24 AM | Permalink

302 Leif
Leif
Thanks for resurrecting that query. I will answer very carefully.

All your queries relate to my markings at the altitude 5-10 Km. The light coloured zone that lies between 5 and 10km has a temperature between 0 and 4°C. It is warmer than the light green zone that lies above and below the light buff coloured zone. The latter has a temperature between 0 and minus 4°C.

I have observed that the light buff coloured zone expands in vertical extent when the tropical lower troposphere (UAH data) is warming. My markings record a vertical expansion (W) or contraction (C ) of that warmer zone. On a Hovmoller diagram showing all longitudes one can see that the zone expands laterally as well as vertically. When the UAH anomaly in the tropical troposphere falls below zero that warm anomaly disappears entirely, as it has at the moment. You don’t need to look at anomalies in fact. The change is clearly apparent in the mean temperature.

The warming zone between 5 and 10 km in elevation has a temperature between minus 5°C and minus 60°C and I believe that the dominant cloud at this level is cirrus and stratus. Most cloud of the rain bearing variety lies well below this level. The release of latent heat of condensation occurs mostly below 700mb or about 2.5 km in elevation at the equator. So, precipitation occurs mainly as water rather than hail. Outgoing long wave radiation from cloud absorbtion of solar energy in this zone assists to warm the cloud layer and drive convection. The zone of greatest anomalies includes the warmest tropical waters and those immediately to the East, slightly less warm.

A part of the ocean that is always cool, like that in the South Pacific adjacent to South America shows up as a cool zone all the way to 12.5 km where a temperature reversal occurs and the zone appears as warm by comparison with its surroundings.

The differentiation in temperature that we observe at the surface of the Earth diminishes rapidly with elevation and is much diminished at 2.5km. Because of the lack of moisture between 5Km and 10Km there is little opportunity for outgoing long wave radiation to be absorbed. One could equally remark that there is little opportunity for the absorbtion of shorter wave infrared. My gut feeling is that the shorter the wave length the greater the heating impact is likely to be but don’t hang me for this. The argument does not depend on gut feelings.

If the anomalous warming of the atmosphere between 5K and 10K were due to long wave radiation we would see a signature in the stratosphere as we do for the East Pacific cold pool. We do not.

The anomalous warming is generally more intense in January when solar radiation is high, OLR low and ocean heating greatest.

This anomalous warming occurs above the hot pool and also partly into the cloudy zone of the central Pacific that emits a lot of OLR but sees much less convection. There is no difference in the intensity of anomalous warming between these parts. The split is about 50/50.

OLR from the Earth is greatest in Northern Hemisphere summer when solar radiation is weakest (aphelion). This signal is very apparent in the tropics despite the lack of land in those parts. By contrast the anomalous warming at 5-10 km is most intense in Southern Hemisphere summer.

There is a greater degree of differentiation in temperature between 5 and 10Km than immediately above or below.

Very frequently heating is simultaneous high in the atmosphere and in the 5-10km zone.

Hence the evidence strongly suggests that the anomalous heating at 5 to 10Km is solar in origin.
It appears that this heating at 5-10K is linked with tropical warming and the generation of many features that move with the solar cycle.

The data that is needed to demonstrate all these observations is so voluminous as to make posting difficult but I am working on it.

310. cba
Posted May 16, 2008 at 7:50 AM | Permalink

BTW,

these graphs are nice and small and don’t extend beyond the width of the screen, but they’re too small for me to read the legends or have any clue as to what they’re about

311. Posted May 16, 2008 at 8:54 AM | Permalink

310 (cba): since you do not refer to a posting number, how can one know which graphs you mean?….

312. Posted May 16, 2008 at 9:00 AM | Permalink

309 (Erl):

All your queries relate to my markings at the altitude 5-10 Km. The light coloured zone that lies between 5 and 10km has a temperature between 0 and 4°C. It is warmer than the light green zone that lies above and below the light buff coloured zone. The latter has a temperature between 0 and minus 4°C.

I have observed that the light buff coloured zone expands in vertical extent when the tropical lower troposphere (UAH data) is warming. My markings record a vertical expansion (W) or contraction (C ) of that warmer zone.

Erl, the above is all that was needed, ALL THE REST IS NOT INTERESTING OR RELEVANT AT THIS POINT. This is the general problem with your postings that the point gets obscured by details that are as yet unsubstantiated.

313. Basil
Posted May 16, 2008 at 9:02 AM | Permalink

#300

Leif,

I brought lunar into it because that’s what the evidence suggests: the period of the shorter cycles is closer to 9 years than 11 years. The period of the longer cycles — which MTM spectrum analysis classifies as a “harmonic,” meaning a very strong and statistically significant signal — is a bit over 20 years, way too long to be the full lunar nodal cycle, but a bit too short to be a full Hale cycle. But it is right where it needs to be to be a beat cycle of the two.

What is your problem with a possible lunar influence on climate cycles, other than it’s lunacy? :) There’s a debate in the literature over whether the drought cycle of the western US is the lunar nodal cycle or the Hale cycle. It was P. R. Bell who suggested that it was a beat combined from the two. Hoyt and Schatten discuss Bell’s theory, and find it attractive except for the lack of any known influence of variation in the Sun’s magnetic polarity on climate. But to repeat the old refrain, absence of evidence is not evidence of absence. Just because we do not know what the connection is doesn’t mean the connection is not there.

Back to lunar for a moment, the full lunar nodal cycle has been shown to play a role in climate dynamics in the North Pacific. Some, who seem to want to be sure that the Sun is not attributed with significant influence on climate would attribute any and all bidecadal signals to either lunar or internal ocean dynamics. But any oscillation of 20+ years is too long to attribute solely to the lunar nodal cycle.

I’m just going where the data is taking me. When I got into all of this, it was simply to compare temperature trends during the 1920’s and 1930’s with temperature trends in the 1980’s and 1990’s. I was totally surprised to find such clear evidence of bidecadal cycles in temperature. I initially used a technique that is unfamiliar to climate science (Hodrick Prescott filtering), but the cycles are there using other more conventional techniques (periodograms, MTM spectrum analysis).

I don’t understand the basis of your skepticism of the Gnevyshev-Ohl “rule.” Yes, there have been exceptions, or violations. But the exceptions are rare enough to be considered unusual, and the norm as usual. Someone who was given even odds, and who bet on the basis of the “rule” would win more often than they lost. That says something. You cannot dismiss something like that as coincidence, and ignoring it is not good science. We should try to find out why it is the usual case. And I suspect that when we do come to understand it, it will help us understand the bidecadal oscillations in terrestrial climate, because as my graph above shows bidecadal oscillations in temperature show a fluctuating amplitude that corresponds to the Gnevyshev-Ohl rule: greater amplitude during odd numbered solar cycles, and lesser amplitude during even numbered solar cycles.

314. pochas
Posted May 16, 2008 at 10:34 AM | Permalink

293 (me):

I revised the graph to lag the sunspot numbers by 36 months. This shows more clearly what is solar and what is “weather”. Apologies to you, Leif – I know you dismiss such nonsense out-of-hand.

315. Posted May 16, 2008 at 11:51 AM | Permalink

First chart, Lean against Dev. of Tropos. Temp.:

Second chart, Svalgaard against Dev. of Tropos. Temp.:

Third chart, Number of Sunspots against Dev. of Tropos. Temp.:

Fourth chart, Total Solar Irradiance against Dev. of Tropos. Temp.:

316. Posted May 16, 2008 at 11:54 AM | Permalink

Well… I love hexics ;)

317. cba
Posted May 16, 2008 at 12:01 PM | Permalink

315 (Nasif):

those are actually pretty readable now . tnx

(Leif):

I was referring to any small ones like those Nasif was doing.

318. Posted May 16, 2008 at 5:17 PM | Permalink

314 (pochas): Use the actual sunspot number instead of the artificial sine wave. Modulate it, if you must, by the lunar cycle, standardize both to same variance then plot. Your ‘absence of evidence’ etc has to used correctly. You said: “absence of evidence is not evidence of absence. Just because we do not know what the connection is doesn’t mean the connection is not there”. The second sentence should have read: “just because we do not have any evidence for the connection, doesn’t mean the connection is not there”. I will contend that if there is no evidence, then one should not assume that there is some. As I have said many times, you can find papers that claim just about anything; that does not mean that those claims are substantiated. Maybe it is because I have read hundreds of such papers over the decades and found them wanting that I have a measure of healthy skepticism. Don’t get me wrong, I would absolutely love that there were a solid connection [it would make my field so much more important and perhaps even commercially valuable], but alas I haven’t seen any yet that I could take to the bank.

319. Sam Urbinto
Posted May 16, 2008 at 5:39 PM | Permalink

As I long-windedly ask in this topic over at the bulletin board, something raised the anomaly overall by basically +.2 twice since 1945, once in 1977 and once in 1992-1995. In other words, if you adjust 1977-2008 -.2 the anomaly is hovering around zero until ~2002 and if you adjust 2002-2008 again by -.2 it hovers around zero again.

Did instruments and gathering the anomaly change? Was it something in the pipeline lagging until then? is it the sun or cosmic rays or Earth magnetic field strength, is it the land-use changes and fossil fuel usage?

If you add .2 and .4 you pretty much get the trend since 1880.

320. Posted May 16, 2008 at 6:05 PM | Permalink

319 (Sam): 1977 and ~2003: nothing special with the Sun, cosmic rays, Earth’s magnetic field, etc.

321. Posted May 16, 2008 at 6:33 PM | Permalink

318 (Leif): pochas, I got you conflated with Basil. Sorry for that.

322. Posted May 16, 2008 at 6:58 PM | Permalink

Erl, this looks like the absolute temperatures:
http://bobtisdale.blogspot.com/2008/05/fresh-look-at-ncdc-absolute-temperature.html

323. Erl Happ
Posted May 16, 2008 at 11:50 PM | Permalink

319 (Sam)
My tuppence worth:

1976 was the turning point where solar cycles as a whole went from ENSO negative to ENSO positive. A change in the timing of increases in irradiance can do this.

In the nineties a series of three strong El Ninos occurred in the decline phase of SC22 and the effect was compounded by the massive 1997-8 event at the start of SC 23. The upswing in solar cycle 23 brought a strong increase in ocean temperatures. However, evaporation from continually warming oceans could not keep up with the trend in irradiance. The oceans continued to warm during the La Nina of 1999-2000. Precipitable water in the atmosphere increased continuously as did relative humidity.

El Nino’s are falsely described as an internal oscillation in the climate system implying temperature neutrality. In focussing on the Pacific people miss the main event, a generalised warming of the entire tropical region.

ENSO events involve a change in the Earths albedo brought on by heating at the 250mb level that evaporates the cirrus cloud that shields the tropics from solar radiation. Without that cirrus cloud the sun has nothing in its way except for the thunderstorm activity that rings the Earth about the equator.

324. Erl Happ
Posted May 17, 2008 at 12:11 AM | Permalink

322 (Leif)
Data is for the globe rather than the tropics. Because parts of the globe experience heating while other parts experience cooling I would not expect to see strong correlations. Irradiance variations that occur during NH summer do not add much energy to the oceans.

For these reasons I prefer to work with data for the tropics. By and large, this is where cloud cover is least and changes in irradiance will produce ocean warming. Ocean warming is important for the PDO and the NAO.

But thanks for the reference. I appreciate the thought.

325. Erl Happ
Posted May 17, 2008 at 12:18 AM | Permalink

319 (Sam)

what happened around 2002-2005?

From about 2003 the oceans began to cool. We are into decline phase for SC 23. The decline in the ocean temperature has been progressive as you see in #284.

326. Erl Happ
Posted May 17, 2008 at 12:34 AM | Permalink

http://climatesci.org/2008/05/13/tropical-water-vapor-and-cloud-feedbacks-in-climate-models-a-further-assessment-using-coupled-simulations-by-de-zheng-sun-yongqiang-yu-and-tao-zhang/

underestimating the negative feedback from cloud albedo and overestimating the positive feedback from the greenhouse effect of water vapor over the tropical Pacific during ENSO is a prevalent problem of climate models

No surprise.

327. lgl
Posted May 17, 2008 at 6:09 AM | Permalink

319, Sam

Have you seen this?
http://icecap.us/docs/change/ocean_cycle_forecasts.pdf
PDO positive in 1977, AMO positive around 1995, and the sum of them probably peaked around 2002.

328. Basil
Posted May 17, 2008 at 6:26 AM | Permalink

#318

Leif,

Let’s try this one:
These are actual temps (red, HadCRUT monthly) and SSN’s (blue), smoothed, and then differenced. Prior to the early 1900’s, the two are negatively correlated but in almost an 180 degree phase. There is one phase reversal after that, ~1950 (cycle 18), and a couple of extra small temperature cycles around the time of solar cycle 16. Here are some comments.

1. Out of phase, or negative correlations are widely reported for the 1800’s I believe. You can see that in this data as well. I’ve seen one article recently speculating about the existence of a “double Gleissburg cycle” to explain such phase reversals on a 80-100 year basis.

2. I don’t think SSN’s per se are the causative agent in driving Earth’s climate or temperature cycles. They are simply proxies for something else that is going on. Whatever that “something else” is, there is a correlation between the frequency of the cycles, as measured by SSN’s, and the frequency of cycles in climate variables (temperature, rain, drought, tree rings, et al.).

3. Issues of phase aside, there is a remarkable (in my opinion) correspondence between the two in terms of something like the Gnevyshev-Ohl rule. Look closely at the temperature signal, and you will see much stronger peaks associated with the odd numbered solar cycles, i.e. a longer “double sunspot” or Hale cycle periodicity in the temperature cycle, with reduced peaks in temperature cycles associated with the even numbered solar cycles.

4. There’s no “lunar” added to anything you see in the figure. I still think lunar may be a factor in modulating the frequency of temperature cycles. And there may be other terrestrial dynamics that would interfere with the phase relations between the Sun-Climate connection. In terms of frequency, amplitude, and phase, we are batting two for three here, i.e. significant associations in frequency and amplitude. And over long periods of time, there is even a good fit with phase, especially if a plausible explanation can ever be found for the two being 180 degrees out of phase.

Basil

329. pochas
Posted May 17, 2008 at 7:44 AM | Permalink

#328 (Basil):

Having just completed a similar exercise, I am persuaded that there is a solar influence on climate. However, I think you will agree that it is important not to be too dogmatic about this. There are numerous significant temperature excursions that simply cannot be associated with solar cycles. In the short term, internal cycles can dominate and even hide the solar influence.
I think that in this area one must not over-analyze.

330. Joachim
Posted May 17, 2008 at 9:05 AM | Permalink

Sorry for lurking, but are not the fabulous charts Leif showed us here?
http://www.solarcycle24.com/trend.htm

331. Joachim
Posted May 17, 2008 at 9:09 AM | Permalink

#330 (me) ah, sorry. I see it was the comparison between spots from different cycles that was the issue.

332. Posted May 17, 2008 at 9:51 AM | Permalink

328,329 (Basil,pochas): People have been looking at this for centuries. Correlations have come and gone. Every proponent is absolutely convinced and unshakable. The ‘debate’ is still ongoing simply because the connection has not been established. This does not mean it is not there. It means that the various attempts to establish such a connection have not been up to a standard that compels the scientific communities [and there are several involved] to embrace the ‘evidence’ and/or the ‘mechanisms’ proposed. The analyses just presented are typical and are just as unconvincing [to me at least]. The usual ‘defense’ by the proponents is that they just go ‘where the data takes them’, but that often in the end turns out to be a lonely place. This does not deter people form going there, but often leads to a certain bitterness. My advice is to be happy in your belief.

333. Basil
Posted May 17, 2008 at 10:20 AM | Permalink

#329

Pochas,

I certainly hope I do not come across as dogmatic about any of this. To paraphrase what Leif often says about the Sun, climate is very messy also. The correlations I’m discussing filter out pretty much everything less than a decade in length, and there are certainly many “temperature excursions” (as you put it) that cannot be attributed solely to the solar cycle. Even some long term — say ~60 years or more — climate cycles probably owe to terrestrial ocean dynamics, rather than the solar cycle.

My focus is primarily on bidecadal periodicities, not short run activity. There’s too much evidence of bidecadal climate variation to dismiss it as coincidence. So what causes it? Broadly, there are three possible agents: the solar cycle, lunar cycles, and endogenously determined climate oscillations. I do not see much evidence for the third of these, at least not for the kind of well developed bidecadal signals I’m talking about (which thereby excludes ENSO, PDO, AMO/NAO oscillations, which are likely endogenous to some extent). That leaves solar and lunar. I’ve seen arguments for one or the other, for this bidecadal cycle or that one. And maybe that’s the way it is — some are driven by solar variations, and some by lunar cycles. And possibly, sometimes it is interaction of the two (which is what I think makes the best sense of the data I’m looking at).

Basil

334. Ian
Posted May 17, 2008 at 12:04 PM | Permalink

332, 333

“The usual ‘defense’ by the proponents is that they just go ‘where the data takes them’, but that often in the end turns out to be a lonely place. This does not deter people form going there, but often leads to a certain bitterness. My advice is to be happy in your belief.”

People have researched this and been shunned but some have gone on to make fortunes using the information gleaned, even if incomplete and deficient (to date) in the provable science department.

335. Posted May 17, 2008 at 12:26 PM | Permalink

334 (Ian):

some have gone on to make fortunes using the information gleaned,

“There is a sucker born every minute” – David Hannum

336. Posted May 17, 2008 at 1:19 PM | Permalink

Quoted from solar Physics (2007) vol. 245, p 247-249:
Did the Sun’s Prairie Ever Stop Burning?
Peter Foukal and Jck Eddy.
“The red flash, observed for almost ten seconds immediately before and after totality, is familiar to solar eclipse observers as a thin red arc on one side of the Moon’s dark edge. It arises from the intense red hydrogen radiation of the chromosphere, which consists mainly of radially oriented ‘spicules’, extending 5000-10000 km above the photospheric limb. Drawings of these spicules by early observers led to the description of the chromosphere by the noted 19th-century Jesuit astronomer Angelo Secchi as a “burning prairie”.

“It is now well known that the spicule jets move upward along magnetic field lines rooted in the photosphere outside of sunspots. Thus the observation of the red flash produced by the spicules requires the presence of widespread solar magnetic fields. Historical records of solar eclipse observations provide the first known report of the red flash, observed by Stannyan at Bern, Switzerland, during the eclipse of 1706 (Young, 1883). The second observation, at the 1715 eclipse in England, was made by, among others, Edmund Halley – the Astronomer Royal. these first observations of the red flash imply that a significant level of solar magnetism must have existed even when very few spots were observed, during the latter part of the Maunder Minimum.

“This historical evidence may bear on reconstructions of total solar irradiance (TSI) variation and its possible forcing of climate over the past few centuries. It has been suggested, for example, that, during the Maunder Minimum, TSI may have decreased sufficiently to help account for the broad minimum in global temperature during the coldest part of the Little Ice Age (Lean, Beer, and Bradley, Geophys. Res. Letters, vol 22, p 3105, 1995). The mechanism suggested for this large TSI decrease is a significant depletion of the bright magnetic regions in the photosphere that are associated with spicule foot points. To obtain a climatically significant TSI decrease, Lean Beer, and Bradly proposed the disappearance not only of the bright network structures associated with spicule foot points but even of the faintest inter-network magnetic elements located within supergranule cell centers. The historical eclipse observations described here seem to require the presence of the bright network structures, and thus of substantial solar photospheric magnetism during at least the last decade of the Maunder Minimum. Hence, the red-flash observations would argue against a climatologically important decrease in TSI during that period of time.”

Note the decrease of TSI advocated by Lean et al. in 1995 was proposed in part in order to explain the Little Ice Age. So it is hard to take that as evidence that TSI was significantly lower. Anyway, the red-flash story here shows the value of old observations when interpreted through modern knowledge.

337. Posted May 17, 2008 at 3:39 PM | Permalink

# 336

Leif,

I didn’t go to the synagogue again… Well, your explanation is quite clear. A low sunspots number doesn’t explain the Maunder Minimum. However, what was the cause if supposedly the CO2 was at their “normal” levels? And what caused the recovering of the atmospheric temperature if the solar activity and the CO2 level in the atmosphere were “normal”? Perhaps Interstellar Cosmic Rays?

338. Posted May 17, 2008 at 4:34 PM | Permalink

337 (Nasif): Why must everything have an exterior cause? Complex systems continuously fluctuate about their mean position and the various feedbacks play against each other.

339. Joachim
Posted May 18, 2008 at 4:31 AM | Permalink

#338: Leif
Is that your personally preferred explanation to climatic fluctuations in the last 100-150 years?

340. cba
Posted May 18, 2008 at 5:09 AM | Permalink

338 (Leif):

Even complex system chaotic variations tend to have causes and just because they’re complex and average out – making them essentially irrelevent in the scheme of things – doesn’t mean important details can’t be teased out. It’s just that the results are usually not worth such effort. In this case, it may not be possible to even accomplish what is already being attempted but since it is so important and relevent to the survival of modern society, it’s just not intellectually acceptable to think that this tangled ball of twine cannot be unraveled. Hey, it makes as much sense as particle people looking for the ultimate symmetry because nature must be as elegant as math.

341. Posted May 18, 2008 at 6:16 AM | Permalink

338,339 (Joachim,cba): what I meant was that the climate system is so complex that I don’t think it pays to look for simple, immediate, and linear external sources, like solar activity the last few years have declined and therefore the Earth’s climate now is cooling, or there was a flare yesterday and therefore the Greenland Ice Cap is falling into the sea. It might in the fullness of time be possible to disentangle the mess. We did that with geomagnetic activity where most [but not all] interactions are now understood. It only took a couple of hundred years to do it. We might fare better with understanding the climate [more data, researchers, etc] although that might be balanced by the climate system being more complex. So, no need to despair, but also little chance of simple solutions.

342. kim
Posted May 18, 2008 at 6:18 AM | Permalink

340 (cba) Nice perspective. It seems ‘Kitten’ Erl is having as much fun as any cat with the ball of twine.
============================================

343. Erl Happ
Posted May 18, 2008 at 8:17 AM | Permalink

338 (Leif)

Why must everything have an exterior cause?

Not everything. Just the temperature of the tropics where the sun shines most brightly and a difference between a little bit of cloud and no cloud at all is absolutely vital.

the climate system is so complex that I don’t think it pays to look for simple, immediate, and linear external sources, like solar activity the last few years have declined and therefore the Earth’s climate now is cooling

342 (Kim)

It seems ‘Kitten’ Erl is having as much fun as any cat with the ball of twine.

Kim, your support is much appreciated. In the end, it’s not that difficult or technical.

I believe that the microwave wavelength 2800 MHz daily radio flux correlates highly with the daily sunspot number and the two databases are used interchangeably. That wave length will also heat water and very likely accounts for the relationship shown in the graph below and consequent fluctuations in the amount of cirrus cloud in the relatively cloud free zone of the tropics.

Salient points to note:
1. Temperature anomalies in the TLT fairly closely follow changes in microwave flux with a fairly consistent lag of two months.
2. Short term changes in irradiance frequently exceed the 7% that is due to orbital influences.
3. Recently, periods of increasing flux have been of short duration and interspersed with much longer periods when irradiance is decreasing. This accounts in part for the decline in UAH temperature anomalies in the lower troposphere since 1995.
4. At the 250mb level periods of anomalous warming occur that are unrelated to what is happening in the atmosphere at lower and higher elevations. These periods are marked not so much by exaggerated temperature differences but by the spread of anomalous light warming so that it becomes much more extensive.
5. The zones of anomalous warming lie above the warmest part of the tropical oceans and are sometimes described as the ‘Hot Pool’ of the Pacific and the Indian Oceans. These zones are characterised by higher evaporation and strong uplift. They are also that part of the atmosphere which has the highest precipitable moisture, highest relative humidity, but generally low precipitation (not shown here but next post) and exhibit relatively low levels of outgoing long wave radiation. The anomalous warming can not be accounted for by absorption of long wave radiation from below. Rather, it must be related to solar flux as the graph strongly suggests. These anomalously warm zones at about 250mb pressure lie above portions of the oceans that cool strongly via evaporation, the release of latent heat of condensation, convection and decompression. Long wave radiation is of low importance in the mix. By contrast the cold pool near South America cools strongly via OLR and it shows up as a cool spot in the Hovmoller diagrams up to an elevation of 150mb (12.5Km) at which point the presence of ozone yields a temperature reversal in that zone.
6. The Southern hemisphere has proportionally much more ocean than the northern hemisphere. Increases in solar flux that occur in SH summer are very effective in increasing sea surface temperature. Over the period from 2005 SH summer (blue circles) has been uniformly characterised by decreasing solar flux. It would be expected that under such circumstances the cirrus cloud of the middle troposphere would extend in its extent and opacity.
7. The resultant increase albedo would cause the surface cooling that is apparent over the period. This has occurred despite the advantage of orbital proximity that yields an extra 7% energy in January over July. Had the situation been reversed it is likely that the decline in temperatures would have been much more modest.
8. It is noticeable that the stratosphere warms strongly in NH summer. This must be due to the emission of Long wave radiation by land masses. Increases in irradiance that occur in the period June to November are never going to be geo-effective.
9. Increases in irradiance in the SH summer period are unlikely to show up in the stratosphere in terms of heating due to increased OLR. The energy goes into the oceans and to the extent that it is returned to the atmosphere it will be via evaporative processes, convection and heat loss due to decompression. The cooling signal due to convection and decompression appears strongly in the stratosphere. This is a good case of influence from below, as is the heating due to enhanced OLR between June and October.

A note on the mechanism:
The variation in irradiance over the entire solar cycle is relevant but only in that the relativity between cooling and warming episodes changes. It is the short term changes (months at a time) much more than 7% due to orbital influences that affects the albedo of the upper atmosphere. The fact that the albedo is soon replenished is important. The Earth system is not a passive player.

One of the most noticeable variations that occur over the solar cycle as a whole is that in precipitable moisture. The solar cycle is one big El Nino event. But don’t get hung up on the Pacific and the notion of internal oscillations. That is a monstrous red herring. It is the entire tropics that cools and warms, some parts more than others according to where the furniture is located. And it is changes in irradiance that drives the changes in temperature.

Forgot to show it as a label but #284 charted the decline phase of Solar cycles 22 and 23.

Finally, without heat from the tropics the northern hemisphere will get decidedly cool. In the SH it is the (slightly cooler) ocean that will keep us warm.

Can’t tell a good story without a few words.

344. Erl Happ
Posted May 18, 2008 at 8:23 AM | Permalink

This accounts in part for the decline in UAH temperature anomalies in the lower troposphere since 1995.
Should be 2005.

345. kim
Posted May 18, 2008 at 8:27 AM | Permalink

343 (Erl) You are welcome. How about ‘A graph is worth a thousand numbers’?
=================================================

346. Posted May 18, 2008 at 8:54 AM | Permalink

343 (Erl):

I believe that the microwave wavelength 2800 MHz daily radio flux correlates highly with the daily sunspot number and the two databases are used interchangeably.

This has been known for fifty years, with the all important caveat that it is only true for the flux adjusted to 1 AU, which is NOT the flux received by the Earth.

Salient points to note:
1. Temperature anomalies in the TLT fairly closely follow changes in microwave flux with a fairly consistent lag of two months.
2. Short term changes in irradiance frequently exceed the 7% that is due to orbital influences.

1: the flux received by the Earth is not the adjusted to 1 AU flux.
2: you have to be more precise. Irradiance is not the radio flux. And if you mean irradiance, then the statement is false.

When ionospheric researchers calculate the impact on the thermosphere of solar radiation and the drag on satellites they, correctly, use the true flux at Earth [not the 1 AU adjusted flux] and so should you.

347. Erl Happ
Posted May 18, 2008 at 9:26 AM | Permalink

346 (Leif)
I deliberately chose the 1 AU flux because I wanted to see the effect due to the sun itself not the orbital/seasonal effect. Yes, the orbital effect will go some way to cancelling the cooling effect of a fall in Microwave flux in SH summer but the declines in flux are relatively greater at that time.

The other thing is that the temperature anomaly is just that. It has a discount for the orbital effect already built in.

Sorry about the confusion of radio flux with irradiance. Nearly but not the same. Neither is sunspot activity the same thing despite my quote from Penticton. And would not the ionosphere react more to very short wave radiation which is a different thing again and also to the solar wind. And that is no doubt part of the reason for the slightly sloppy fit.

Another mistake. I forgot to retard the periods of negative temperature anomaly when I marked the Hovmoller diagrams.

But, you seem to be happy with the notion that microwaves heat water!

348. Posted May 18, 2008 at 9:32 AM | Permalink

302 (Erl):

All your queries relate to my markings at the altitude 5-10 Km.

That does not mean that there are not problems with the others. Simply that I started from the bottom.

The light buff coloured zone that lies between 5 and 10km has a temperature between 0 and 4°C. It is warmer than the light green zone that lies above and below the light buff coloured zone. The latter has a temperature between 0 and minus 4°C.

I have inserted the word buff in the quote above to reflect how I read the [unclear] text.

I have observed that the light buff coloured zone expands in vertical extent when the tropical lower troposphere (UAH data) is warming. My markings record a vertical expansion (W) or contraction (C ) of that warmer light buff coloured zone.

Again I have inserted the bold words to indicate how I read the text.
The two green/blue circles around the W and C markings have the same extent of the light buff coloured zone, so why do they have different markings?

349. Erl Happ
Posted May 18, 2008 at 9:49 AM | Permalink

348 (Leif)
Confused. In relation to # 302 are you referring to my squares or your circles?

Is it worth while carrying on with a discussion of the 2007 data where trends are so slight as to appear insignificant? Can we look at 2008, or 1997-2000 or to 2005 onwards?

350. Posted May 18, 2008 at 9:49 AM | Permalink

347 (Erl):

But, you seem to be happy with the notion that microwaves heat water!

Nonsense in this connection. The f10.7 cm radio flux has an energy that is 10,000,000,000,000,000,000 times smaller than that of green light.

351. Erl Happ
Posted May 18, 2008 at 9:58 AM | Permalink

350 (Leif)
Very specific. But there is quite a range of frequencies that heat water.

352. Posted May 18, 2008 at 10:25 AM | Permalink

351 (Erl):

But there is quite a range of frequencies that heat water.

True, but the microwaves used to heat food in microwave ovens have a wavelength of 12.2 cm which is optimal [and close to 10.7 cm] for heating water. And the heating is less at other frequencies. If you integrate over a large range of microwaves you might knock 3 to 5 zeroes off the large number I gave, but there are enough to take from. The solar radio waves do not heat the water in the Earth’s atmosphere. Period.

353. Posted May 18, 2008 at 10:30 AM | Permalink

346 (Leif):

When ionospheric researchers calculate the impact on the thermosphere of solar radiation and the drag on satellites they, correctly, use the true flux at Earth [not the 1 AU adjusted flux] and so should you.

And just to clarify that point: it is not the radio flux that heats the thermosphere, but the radio flux is just a proxy for the UV that does.

354. See - owe to Rich
Posted May 18, 2008 at 1:19 PM | Permalink

199 (Basil)

I found your posting most interesting because of my article on the BB about temperature variations within the Hale cycle. Had you seen this?

Anyway, you will see that I had been asking what other information there might be on the Hale cycle and climate, so your reference to “a Gnevyshev-Ohl rule” is fascinating, especially since it appears I may have rediscovered it or something like it. But could you give a more specific reference, please?

I have a couple of questions about your graph. First, you talk about cycles of lengths 258 and 111.6 months, but then say that a best fit is with 248 and 110 months. Which cycle lengths were used in the graph?

Second, whereas the lunar cycle is fixed, so only the phase (and not the period) should be deduced from the data, the Hale cycle is not constant. Have you considered fitting it separately from the minimum of each cycle, so as to allow for differing periods, to see whether this helps or hinders the alignment of your function with the HadCRUT3 data?

Anyway, the lunar precession connection is new and interesting to me – it affects oceanic tides and so (presumably) atmospheric tides. But I should like to see a proper analysis of variance applied to the inclusion/exclusion of this effect.

Rich.

355. pochas
Posted May 18, 2008 at 2:03 PM | Permalink

Leif,

Can you point me to a data file of sunspot numbers adjusted to your specification?

Thanks,

356. Posted May 18, 2008 at 2:19 PM | Permalink

355 (pochas): Are yearly values good?

357. pochas
Posted May 18, 2008 at 2:35 PM | Permalink

356 (Lief):

Yes, thanks again.

358. Basil
Posted May 18, 2008 at 3:52 PM | Permalink

#354 Rich,

I had not seen your analysis on the BB. But having looked at it quickly, if I understand it, then you are showing something similar to what we see in my graph. But just to be sure you understand what you are looking at when you looking at my graph, I’m plotting the rate of change in temperature, not temperature itself. It might be interesting to try to replicate your analytical procedure on my data, especially since I have the data in monthly format, which should allow for greater precision. I think you are using annual data, right?

As for the Gnevyshev-Ohl “rule” it is simply the observation that the sum of the number of sunspots for a odd numbered solar cycle tends to exceed the number of sunspots for the even cycle that proceeded it. Goggle it for more information.

In the graph, the sinusoidal fit was done with 258 and 111.3. Visually, the difference between that fit, and the least squares fit would not be apparent unless you view both at the same time. The difference in R-squared between the two is about 0.03, if I recall. E.g., the R-squared for the model shown in the graph is 0.33, and with the best fit it is 0.36. FYI, that’s just with two parameters. I can get the R-squared up to 0.7, though after 4 parameters I think the additional cycles are pretty meaningless. The other two periods that I think meaningful in some way are ~180 months (15 years) ~655 months (54.5 years). These four take the R-squared up to about 0.6.

I really don’t see the utility of trying to improve the fit by adjusting it to the actual length of individual Hale cycles. I think any improved fit will quickly be overtaken by the loss in degrees of freedom. And my thinking here is influenced by a lot of analysis that is not presented in the above graph or accompanying posts, such as MTM spectrum analysis of the data.

Obviously, I don’t share Leif’s skepticism of a role for lunar cycles in climatic oscillations.

359. Posted May 18, 2008 at 4:39 PM | Permalink

357 (pochas): You can find my series [since 1841] at http://www.leif.org/research/Corrected%20Sunspot%20Series%201841-Present.xls
Column E is what you want. Here is a figure constructed from the data:

The bottom panel shows the official SSN [black] and my reconstruction [red].
The IMF strength [black curve, upper panel] is the interplanetary magnetic field strength estimated from the sunspot number series.
Warning: this has still to go though peer-review [which will be an uphill struggle, because nobody wants their boat rocked too much].

360. Posted May 18, 2008 at 4:50 PM | Permalink

349 (Erl):

Confused. In relation to # 302 are you referring to my squares or your circles?

My green/blue circles each of which has one of your squares in it. So, I’m referring to your squares, of course.

Is it worth while carrying on with a discussion of the 2007 data where trends are so slight as to appear insignificant? Can we look at 2008, or 1997-2000 or to 2005 onwards?

The 2007 was the one you selected to show. And, yes, either we continue with EVERY single square on 2007 [in due time, one or two or three at a time] or you tell me that they are ALL not worth looking at [which begs the question why you posted that Figure in the first place] and you select another year.

361. Posted May 18, 2008 at 5:12 PM | Permalink

358 (Basil): Even-odd rule: If you look at the graph in 359, you will note the alternations as follows, where the first letter D or U says what the expected change should be, Down for odd-even, Up from even-odd; and the second letter shows what happened:
10-11 U U ok
11-12 D D ok
12-13 U U ok
13-14 D D ok
14-15 U U ok
16-17 U U ok
18-19 U U ok
20-21 U U ok
21-22 D D ok
thus 9 ok and 5 bad. This you can easily get by pure chance. If NASA is correct and the next cycle is big, you will have 9 ok and 6 bad. If I’m correct, it will be 10 ok and 5 bad. this is too flimsy to build anything on.
Going further back in time the record becomes unreliable.

362. Posted May 18, 2008 at 6:16 PM | Permalink

361 (Basil): And I apologize for getting 19-20 wrong:
19-20 D D ok
The point still stands: it could easily be a fluke, especially since we have no mechanism that would do this. Often the two hemispheres show quite different activity and the transitions 21-22 is different between hemispheres:
North: D D ok, but South: D U bad. But I also realize that it is not for me to tell you what to believe. I just tell you why I don’t believe in the even-odd rule.

363. Basil
Posted May 18, 2008 at 8:14 PM | Permalink

Leif,

I’m reluctant to question someone who’s far more the authority in these matters than I am, but are you certain that you are applying the “rule” correctly? I think you should only be counting Hale cycles, i.e. pairs that end in odd numbered cycles. Doing it that way:

22-23 exception
20-21 ok
18-19 ok
16-17 ok
14-15 ok
12-13 ok
10-11 ok

Assuming that is the correct way to do the count, now what are the chances that it is coincidence?

As for “no mechanism,” there seems to be a clear difference in cosmic ray behavior between even and odd numbered solar cycles. Would you care to comment on Mavromichalaki, et al., Hale Cycle Effects in Cosmic-Ray Intensity During the Last Four Cycles?

Basil

364. Erl Happ
Posted May 18, 2008 at 8:47 PM | Permalink

360 (Leif)

The 2007 was the one you selected to show.

Let’s get this in perspective. I have shown 1997-1999 where the loss of the warm anomaly during the period of negative TLT anomalies at 5-10km altitude is very plain. The data for that period is at Svalgaard 5/456. At #103 I show 2008 where it disappears entirely as it did in 1999 at the commencement of the La Nina. I have shown the graph that directly links 10.7 radio flux to TLT temperatures in #343. I have shown the horizontal extent of the temperature anomalies at 250mb in #343 and demonstrated that they are clearly related to solar influences rather than heat returning from the surface of the Earth and also to areas with warm ocean and high atmospheric moisture content.

In #103 I acknowledged your query in relation to 2007: The progression towards extremes of cooling in sea surface temperatures and very low sunspot activity that has occurred in 2008 has created a picture that is relatively easy to read. That was not the case in 2007 when the tropics slowly transitioned from mild warming to mild cooling over nine months, but the signal is still there, if less readily apparent.

The vertiocal depth of the anomaly reduces from left to right.

If you want to query every marking on that graph we should do it over the phone.

You choose to challenge my interpretation of the 2007 data and disregard the rest. I get the distinct feeling that my seed is falling on barren ground. Are you interested in a solution or is it just the debate?

We need to look at the historical sweep of data and not get caught up in minutae.

Thanks anyway for being dutifully present through all the ups and downs and setting my right when I patently needed it. Just occasionally it would be good to hear you say: There might be something in this, Perhaps if we did this or that etc

At the moment you are looking like Leif the diehard conservative. To be helpful we should be looking at the energy that can interact with water vapour and trying to quantify all sources and relate it to the quantity of water vapour and air at the 9km level to see whether a 4K rise in temperature is conceivable. Microwave, infrared, visible, whatever! Patently, the anomaly is a fact of life. It will not go away. It is very likely related to cloud cover dynamics. Radiation is plainly related to temperature dynamics in the lower troposphere. It would be good to relate it to temperature at vthe 250mb pressure level too.

Last step is to relate radiation dynamics and or 250mb (9km) temperature anomalies to cloud cover but there is some interesting stuff to look at in the interim.

365. Posted May 18, 2008 at 9:22 PM | Permalink

363 (Basil):

I think you should only be counting Hale cycles, i.e. pairs that end in odd numbered cycles

The Sun doesn’t know about our counting convention. The situation should be symmetrical, counting from odd to even, i.e.
11-12 D D ok
13-14 D D ok
19-20 D D ok
21-22 D D ok
Now with this you could argue that the ‘bad’ are as numerous as the ‘ok’ so one has a significant ‘inverse’ Hale cycle. I hope you can see that this is just numerology [although I would put money on the table betting that you cannot/will not]. Part of the reason for the different behavior for even-odd, odd-even transitions is that there are longer period [~100 year] ‘cycles’ in the sunspot series. This gives you a couple of ‘free’ even-odd hits [16-17, 18-19] where you chose to interpret the rising trend as an even-odd effect. You should detrend the series first, which will have the effect of removing almost all of the even-odd effect for those cycles..
Anyway, the cosmic ray difference [and there is one] is such that the cosmic ray ‘cycles’ have alternately ‘peaked’ or ”rounded’ maxima. but these cosmic ray maxima occur at solar minimum, so the cosmic ray ‘peaked’/’rounded’ difference goes from sunspot maximum to sunspot maximum and are 1890 degrees out of phase with the solar cycle. The paper you cite, is confused about this.
This has been discussed many times here. ‘Kitten’ kim can relate to that.

366. Posted May 18, 2008 at 9:49 PM | Permalink

364 (Erl): I have tried to impress on you that an important part of the scientific method is ‘the crucial experiment’. Now, in Earth Systems and astonomy we cannot do any experiments, so we must be contents with that ‘crucial example’. In Geology, the most famous example is ‘Siccar Point’ [Google it]. There are thousands and thousands of rock formations all telling a jumbled and confused story that can be interpreted many ways [that’s your ‘perspective’ – the totality of the evidence], but Siccar Point was the clear simple example that cut through the fog [and your missives are just like that – an impenetrable fog]. So we need to find your Siccar Point. Select the single best example sequence of events that shows what you are after, and let us discuss that. As I have said before, the onus is on you to do that. Most people [including me] are not inclined to look through the mass of graphs and data that you present.
About being conservative: that is part of science too. We know, from experience, that of a 100 ideas or theories or explanations, 99 are just wrong, so we have learned to be ultra-conservative. Even to the point that we will cling to an idea or theory that we know is wrong if there is not another one with a better of more compelling mechanism. Wrong explanations then serve simply as mnemonics that help us remember what the phenomena are about.
So, where is your Siccar Point?

367. Posted May 18, 2008 at 10:00 PM | Permalink

364 (Erl):

air at the 9km level to see whether a 4K rise in temperature is conceivable.

Is this your Siccar Point? A 4K rise in temperature? Rise above what? When? And why is that something that need an explanation other than the ordinary ‘head-in-the-sand’ standard textbook explanation? the next question us then: if that 4K rise is not there, does all the rest then simply fall away? Is the 4K a sine qua non?

368. Posted May 18, 2008 at 10:04 PM | Permalink

365 (Leif): for the nit-pickers: 1890 degrees should be 180 degrees.

369. Basil
Posted May 18, 2008 at 10:38 PM | Permalink

#365 “[although I would put money on the table betting that you cannot/will not]”

Leif,

You’ve made at least two or three demeaning remarks like this in our exchange. I’m able to keep a more open mind than you seem to think, and see the point of your response to the G-O “rule.” Other than my belief (if I can have one without snide remarks) that there is evidence of bidecadal variations in climate that appear related to similar periodicity in solar activity — a belief you do not share — have I actually given you any reason to lace your remarks with personal comments like this?

On the inverse correlation between cosmic ray flux and sunspot activity, I’m aware of that.

Basil

370. Posted May 18, 2008 at 10:52 PM | Permalink

369 (Basil): This was not meant as a ‘snide’ remark, just that some people that I know also have such strong feelings about this that is very hard to rock. My good friend and colleague Ed Cliver is also an even-odd guy, and nothing I have tried can shake him off that. So, if it came across as demeaning, I apologize for that; that was not my intention.

The correlation between cosmic ray flux and sunspot activity is inverse as you point out, but that was not the issue. The issue is that the cosmic ray even-odd difference is 180 degrees out of phase with the solar cycle and therefore has nothing to do with the Hale cycle as such. So you cannot invoke the cosmic rays as something that is different between even and odd cycles.

371. Posted May 18, 2008 at 11:08 PM | Permalink

369 (Basil): Here are some other even-odd proponents: http://cc.oulu.fi/~usoskin/personal/SolPhys_22_2001.pdf so you are in good company.

372. Posted May 18, 2008 at 11:23 PM | Permalink

363 (Basil): My reluctance to ascribe anything to the Hale-cycle and the G-O rule is that these are second-order effects. We can hardly detect if there is the first-order effect of the 11-year cycle, which makes the second-order effect even more unlikely. In geomagnetism, ionospheric research, aurorae, etc, where the 11-year cycle is so strong, there is no accepted trace of the 22-year Hale cycle [nor of the 19-year lunar cycle], and this is my main reason for not ascribing any bidecadal cycles [which may be there] in climate to the sun. And you can put money on the table that I won’t change my mind on that :-) unless, of course, someone shows me a mechanism.

373. Erl Happ
Posted May 19, 2008 at 1:21 AM | Permalink

367 (Leif)

Is the 4K a sine qua non

From Wikipedia It refers to an indispensable and essential action, condition, or ingredient.

It is not the magnitude of the anomaly that is important even though it exceeds that in the atmosphere beneath the 5Km level. It is the spread of the temperature anomaly in the vertical and horizontal dimension that is telling. The extent corresponds with the likely distribution of moisture at the 250mb level providing a dependent variable that governs cloud dynamics. The lack of any cloud but cirrus in the tropical atmosphere and the susceptibility of that cloud to warming is also important. The water holding capacity of the atmosphere is so small at the 9Km level that small changes in incident energy should be influential. So, here is a plausible mechanism for albedo change. Changing the volume of cloud in the lower troposphere would demand much more energy. That energy is available as OLR in NH summer but is not available from incoming solar energy. The absence of similar phenomena below and above the 250mb level suggests a solar driver. The absence of a signature in the lower levels and in the stratosphere rules out OLR.

But perhaps all of this is wrong and I am missing the real mechanism entirely.

Whatever the mechanism is, we need to account for relationship between this anomaly at the 9Km level and temperatures in the TLT.

And there is a relationship between radio flux and temperature in the TLT that demands elucidation.

If you want to contest the strength of these relationships or develop a better theory to account for the phenomena go right ahead.

As for solar cycle influence on the atmosphere ‘temperature’ is just one of the many dependent factors. Perhaps before you get too excited about contesting this one, wait for an exposition of the rest.

374. Erl Happ
Posted May 19, 2008 at 1:30 AM | Permalink

367 (Leif)
Geology was also influenced by the charismatic German mineralogist Abraham Werner, who contended that most rocks were sediments deposited during the biblical Flood, and who dismissed active processes like river erosion and volcanism as recent and insignificant phenomena.

In 1785, before the Royal Society of Edinburgh, he (Hutton) presented his principle of uniformity: the idea that the processes we observe shaping Earth today are the same as those that shaped it in the past.

Siccar Point, exhibiting new sedimentary rocks overlying very much older sedimentary rocks that were tilted to the vertical illustrated violent upheaval and subsequent alteration that would have taken a great length of time…….rather longer than since the biblical flood.

The sun puts out enough radiation to fry us all. That it does not fry us proves the importance of atmospheric processes in shielding us from its rampant energy. Look up, and the cloud cover is forever changing. In the afternoon and in the summer time the clouds disappear. It is the interaction of solar energy and water vapour that controls temperature in the troposphere together with humidity and rainfall.

That’s my Siccar Point.

375. Basil
Posted May 19, 2008 at 8:01 AM | Permalink

#370

Leif,

Help me understand your point better about cosmic rays being 180 degrees out of phase with the Hale cycle. I think I’ve been confusing the Hale cycle with the reversal of polarity that occurs at the time of solar maximum. There is also a ~22 year cycle in the latter, which captures the two types of cosmic ray flux patterns, and which occur at the time of solar maxima, right? Is there a “name” or descriptive phrase or expression to refer to that 22 year period, to distinguish it from the Hale cycle?

I think I can better understand, now, your “negativity” about relating bidecadal variations in climate to the Hale cycle. In my novice attempt to understand the solar physics of all this, what I’ve been thinking of as a possible sun-climate connection is to the 22 year variation we can see in the pattern of cosmic ray flux. Now don’t assume that I’m jumping on the Svensmark-Marsh bandwagon here. At first, I thought maybe there was some connection to their theory in the temperature pattern I’m seeing, but not any more. If I had to choose a connection between a cosmic ray-cloud interactions, and the temperature data I’m seeing, it would be to cirrus cloud cover. But I know there is too little data to make that connection. But I still think that there may be something to the 22 year cycle of variation in cosmic ray flux.

Or it could be interstellar duest. :)

And accept my apologies for misreading your intent. As we both know well, I’m sure, this is a medium that lacks some of the cues that we’d have in FTF encounters.

Basil

376. kim
Posted May 19, 2008 at 8:02 AM | Permalink

372 (Leif) et al. Out of phase works if the sun’s effect on climate is expressed through a diminution of effective insolation, perhaps via CR and clouds, rather than the magnification of intensification we’ve all been searching for.
=============================

377. Posted May 19, 2008 at 9:43 AM | Permalink

375 (Basil):
Look at this first:

This is the cosmic ray flux over the last several solar cycles [for several stations]. Two things to note:
1: cosmic ray flux minima at solar maxima [see green solar cycle curve]. This is the ‘inverse’ correlation you mentioned.
2: but, more importantly, note that there is an alternation is the shape of the cosmic ray flux variation, with ‘peaked’ shape alternating with ’rounded’ or ‘flat’ shapes at minimum. This is the 22-year cosmic ray variation. Its cause is well understood and is related to the Sun’s polar field, the sign of which determines how CRs ‘drift’ through the solar system. Note that it goes from solar max to solar max [actually polar field reversal to next polar field reversal]. There is a ‘name’ for this [lousy though] namely ‘qA<0′ and ‘qA>0′ respectively for the two shapes. We are right now in a ‘peaked’ [qA<0] cycle. [be careful with the “<” and “>” symbols; you have to enter them as “&lt;” and “&gt;”].
But, as I said, the 11-year inverse flux variation is the main variation, the 22-year shape changes are but a minor second-order effect. I know of no phenomenon [other than direct observation of CRs] where this second-order effect is detectable.
Now, there are [second-order] 22-year cycles in geomagnetic activity, aurorae, etc, which are not caused by the 22-year variation of CRs, but which are also related to the reversing solar polar fields. Section 9 [page 53ff] of a very old paper of mine explains how that works: http://www.leif.org/research/Geomagnetic%20Activity:%20Dependence%20on%20Solar%20Wind%20Parameters.pdf
This effect is so small [and second-order] that as long as no compelling relation is found between the first-order geomagnetic/aurora activity, we shouldn’t invoke the second-order effects. [This will not stop some people from doing it anyway – or from using ‘third-order’ effects like the difference from year to year of the second-order effects and the like].

378. pochas
Posted May 19, 2008 at 10:21 AM | Permalink

375 (Basil):

I am sure that Leif would not want us all to be sincophants of his. That wouldn’t be any fun, so just to keep the pot boiling there have been charts posted on this thread showing that alternate solar minima have pointy-topped CR count profiles alternating with flat-topped ones, so there is a difference. See Fig 3 of Usoskin’s review which illustrates this:

http://cc.oulu.fi/~usoskin/personal/CRAS2A_2712.pdf

The climatic effect seems to be evident mostly in the mid-latitudes, as this South-African writes:

http://nzclimatescience.net/images/PDFs/alexander2707.pdf

I think Leif is a little too strong in his no-effect-from-solar stance.

379. Posted May 19, 2008 at 11:45 AM | Permalink

378 (pochas): When I see papers cited I always check if they pass my ‘crank filter’ and my ‘sloppiness filter’ and my ‘solid analysis filter’. These checks are usually quick [seconds]. Usoskin’s is a review so no new analysis is presented. We can discuss some of the papers he cites, but let me first dispose of the South-African paper.
Here is their Figure 10:

The authors are ‘barycenter people’ and calculate TSI received from the Sun which is supposedly wobbling around the center of mass [SSCM – the cross] of the Solar System. They calculate four values [perihelium, aphelium, two equinoxes – interesting enough these Southern Hemisphere people call the March equinox the ‘Spring’ equinox]. Here I show the actual observed TSI. I have plotted their values as the red dots:

Since they don’t match up and don’t make sense, I can now dismiss the paper, which I hereby do. It does not pass muster.

380. Posted May 19, 2008 at 11:55 AM | Permalink

379 (Leif): Somehow the Figure caption didn’t make it. Here it is; the data is for 1993:

Figure 10 Received Wattages per square metre for the year 1993, based on the calculated 0.63×10^6 km solar displacement from the SSCM

381. Sam Urbinto
Posted May 19, 2008 at 12:34 PM | Permalink

Indeed, the sun being stronger or not shouldn’t be an issue, because the planet can control what percentage of whatever energy getting here gets in.

If that percentage was 85% what would happen if it changed to 84% or 86% even if the same amount and strength of sunlight was being received, that’s the question.

382. pochas
Posted May 19, 2008 at 1:24 PM | Permalink

379 (Leif):

I didn’t cite the Alexander paper because of the barycentric business, which isn’t Alexander’s work (and which is an unwelcome topic here anyway per Steve M), but because of the rainfall patterns he observes and correlates to the Hale cycle. So yes, let’s leave the barycentric stuff aside.

383. Jim Arndt
Posted May 19, 2008 at 2:10 PM | Permalink

375 (Basil),

This paper by the USGS surmises a correlation between solar and the Mississippi River flow. Very interesting because it sights a 22 year cycle also.

384. Posted May 19, 2008 at 2:47 PM | Permalink

381 (Sam): you lost me completely. I have no idea what you are tlaking about; must be deep…
382 (pochars): I do not cherry pick from a paper what I like. If there is any part of the paper that is wacky, then the rest is discarded too, as far as I’m concerned.

385. Posted May 19, 2008 at 2:55 PM | Permalink

383 (JimA): there is a strong selection effect. Negative results often do not get published. If you try to submit a paper that “shows that the Amazon river does not have a 22-year period, nor a 14-year period, and certainly not a 31-year period, not to speak about the non-existent 57-year period”, you’ll find it hard going getting through peer review; the editor would likely reject the paper before even sending it out for review; or you yourself might not want to go to the trouble [after all you have to pay publication cost, etc]. Now, 11 or 22 years are respectable numbers, so you, the editor, and the reviewer might look at it a bit more favorably…

386. Posted May 19, 2008 at 3:13 PM | Permalink

382 (pochas):

I didn’t cite the Alexander paper because of the barycentric business, which isn’t Alexander’s work

Yes, it is used by him, and his Figure 10 disqualifies the whole pare, IMHO. It is like if he had said “We’ll start with the fact that 2+2=5 and see where that takes us”.
From the paper:
(“Earth to sun chord distance: As a result of the sun’s wobble, the chord length between the earth and the sun and the amount of energy received by the earth will change accordingly. The next exercise is therefore to determine the corresponding changes in the distance between the earth and the sun and thereby the changes in the rate of solar energy reaching the earth. This is amenable to precise calculation.”
And he then gets Figure 10. Enough to qualify for the round file cabinet.

387. Basil
Posted May 19, 2008 at 3:51 PM | Permalink

#378

pochas,

I don’t think Leif will imagine that I’m a syncophant of his any time soon. But I have my own problems with the supposed cloud-cosmic ray connection shown in Figure 3 of the article you linked. The plot of cloud cover does track the cr flux, but it doesn’t track with temperature. The plot is shown as rising in the latter half of the last decade, and then falling into this decade. Since the supposed connection is to low cloud cover, the rise in cloud cover should have been reflected in cooling then, while the decline of cloud cover into this decade should be causing warming now. But general trends in temperature are just the opposite.

That’s why I mentioned cirrus clouds in one of my posts above. That would make more sense of the general pattern of cr flux and recent temperature trend: if cr flux was rising, and seeding cirrus clouds then temperature would rise with the rise in cr flux, and fall with the decline in cr flux. But my understanding is that we really do not have good enough cirrus cloud cover data to test that hypothesis.

388. Posted May 19, 2008 at 4:01 PM | Permalink

378,387: συκοφάντης please, if you must be fancy.

389. Sam Urbinto
Posted May 19, 2008 at 4:53 PM | Permalink

Leif #384 Deep or too simple. :) I meant if the atmosphere et al are letting in 85% of the sunlight reaching the planet, even if the strength of the sun and everything stays the same, it would be more energy if that let in 86% or less if 84%. Meaning you don’t have to explain a greater or lesser effect of the sun using the sun itself as the explanation.

390. Posted May 19, 2008 at 5:23 PM | Permalink

389 (Sam): I’m still lost. Maybe it is because you didn’t post in 381 a reference number to what the topic was.

391. pochas
Posted May 19, 2008 at 6:55 PM | Permalink

387 (Basil):

The El Nino of 1998 occurred in this interval. It could well have scrubbed out any cooling from the cycle 22/23 minimum, and then with the upswing of cycle 23 the warming resumed. Or, if you prefer, the AGW resumed. :-)

It might be impossible to tie CR directly to temperature during this time period, so close to a major ENSO event, or to any other time period for that matter. But one can make the connection to low-level clouds, which theoretically should be a cooling influence which would contribute to the overall schema. Usoskin does not oversell the CR-cloud connection.

392. Jim Arndt
Posted May 19, 2008 at 7:17 PM | Permalink

385 Leif);

I’m sorry I miss the point??

393. Jim Arndt
Posted May 19, 2008 at 7:22 PM | Permalink

Leif,

EIT and MDI are showing most of the activity is on the middle latitudes of the sun. Do you think we are in for a prolonged SC23 at this point or the jury is still out?

394. Posted May 19, 2008 at 8:06 PM | Permalink

392 (JimA): The point is that since papers that proclaim a 22-year period in a certain locality of a certain phenomenon has a better chance of being published than a paper that claims that there is no effect. Thus we amass many ‘positive’ papers and few ‘negative’ ones. So the fact that one can pull out many papers claiming some effect does not mean that there are places that do not show any effect, as these are not publicized as much.

393 (JimA): That there still are low and mid-latitude SC23 regions is not unusual. Many solar cycles show a similar behavior. SC23 will hang in there another year. The big question is when SC24 really takes off. That could be tomorrow or a year from now. Nobody knows. See also #175.

395. Basil
Posted May 19, 2008 at 9:48 PM | Permalink

#394 (Leif)

I’m not so sure there is as much of a selection bias as you contend. If a specific claim has been made, and someone challenges that specific claim, that has a good chance of getting published.

Example: Elsner and Tsonis (1991), “Do bidecadal oscillations exist in the global temperature record.” From the abstract:

…Recently Ghil and Vautgard … identified a secular warming trend and a small number of oscillatory modes. … Our results offer no support for the presence of bidecadal oscillations in the global surface temperature records.

This happens regularly.

What I think is interesting, though, is how the claims of periodicities tend to cluster around a few specific numbers. Thus we have lots of reports of 20 or 22 year periodicities, compared to, say, 16 or 28 years. There is probably a variety of reasons that can be advanced for why studies might turn up showing no periodicity. But if the periodicities that do turn up were just happenstance, then they would not cluster they way they do around a few specific periods.

I don’t doubt that sometimes the evidence for a claimed periodicity is weak or strained. But sometimes the evidence is pretty strong. The 20-21 year period in HadCRUT is strong enough to get classified as a harmonic in MTM spectrum analysis. The evidence for a shorter cycle of 9-10 years is weaker, and more variable, but still stands out from background noise as “statistically significant.” One might reasonably question what this “proves” (i.e., what causes it, or what good it does us to know this), but it is not reasonable to dismiss evidence of this nature because one thinks evidence to the contrary is being ignored, or because the evidence that was presented on some other occasion involving something similar was unconvincing or inconclusive. Every claim deserves to judged on the basis of its own merits, and not on the basis of such broad generalizations as I think I see in some of your comments.

396. Sam Urbinto
Posted May 20, 2008 at 9:20 AM | Permalink

Leif 390 etc I was probably mainly commenting on kim’s 376 about effects stemming from a change in effective insolation rather than a change in intensification. So 381 was mostly a generic comment in regards to some of the earlier discussion of what effects the sun has on the anomaly the weather climate. I was just mentioning the obvious, that the sun doesn’t have to be doing anything special in and of itself for it to be having a greater or lesser effect upon ‘the spheres’ (atmosphere, lithosphere, cryosphere, hydrospere, biosphere).

397. Posted May 20, 2008 at 9:24 AM | Permalink

396 (Sam): Ah, I see, a generic comment to a generic post stating the obvious in a generic way…

398. Posted May 20, 2008 at 10:30 AM | Permalink

http://www.redorbit.com/news/space/1391817/new_solar_cycle_not_packing_much_punch/index.html#

399. Pete
Posted May 20, 2008 at 12:38 PM | Permalink

398(Leif): Is this the paper that Science would not publish?

2006 Penn and Livingston

400. Posted May 20, 2008 at 12:47 PM | Permalink

399 (Pete): Yes and No. The Science paper was shorter [less detail] but had the extra speculation that we may be in for a Maunder Minimum. With cooling, and that is against the science that ‘is settled’ and so requires more proof [like the Potomac freezing over in July]. The detailed paper you refer to was accepted by Astrophysical Journal (Letters) and is solid work. I have his data up through March 2008, confirming the trend. It may be that the spots were are seeing for SC24 are so small and weak simply because Livingston is correct. That all spots will disappear in 2015 may be a bit of a stretch, but, hey, the Sun has done it before.

401. Sam Urbinto
Posted May 20, 2008 at 2:27 PM | Permalink

Leif #397 Right, generically speaking. :) Sometimes though we have discussions where everyone involved thinks they are thinking of the same obvious thing (but are not) and nobody bothers to clarify. I’m more making an observation that we can perhaps “blame the sun” but from not what it’s doing, but how the “Earth is reacting” to “business as usual”. An “effect-from-solar” that is on our side. Which seems more likely anyway rather than making up odd behaviors for the sun that (seemingly obviously) can’t be shown to any extent. Rather like assuming carbon dioxide is driving the anomaly. :D

402. lgl
Posted May 20, 2008 at 3:21 PM | Permalink

There’s a decadal variation in volcanic activity too.

403. Posted May 20, 2008 at 4:02 PM | Permalink

402 (lgl): and with appropriate smoothing and filtering I’m sure somebody can show that there is a lunar cycle or a Hale-cycle in the VEI as well.

404. Basil
Posted May 20, 2008 at 8:20 PM | Permalink

#402 Leif writes:

and with appropriate smoothing and filtering I’m sure somebody can show that there is a lunar cycle or a Hale-cycle in the VEI as well.

If that’s an aspersion to anything I’ve been posting, you may be letting your prejudice get in the way. Here’s an MTM spectrum analysis of the raw HadCRUT monthly temperature anomaly: No smoothing or filtering here, Leif. Or, if you prefer a more traditional periodogram:
Again, raw, unfiltered HadCRUT monthly anomalies. (The red lines are p-factor thresholds for .05 and .01 levels of confidence.)

Now my earlier charts were of first differences in the data after smoothing with Hodrick-Prescott filtering, but that is not creating cycles that are not there. The cycles are there, and are not an artifact of the smoothing process. If you want deny that they are solar or lunar in origin, that’s one thing. But don’t imagine that you can dismiss them as the artifacts of smoothing or filtering.

405. Posted May 20, 2008 at 9:23 PM | Permalink

404 (Basil): No prejudice. I do not say that there is no 20-year power in the data, just that I don’t think it has anything to do with the Sun. Now, as an example of the state-of-the-art I will note that ‘lgl’ claims in #402 that there is a ~10-year period in VEI, and since volcanoes clearly have an effect, there should also be a similar period in the temperature and, indeed your graph shows that there is, possible at the same [low] level of significance as in the VEI.
Your graph shows a much stronger 60-year period [probably related to the PDO]. Just a couple of weeks ago over Chez Tammy I also advocated a 60-year period and was lectured on that it was not statistically significant. What do you get if you apply your methods to the VEI-data? Show us. Then bandpass filter it, say to the band 15-25 years, “to bring out the power around 20 years”. And show what you get. Why would one do such a silly thing? Because I have seen scores of papers that do exactly that.
My problem is that there are lots of 5-60 year periods and cycles in the climate, and I am skeptical of ascribing any of them to the Sun without any plausible mechanism. Here is a typical example:

Evidence of Drought in Western Sicily during the Period 1565–1915 from Liturgical Offices
Journal Climatic Change
Issue Volume 49, Numbers 1-2 / April, 2001
DOI 10.1023/A:1010746612289
Pages 225-238
E. Piervitali and M. Colacino
Abstract Many recent climatological studies aim to reconstruct past climate on the basis of different data sources. In the present work, the drought events that occurred in western Sicily during the period 1565–1915 have been analysed, using historical information. Inside the cathedral of Erice, a small and ancient town in western Sicily,there is a marble plate reporting the dates of the processions held during that period. This information, together with other documents and manuscripts found in the city library, has allowed us to reconstruct the drought chronology. The comparison between these series and the quantitative data on precipitation indicates some agreement. The application of the operator ARP (Automatic Research for Periodicities) shows significant periods of 11, 17, 22, 32, 73, 90 years. Finally, the comparison between ENSO and drought chronology shows that in periods of many drought events a reduction of ENSO events occurred and vice versa.

These sorts of papers have been showing up for longer than any of us have been alive, here is another one:
http://docs.lib.noaa.gov/rescue/mwr/047/mwr-047-09-0637.pdf
with this abstract:
Volume 47, Issue 9 (September 1919)
Monthly Weather Review pp. 637–638
HISTORICAL DATA ON THE VARIATION OF RAINFALL IN CHILE
C. E. P. BROOKS
Meteorological Office, London
DOI: 10.1175/1520-0493(1919)472.0.CO;2
ABSTRACT
A book published by B. V. Mackenna in 1877 contains numerous researches into the municipal archives at Santiago. These are summarized, and shown to suggest two periodicities in rainfall: 1. A variable “Brückner period.” 2. A period of about 108 years. 3. In addition, the eightheenth century was drier than the nineteenth.

A Brückner period is ~36 years. [ = 2 lunar periods – if one is of that bent].

And on and on.

To re-iterate: There are many, many claims of many, many different periods and cycles in climate. All of them statistically significant [or claimed to be]. Some of these coincide with the 11-year, 22-year, and perhaps 100-year periods seen in solar parameters but most do not.

406. Posted May 20, 2008 at 10:05 PM | Permalink

The Bruckner period [~35 years] was popular a hundred years ago. Just searching Science Magazine yields these gems:

CURRENT NOTES ON METEOROLOGY AND CLIMATOLOGY R. Dec. Ward
Science 15 March 1907 25: 434-435 [DOI: 10.1126/science.25.637.434] (in Articles)
THE BRUCKNER PERIOD OF RAINFALL AT RIO DE…fall in line with the Bruckner period. An important difference…however, this: that Bruckner started on his quest…favor of any particular period, while the more recent……

SOLAR RECORDS IN TREE GROWTH A. E. Douglass
Science 4 March 1927 65: 220-221 [DOI: 10.1126/science.65.1679.220] (in Articles)
…well-known thirty-five-year Bruckner cycle. But the stronigest…receipt of the letter this period was immediately recognized…tree groups. The triple period, essentially Bruckner’s cycle, has operated…certain interfering periods, is more easily traced……

THE SIGNIFICANCE AND SCOPE OF THE IDEA OF FREQUENCY IN PHYSICS Wm. S. Franklin
Science 4 March 1927 65: 221-223 [DOI: 10.1126/science.65.1679.221] (in Articles)
…tracing the changes of the solar period. When these conditions were allowed…by our tree groups. The triple period, essentially Bruckner’s cycle, has operated in Arizona…absences of certain interfering periods, is more easily traced through……

SCIENTIFIC NOTES AND NEWS
Science 6 June 1902 15: 916-919 [DOI: 10.1126/science.15.388.916] (in Articles)
..considered seismic disturbances within that period. I find beyond question that the most…minimum. More than this, the 35-year solar period established by Dr. Lockyer, which corresponds approximately with Bruckner’s meteorological cycle, can also be……

Current Notes on Meteorology R. DeC. Ward
Science 27 February 1903 17: 353-354 [DOI: 10.1126/science.17.426.353] (in Articles)
….States, it may be recalled, have lately been studied by Bruckner, who finds that they correspond very well with his thirty-five-year climatic period. Mr. Murdoch makes no reference to Bruckner’s work along these lines.

SECTION A, MATHEMATICS AND ASTRONOMY Laenas Gifford Weld
Science 3 February 1905 21: 174-178 [DOI: 10.1126/science.21.527.174] (in Articles)
…..art extension of Professor Bruckner’s researches on the 35-year…elements and the results of Bruckner have been supplemented by…The probable value of the period length is found to be 36.2 years, instead of 34.8 years, Bruckner having used in calculating……

THE WEST INDIAN ERUPTIONS AND SOLAR ENERGY
Science 6 June 1902 15: 915-916 [DOI: 10.1126/science.15.388.915] (in Articles)
…..con-sidered seismic disturbances within that period. I find beyond question that the most…minimum. More than this, the 35-year solar period established by Dr. Lockyer, which corresponds approximately with Bruckner’s meteorological cycle, can also be……

Sex and Consumption G. W. HAMBLETON
Science 2 March 1888 ns-11: 108 [DOI: 10.1126/science.ns-11.265.108-b] (in Articles)
……interest, as it shows the alternating periods of humidityand dryness. It is partic-ularly…important in connection with Seibt’s and Bruckner’s studies of similar changes in the…northern hemisphere passed through a dry period between 1830 and I840. This was……

THE PHYSICAL SCIENCES AT THE BRITISH ASSOCIATION A. LAWRENCE ROTCH
Science 15 November 1901 14: 757-767 [DOI: 10.1126/science.14.359.757] (in Articles)
……Atmosphere and the Causes of Glacial Periods.’ ‘ The Systematic Exploration…evidence of a thirty-five-year period in the occurrence of sunspots…variations indicated by Professor Bruckner, of Berne, and with the frequency…open to all the world. The period of academic study should be……

LITHOLOGIC EVIDENCE OF CLIMATIC PULSATIONS C. E. Vail
Science 27 July 1917 46: 90-93 [DOI: 10.1126/science.46.1178.90] (in Articles)
……ably demonstrated by Bruckner, Huntington and others. The ” Bruckner cycle,” about thirty-five…observed an eleven-year period to be about the average…the tables. These periods correspond rather…number of years in the Bruckner cycle, as from the……

THE RELATION OF BIOLOGY TO PHYSICS Thomas H. Morgan
Science 4 March 1927 65: 213-220 [DOI: 10.1126/science.65.1679.213] (in Articles)
……is familiar with the literature of that period will doubt that it is essential to know…perhaps in the well-known thirty-five-year Bruckner cycle. But the strongest bit of evidence…the trees. On receipt of the letter this period was immediately recognized as the interval……

———————

Surprisingly, the 35-year Bruckner cycle is still with us, but now in the Sun:
AGU Fall meeting 2000:
SH21D-09: Operational Ap Forecast for the Solar Cycles: Status of Cycle 23 Activity
Ahluwalia, H S
Planetary index Ap was designed by Bartels to assess the geoeffectiveness of the solar corpuscular emission emanating from the M regions on the Sun. Quite surprisingly, Ap has found a new operational use as a precursor for forecasting the amplitude of the sunspot number cycle; due to new insights acquired very recently about its intrinsic characteristics (Ahluwalia, 2000). In particular, it has been shown that Ap index exhibits a three cycle quasi-periodicity. In addition, it shows a correspondence with the fine structure observed in imf intensity measured in situ at Earth’s orbit which in turn represents the temporal variations of the solar global field. A new method was proposed, based on the observed Ap minimum in 1997 for cycle 22 (Ahluwalia, 1998),to predict (correctly) that cycle 23 will be quite modest compared to cycle 22. A team of solar astronomers crticized our methodology in very strong terms; they were expecting cycle 23 to, “potentially be one of the greatest cycles in recent times, if not the greatest.” The attack was defended (Ahluwalia, 1999); we questioned the reliability of the little understood and unproven rule that odd cycles of the even-odd pairings were always observed to be more active. Our work supports the speculation that the solar polar “seed” field contributes to the toroidal field (within the frame work of Babcock’s self-excited dynamo) which is responsible for the activity in the new cycle (Ahluwalia, 2000). We comment on the several other misconceptions that limit the forecast of the cycle activity by the seasoned solar astronomers. We will summarize the current status of our prediction as of the end of November 2000.
Eos Trans. AGU, 81 (48), Fall Meet. Suppl., 2000

and on and on.

407. lgl
Posted May 21, 2008 at 12:11 AM | Permalink

#403 Leif,

Yes, you can clearly see the Saros/2 (9yrs) in Basil’s spectrum.

408. bender
Posted May 21, 2008 at 12:19 AM | Permalink

Leif, see if you can debunk the solar-porcupine link.

409. pliny
Posted May 21, 2008 at 1:20 AM | Permalink

#404
Just looking at Basil’s periodogram, you can’t expect a sharp 60 yr peak from only 2-3 cycles of data. There just isn’t the information to identify it. If you really had a 60 yr sine function, it would be multiplied by, say, a 120 yr gate function, and in the fourier domain, that means the peak is convolved with a sinc function (of form sin(x)/x). And, squared, that does look like a bit what has happened here. There’s a harmonic at about 30 yrs and one at 20 yrs, and a continuing progression in frequencies.

On that interpretation, if the 60 yr peak is real, the rest are likely artefacts.

410. kim
Posted May 21, 2008 at 3:01 AM | Permalink

396 (Sam) Generic it may be, but you got it. Insofar as the the effect of the sun on the climate, there is a lot more room for variability on the downside than the upside. With the narrow range of changes in the sun that Leif insists we acknowledge, the downside is probably where the mechanism is. And it seems clouds could well moderate whatever stimulus is solar related.
=====================================================

411. Posted May 21, 2008 at 3:49 AM | Permalink

Re kim 410

And it seems clouds could well moderate whatever stimulus is solar related.

If cloud modulation is the cause of warming, then the signal would be different from the expected CO2 warming signal. In one the atmosphere heats the ground, in the other vice-versa. If, as UKJohn said some time ago on a different thread, ‘it’s getting warmer because it’s getting sunnier’, there should be a different atmospheric temperature profile. Where is it?

JF

412. Erl Happ
Posted May 21, 2008 at 3:49 AM | Permalink

Leif

Re Kims

And it seems clouds could well moderate whatever stimulus is solar related.

And Sam’s

Indeed, the sun being stronger or not shouldn’t be an issue, because the planet can control what percentage of whatever energy getting here gets in.

If that percentage was 85% what would happen if it changed to 84% or 86% even if the same amount and strength of sunlight was being received, that’s the question

And my own in response to Leif’s ‘Siccar Point’ query

The sun puts out enough radiation to fry us all. That it does not fry us proves the importance of atmospheric processes in shielding us from its rampant energy. Look up, and the cloud cover is forever changing. In the afternoon and in the summer time the clouds disappear. It is the interaction of solar energy and water vapour that controls temperature in the troposphere together with humidity and rainfall.

And Leifs response to Sam

396 (Sam): Ah, I see, a generic comment to a generic post stating the obvious in a generic way…

Can we elucidate the obvious? Can we put some numbers on it? Can we show how the Earth’s atmosphere controls the amount of sunlight reaching the surface? Can we show how it adds heats to the oceans on the upswing of the solar cycle and takes heat away on the downswing? Can we show which places warm, which places get more humid, where the rain goes, which regions lose heat convectively and which ones lose it as OLR? And what stimulates the atmospheric thermostat to open and close? Can we show where is the thermostat located and what accounts for the changing opacity of the atmosphere to solar radiation?

Has Leif anything to offer in relation to these crucial questions?

413. Posted May 21, 2008 at 3:52 AM | Permalink

[delurk]
Basil may find the graph below of some interest

With due deference to #403 (Lief)”and with appropriate smoothing and filtering I’m sure somebody can show that there is a lunar cycle or a Hale-cycle in the VEI as well.”

Please excuse some blatant wiggle matching

The values of sunspots in green and the surface temperatures data of the GISS series in red are treated to fourier analysis and a low pass harmonic filter of a value which matches a 21.5 year cycle. Temperature is further detrended to ‘deHansenitise’ :) the data by an amount which roughly corresponds to a 0.4C warming over the C20th.
Clearly the ratios of wiggle heiht would be closer if Lief’s corrected sunspot number series were available (please suggest it to Paul Clerk at woodfortrees.org Lief) but what particularly interested me in the light of Basil’s posts and Erls comment that the PDO changed phase around 1970 was the apparent shift from the earth’s temp anomaly preceding the solar changes from around 1940 to the reverse situation around 1970.
I don’t have the facility with software or stats to extend this to an earlier period to see if the ‘phase reversal’ is periodic but wondered if it may be of sufficient interest to merit further examination by interested parties. Is there an interplay of a solar cycle of around 80 years (Gleissberg?) and a terrestrial cycle of around 60 years here?

I’m not a mathemetician or statistician, my qualification is in philosophy of science, so please disregard if I’m too far OT here.

[/delurk]

414. Posted May 21, 2008 at 4:19 AM | Permalink

408 (bender): porcupine link. I don’t think the animals react to any solar cycles. They may very well react to climate and weather, so we are led into the familiar circus: there are decadal, bi-decadal, tri-decadal, etc climate cycles and they are claimed to be controlled by solar cycles. As we have discussed, the 22-year solar differences are second-order effects, and the ‘sunspot’ and ‘polar’ 22-year cycles are out of phase, so, as the 11-year first-order effects are weak, the 22-year second-order effects will be weaker still. This is not so much a debunking as just an appeal to common sense [which as someone has remarked is not so common].

415. Posted May 21, 2008 at 5:07 AM | Permalink

412(Erl):

Has Leif anything to offer in relation to these crucial questions

I’m afraid not, if offer is to be taken positively. First, you have misunderstood my Siccar Point issue. It does not refer to the total view but rather to that singular, obvious observation that makes the whole thing immediately clear without further ado. And that you have not provided; on the contrary, you have appealed to increasingly complex and obscure [what you call ‘plain’] points, events, and relations.
Second, I can offer this perspective:

The graph shows the last 11 years TSI plotted every day as a function of the day within the year, so there are 11 overlapping curves. Had I plotted the last 400 years worth of TSI data, they would still fall within the width of the envelope of the black curves, i.e. you couldn’t tell the difference. Occasionally you can discern a tiny wiggle departing ever so slightly from the envelope, but any such is short-lived, tiny, and inconsequential to a climate system whose ‘time-constant’ may be of the order of 10 years [e.g. see lucia’s post at http://rankexploits.com/musings/2008/what-is-the-climate-time-constant-refining-the-estimate-i/ ].
Third, direct measurements of the albedo [including Palle’s earthshine data] may soon be able to shed light on some of these issues. We shall see. There is no doubt that the albedo, clouds, land-use changes, all that, are important. My ‘relentless’ resistance to dumping it all on the Sun may at times come across as ‘reactionary’, and I can only offer an anecdote in my defense. In my native language the acronym FIR means ‘Error In Accounting’. By including a FIR column, you can effortless balance any account. I don’t like to have the Sun be that FIR column, as that hides the problems of climate change rather than solving them. Now, the neutrino was once such a FIR particle, and yet it actually exists, so …

416. Posted May 21, 2008 at 5:22 AM | Permalink

Yet some more cycles [incl. 22-years]:
Trends and cycles in long Irish meteorological series
C.J. Butler, A. Garcia-Suarez, E. Palle,
Biology and Environment: Proceedings of the royal Irish academy, 107B(3), 157-165, 2007.
ABSTRACT
We have analysed the trends in four long meteorological time series from Armagh Observatory and compared them with series available from other Irish sites. We find that although maximum and minimum temperatures have risen in line with global averages, minima have risen faster than maxima, thereby reducing the daily temperature range. The total number of hours of bright sunshine has fallen since 1885 at the four sites studied, which is consistent with both a rise in cloudiness and the fall in the daily temperature range. Over the past century, soil temperatures at both 30cm and 100cm depths have risen twice as fast as air temperature. Wavelet analysis has found significant cycles with periods of 7-8 years, 20-23 years and 30-33 years in the seasonal and annual meteorological series from Armagh. Some of these cycles are clearly linked to the North Atlantic Oscillation.

417. Posted May 21, 2008 at 5:28 AM | Permalink

415(Lief) “you can discern a tiny wiggle departing ever so slightly from the envelope, but any such is short-lived, tiny, and inconsequential to a climate system whose ‘time-constant’ may be of the order of 10 years”

Does this mean the Earth in effect “smoothes and filters” the suns activity?

418. Posted May 21, 2008 at 5:33 AM | Permalink

417 (Stroller):

‘time-constant’ may be of the order of 10 years”
Does this mean the Earth in effect “smooths and filters” the suns activity?

That is how I would interpret it, yes. I would again interpret that as saying that variations on a timescale much less that the time-constant have no effect.

419. Posted May 21, 2008 at 5:45 AM | Permalink

418 (Lief)”variations on a timescale much less that the time-constant have no effect.”

If the time constant is around 10 years, that would also have quite a smoothing effect on longer term trends in sunspot activity wouldn’t it? It would introduce a significant lag of a terrestrial reaction to a reversal in trend. Does that mean my graph at #413 may have some significance after all?

420. Posted May 21, 2008 at 5:52 AM | Permalink

419 (Stroller): I’ll let others pitch in on this as it also impacts their analyses.

421. Posted May 21, 2008 at 5:58 AM | Permalink

420 (Lief)Thanks for your time Lief, I’ll check back later to see if my graph has generated further interest or comment.

422. Posted May 21, 2008 at 6:36 AM | Permalink

423. Posted May 21, 2008 at 7:09 AM | Permalink

422 (Srtoller): no problem. In my wife’s native language your misspelling of my name means ‘dear’.

424. Basil
Posted May 21, 2008 at 8:09 AM | Permalink

Leif,

My remark about “prejudice” was directed at what seemed to be the presumption that the cycles I’ve been presenting were the result of smoothing and filtering. There has certainly been no dearth of that in this matter of trying to divine cycles that would make sense of climate variability. I have a mantra that teach students whenever I teach anything that involves statistics:

If you torture the data long enough, it will confess, even to crimes it did not commit.

I think you are perhaps a bit jaded by what you see as the sheer volume of this in climate studies, and I can perhaps understand that. Especially, given your background and expertise, how aware you are of the lack of any plausible physical (solar) mechanism to account for these cycles, even when they can be shown without subjecting the data to excessive torturing. And as long as that remains true — no consensus of what physical mechanism could account for such cycles — the debate about them, and what they mean, will rage on ad nauseam.

However, there is a usefulness or utility to recognizing them — cycles — where they exist, even if we do not fully understand what causes them: they can help prevent the kind of hubris we’ve seen in the more excessive or alarmist AGW claims. We’re in a current cooling phase that could have been predicted with considerable “skill” from the regularity of the bidecadal and decadal cycles I’m seeing, regardless of their cause. That AGW alarmists seem to have been caught by surprise at this, coupled with increasingly shrill claims about how this or that year is the hottest ever, simply shows how they’ve been looking at the temperature record, not for what it is, but as something to torture until it confesses to what they want to hear.

425. Posted May 21, 2008 at 8:51 AM | Permalink

424 (Basil):

However, there is a usefulness or utility to recognizing them — cycles — where they exist, even if we do not fully understand what causes hem

Cycles is too strong a word. But many systems oscillate about an average state and when we have no knowledge [or idea] of what causes the changes that so clearly are there, I agree that the next best thing is simply to extrapolate the last few oscillations. A good example is the PDO. It seems to switch between a warm and a cold state every 30 years or so. Since we have had about 30 years of warmth I think it is legitimate to assume that we may be in for 30 years of cold [in lieu of knowing what will happen]. Similarly, there seems to be a ~100 year variation of solar activity, with minima near 1700, 1810, 1900, and maybe 2015 [in lieu of knowing what will happen]. We do not know what causes these ‘cycles’ and so cannot be taken to task for assuming them, but we must not be surprised if they don’t pan out. All this is clear enough. It is also clear that we should not take them seriously. An example of ‘misuse’ or ‘overuse’ of ‘cycles’ is your mixing of two cycles, the Hale cycle and the lunar nodal cycle, because that presumes that we have identified the physics [even if rudimentary or suggestively] of the situation and are now trying to explain, or substantiate, or whatever, the situation in terms of those known cycles, with the potential for prediction based on them, and the overconfidence in our prediction we get from believing that we know what is going on. As long as our prediction is simple cyclomania, I’m comfortable and might even go along with such one, e.g. the PDO.

426. Posted May 21, 2008 at 9:31 AM | Permalink

424 (Basil): I will get one in on your

presumption that the cycles I’ve been presenting were the result of smoothing and filtering.

When you calculate the FFT or the power spectrum that is a form of smoothing or filtering, namely to a set of sine waves. Wavelets are slightly better as they don’t assume that the phases of the waves are constant, but in either case it is filtering. It could be worse, namely that you pre-filtered the data to already suppress certain frequencies [or to be kind to the people that do that – suppress some ‘noise’]. That you didn’t do, which is good.
1. Worn out; wearied
2. Dulled by surfeit; sated
3. Cynically or pretentiously callous.
I don’t know which one you were thinking of, but actually none of them apply, as my attitude is dictated by my knowledge of the history of science, and what one extracts from that is that 99% of everything is wrong [and that 82% of all numbers are made up – even that number :-) ]. I was just trying to put into perspective why yet another cycle analysis does not make me tingle with excitement.

427. Posted May 21, 2008 at 9:37 AM | Permalink

If the PDO cycle has been around 30 years long in recent history, and assuming the phase change in 1970 mentioned by Erl was from cool to warm, it would seem to be a candidate for one the causes of the cooling period from 1940 to 1970 which bucks the tendency for earth temp to follow sunspot trends with a varying degree of lag apparent in my graph at #413 (which is filtered at the length of the Hale cycle Basil!). Maybe this was reinforced by the by the additional factor of sulfate emissions as noted by Jim Hansen:

Although I don’t know how much effect they could have had, I suppose it’s also worth noting that the atmospheric testing of nuclear weapons also fell in this period. Other candidates for aerosols include postwar reconstruction, and overlapping with albedo effects, modern agricultural practices.

I realise these factors are cited by Co2 driven climate change proponents to save their theory, but maybe we shouldn’t throw the baby out with the bathwater if they equally serve to strengthen a solar correlation with temperature anomaly.

428. Posted May 21, 2008 at 9:46 AM | Permalink

Correction: The PDO cycle would be 60 years as Leif notes at #425: 30 years cool, 30 years warm.

429. Posted May 21, 2008 at 9:47 AM | Permalink
430. bender
Posted May 21, 2008 at 9:58 AM | Permalink

#428/#429 So, that makes n=1.5 cycles for this eigenthingie we call PDO. Gosh, that’s robust.

431. bender
Posted May 21, 2008 at 10:09 AM | Permalink

#427 is, Andrew, yet another example of the sort of argument that I predicted would become increasingly common around here. Internal variability as an alternative explanation for a warming trend. Folks, internal noise and external signal do not divide so cleanly as you suppose.

For example: who can prove to me that PDO flip to positive in 1976 was not a product of GHG warming? (The answer is “no one”. Your only rebuttal is to unask the question and try to switch the burden of proof back onto me. Which is, of course, not a proof, and not how science works.)

Sorry for the interruption, Dr. Svalgaard. But I am sensing a trend: if it’s not the sun, it must be PDO. And if it’s neither, then perhaps it’s both.

432. jae
Posted May 21, 2008 at 10:30 AM | Permalink

431, bender:

For example: who can prove to me that PDO flip to positive in 1976 was not a product of GHG warming? (The answer is “no one”. Your only rebuttal is to unask the question and try to switch the burden of proof back onto me. Which is, of course, not a proof, and not how science works.)

But the flip to positive in 1920 casts doubt on the AGW exlanation, no?

433. Posted May 21, 2008 at 10:36 AM | Permalink

431 Bender: I don’t know Andrew and I was talking about a cooling period between 1940 and 1970, not a warming trend, but 427 is mine, so I’ll respond by saying that I agree with Leif that cycle is too strong a word for the oscillation. For all I know, the 1940-1970 anomaly in the lag of temperature to sunspot count correlation noted in 413 is a one off. I was merely listing possible contributing factors, little cuts if you like.

434. Posted May 21, 2008 at 10:43 AM | Permalink

431 (bender):

But I am sensing a trend: if it’s not the sun, it must be PDO. And if it’s neither, then perhaps it’s both.

You are [wrongly] assuming an agenda. What I said was that presented only with the PDO graph and having no other knowledge that it would not be a bad strategy to go with a cooling the next 30 years. Various people claim they know what is going on, and for them, it would be correct to view the PDO in terms of what they think they know. I do not know that solar activity will be low the next 30 years. I think so [but, for example, only managed to convince half of the NASA panel on prediction of the solar cycle]. I think we are closer to knowing that the solar variations are minute and therefore [unless the climate sensitivity is extreme – which I don’t think – runaway and all that] that it doesn’t matter what the sun is doing. That leaves a plethora of other things that could control the climate, and I have no ax to grind there – except that I welcome a bit of global warming – coming from a cold country :-) ]

435. bender
Posted May 21, 2008 at 11:13 AM | Permalink

#432

the flip to positive in 1920 casts doubt on the AGW exlanation, no?

No. According to this theory, as GHGs build their role in flipping circulatory modes would build. As observed.
A proper rebuttal must be grounded in physical models, not statistical retrospectives.

#434
For the record, I’m not assuming an agenda. My remark was directed in anticipation of comments such as jae’s. i.e. The cause for the trend must be anything but GHGs.

436. duwayne
Posted May 21, 2008 at 11:15 AM | Permalink

424(Basil):
Your comments hit home with me. From observation of recent climate history (100+ years), I’ve derived a model which can simulate past and possibly future trends. It’s this simple.
1 There is a general upward trend of 0.7C per century or 0.07C per decade
2 There is a 60-year cycle (probably PDO-related) which totally offsets the general upward trend on its cooling half-cycle and doubles the general upward trend on its warming half-cycle. Thus there is a 0.07 per decade downward “push” during the cooling leg and a 0.07C per decade upward “push” during the warming leg. The closely fits with last warming period (1977 to 2007) and the previous cooling period 1947 to 1977).
3 There are ENSO (El Nino/La Nina) events which cause significant excursions from the trend on a month-to-month basis. A moderate El Nino/La Nina causes a 0.3C excursion up or do respectively from trend and a very strong El Nino/La Nina causes a 0.7C temporary swing. These swing temperatures follow ENSO by about 3 to 4 months.

From these I can predict that the 2007 to 2037 period will have a flat Global temperature trend because the 60-year cyle downtrend will offset the general long-term uptrend. And that individual months will be as much as 0.7 above or below trend due to ENSO swings.

The problem is that I have added nothing to the scientific understanding of what causes the general uptrend, the 60-year cycle or ENSO. But in 1998 when the NOAA director said temperature trends were “consistent with a 3 to 8C Global warming by 2100″, I could confidently say “I don’t think so” because I knew the general warming trend, the 60-year cycle and ENSO were in conjunction and were clouding his crystal ball. And today when the IPCC with all its sophistication with respect to scientific theory of greenhouse gases and models says they expect warming of 0.2C per decade during the next 30 years, I again say “I don’t think so” since its very likely we will be in the cooling leg of a 60-year cycle during that period. Basil, I think there’s a real place for observation-based models in the climate arena since the science is so far from being settled and its easy to run amok.

437. lgl
Posted May 21, 2008 at 11:24 AM | Permalink

Here’s my suggestion. First a few assumptions:

1. Low SN or high VEI -> more clouds -> cooling
2. Low VEI or high SN -> less clouds -> warming
3. Low VEI and high SN -> not less clouds than 2. above, meaning if already less clouds by low VEI, high SN will not have much impact.

Then the combined SN and VEI (inverted) get quite interesting. pre-1977 they are in phase (VEI fliped), post-1977 they are in counter-phase. We see that much of the 50s 60s and 70s get increased cloud cover, and nearly all the post-1977 will have less clouds either by low VEI or high SN.

Since Leif is not fond of Saros we can do it another way, using the 22 yr and 18,6 yr cycles. Added 22/2 and (18,6)/2 will produce an envelope 55 yrs cycle (and 22 and 18,6 will give a 110 yrs envelope, both very close to observed cycles). Leif, if you don’t agree the 22 and 18,6 are real cycles there is not much hope for you – or was it me :-)

438. bender
Posted May 21, 2008 at 11:34 AM | Permalink

#437
What’s with the wiggle-matching? Don’t you know how to compute a correlation? Some of your cycles coincide (the first four). Others do not (the last three). Why is no one learning from Dr Svalgaard’s lectures on how to do analysis? Graphics are inadequate.

439. Basil
Posted May 21, 2008 at 11:41 AM | Permalink

Leif (#426),

Touche — on FFT/spectral analysis being a kind of filtering. But no more than, say, regression analysis to fit a trend line to some data. As you observed, my point was that that the data (in the latest graphs) was not pre-filtered to show the “cycles” we’ve been discussing.

Re #425 and the point about “cycles” being too strong a word. A nice feature of MTM spectrum analysis is the ability to distinguish between a “harmonic” (which ought to qualify as a “cycle”) and “quasi-oscillations” (oscillations which rise above the noise level, but in which the period is too variable to qualify as a “cycle” (or harmonic). In the data that I’ve presented, the ~20 year period qualifies as a cycle/harmonic, but not the ~9 year period. I have a “theory” about that, but in the absence of a “first order” effect to account for it, I wouldn’t expect you to be impressed.

pliny (#409)

I only annotated the ~60 year peak because I figured that if I didn’t, somebody would ask. You are right, though, that with only enough time for ~2-3 cycles, caution is in order. However, I believe that there are proxy records for an oscillation of this magnitude. And for what it is worth, if you combine 11 and 9.3 year signals into a beat cycle, you’ll get a beat wave approximately every 60 years.

440. bender
Posted May 21, 2008 at 11:57 AM | Permalink

Your second-order CRF “peaks” (buried in the CRF troughs) coincide with the TSI peaks. If there is an interaction between the two processes (sunniness coupled with lack of cloudiness, amplified terrestrially by January summer effects in the S. Pacific), then this “second-order” effect may be amplified to first order.

This is more or less what Erl and Raven et al have been arguing. But every time they come close to expressing this idea, you seem to dodge it. Tell me I’m wrong. Ask me for proof. But don’t dodge me.

441. lgl
Posted May 21, 2008 at 12:15 PM | Permalink

#438
Read it one more time then maybe you will understand it. “Some of your cycles coincide (the first four). Others do not (the last three)” is exactly the point and what I think is the cause of the climate shift around 1975.

442. Sam Urbinto
Posted May 21, 2008 at 12:49 PM | Permalink

Wow. Well, Erl in 412 brought up points that rather revolve around the discussion since then (a cycle if you will) :) Can we put some numbers on the obvious? What controls the atmospheric thermostat?

I’ll answer generically. Sort of. When it gets too hot, clouds form and it rains.

The thermostat is the planet regulating itself by doing work with the extra energy. Trying to remove the parts and seeing “which one causes it” is like taking apart a car to the smallest washer (including the engine and transmission and exhaust and suspension and body…) and trying to figure out how it goes 200 KPH. Or another way to put it, sort of, we don’t have to understand how exactly the engine develops x pounds of torque to know the pieces do various things and it develops the torque. Maybe a bad analogy, since we know or can find out what each part does how and we can control which parts are what kind and make substitutions.

I’ll just say: The planet is one big negative feedback loop.

Bender, #431 “who can prove to me that PDO flip to positive in 1976 was not a product of GHG warming” Nobody. Who can prove it was? Nobody. So the answer is we don’t know.

However, as I demonstrated, if you lower the years after 1976 by .2 the graph smooths out (the variability is about the same around the zero line) until about 1995. Subtracting .2 again moves the years around 0 (if you include 2008 to date) although 1999-2007 is still fairly “high” compared to some of the other years (I used 1945-2008 to catch about 30 years before and after 1976 and the first jump ahead of time before seeing the results).

The graphs are here on the BB at How random is the historic temperature record? (also some interesting samples of years by month).

So the question then becomes; can we can look at the anomaly and ask “What offset the anomaly by .2 in 1976 and again by .2 in 1994ish?” If we are working from the operational assumption that anomaly=temperature and temperature=overall energy levels, we have to look for the most likely “cause” of what we think is an additional overall .2 rise for ~18 years and then another overall .2 rise the last ~14 years (and will it do it again in 4 years or so or not.)

So, what is the most likely explanation of this, logically? It’s not the sun itself; as Leif has demonstrated, it’s not enough to answer the question. Is the answer the broad brush of “The planet is one big negative feedback loop.” as we could say “The engine produces power.”? Perhaps.

Or phrasing it another way and trying to answer in more detail: “If the .2 thing in the anomaly isn’t just coincidence, or measurement error, or wrong, what could caused it to move to a new equilibrium in 1977 and 1994?”

Then I could look at long term weather events (or anything else I wanted) and see if there’s a correlation of some kind that makes sense. Weather patterns, the rise of carbon dioxide, the rise of GHG levels, the rise in population, the sun, urbanization and industrialization, the “population <—> technology cycle”, cosmic rays, magnetic field, fuzzy bunny slippers, number of cows, the DJIA; what?

PDO, IPO and AMO make more sense to answer that particular question than anything else does. Perhaps it’s the wrong question to ask. Perhaps it’s the wrong answer. But what’s wrong with developing an operational assumption or operational definition to ‘what if’ with.

443. Posted May 21, 2008 at 12:50 PM | Permalink

437 (lgl):

Leif, if you don’t agree the 22 and 18.6 are real cycles there is not much hope for you

The 22 [vary 20-25] is not, but the 18.6 is. It is just that they have no effect. Or maybe the graph is supposed to show how the Moon’s gravity combined with the Sun’s heating [and thus slowing the Earth’s rotation] is creating large volcanic eruptions close to solar minima in the first half of the 55-year cycle starting in 1940 and close to solar maxima in the last half of the 55-year cycle? You see, without a mechanism [or with the flimsy one I just expounded] anything goes.

444. jae
Posted May 21, 2008 at 1:01 PM | Permalink

For the record, I’m not assuming an agenda. My remark was directed in anticipation of comments such as jae’s. i.e. The cause for the trend must be anything but GHGs.

That is not what I’m saying. I’m saying we can’t tell. Why don’t you prove it IS GHGs?

445. bender
Posted May 21, 2008 at 1:12 PM | Permalink

#444

Why don’t you prove it IS GHGs?

As I predicted, this will be your only defense: to try to shift the burden of evidence on to me. That’s a fair tactic in debating, but it’s not science.

But to answer your question: Why don’t YOU read the literature? Maybe the proof is there already? What makes you think it’s not?

446. Posted May 21, 2008 at 1:15 PM | Permalink

440 (bender):

Your second-order CRF “peaks” (buried in the CRF troughs) coincide with the TSI peaks.

I don’t think so, unless you can show me something I didn’t know. The troughs generally correlate fine with the sunspot number and thus with TSI. So without us seeing eye-to-eye on the above, the answer to your question: “If there is an interaction between the two processes…” will have to wait.

This is more or less what Erl and Raven et al have been arguing.

After hundreds of posts I still do not know what Erl is claiming. I have tried to get him to tell me, but I must be to dense for his expositions.

But every time they come close to expressing this idea, you seem to dodge it. Tell me I’m wrong. Ask me for proof. But don’t dodge me.

At least you say ‘seem to‘. I never dodge anything no matter how profound, weird, or wacky? About wacky, may I quote Niels Bohr who, when asked about his opinion about someone’s theory, answered “it is crazy, but not crazy enough to be true”. That I demand clarification of [at times] seemingly trivial points is simply that the devil lurks in the details, and I have learned the hard way not to skim and skip over the details. Anyway, since they are trivial details, they should have trivial and easy answers; but we are not there yet.

447. Posted May 21, 2008 at 1:17 PM | Permalink

444 (jae):

That is not what I’m saying. I’m saying we can’t tell. Why don’t you prove it IS GHGs?

I think there must be another thread [or blog. Try Tammy’s]for this.

448. Posted May 21, 2008 at 1:18 PM | Permalink

445 (bender): getting a little on the civility there…

449. Sam Urbinto
Posted May 21, 2008 at 1:22 PM | Permalink

bender

#435

No. According to this theory, as GHGs build their role in flipping circulatory modes would build. As observed.

This is indeed possible. Do you have any evidence that steadily building GHG hit a point in concentration that causes a circulatory mode to flip? Perhaps a GHG, industrialization/urbanization, population ensemble factor that causes them to flip?

My theory is that it’s caused by population hitting a certain point, the extra weight on the planet causes the modes to flip. Sadly I have no data, it just sounds as possible as yours, so I thought what the hell.

My remark was directed in anticipation of comments such as jae’s. i.e. The cause for the trend must be anything but GHGs.

You’re not being fair, he said “cast doubt on the AGW explanation”. Asking a fair question, what happened in 1920 that happened in 1976. What correlates to all the flips? Certainly that at least “casts doubt” on AGW as a causal agent.

Although I do understand your point, you want data. There’s two problems with it. None of the explanations can be quantified. You haven’t shown any data either. :) But that’s probably your point too. :D

#440

If there is an interaction between the two processes (sunniness coupled with lack of cloudiness, amplified terrestrially by January summer effects in the S. Pacific), then this “second-order” effect may be amplified to first order.

I’d be interested in seeing any answer to this also; CRF and TSI. Although I’d hazard a guess that whatever the correlation, the effects wouldn’t be strong enough. It would be nice if Leif could should that in a graph and a paragraph on it though.

But I like that idea, a lack of cloudiness (due to whatever reason, it’s immaterial) amplifying the Earth’s summer effects in the South Pacific.

As I mentioned earlier in my “spot the obvious!” exercise, what if it’s just a random sequence of events that lowers cloud cover long enough, we get a flip. Instead of ‘random’, maybe some combination of forces converging; possibly involving GHG, waste heat, population, number of cities, etc (or not); might be a better way to put it.

Then the question is: If the sun’s strength (or TSI/CRF, or whatever) is at 98% of nominal or at 102% of nominal (or whatever range, +/- .01% +/- 5%), would that flip happen anyway? If the flip happens anyway if you are at 75% of normal or 125% of normal, then discussions of any effect under 25% becomes immaterial.

Isn’t that really what we want the answer to, is there anything specific being stronger or weaker that would cause or inhibit the flip? Or, as I believe, are they uncontrolable: The cause is a random chaotic sequence of events that is not qualifiable or quantifiable.

450. bender
Posted May 21, 2008 at 1:26 PM | Permalink

#449 You think I start arguments without having backing ready for you? You think your “argument” is as weighty as mine? Meet you in unthreaded.

451. bender
Posted May 21, 2008 at 1:33 PM | Permalink

440 (bender):

bender: Your second-order CRF “peaks” (buried in the CRF troughs) coincide with the TSI peaks.
Leif: I don’t think so, unless you can show me something I didn’t know.

Please look again. Bottom frame. Blue and red curves. Sub-peaks at 1958, 1969, 1981, 1991, 2001 coincide with decadal TSI peaks (green). Let’s start there. If we can’t agree on the pattern, there’s not much hope for communicating the subsequent argument.

As I see it you and Erl are dancing around a subject that I want to see addressed. I’m trying to squeeze you into that arena. Maybe you don’t see it that way becasue you can’t guess what Erl’s up to.

452. bender
Posted May 21, 2008 at 1:42 PM | Permalink

#449, part 2:
Regardless whether or not there is data to estiamte the strength of such a mechanism, my immediate question is whether the GCMs are even capable of representing such an interaction, qualitatively. That’s an easy auditing question.

I hope Dr Svalgaard will see the pattern I am pointing to and address this mechanism. Although I note it is neither “solar” nor “terrestrial”, but solar-terrestrial interaction.

Erl, chime in if this is part of your argument.

453. Raven
Posted May 21, 2008 at 1:47 PM | Permalink

Leif Svalgaard says:

At least you say ‘seem to‘. I never dodge anything no matter how profound, weird, or wacky?

I appreciate your efforts on this thread. It has certainly educated me about the difference science based on an understanding of the physical mechanisms and ‘wiggle matching’. My personal opinion at this point in time is there is likely a solar-cloud-enso link that may or may not include GCRs.

My opinion is based on the following:

1) There appears to be a 0.1 – 0.2 degC correlation with the solar cycle in the temperature record (Camp and Tung, IPCC).
2) This effect is too large to be explained by direct TSI (guessimates using the 90 w/m2 annual forcing put the TSI effect at 0.01 to .03 degC).
3) Small changes in cloud cover can have significant effect on the temperature (Spencer et. al.)
4) GCRs offer a plausible link between the sun and cloud cover.

Unfortunately, we will likely never be able to demonstrate it conclusively since we have no proxies for cloud cover and it would take a century or more to collect enough data to extract any meaningful pattern from the noise. I suspect the CO2 hypothesis would have attracted an equal amount of criticism if we did not have paleo-proxies for CO2 and temperature (the 20th record does not support the CO2 – temp link unless unverifiable aerosol forcings are introduced into the models).

454. Posted May 21, 2008 at 2:07 PM | Permalink

442 (Sam):
[position]
The Earth has had a stable climate for 3,500 million years, probably will have that for another such interval [before evolution of the Sun overwhelms us]. By stable, I mean one that has allowed life to survive and at times flourish. There has been changes [some even catastrophic – impacts, super volcanoes, …], but the system has absorbed them [this is not necessarily so, as the case of Venus may show], so the negative feedback loops have worked fine so far. Some changes may not be to the liking of everybody [my ice is melting and I can grow crops and live, your forests turn into deserts and you die]. Adaptation to such is a political problem [crop policy, war, mass-migrations, famine] that has a long history and there does not seem to any unknowns there [perhaps with the exception that our destructive power has increased tremendously, which may shorten some of the struggles]. Science has brought us a long way with no end in sight. This has been a positive feedback loop with a positive result. If politics interferes with or dictates science, the positive feedback may lead us down a road that it will take a long time to recover from. This has also happened before [at least three times] and is a well known and well-studied issue. The polarization [with associated hostility, attacks, and general crap] that I have observed [as I have lived on other blogs too] – even on this blog – at times interferes with understanding [and at other times is just pointless prattle to be ignored]. It may take us a hundred years to sort out how the system works if we work together; if we spend all the energy fighting each other it will take longer [a lot longer]. The problem is hard so there will be lots of false turns and plenty of crackpot theories [one might just be “crazy enough to be true” – Niels Bohr] and lots of cyclomania and theories that seek to explain everything , ending up explaining nothing, but we’ll get there.
[/position]

455. bender
Posted May 21, 2008 at 2:11 PM | Permalink

#446

At least you say ‘seem to‘.

Carefully chosen words.

I never dodge anything no matter how profound

Not carefully chosen words. Instead of “dodge”, try “steer away from”.

As lead proponent in this thread it is, of course, your right to “steer” conversation as you choose. The question I ask is why do you steer as you do, and not some other way. Respectfully, there is ALWAYS an agenda.

You are somtimes directing. Sometimes probing. Sometimes lecturing. Sometimes learning.

Please evaluate the hypothesis that I have put before you. Either it bores you or it intrigues you. Just don’t tell me, as you have told Erl, that it has no worth because there is no data backing it. You know very well that science advances most through insightful guesses made by critical minds, not humdrum data collection by disengaged technicians.

456. lgl
Posted May 21, 2008 at 2:16 PM | Permalink

#443
Leif, again, this isn’t Swiss clockwork. Is the solar cycle on average close to 22 yrs or not?
There was “Lunar standstill” in 2006 and 1997 http://en.wikipedia.org/wiki/Lunar_standstill so with 9,3 years spacing there should be one in 87, 78, 69, 59 and 50 too, and there were no VEI>=4s any of those years, and there were several large eruptions between all those years. Isn’t that an indication that the moon’s gravitation has significant infuence on the volcanism?

I’m not saying the link to the sun is the direct heating. The IR Low-cloud amount peaked around 86, 95 and 04. You know better than I how that fits the solar activity but more clouds those years does not surprise me. http://isccp.giss.nasa.gov/climanal7.html

457. bender
Posted May 21, 2008 at 2:20 PM | Permalink

#454
A position I agree with.

Just do not tell me that synergy is not valid mechanism for amplification of an effect. It seems reasonable to a non-physicist to postulate GCR-TSI synergy through a solar-terrestrial connection involving oceans and clouds – arising specifically in the S. Pacific summer, and then delivered to the people on the continents through ENSO and PDO pathways.

[None of this denises a role for GHGs.]

If this is unreasonable, please explain why, with references to the primary literature.

458. Sam Urbinto
Posted May 21, 2008 at 2:34 PM | Permalink

bender #450 My quip about population and their weight causing a “flip” was a joke.

I didn’t see you at unthreaded…. Of course I figure you have something (I’m not arguing with you by the way) but it escapes me what it is or why it’s important. I can’t figure out what would exist that can take a minor part of the system (let’s say 10% of the greenhouse effect, what ModelE says it adds) which would be about 7% of the atmospheric energy. It’s been steadily rising. Wait. Another track.

You’re fooling with me, aren’t you? You said GHG. Water vapor is halfish of the 30% down and most of the 70% up. Since water drives the climate and all. So if your point that GHG (which water vapor forms the gradients that along with sun cause wind and everything else) are the major portion behind a flip, I agree. But the GHG don’t control themselves, water vapor reacts to the sun, the temperatures, the altitude, the wind, the rest of the atmosphere, the other phases of itself in the hydrosphere. So, if I include clouds and how the atmosphere functions along with GHG, sure.

Leading back to my contention that the energy levels are fairly stable because “The planet is one big negative feedback loop.”

#451 You mention “1958, 1969, 1981, 1991, 2001″. What happened in those years with PDO IPO and AMO? Anything?

#452 “neither ‘solar’ nor ‘terrestrial’, but solar-terrestrial interaction.” Of course it is. The question is what part the solar plays. Oscillations switching direction; how much does solar have to change to start or stop them? .001%? 1%? 10% We have to figure out the point where it makes a difference and find out what the current difference is (if any) both. Can we find both?

And that’s the first part of that post. My immediate question is also if the GCM can handle it, and my answer is not until they do the hydrosphere a lot better as well as the atmosphere a lot better.

—————————–

I’m still not sure what you’re driving at that you want Leif to answer. That it’s a solar-terrestrial interaction? But here’s the argument, I think, on the other issue.

“If the two .2 rises in the anomaly isn’t just coincidence, or measurement error, or wrong, what could have caused the anomaly proxy to move to a new equilibrium in 1977 and 1994?”

PDO, IPO and AMO make more sense to answer that particular question than anything else does.

What causes behaviors in PDO, IPO and AMO and the rest of the weather short- or long-term? Mostly water in the atmosphere and hydrosphere, along with the rest of the atmosphere, powered by sunlight.

Or in other words, GHG transform the energy from the ground and create situations where phases can flip.

Of course “GHG cause long-term weather patterns to shift direction” is too hopelessly simplistic as a simple statement, unless, perhaps, one is talking about a model.

—————————-

BTW, speaking of the models, the CMIP control runs between ~11-17 C being taken as the median of the highest and lowest of 14 C seems to be our much vaunted base period which appears to be almost exactly the same for any 30 year period. So when the base period is seems unimportant.

I also want to say; if we are tracking an anomaly based upon model runs covering 14 C +/-3, what exactly are we worried about a trend of .7 up over 130 years?

459. bender
Posted May 21, 2008 at 2:37 PM | Permalink

#458
You are full of such quips. One day they will get you in trouble. Today perhaps.

460. bender
Posted May 21, 2008 at 2:41 PM | Permalink

#458
The thing you are conveniently forgetting, sir, is that phase changes can hide changes in energy. Your favorite topic. What makes you think the GMT rise would not be higher were it not for these flips? This is a common skeptic double standard. Internal climatic variability cuts both ways.

461. Posted May 21, 2008 at 2:43 PM | Permalink

455 Bender: Everyone remembers about ‘paradigm shifts’ in Thomas Khun’s ‘structure of scientific revolutions’ but few read him closely enough to remember what he had to say about instumentation and the circularity of self certifying theory within the framework of ‘mainstream science’.

If we are to permit Erl a way to formulate his ideas and quantify effects, he may need some climatic equivalent of ‘dark matter’ or ‘aether’ to work with.

Or maybe we need to permit him to ‘bend space’.

462. Posted May 21, 2008 at 2:55 PM | Permalink

456 (lgl): http://www.jupitersdance.com but wise monkeys don’t discuss this on CA :)

463. lgl
Posted May 21, 2008 at 3:36 PM | Permalink

#456
Thank you Stroller. I guess I will find your link more interesting than Leif will.
Since I’m only a monkey and not wise I can add a few more words. The nice thing about this is that we will probably know in a few years so we don’t have to discuss it. Volcanism will peak around 2010 but sunspots will probably peak a few years later, so SN and VEI (not fliped this time) will already then be almost out of phase and we should see cooling like the 50s.

464. Sam Urbinto
Posted May 21, 2008 at 3:53 PM | Permalink

bender #460 I’m not forgetting anything, I’m fairly sure phase changes do hide energy changes. That’s one of my points, how do you unlink the two from each other, much less the system. No double standard, the global mean temperature anomaly could very well be higher if it wasn’t for the flips. Or not. I’m well aware internal climate variability “cuts both ways”, I’d say it’s why this is so difficult to nail down. You should note that I said “create situations where phases modes can flip” (I meant some oscillation can change direction or sign, specifically warm mode (above zero line) versus cold mode (below zero line))

In that respect I am a skeptic, but of both viewpoints. As far as I’m concerned, there is no way to state either way what exactly water vapor is doing in the atmosphere, and it and the other phases are doing in the hydrosphere et al. Or that the anomaly reflects energy levels well (or doesn’t). Or that increased levels of non-water vapor GHG in conjunction with the effects of water in any form along with lower or higher levels of sunlight don’t result in higher or lower net energy levels. Or exactly what the combination of factors does to temperature, humidity, and lapse rate at any given altitude, plus or minus, to drive the negative feedback loop for energy levels there seems to be.

I’m still interested in seeing what you were working at for your question “who can prove to me that PDO flip to positive in 1976 was not a product of GHG warming?” Were you interested in somebody trying to prove a causal link between any rise in energy levels from GHG and a flip in the PDO in 1976? Did you want somebody to show a sudden jump of .2 in the anomaly caused PDO to flip and not the other way around? Do you have a paper not based on a model that proves:

A) The anomaly reflects energy levels.
B) A steady rise in GHG levels over time caused energy levels, and therefore the anomaly, to jump suddenly in 1976.
C) The jump in energy levels caused the PDO to flip.

Seems far too complicated a string of proofs.

As far as your “if it’s not the sun, it must be PDO. And if it’s neither, then perhaps it’s both.” I also fail to see what you’re getting at. Weather patterns change for some reason (I take it your reason is a postulation of a GCR-TSI synergy through a solar-terrestrial connection involving oceans and clouds in which GHG probably play a role), and this mix of events causes a PDO shift? Seems rather obvious that would be the case. What’s your point?

#457 “A position I agree with.” If you agree with Leif on the idea of a negative feedback loop that keeps the anomaly, whatever it reflects, close to the median of the CMIP control run min and max, what are we discussing?

It seems reasonable to a non-physicist to postulate GCR-TSI synergy through a solar-terrestrial connection involving oceans and clouds – arising specifically in the S. Pacific summer, and then delivered to the people on the continents through ENSO and PDO pathways. [None of this denises a role for GHGs.]

Is that retorical? Why wouldn’t it seem reasonable, isn’t this what we’re all saying, that it’s a convergence of factors. Call it whatever you want. Although I’m not sure if you asking if it’s ‘not denying a role’ means anything, because quite clearly they do, so are you trying to be vague if they do or not, or do you mean the increased concentration of non-water GHG might be a causal factor? That seems like a stretch (although of course possible).I’d be surprised to find any primary literature that shows increased amounts of methane, CFCs, carbon dioxide and nitrous oxide on a steady climb cause PDO shifts…

What exactly are you trying to get at here? I’m confused. Isn’t this all just we can’t see “the cycles” (however you want to phrase it) because there are too many overlapping ones on different time scales?

You are perplexing. Have I ever said it can’t be “the GHG” (however you mean it). How do you mean it. If you mean with water vapor, too obvious. If you mean without, unlikely. I’m also wondering when you say “be” you mean “be partly from” and just failing to qualify/quantify it.

465. Posted May 21, 2008 at 4:02 PM | Permalink

464 (Sam):

#457 “A position I agree with.” If you agree with Leif on the idea of a negative feedback loop that keeps the anomaly, whatever it reflects, close to the median of the CMIP control run min and max, what are we discussing?

I think this is a bit stronger than I said. Well within my range fall the deepest glaciations and the Permian desert-climate, which I think are somewhat outside of the CMIP control runs. But that is a detail, there is enough stability and resilience in the system to keep it reasonable [i.e. not Venus or Mars like] possibly because we are better positioned as far as distance to the Sun is concerned.

466. Posted May 21, 2008 at 4:05 PM | Permalink

461 (Stoller);

If we are to permit Erl a way to formulate his ideas and quantify effects, he may need some climatic equivalent of ‘dark matter’ or ‘aether’ to work with.

If I were Erl, I would respond that I do just fine with ‘gray matter’.

467. Sam Urbinto
Posted May 21, 2008 at 4:21 PM | Permalink

Leif #465 Sorry, I meant within the since 1880 thermometer/sensor readings. I didn’t mean the last few hundred thousand or anything!

468. Joe Solters
Posted May 21, 2008 at 4:44 PM | Permalink

Leif #454 Re: Politics of AGW and the “why can’t we all get along” syndrome. Unfortunately, climate science has been a highly politicized issue for over 10 years. Pending now before the US Congress is massively intrusive, mandatory legislation specifically designed to reduce CO2 emissions, that will affect deeply the lives of 300 miilion people. If “It may take 100 years to properly understand the earth’s climate”, why are AGW proponents rushing these laws before the science is complete and credible? If the search for provable scientific conclusions was purely a scientific venture, the vitriol surrounding this subject would be minimal. The AGW political push has become part of the science and rational people are pushing back. Politics has always been a contact sport.

469. Posted May 21, 2008 at 5:07 PM | Permalink

451 (bender):

440 (bender): your second-order CRF “peaks” (buried in the CRF troughs) coincide with the TSI peaks.
Leif: I don’t think so, unless you can show me something I didn’t know.

bender: Please look again. Bottom frame. Blue and red curves. Sub-peaks at 1958, 1969, 1981, 1991, 2001 coincide with decadal TSI peaks (green). Let’s start there. If we can’t agree on the pattern, there’s not much hope for communicating the subsequent argument.

Because of the small scale it is hard to see anything, so here is the data on a more reasonable scale, with the times you have singled out as having a pattern:

The blue curve with a pink 27-day running mean is the cosmic ray count, the pink curve [bottom of each panel] with the blue 27-day running mean is the sunspot number [proxy for TSI]. Below the main panel are panels for each of the times you mentioned: 1958, 1969, 1981, 1991, and 2001. Tell me what the pattern is, because I cannot see it.

As I see it you and Erl are dancing around a subject that I want to see addressed. I’m trying to squeeze you into that arena. Maybe you don’t see it that way because you can’t guess what Erl’s up to.

I don’t need to be squeezed into anything: if there is gold in them thar hills, I’ll go diggin’. So, show me the gold.

470. Posted May 21, 2008 at 5:16 PM | Permalink

463(lgl) The volcano in Chile has already smoked out parts of Argentina and it hasn’t even errupted yet. Potemtially a convenient escape for AGW proponents which will muddy the waters as well as the skies.

471. Posted May 21, 2008 at 5:18 PM | Permalink

468 (Joe): in a democracy people have the government they deserve; if you don’t like the way it is going, go to the ballot box and show it, or, if that doesn’t help, take to the streets and the barricades. The AGW people want action now because they think it will be too late in a hundred years, or twenty years, or tomorrow, or whatever. For them it is a question of morality, not of science. Therefore a scientific discussion is not useful. That said, let’s hold of any further discussion of politics.

472. Posted May 21, 2008 at 5:24 PM | Permalink

456,462 (lgl,Stoller): The argument in #379 effectively demolishes the barycenter nonsense.

473. Sam Urbinto
Posted May 21, 2008 at 5:49 PM | Permalink

One more thought for bender. Look at the anomaly trend as an average of all the readings, side by side with the population growth trend, as an average of all the people. Which is more likely, the anomaly trend causes more people, or more people causes the anomaly trend, or neither causes the other, or it switches around. To me, saying it’s more likely “non-water vapor GHG warming” causes a long term weather pattern of some sort to do something or another is like me saying it’s more likely the anomaly causes population. I don’t need literature to tell me the chances of that relationship are just about nill, but it won’t make me say population causes the anomaly is a proven fact either. What are you shooting for? By the way, if Leif is not guessing what Erl is up to, that makes two of us. Not that I always understand anyone here, even Leif or Steve or Tom or David or, well, myself! :)

Leif #471 “For them it is a question of morality, not of science. Therefore a scientific discussion is not useful.”

Then it’s not politics either. It’s social studies or perhaps religion. Does it really matter what it is, as long as we know what it’s not — It’s not science. People don’t get this polarized about a simple science matter, only on interpretations of results, meaning, future actions, or perceived viewpoints of others due to improper communication, etc.

474. Mark_T
Posted May 21, 2008 at 8:50 PM | Permalink

Wow, I just found this thread and so far, I’ve read a lot of interesting things. I’m about at the 169 post mark on the following page which I think is the very first page.

I’m not yet convinced that the sun isn’t responsible for the warming… There are quite a few papers that show a link between the sun and earth. But I will read through the entire set of 6 pages to see what Leif and others have to say. I’m sure my hrair limit will be reached many times in the process.

Mark

475. Posted May 21, 2008 at 10:37 PM | Permalink

474 (MarkT): welcome to the thread and thanks for slogging through it.
Regards, Leif

476. Erl Happ
Posted May 21, 2008 at 11:08 PM | Permalink

When you guys get active I try to sleep.

414 (Leif)

Yes TSI as measured in space varies by .01% and measured at the limits of the atmosphere varies by 7.6%. So we have about 7% more irradiance in January when the Earth is in best position to absorb and store heat due to the extent of ocean exposed.

BUT

TSI is a mixed bag of goodies that exhibits disparate trends. For example, 1280MHz follows the sunspot number and if you were to produce the same graph showing variations from year to year in the monthly figure for this particular component of TSI it will vary between 650 and 2300 units in any month. This wave length is capable of heating water. It’s part of a much larger parcel that can do so. We need to know what proportion of TSI can be absorbed by water as it comes in and where it tends to be absorbed. The physicists may have an answer to this question. I don’t know the physics but I am a diligent searcher accustomed to looking for secondary signs and signatures.

By looking hard I have discovered that the vertical extent of the temperature anomaly that is best expressed at 250mb or 9km elevation also varies in horizontal extent expanding and shrinking in the cycle. At is level the contrast between warm (Atlantic, Indian and Western Pacific to 100°W) and cold parts (Eastern Pacific) is most marked. The highest cirrus cloud is located here and there is very little water. It’s an interface zone between solar radiation and water. Critically, the difference in temperature across this layer is not due to differences in outgoing long wave radiation even though the warm zone is co-extensive with the warm zone at the surface. The atmosphere over the warm zone (its all ocean and tropical rain forest) cools by convection and decompression and that over the cool parts of the ocean cools by long wave radiation emission. So at 12 Km the atmosphere over the cool zones becomes warmer due to ozone excitation and that over the warm zones does not. The warm zone signal gets weaker as the cold zone signal strengthens. This indicates a solar origin to the heating in the warm zone. If it were heated from below by other processes the anomaly that exists at the surface would become weaker with elevation. This anomaly becomes stronger. It is not a good idea to think of the Earth as if it were made of the same stuff at all latitudes and longitudes. If you do so, you never se the subtleties.

Things that change in parallel with this high altitude temperature anomaly, exhibiting the same spatial distribution, include surface temperature, total atmospheric column precipitable water, precipitation, surface atmospheric pressure and relative humidity (up to the 300mb or 8km) level. All these characteristics exhibit the mark of the solar cycle.

The Earths surface and the atmosphere show a Solar Cycle fluctuation, an annual fluctuation and an ENSO fluctuation. In all aspects temperature, humidity, precipitation and surface pressure there is an expansion of the zone affected.

The annual fluctuation is strongest followed by solar cycle followed by ENSO. If ENSO disappeared (and its presence is highly variable) there would still be a solar cycle fluctuation but the presence of ENSO makes the solar cycle fluctuation more pronounced. The biggest fluctuations in those components of irradiance that can heat water vapour occur at the top of the solar cycle. The strongest La Nina cooling event is frequently found when solar activity is very low as at present. The big El Ninos occur when irradiance varies most strongly, at the start of the cycle and in the decline phase. The ENSO event signature is strong at 250mb pressure level. ENSO is not an internal fluctuation. I am not an adherent of the notion that when a butterfly flaps his wing in Buenos Aries a storm is brewed up in the Indian Ocean. The Earth system loses energy continuously and without continuing stimulation would come to rest.

The outstanding problem is this. That proportion of TSI that is intercepted by water vapour probably accounts for the temperature differences at 250mb. The affected zones change albedo. This needs to be verified in terms of the physics and mathematics and also as observations of cloud density and extent. Inference is one thing. Proof is another.

And what happens in the tropics is vital, not just for the tropics but for higher latitudes.

I am disappointed that people are having trouble following this line of reasoning. I presume that they are suffering from the common misconception that ENSO, PDO etc are due to internal oscillations of the climate system. Let’s get real. The surplus of disposable energy comes into the tropics and is reflected in very high sea surface temperatures. Other parts of the globe depend on that energy. Other parts are net emitters. It is the Northern Hemisphere alone that has warmed and the change has been in minimum temperatures. The NH is a strong net emitter in all seasons.

477. Posted May 22, 2008 at 1:57 AM | Permalink

472(Leif)The page I linked doesn’t contain any barycentric nonsense. It does however contain a fair bit of numerological nonsense based on the coincidence of the solar cycle length and the Jovian orbital periodicity. The reason I linked it for lgl in the context of volcaanism is that right at the top in the intro there is this statistical gem:

“Research shows that large earthquakes occur at low sunspot frequencies. A sample of global earthquakes greater than Magnitude 6 for the period 1973-2005 (USGS) compared to smoothed monthly sunspot figures (SIDC) show that 71% of earthquake energy is released and 59% of earthquake events occur at lower than average sunspot activity.”
Not a very strong correlation on an insufficient time period in my view, but it goes on to say that the top 12 largest earthquakes over the past hundred years (USGS) on average occurred at lower than average sunspot levels.

Average sunspot count of 17 for top 12 earthquakes in the last century

Average sunspot count for period is 22.5

Which although it doesn’t strengthen the correlation, at least extends the sample period. Seems to me it could equally be an internal (to the earth-moon tidal system?) rhythm which happens to overlap the sunspot cycle frequently enough for an apparent (weak) correlation to appear. I guess someone interested enough might plot earthquake location against the saros cycle to see if the moon’s declination is statistically significant.

Since you brought up the subject of the barycentric nonsense, I’ll see if I can get away with making a couple of observations about it without getting myself audited by our web master. :)

I can’t really comment on your demolition of it in #379 without doing the numbers to see if there is some other error in their calculation (I’ll trust yours) which would introduce the lag and skew into their TSI plot, and like you, I’m not prepared to spend too much time on it.

I would however be interested to know your a priori reasons for rejecting the possibility that the more massive elements of the solar system might in concert, move the sun around enough to cause changes in earths orbit and affect the sun’s angular momentum sufficiently to cause ripples in the motion of the plasma and the ‘solar conveyor belts’ where most sunspot activity seems to occur.

If I understand relativity well enough (!), gravity is a notional construct which belongs to a Newtonian universe. Nobody looks for gravity particles anymore. If we consider the solar system as being within the rather squishy Einsteinian ‘mollusc of reference’ we are unable to determine it’s internal ‘many body problem’.

It seems to me that the complexity of the ‘many overlapping cycles and oscillations’ of the climatic system isn’t going to be resolved solely by looking at the ripples in the oceans which are confused by reflected waves, crosswinds and topography. These ripples are pale reflections of externally applied forces, the shadows on the platonic cave wall.

Erl’s powerfully oscillating grey matter is attempting a herculean task, we need to assist by easing the force of Newtonian gravity, by energising the dead Daltonian atomic particle, or by supplying him with an archimedean lever with which he can move the world.

478. Posted May 22, 2008 at 2:55 AM | Permalink

476(Erl)I think Leif will take issue with you on the relative strengths of ENSO and solar cycle effects on the anomaly.

436(Duwayne) said:

1 There is a general upward trend of 0.7C per century or 0.07C per decade
2 There is a 60-year cycle (probably PDO-related) which totally offsets the general upward trend on its cooling half-cycle and doubles the general upward trend on its warming half-cycle. Thus there is a 0.07 per decade downward “push” during the cooling leg and a 0.07C per decade upward “push” during the warming leg. The closely fits with last warming period (1977 to 2007) and the previous cooling period 1947 to 1977).
3 There are ENSO (El Nino/La Nina) events which cause significant excursions from the trend on a month-to-month basis. A moderate El Nino/La Nina causes a 0.3C excursion up or do respectively from trend and a very strong El Nino/La Nina causes a 0.7C temporary swing. These swing temperatures follow ENSO by about 3 to 4 months.

1. Gets a resounding ‘maybe-maybe not from me.
2. I’d like to know what you think of the PDO 30 year cool-30 year warm oscillation and the likelihood of it’s continuing in a regular manner while the anomaly stays high, and stable. And if it falls?
3. Would indicate the effect of ENSO on the anomaly is a lot higher than that of the solar cycle variation, though I take your point that they are coupled anyway. Does the coupling and amplitude of the effect of ENSO hold for historical periods when the anomaly is low? E.G. Daltom minimum, and 1940-70 cooling period?

The main sticking point between you and Leif seems to be the amount of energy you require to be absorbed in the upper levels of the troposphere from outside. Can you or anyone quantify it?

The stratosphere has been cooling for a long time, could your missing energy in the mid-upper troposphere be coming from there, rather than it being lost to space? Can heat head down rather than up more easily in rarified regions of the atmosphere? Presumably, to satisfy the second law of thermodynamics, this would have to be a pulsing process?

Excuse the barnstorming, just trying to help break a circle of negatives here.

479. Erl Happ
Posted May 22, 2008 at 6:03 AM | Permalink

478 (Stroller)

I think Leif will take issue with you on the relative strengths of ENSO and solar cycle effects on the anomaly.

Pretty hard to disentangle the two because in my view the atmosphere is responding to elements of solar radiation in both cases. But, hopefully, I will present some pictorial data soon that will give an impression of the strength of the solar cycle as a separate phenomenon. I have upper atmosphere data on three solar cycles. In each case the annual swing and the ENSO swing are features within the solar cycle swing. In each case precipitable moisture advances with sunspot activity and peaks with sunspot activity.

ENSO does not disappear when sunspot activity is low. Evaporation from the oceans and the appearance and disappearance of clouds goes on regardless of the absolute moisture content of the atmosphere. It does not take a big change in solar activity to change the relative humidity of very cold air at the margins of the troposphere where moisture levels are always very low. That in turn changes the cirrus cloud in the tropics. By and large the tropic has only cirrus except where there is thunderstorm activity at the so called Intertropical Convergence Zone. The area involved is relatively very small. The ITCZ is an intermittent thing that gains and loses intensity and at the moment it is very weak.

I live on the west coast with a vast ocean to the west, no mountains to disturb the flows and the continent and the oceans alternate in dictating the local weather. At latitude 33°S you can see the tropical dynamics dictating where the wind will come from. Up till a few weeks ago tropical moisture was being swept south. Now it is going north and the fronts from the south are bringing our winter rain.

480. Basil
Posted May 22, 2008 at 6:46 AM | Permalink

Erl, if #476 is a brief of what you’ve been trying to say in your exchanges with Leif, I may print it out and use it as a roadmap to go back and read the previous exchanges more carefully. To be honest, I’ve skipped over them because the discussion of atmospheric processes has been over my head. Though I got a small dose of climatology while majoring in physical geography as an undergrad, I eventually ended up with degrees in economics. So what little expertise I currently have that is relevant to the discussion has to do with the temporal behavior of temperature, looking at it like I would any other time series.

On that, you wrote:

The Earths surface and the atmosphere show a Solar Cycle fluctuation, an annual fluctuation and an ENSO fluctuation.

If you look back at the figures I posted in #404, if there is a solar cycle fluctuation, it appears to be the 22 year cycle, not the 11 year cycle. Actually, I think there is still something going on related to the 11 year cycle, but the influence is much weaker during even numbered cycles than during odd numbered cycles (see the figure in #328), so that (as a working hypothesis) the frequency of the decadal signal is dominated by lunar tidal forces, while the frequency of the bidecadal signal can represented more or less exactly as a beat cycle from combining the 22 year solar cycle and the 18.6 year lunar nodal cycle.

Which brings me to my question. Is there anything in your view of the solar cycle connection that could explain why the dominant signal in temperature is bidecadal, rather than decadal, i.e. has something to do with the 22 year solar cycle? One thing I’ve learned from Leif is to distinguish the “Hale cycle” from the 22 year cycle in polar magnetic polarity (which occurs at the time of solar maxima). If there is a solar cycle connection to the bidecadal variation in temperature, I think now that it may be to the latter, rather than the Hale cycle per se. With apologies to Leif, here is a chart which filters out all but the strongest temperature cycles (though the spectrum of this plot is the same as the spectrum of the raw unfiltered data), plotted along with solar cycles.
It should be noted that the lines indicating the odd numbered solar cycles are drawn with reference to their peaks, not when they began. At least since solar cycle 15, the peak in the temperature cycle has led the peak of the solar cycle by 2-4 years which I think may correlate with something Leif has said about the cycle in polar magnetic field stabilizing about three years before it peaks.

Summing up, I think any theory of the sun-climate link has to explain why the 22 year signal is stronger than the 11 year signal, and I cannot help but wonder if the answer will somehow involve the 22 year cycle of solar magnetic polarity.

481. Erl Happ
Posted May 22, 2008 at 8:18 AM | Permalink

480 (Basil)

I started off this experience with a similar view to yours looking at the aa index and comparing it to temperature data but I have learned a lot in the process. Others may disagree but there you are.
Re

At least since solar cycle 15, the peak in the temperature cycle has led the peak of the solar cycle by 2-4 years

That’s the El Nino signal in the early stage of the upswing.

Is there anything in your view of the solar cycle connection that could explain why the dominant signal in temperature is bidecadal, rather than decadal

Only the relative height of the cycles with odd and even having a relationship that occasionally breaks down.

When I look at your data I see peaks all over. Sometimes in the trough. Sometimes at the peak of sunspot activity and some at the rising stage when the big El Nino like 1978 kicks in.

Global data suffers from offsetting factors. SH versus Northern and high latitudes versus low. So, I like to look at the data for a region that avoids those offsetting trends. The Equator shows little variation in sea surface temperatures but good differentiation in temperatures in the lower troposphere. Surface data for the tropics is iffy. For reliability and lack of bias you can’t beat the satellite record of temperatures in the troposphere. For years earlier than the satellite record the SOI inverted proved a rough guide. When you look at the temperature data for the tropics, it leads the global figures and has greater amplitude. That tells us something. Even in the tropics temperatures in the Eastern Pacific fall dramatically as the temperatures in the rest of the tropics rise strongly. We take no account of the changing calorific status of the ocean even close to the surface. Basic requirements of careful measurement are nowhere met.

482. kim
Posted May 22, 2008 at 8:36 AM | Permalink

481 (Erl) I particularly like the leading, higher magnitude, tropical temperatures. That’s piston-like. As I prattle, I can feel the water ebbing under my feet for the wave of your understanding coming.
==========================

483. Erl Happ
Posted May 22, 2008 at 8:47 AM | Permalink

482 (Kim) Should have been a poet. But, there is a long way to go yet. It’s still a guessing game.

484. lgl
Posted May 22, 2008 at 10:56 AM | Permalink

#470 Stroller

muddy the waters – I didn’t think of that

“iron in the volcanic aerosols fertilized the equatorial Pacific” &
“Plankton Cool Off With Own Clouds, These compounds collect as dust particles that promote water condensation, which, finally, leads to cloud formation.”

But Leif do not like speculations and this is a bit OT.

485. Sam Urbinto
Posted May 22, 2008 at 10:59 AM | Permalink

This any help? Insolation values for the climate of the last 10 million years. A. Berger and M.F. Loutre Quaternary Science Reviews, 10: 297-317. Or this graph from the data?

Wikipedia gives this example using the same latitude.

The Earth’s orbit and obliquity change with time (over thousands of years), sometimes forming a nearly perfect circle, and at other times stretching out to an orbital eccentricity of 5% (currently 1.67%). The total insolation remains almost constant but the seasonal and latitudinal distribution and intensity of solar radiation received at the Earth’s surface also varies.[5] For example, at latitudes of 65 degrees the change in solar energy in summer & winter can vary by more than 25% as a result of the Earth’s orbital variation. Because changes in winter and summer tend to offset, the change in the annual average insolation at any given location is near zero, but the redistribution of energy between summer and winter does strongly affect the intensity of seasonal cycles. Such changes associated with the redistribution of solar energy are considered a likely cause for the coming and going of recent ice ages (see: Milankovitch cycles).

I’m just bringing it up, have no idea if it means anything or not.

486. Sam Urbinto
Posted May 22, 2008 at 11:48 AM | Permalink

Let me see if I can verbalize my own cartoon (executive summary?:)). Don’t get wrapped up in the details.

Sunlight comes in.

At TOA it’s most strongly irradient in the visible. In the stratosphere, oxygen and ozone absorb UV, the visible gets through, water vapor and a tinge of carbon dioxide get some of the IR, in the shortwave area, where the two exist. Mostly of course in the 80% of the atmosphere that is the troposphere, and mostly water vapor. This powers some chemical reactions and heats the atmosphere about 30% overall.

However, this happens different ways at different altitudes which have different pressures, mixes of substances, variations in clouds (or not) that block it (or not) in various ways. The atmosphere also is differently segmented at different sections of the Earth, and the boundry between troposphere, tropopause and stratosphere changes. The strength of the sun is also affected by the tilt (which half of the half of the planet that’s in the light), and how directly the sun hits various sections by latitude and longitude.

The visible light heats the ground and water (and the substances on them) in various ways at various times.

Heat goes out.

At the same time (or only, for the other half of the planet) the longwave IR is going back out towards the sun since it’s warmer, and also heating the atmosphere, the other 70%. Various substances, mainly water vapor, grab the heat and send it up and down, and there’s a bunch of other reactions and physical processes happening all at once, in random and unpredictable ways, over what is basically an infinite time/space continum.

Weather happens.

All of this causes temperature and humidity differences that all react and interact with each other, the lapse rate moves around, and there’s some equilibrium of energy in and out, but because the air is sending the ground’s heat back down, it’s warmer than it would be. This energy moving around and such creates wind, clouds, rain and other effects, and the phase changes of water move this energy around, and power the lifting of the air up so the energy can get out also.

Catalyst occurs.

Eventually these processes make some things hot enough and some things cold enough in and between land and water, in and between the hemispheres, and according to season, where whatever energy is there starts a chain reaction. This seems to be time dependent. The cause is unknown. Various theories abound. Many seeming correlations of various degrees of correlation exist.

487. Erl Happ
Posted May 22, 2008 at 11:51 AM | Permalink

486 (Sam) Got to remember that the Northern Hemisphere is a great heat waster. If the orbit changed so that most insolation was recieved in July the Earth would cool. Then, the bigger the difference between January and July the more that would get wasted.

488. Sam Urbinto
Posted May 22, 2008 at 12:07 PM | Permalink

I tried to stay away from any technical stuff, even things like solar radiation, thermal radiation, transmission, emission, absorption, watts per meter squared, and the like.

But I probably should have mentioned sunlight in is about .2 to 2.6 micrometerss (UV, visible, shortwave IR) and heat out is 2.6 to 70 micrometers.

Anyway, this is interesteing, a chart of US insolation.

You can generate your own here at NREL using various criteria.

Interesting stuff here too. http://www.nrel.gov/csp/maps.html

489. Posted May 22, 2008 at 12:18 PM | Permalink

476 (Erl):

1280MHz … This wave length is capable of heating water.

I have already pointed out [with 19 zeroes] how small the energy in radio waves is. Some people have problems with numbers, so let me try this way: The sum total energy of all the radio flux that has been picked up by all radio telescopes in all the world in all the time we have been observing [more than 60 years], is less than the kinetic energy of a single snow flake falling to the ground.

490. Sam Urbinto
Posted May 22, 2008 at 12:20 PM | Permalink

2.6 to 70 micrometers IR only (long wavelength in comparison to under 2.6)

I like what it says up top, all the info on what we’re really looking at.

Model estimates of monthly average daily total radiation using inputs derived from satellite and/or surface observations of cloud cover, aerosol optical depth, precipitable water vapor, albedo, atmospheric pressure and ozone, resampled to a 40km resolution.

Erl #488 I would think more land would do things differently in the NH. I have no idea what would happen if the Earth changed orbit enough for that to happen! :)

491. Sam Urbinto
Posted May 22, 2008 at 12:26 PM | Permalink

Me #491 I meant that if the Earth’s orbit changed radically we’d have more problems to worry about than global warming. I think.

Leif #490 How hot will a cup of water get if you gave it the kinetic energy of a snowflake falling? :D

That is a tiny amount.

492. jae
Posted May 22, 2008 at 1:12 PM | Permalink

new cosmic ray study reviewed on CO2science here:

493. Posted May 22, 2008 at 1:24 PM | Permalink

We’re looking for TSI for explaining global warmings. However, the annual average of incident solar radiation reveals a very strong correlation between SI and CC:

494. Posted May 22, 2008 at 1:44 PM | Permalink

A more revealing graph:

495. Posted May 22, 2008 at 1:47 PM | Permalink

493 (jae): Whatever the verdict may be on Kirkby’s review, at least he has seen the light about one thing [page 3]:

In earlier studies, long-term solar magnetic variability was assumed to be a proxy for irradiance variability [5]. However this assumption lacks a physical basis, and more recent estimates suggest that long-term irradiance changes are probably negligible [6, 7, 8].

How can I disagree with that? :-)

496. Posted May 22, 2008 at 1:56 PM | Permalink

# 494 and 495

Me,

We considered only the insolation during clear-sky days for plotting those graphs.

497. Sam Urbinto
Posted May 22, 2008 at 2:11 PM | Permalink

#497 Nasif

Interesting. Did you also calculate what went on for non-clear sky days and then all days? Something like that along with temperature and humidity readings for all three might ‘shed some light’ on this. :) But it’s pretty clear there’s no pattern. Not really. Unless the climate is far more sensitive than it logically should be, it must be “something else”.

498. See - owe to Rich
Posted May 22, 2008 at 2:34 PM | Permalink

After I published this article at the end of January, I promised Leif that I would consider whether size of solar cycle could be used in place of length of solar cycle. I believe that others have looked at this, and though I may possibly have looked at it in a novel way, I cannot say that I have succeeded where they failed.

My “novel way” is to look for a highly non-linear function of sunspot number (SSN), strongly correlated with solar cycle length (SCL). The logic here is that if, as I believe, climate is correlated with SCL, it might be even more strongly correlated with some function of SSN, but most likely that function needs to be correlated with SCL. I used the NOAA graph for SSN.

There seems to be no evidence of the maximum smoothed SSN having any effect on HadCRUT3 temperatures. For example, from cycles 14 to 17 there was a period of warming, even though those cycles were small to moderate in size, and yet the larger cycles 18 and 19 (the biggest of all on that graph, at 180) were associated with a slightly cooling period. Therefore my non-linear function would take no significant account of cycle maxima.

It would, however, take account of the shape of cycle minima, for example by using the total number ‘DS’ of days*spots in the period bounded by a certain value of the smoothed SSN. The hypothesis for which this would make sense would be if high numbers and medium numbers of sunspots have virtually the same effect on temperatures, but that low spot numbers have a significant (cooling) effect.

But the following observations appear to argue against this hypothesis.

a. The warming in Cycles 14-17 (-.50, -.37, -.25, -.08) was accompanied by small cycles (60, 100, 80, 110) with large (to the eye) values of DS; but the warming is well explained by short cycles (except for the first one, which was very cool): 11.9, 10.0, 10.2, 10.4 years.

b. The post-18 DS looks small, which should mean warming, and yet the flat 18-20 sequence of temperatures (-.12, -.12, -.11) is well explained by increasing CO2 warming matched against lengthening solar cycles: 10.0, 10.7, 11.6 years.

c. The post-11 DS looks big, suggesting cooling, but the 11-12 sequence (-.33, -.28) is slightly warming, and Cycle 12 at 10.7 years was the shortest (=warming) since Cycles 7 and 8.

The conclusion is that if SCL affects temperatures, as I believe it does, I have not, alas, found anything in the actual distribution of spot numbers to help explain the effect. Perhaps someone else can?

To conclude, I wish to note that there may be a non-linear effect from the SCL, in that the coolest cycle, #14 which ended in 1913, was preceded by the longest cycle (in the 10-22 range) at 12.1 years. We need more data to judge whether this effect is valid – how about the data we shall get from Cycle 24 following the current cycle, which in a couple of months should break that 12.1 year barrier?

Rich.

499. Posted May 22, 2008 at 2:45 PM | Permalink

# 498

Hi, Sam!

We are working also on non-clear sky days and all days, but it will take us more time. Yeah, there is none pattern… Heh! ;)

500. Sam Urbinto
Posted May 22, 2008 at 2:57 PM | Permalink

#500 Nasif Cool!

#499 Rich Does any of that correlate to whatever happened in 1977 and 1994?

501. See - owe to Rich
Posted May 22, 2008 at 3:05 PM | Permalink

436 (duwayne)

Aha! A competitor to my CO2-solar-cycle-length model. WOuld it be possible for you to fit your model to the HadCRUT3 data in the same way that I did, and see if you can beat me on sum-squared-errors/degrees-of-freedom? You would have to concentrate just on the trend and the 60-year cycle if you are going to get anything from the 13 data points I used.

Rich.

502. Basil
Posted May 22, 2008 at 3:13 PM | Permalink

#481 Erl

That’s the El Nino signal in the early stage of the upswing.

I don’t think so. That’s another signal entirely that’s filtered out of what you are looking at.

503. Erl Happ
Posted May 22, 2008 at 5:14 PM | Permalink

503 (Basil) If you filter out ENSO you lose a vital determinant of terrestrial temperature that is an inherent part of the way the solar cycle interacts with the atmosphere to drive climate. ENSO is driven by changes in irradiance, just like the solar cycle and the annual cycle at the Equator. You are talking virgin birth or four legged animals walking on just two.

490 (Leif) Which components of electromagnetic energy derived from the sun are capable of imparting energy to the water molecule as liquid, gas or solid? What fraction of the energy available from irradiance will impart energy to water? Have you not noticed that open water warms when exposed to sunlight? Have you ever seen a map of sea surface temperatures? Your body is mostly water. Have you ever stood outside in the sun?

504. Sam Urbinto
Posted May 22, 2008 at 5:44 PM | Permalink

As far as I can tell, the main cycles of PDO, IPO and AMO make the rest do wacky things.

505. Posted May 22, 2008 at 5:57 PM | Permalink

504 (Erl):

Which components of electromagnetic energy derived from the sun are capable of imparting energy to the water molecule as liquid, gas or solid? What fraction of the energy available from irradiance will impart energy to water?

All this was gone over in great detail in 92-113. A one sentence reminder: Water is heated by the near infrared. The fraction of TSI is not important; the product of TSI and the fraction is, and that number is very small outside the visible and the near infrared. Even though 90 cents is a much larger fraction of one dollar than $10 is of$100, I would rather have the latter.

And this should have been beneath you:

Have you not noticed that open water warms when exposed to sunlight? Have you ever seen a map of sea surface temperatures? Your body is mostly water. Have you ever stood outside in the sun?

Maybe you stood outside too long in the Sun :-) Here in Houston we know better, not to do that.

506. Erl Happ
Posted May 22, 2008 at 7:26 PM | Permalink

506 (Leif)
Perhaps that comment should have been beneath me. But I took umbrage about your ‘Some people….’ statement. Too dismissive.

I see a relationship between the two variables graphed below. I have seen this relationship for a long time and it has been a battle to get to this point.

You and I know that the amplitude of the variation in sunspot activity is much greater than is shown on any of these figures. We also know that when we plot the series over any more than a few years the relationship seems to disappear. It is quite apparent that there is a strong relationship between the two variables when we look at shorter time intervals.

Why does the relationship appear to break down when plotted over long time scales? The relationship is conditioned by a thermostatic action in the Earths atmosphere. When the thermostat (the clouds) so determines, the effect of increasing irradiance goes for nought. That’s when the two series diverge as in 1997-98. Elsewhere the relationship is pretty good. A one month lag is plausible.

My queries to you:
Would an impartial observer need a statistical test of significance to agree that sunspot activity drives temperature?
Does it appear as if the thermostat is generally working alright so as to keep the climate equable?
What do you think the thermostatic action could be?

My guess is that it is a component of electromagnetic energy acting on water in the atmosphere. If the component is very small then the atmosphere is hypersensitive to it. My guess is that the thing is driven by a not insignificant portion of the spectrum acting upon a relatively small but very influential amount of water vapour at the 250mb pressure level. I will post on that in due course.

By the way, it’s raining cats and dogs here. It is the first big front of the winter. In the East of Australia records for cold temperatures are being set. Here is a report http://www.eldersweather.com.au/breakingweather.jsp

507. Basil
Posted May 22, 2008 at 8:01 PM | Permalink

#504

Erl,

I wouldn’t say I’m ignoring ENSO entirely. I have a couple of things going on in the procedures I’m using. If you look at the MTM spectrum in #404, you’ll see frequencies of 3.63 and 4.74 years that I might imagine are related to ENSO. That’s been filtered out in the figure #480 so I can highlight the bidecadal signal, which just happens to be the focus of the research these charts represent. I actually agree strongly with statement of yours I quoted:

The Earths surface and the atmosphere show a Solar Cycle fluctuation, an annual fluctuation and an ENSO fluctuation.

It just happens that the focus of my analysis has been on the first of these — the solar cycle — and it leads me to the conclusion that the dominant cycle is the 22 year cycle, which is why I asked about it.

508. Posted May 22, 2008 at 8:23 PM | Permalink

507 (Erl): “problems with numbers”. I was not being dismissive at all. Just realizing that you have said many times that putting numbers to correlations etc did not work for you. Even in this very 507 you question the “need for a statistical test of significance”. Anyway, I was trying to convey in a striking way the minuteness of the radio flux, and all I got for my trouble was being accused of being dismissive. Ah, well, that’s life, I guess. If you right-click on the graph below, select ‘properties’ and look at its filename you’ll see how non-dismissive and patient [stubborn?] I have been with you [and everybody else for that matter].

Why does the relationship appear to break down when plotted over long time scales?

This often happens in this type of search for correlations. Many, many claimed relationships have been ‘shipwrecked’ on the those shoals. It most often means that the relationship is not real, is spurious, is in the eye of the beholder, and that it is time to scratch that one off, and move on.

It is quite apparent that there is a strong relationship between the two variables when we look at shorter time intervals.

No, that is not apparent at all. Now, if you have two series each varying randomly, you can also find such visual match-ups. I just generated two series of random numbers [each data point is the sum of five RAND() function calls]:

Green ovals mark a strong apparent relationship. But if you submit the series to statistical significance tests you’ll find that there is no relation [as there shouldn’t be with random numbers].
So, the answer to the first query:

Would an impartial observer need a statistical test of significance to agree that sunspot activity drives temperature?

should be a clear yes, unless the observer doesn’t care about the science, and simply wants to subscribe to a viewpoint.
If the significance tests turn out not to support a strong correlation, then the two other queries are moot.

509. Posted May 22, 2008 at 8:30 PM | Permalink

510. Erl Happ
Posted May 22, 2008 at 8:49 PM | Permalink

509 (Leif)
As I have said before, where the relationship is dependent on complex interactions amongst a number of variables you will throw the baby out with the bathwater. That sort of science would paralyse you. You would never make a farmer.

511. Posted May 22, 2008 at 8:58 PM | Permalink

511 (Erl): Analyzing “relationships dependent on complex interactions amongst a number of variables” is what I have been doing for some 40 years, and while I may never make a farmer, I am pretty good at what I do. I know that you don’t care much for math and such, but this old paper of mine is a very good illustration of how to unravel the complex interactions:
http://www.leif.org/research/Geomagnetic%20Activity:%20Dependence%20on%20Solar%20Wind%20Parameters.pdf
There is only minimal math and simple physics involved and it should be accessible. Please do me the favor of seeing how such an analysis can be done without too much paralysis.

512. Posted May 22, 2008 at 9:31 PM | Permalink

The interaction between the magnetized solar wind and the Earth’s magnetic field is very complex and involves quantities that are normally not encountered in climate, such as conducting plasma and electrical currents. Here are the principal players:

1. The interplanetary magnetic (B) flux per unit time and area, F = B V
2. The solar wind momentum (n V) flux per unit time and area, P = (n V) V
3. The angles between the Earth’s magnetic field (M) and the IMF direction (α)
and flow direction (ψ)
4. The time scale of interest (hours to days) and the variability within that

A total of eight variables, producing a bewildering array of effects that it took 250 years to understand the basics of. I have shown it before, but it is useful to show it again:

What you see is geomagnetic activity [am-index] calculated [red curve] using the analysis I referred to compared to actually observed [black curve]. The disagreements for small values of am are mainly due to the problems measuring the small values of am. So, it is possible to simultaneously juggle seven or eight parameters and unravel the complex relationship. The human brain cannot handle more than seven things at the same time, so cannot see relationships involving more parameters than that. So, 1: it is possible, 2: I can do it without paralysis and without losing the baby, and 3: you have to give me some credit for being able to handle complexity.

513. Posted May 23, 2008 at 4:25 AM | Permalink

513(Leif) certainly gets a “wow! that’s impressive” from me.

Answering my question at #477 should be childs play by comparison.

I would however be interested to know your a priori reasons for rejecting the possibility that the more massive extra-solar elements of the solar system might in concert affect the sun’s angular momentum sufficiently to cause ripples in the motion of the plasma and the ‘solar conveyor belts’ where most sunspot activity occurs.

Especially considering how patient and non-dismissive you are.

I know it’s barycentric nonsense but please indulge me anyway. :)

514. Posted May 23, 2008 at 4:35 AM | Permalink

514 (Stoller): For the last time: the Sun is in free fall about the barycenter and just like an astronaut in free fall about the Earth feels no forces, hence no effects on the ‘conveyor belt.

515. Posted May 23, 2008 at 6:49 AM | Permalink

515(Leif) Thanks for that. Is the degree of movement about the barycenter sufficient to have any effect on Earth’s orbit? What cause the changes in the eccentricity of earth’s orbit described by Milankovitch?

516. Posted May 23, 2008 at 7:07 AM | Permalink

516 (Stroller): The Earth is also in free fall so no changes from that. The slow changes in the orbital elements of the Earth, is caused by gravitational perturbations from the other planets and the Moon. As so often, Wikipedia has good information: http://en.wikipedia.org/wiki/Milankovitch_cycles

517. Posted May 23, 2008 at 7:19 AM | Permalink

# 515

Leif, Stroller,

I will take advantage of these brief moments before I leave to the station. I’d like to make a small detailing from the Leif’s paper which is very clear. Leif says, “Thus, the magnetosphere acts like a magneto-hydrodynamic generator, converting kinetic energy of moving plasma into electrical energy which then ultimately is dissipated within the magnetosphere or in the boundary layer.” This assertion could make the reader thinks that the electrical energy goes back to the magnetosphere and recycles. Actually, the energy that is dissipated within the magnetosphere or the boundary layer is transferred to the gravitational field or it is transferred to the void, where probably it is packaged in muons and other quanta. I know that Leif can give a better explanation for the apparent “lost” of that enormous load of energy.

518. kim
Posted May 23, 2008 at 7:42 AM | Permalink

515 (Leif) You claim that an object in free fall feels no forces, but invoke forces to explain the changes in free fall of the Earth.
=========================================================

519. Posted May 23, 2008 at 8:23 AM | Permalink

517(Leif) Thanks again. I’ll take some time out to do some numbers and think about the components of the force the sun exerts on the planets to keep them in their orbits. And the equal and opposite force they exert on the sun. I suppose the key issue is purterbance to trajectory and whether it is non-negligible under certain circumstances.

I would suppose the the fluid dynamics of the sun is key too, considering that it’s motions about the barycentre happen in much tighter arcs relative to it’s diameter than the planets in their orbits. Anyone got a modeled viscosity? :)

520. jae
Posted May 23, 2008 at 9:51 AM | Permalink

Leif, 509:

This often happens in this type of search for correlations. Many, many claimed relationships have been ‘shipwrecked’ on the those shoals. It most often means that the relationship is not real, is spurious, is in the eye of the beholder, and that it is time to scratch that one off, and move on.

Yeah, ceteris paribus (bender taught me this term). Then you add chaos, and things really get confusing.

521. Posted May 23, 2008 at 1:41 PM | Permalink

519 (kim):

invoke forces to explain the changes in free fall of the Earth

where did I say that?

522. Posted May 23, 2008 at 1:52 PM | Permalink

518 (Nasif): the electric currents close through the ionosphere and dissipates some of their energy there; does not disappear into the void.

523. Mark_T
Posted May 23, 2008 at 2:22 PM | Permalink

I just read from NASA that there are now 3 red spots on Jupiter and they say the likely cause is warming at the equator. So now there are 3 planets that show warming. I’ve been reading through this thread (still on page 1) and I am in no way qualified to question Leif. But I find it hard to believe that the sun isn’t the factor in the recent warming of Earth, Mars, and now apparently, Jupiter.

Does anybody know if NASA tracks the temperature of Mercury and Venus (I can’t find anything)? It’d be interesting to see if the temperature of those planets increased in recent years.

Mark

Source

524. Posted May 23, 2008 at 2:27 PM | Permalink

520 (Stroller):

I would suppose the the fluid dynamics of the sun is key too, considering that it’s motions about the barycentre happen in much tighter arcs relative to it’s diameter than the planets in their orbits.

So, if you move the barycentre closer to the sun, e.g. so close that it were only 1 inch from the centre of the Sun [e.g. by moving the planets away] the arc would soooo incredibly tight that maybe it would disrupt the Sun? As I said before: no more barycentric, tidal, gravitational, etc, ruminations in this thread.

525. Posted May 23, 2008 at 2:43 PM | Permalink

524 (Mark):

find it hard to believe that the sun isn’t the factor in the recent warming of Earth, Mars, and now apparently, Jupiter.

And this in spite of the fact that the Sun has been ‘cooling’ the last 50 years? I say ‘cooling’ in quotes, because it is not by much, but, the Sun has not been warming.

526. Posted May 23, 2008 at 3:30 PM | Permalink

469 (bender,me): I’m still waiting on enlightenment…

527. Mark_T
Posted May 23, 2008 at 4:33 PM | Permalink

526 Leif:

Hard for me to let go of all the scientific sources of TSI and sunspot data that I have read which show the opposite of what you are now saying, especially if Jupiter has started to warm up. As a layman, it just seems to me that the graph of TSI and sunspot data from the NOAA is the reason that the 3 planets are showing signs of global warming. Not to mention that the 3 planets all have the same common energy source.

And I’m still reading through the 1st page (when I have time). I don’t recall seeing anything in it yet that showed a revised TSI graph of the last 50 years. I may have overlooked it though…

Mark

528. Posted May 23, 2008 at 4:56 PM | Permalink

528 (Mark):
Don’t just take my word for it. At the latest SORCE [SOlar Radiation and Climate Experiment] meeting in 2008, Judith Lean [the foremost expert on TSI] presented this slide:

It shows the observed TSI since 1978, and a sunspot record. First note that the sunspots have declined since the 1950s [other solar indices have too], and that perhaps TSI has a well. In the lower right-hand corner Judith summarizes the state-of-the-science: “longer-term variations not yet detectable – do they occur?”.
So, the available evidence is that solar output has not increased in any detectable way, and that it is therefore hard to ascribe warmings on three planets to a solar cause that has not been observed.

529. Mark_T
Posted May 23, 2008 at 5:12 PM | Permalink

528 Leif:,

Thanks for the graphs. I noticed from the one in the upper right that the TSI is about 1361 w/m2 while the one to the left of it shows the TSI around 1365.5. What am I missing here? There is about a 5w/m2 delta between the two graphs. Maybe I’m incorrectly assuming the green graph is TSI? I apologize if you already answered this in an earlier post.

Mark

530. Sam Urbinto
Posted May 23, 2008 at 5:21 PM | Permalink

I just thought of something. Steel is not mostly water, but it gets hotter in the sun than your skin does.
:D

Leif #527 I’m with you, I don’t see a pattern either. (second-order CRF ‘peaks’ (buried in the CRF troughs) coincide with the TSI peaks.)

Leif #510 What’s about it about my posting the thing showing that a lot of stuff that’s inter-related going on in a complex random dynamic manner that can’t be decoupled and shows nothing much of interest except that a lot of stuff that’s inter-related going on in a complex random dynamic manner that can’t be decoupled are you talking about? :)

531. Sam Urbinto
Posted May 23, 2008 at 5:27 PM | Permalink

Maybe I should have said:

What’s “about it” that you are talking about? Regarding my posting the thing showing that a lot of stuff that’s inter-related going on in a complex random dynamic manner that can’t be decoupled and shows nothing much of interest except that a lot of stuff that’s inter-related going on in a complex random dynamic manner that can’t be decoupled?

532. Posted May 23, 2008 at 5:40 PM | Permalink

530 (Mark): the 5 W/m2 difference is the difference between different instruments. It is VERY hard to measure the ‘absolute’ level, and the uncertainty is 5-10 W/m2. For each instrument, the ‘relative’, i.e. day-to-day accuracy is a LOT better, less than 0.5 W/m2, and the ‘precision’ which is what matters is 0.007 W/m2, so they can quote [a reproducible] result to three decimal places, e.g. for May 15th it was 1360.949 W/m2. So don’t worry about the constant 5W/m2 difference. What matters is the variation on top of that.

533. Posted May 23, 2008 at 5:48 PM | Permalink

# 523

Leif,

the electric currents close through the ionosphere and dissipates some of their energy there; does not disappear into the void.

I didn’t say energy disappear in the void; I said:

Actually, the energy that is dissipated within the magnetosphere or the boundary layer is transferred to the gravitational field or it is transferred to the void, where probably it is packaged in muons and other quanta.

Probably it is used to form particles. That’s what I said.

534. Posted May 23, 2008 at 6:11 PM | Permalink

531 (Sam): the semi-philosophical color of your missive didn’t do anything for me, but I agree fully that there are many interrelated aspects and effects, and causes, that feed on each other. I’m have given an example [geomagnetic activity] that also exhibits those characteristics, yet have [finally] become understood by careful analysis, once we had enough good-quality data [and that was key!]. Erl’s assertion that he by eye can discern complicated relationships that would paralyze a scientist [or is it just me, he means] is just plainly false. Maybe the same is the case with bender’s. I have little patience with claims of effects that are ‘plain’ but that I cannot see.

535. Posted May 23, 2008 at 6:14 PM | Permalink

534 (Nasif): what you said makes as little sense as what I thought you said. But I don’t think it is important enough to continue to quibble about.

536. Posted May 23, 2008 at 6:20 PM | Permalink

531 (Sam): You might like this quote from an 18th century ‘geomagnetician’ [Olaf Hjorter] that complained about the headache he almost got from trying to make sense of, or even just describe, the ‘irregular regularities and the regular irregularities’ that the record presented him with.

537. Posted May 23, 2008 at 6:25 PM | Permalink

# 536

Leif,

what you said makes as little sense as what I thought you said.

Then you have to demonstrate it.

538. Sam Urbinto
Posted May 23, 2008 at 6:36 PM | Permalink

#523 and #534 Leif and Nasif

Maybe we can say that energy is neither created nor destroyed, it just moves from one substance to another, with some ending up in the void of non-relativistic quantum dynamics?

Next, we need mathematical proof that 2+2 != 5 and that a lizard is not a dog. :D

Leif

#535 Yes, I likewise have issues with trying to act like there’s been any kind of lucid type of anywhere near proof that anomaly=temperature=energy. But I do appologise for my semi-philosophically colored missives.

#537 Nice quote! But are “irregular regularities and the regular irregularities” anything like Wahl and Ammann 2007 versus Ammann and Wahl 2007 by chance? :)

539. Posted May 23, 2008 at 6:50 PM | Permalink

538 (Nasif): the electric currents are not ‘energy’ as such. When a conductor [solar wind plasma] moves into a magnetic field [the Earth’s] an electric field is set up [E = v x B]. So there will be this electric field across the Earth’s magnetosphere directed from the dawn side to the dusk side. This field will move charged particles [and there are many of those out in space] in the direction of the field [that is the definition of an electric field, BTW]. Moving charges are called electric currents. These currents will flow until the charges bump into something that doesn’t move readily, like the neutral atmosphere, then they die out [dissipate] and the energy of the moving charges is converted into heat [this is what happens in an electric heater or cook-top]. No relativistic muons or other exotic animals involved.

540. Posted May 23, 2008 at 6:53 PM | Permalink

# 539

Sam,

Yes… We can deform quantum dynamics into classical dynamics without altering unstable lizards and maintaining Hilbert space dimension and constants fixed . :)

Anyway, if the candle doesn’t warm the room, what is doing it?

541. Sam Urbinto
Posted May 23, 2008 at 6:57 PM | Permalink

Leif #540 “No relativistic muons or other exotic animals involved.” Is a relativistic muon a kind of cow that lives in magma, or a car that goes 1000 miles an hour as long as it’s going downhill in a 1100 mile an hour wind?

Nasif #541 If the candle is not warming the room, what it’s doing is heating air that is being taken out of the room.

542. Posted May 23, 2008 at 7:03 PM | Permalink

541,542 (Sam,Nasif): enough of that 2+2!=5 stuff. #540 ought to be sufficient.

543. Posted May 23, 2008 at 7:14 PM | Permalink

# 540

Subatomic particles enter the Earth’s atmosphere carrying energy, that’s correct; however, we ignore the origin of many cosmic particles that enter the Earth’s atmosphere. Not all is solar under the Sun (Gulp!). The photons that escape towards the void are not “destroyed”, but absorbed by other particles which are plaguing the classical void; that’s why I said that they could be forming particles, BTW. OTOH, ultrahigh-energy cosmic rays that are accelerated by the solar wind cause also a energetic “rain” of particles which energy is transferred to the molecules of water vapor, liquid water and land.

544. Posted May 23, 2008 at 7:16 PM | Permalink

# 543

Leif,

It’s ok. Sorry…;)

545. Posted May 23, 2008 at 7:23 PM | Permalink

# 544

Me,

that’s why I said that they could be forming particles

Leif, just give me a chance to say that there are incoming protons to Earth that carry energy 100 million times higher than the “usual” protons with the solar wind.

546. Posted May 23, 2008 at 9:00 PM | Permalink

546 (Nasif): you can say that [once :-) ] as long as you realize that there so few of these high-energy particles that their contribution to the heat budget of the atmosphere is nil.

547. Steve McIntyre
Posted May 23, 2008 at 9:06 PM | Permalink

Nasif, please. No advocacy here of personal theories of electric fields, quantum mechanics or such things. I’ve been leaving these threads alone, but you know my attitude to persona

548. Posted May 23, 2008 at 9:22 PM | Permalink

# 547

Steve… Yes I know it, and it’s ok. I’m not playing against rules; however, those are not my personal theories, but observed and observable phenomena. It has to do with Leif’s hypothesis, although it had seemed to be odd by that so frequent phrase “no-sense”.

549. kim
Posted May 23, 2008 at 9:36 PM | Permalink

522 (Leif) Are ‘gravitational perturbations’ not forces?
===============================

550. Posted May 23, 2008 at 10:11 PM | Permalink

550 (kim): this is not the place for gravity 101. But, consider the astronaut in free fall around the Earth. He will feel no force, yet, the gravitational force of the Earth bends the path he is following from a straight line to a circle around the Earth, so the astronaut is continually acted upon by the gravitational force, yet because he is in free fall, does not feel any forces. There is no contradiction in this. But please, no more on the barycentre/gravity stuff.

551. duwayne
Posted May 23, 2008 at 10:46 PM | Permalink

478(stroller):
You ask for my view on whether the ENSO and 60-year cycle influences on Global temperatures will change significantly if the future temperature trend anomalies are either much higher or much lower than present. For what it’s worth, I’m projecting a flat Global temperature trend through 2067 (0.3C UAH anomaly) and then a trend rise to 0.7C by 2097 on the basis that ENSO and the 60-year cycle will continue to function similarly to the past 60+ years. If the future trend anomalies differ greatly from my projection then either ENSO or the 60-year cycle or both may have changed patterns but I’m not going to pretend that my scientific understanding of the basic drivers of these phenomena is sufficient to make a useful prediction of what these pattern changes would likely be.

502(See – owe to Rich):
Can you point me to the details of your CO2-Solar-cycle-length model?

552. maksimovich
Posted May 23, 2008 at 11:01 PM | Permalink

43(Bender)

the flip to positive in 1920 casts doubt on the AGW exlanation, no?

Conversely the return to its previous state does,As Schmidt and Humbert tell us the Climate state “forgets” its initial conditions(They also forgot Kolmogorov)

P+=P- = 0.5

553. Erl Happ
Posted May 24, 2008 at 12:59 AM | Permalink

Leif
When I showed this graph to my wife she said ‘Shut up and let me work out what is happening for myself’.

TLT data relates to the tropical lower troposphere over the ocean. http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt.

It seems that correlation coefficients have their uses. The correlation relates 12 months sunspot data to 12 months temperature data and the plot is centred on the 7th month of the twelve. So, its a running correlation. Here is one case where low correlation coefficients are just as important as high correlation coefficients. Every one is important to the story.

My comment at this stage is encapsulated in the two ovals marked on the graph. I am very interested in what people have to say about this data.

554. Andrey Levin
Posted May 24, 2008 at 3:20 AM | Permalink

Re#516 (Stroller):

Leif wrote:

(#515) For the last time: the Sun is in free fall about the barycenter and just like an astronaut in free fall about the Earth feels no forces, hence no effects on the ‘conveyor belt.

(#517) As so often, Wikipedia has good information

Wikipedia has another “good information” on the subject:

http://en.wikipedia.org/wiki/Tidal_force

I hope it will help.

555. lgl
Posted May 24, 2008 at 6:08 AM | Permalink

#554 Anrdey
If not forbidden by Leif I would have asked him, if tidal resonance over millions of years can lead to tidal locking between the Earth and the Moon for instance, why isn’t it then plausible that there is a similar locking between the Jupiter-year and the ‘Sunspot-year’? with some wobbling of course from the other planets.

556. Posted May 24, 2008 at 6:47 AM | Permalink

556 (lgl): it is a question of numbers. The tidal forces are proportional to the mass divided by the distance cubed. Venus actually exerts just as big a tidal influence on the Sun as Jupiter. Here are the numbers [in units of Earth’s]:
Planet Mass Distance Tidal effect
Venus 0.82 0.72 2.20
Jupiter 318 5.2 2.26
In any rate the forces are so small to not have any effect. To take your example, it is the Moon that gets locked into a rotation equal to its orbital period, not the Earth, so your question should have been: is it not plausible that Jupiter gets locked into a rotation equal to its orbital period [12 years]? Clearly the answer is no.
Had Jupiter been a lot closer, it would have a big tidal effect and would surely influence solar activity. There are actually stars with large planets close in, and their stellar activity is indeed observed to be modulated by their planet. But, Jupiter’s [and Venus’] effect has been looked for since the solar cycle was first recognized more than 150 years ago, and no effect has been detected.

557. kim
Posted May 24, 2008 at 7:47 AM | Permalink

556 Leif) Do we know what is the mechanism for the modulation of stellar activity by the large, close in, planets?
=============================

558. Posted May 24, 2008 at 7:53 AM | Permalink

558 (kim): No, we do not know, but we can speculate that rotation and circulation [and hence dynamo processes] are involved.

559. lgl
Posted May 24, 2008 at 8:08 AM | Permalink

Leif, You are of course right, but I actually mean solar activity locked to Jupiters motion (not to use the b-word), and it’s of course not a rotational locking. A small portion of the solar mass in phase with Jupiter if you like, we are dealing with a very small variations compared to the total TSI. I assume the stars with large planets close in you mention show a much larger variation. The main point is tidal resonance. Your argument is something like: The Earth/Moon tidal force is far to small to stop the moon’s rotation in one year. That is of course true, but what about in 1 bill. years.

560. Posted May 24, 2008 at 8:50 AM | Permalink

560 (lgl): By that argument we night have to wait a billion years to catch the first sunspots caused by Jupiter. Enough, now, of “could it not be, might it not be, is it not…,” etc. There is a lunar [tidal] influence on geomagnetic variations. It is very small, but with long enough and good enough data it has been possible to tease it out of the noise. There is absolutely also a planetary tidal influence on solar activity, but it is even smaller and not detectable with current technology and the time span of observations we have. Perhaps in a thousand years we’ll have enough data and so much better data, that we’ll be able to tease that influence out of the noise. For now, let it be.

561. kim
Posted May 24, 2008 at 9:22 AM | Permalink

560 (Leif) Thanks. I know you’ve said it repeatedly, but this last comment is the most clear to me.
==========================================

562. Posted May 24, 2008 at 1:28 PM | Permalink

562 (kim): just that you [and others] really understand this issue: the tidal forces have absolutely nothing to do with the orbits or the barycenter. The tidal force arises solely because the Earth [or the Sun] has a finite size. If you shrunk the Earth [or the Sun] to zero size, the tidal forces would completely disappear. Consider three small touching and lined-up spheres S1, S2, and S3, and a fourth sphere J at some distance along the same line out beyond S3. Then, since the gravitational force decreases with distance, the force F1 between S1 and J is smaller than the force F2 between S2 and J which in turn is smaller than the force F3 between S3 and J: i.e. F1 < F2 < F3, and therefore the differences: F1 – F2 < 0 and F3 – F2 > 0. this means that seen from S2, the spheres S1 and S3 are both pulled away from S2, hence a rising tide on both the side towards J and on the side away from J. If the spheres are made smaller and smaller, the differences become smaller and smaller and will disappear in the limit of zero sizes. And, again, this has nothing whatsoever to do with the orbits and free fall and barycenter motion or any of that.

563. Andrey Levin
Posted May 25, 2008 at 5:03 AM | Permalink

Re#560 (lgl):

There are plenty of speculative theories trying to connect solar activity with orbital tidal forces (undeniably existent and possibly important) of planets. For example:

http://www.tkcs-collins.com/truman/conjunct/conjunct.shtml

http://wefooleffect.com/WFE%20on%20Solar%20system%20formation.htm

However, I agree with Leif on this one: when I read such web sites my BS filter redlines.

564. lgl
Posted May 25, 2008 at 7:59 AM | Permalink

#564 Andrey
(please note Leif, I had actually stopped this non-sense but now you and Andrey got me started again :-)
This is about tidal force, not gravitation. When Jupiter is on one side of the Sun and the rest of the planets are on the opposite side, the planets’ center of mass is much closer to the Sun compared to when all planets are on the same side of the Sun, giving a tide on the Sun. Since you both are so sure this is non-sense I guess you can easily show that this difference in tidal force is minimal. And the hugh difference in the Sun’s mass compared to the individual planets or their total mass is not that relevant here, it’s the distance between the planets center of mass and the Sun that’s important.

565. pochas
Posted May 25, 2008 at 9:50 AM | Permalink

Lief:

Since tidal forces tend to distort an affected body into an ellipse, I thought I would look for papers that
have noted distortion in the Sun’s shape, and there are some:

Sveshnikov (2002):
http://www.maik.ru/abstract/letters/2/letters0115_abstract.pdf

Barkin (2007): (This one should really draw catcalls)
http://www.cosis.net/abstracts/EGU2007/09127/EGU2007-J-09127.pdf

566. jae
Posted May 25, 2008 at 11:09 AM | Permalink

Leif: Welikerocks flagged this over on the BB. Any comments?

567. Sam Urbinto
Posted May 25, 2008 at 5:03 PM | Permalink

Leif #543

Sorry if I’m being too bovine obvious!

568. Posted May 25, 2008 at 7:18 PM | Permalink

565 (lgl) and all the rest: instead of me going through a long explanation, I’ll just refer to a good one at http://mb-soft.com/public/tides.html
Working through the math one gets that the tides by the Moon on the Earth is 367 mm high [mm = millimeter = 1/25.4th of an inch]. Inserting values for the Sun and Jupiter one gets a tide 0.47 mm high [if I have done the math correctly – you might want to check this !]. [I think I have said this several times by now]. Put all the other planets where you want, their individual times will be less that this. Venus’ is almost as high as Jupiter’s. All together, the tidal effects are of the order of 1 mm. Compare this to the convective overturning of the photosphere in Texas-sized granules moving at 1-2 km/sec [that is 1000000-2000000 mm/sec] and you might be able to see that planetary tidal effects can be ignored. Of course, there are always people that have problems with numbers. You might be one of them; if so, think of a large truck running over an ant at 100 miles/hour. The effect of the ant on the trajectory of the truck is relatively much larger than the tidal influence of the planets on the matter of the Sun. This much was known to Isaac Newton in the 1700th Century. Have we not progressed since then? Can we not stand on the shoulders of that giant? And can we not get off this topic, once and for all? BTW, the tides by Jupiter on the Earth is 1/500 of a millimeter [at closest approach]. So, NO MORE TIDAL STUFF! Unless to demonstrate an error in my calculation.

569. Raven
Posted May 25, 2008 at 11:36 PM | Permalink

Leif,

Do you have any thoughts on the expanded storms on Jupiter:

[quote]Marcus bases his prediction in part on the observed appearance of three large white ovals, all thousands of miles across, on Jupiter in 1939, and the unexpected disappearance of two of them between 1997 and 2000, during which time they all merged into one.
Other smaller spots have gone away since.
Marcus says the demise of still more spots over the next seven years will mark the end of a newly proposed, [b]70-year climate cycle.[/b][/quote]

The timing and the length of the cycle got my attention.

570. Erl Happ
Posted May 26, 2008 at 12:40 AM | Permalink

512, 513 (Leif)
Rest assured that I have the greatest admiration for your efforts in untangling the phenomena by which the magnetic (and other) influence of the sun impacts the Earth system. I would not be spending my time here otherwise.

This long post is devoted to exploring a rationale for the pattern of changing correlation coefficient of TLT temperature and sunspot activity that I showed in # 554 and show again in an improved (I hope) presentation below. The correlation is good at some points and not at others? Apparently sunspot activity can promote cooling. At some times there is little response either way? But the pattern is far from random.

The temperature variable relates to the mid tropical troposphere over ocean between 20°N and 20°S centred at about 5Km in elevation as described at http://www.remss.com/msu/msu_data_description.html.

If the temperature data for a year is compared to the sunspot data for the 12 month period retarded by one month, the average correlation improves by 5% but for the sake of clarity I have adopted the more direct approach.

The TLT temperature response to sunspot activity is most often positive in SH summer (November to March- pink rectangles). Facilitating this expression of solar activity may be the 7% greater irradiation (as watts per square metre) available in January over July.

Why is the response poor in July? In NH summer, temperatures in the tropical atmosphere are driven primarily by the response of the atmosphere to the arrangement of land and sea rather than sunspot activity. When the sun is over the tropic of Cancer tropical temperatures at 5Km in the atmosphere are 5°K greater (average) than when the sun is over the tropic of Cancer. The Brightness Temperature Climatology maps show the origin of this effect. There is a fall in cloud cover in the Northern hemisphere associated with atmospheric heating and a higher proportion of available radiation therefore reaches the surface of the Earth. This heating effect is unrelated to changes in solar activity. It is possible that the surface intensity of solar radiation might approach or even exceed that for the Southern Hemisphere despite the irradiance advantage in January due to orbital influence. The upshot is that warm air is drawn towards the Equator via the North East Trade winds and a 5K rise in temperature occurs in the TLT at 5Km elevation.

This enhanced radiation enjoyed at the surface of the Northern Hemisphere can not be stored as kinetic energy in the oceans, to the same extent as it is in the Southern Hemisphere. The heat is dissipated via surface contact, evaporation, convection, release of latent heat of condensation, decompression and long wave energy emission. Of these processes surface contact with warm land masses is obviously important, evaporation relatively less so due to the relative dryness of land, and Long Wave Radiation therefore relatively more important. The signature of enhanced OLR in NH summer can be seen in hovmoller diagrams of the atmosphere at 15km and above due to the interaction of OLR with ozone. In view of the above, the response of TLT temperature to sunspot activity is swamped in NH summer by land mass heating and marked atmospheric warming that is associated with factors other than sunspot activity. This accounts for a better correlation between sunspot activity and TLT temperatures in January but it does not account for the negative correlation in mid year. More of that later.

When sunspot activity is increasing fast at the start of a new solar cycle atmospheric conditions are ripe for expression of sunspot activity in TLT temperatures. Precipitable moisture in the atmosphere rises during the upswing of the solar cycle and falls in the decline. (Hovmoller diagrams below). A period of negative TLT anomalies and generalised ocean cooling frequently precedes the upswing in solar activity. This slows evaporation from the oceans and tends to precipitate moisture from the atmosphere. Slight warming of an atmosphere that is already dry can readily reduce relative humidity and cloud cover admitting more solar radiation. The process of ocean warming in the tropics gets underway. Humidity then increases until cloud cover is restored, but the process is retarded by the low level of atmospheric moisture at solar minimum and a constant loss of moisture to higher latitudes. The period of tropical warming that ensues is frequently the strongest experienced during the solar cycle. This is the first big El Nino of the solar cycle.

However the process of warming can not go on for ever. The oceans tend to retain heat much more than the land and this increases humidity against the further pressure of advancing sunspot activity. Such was the degree of ocean warming that occurred in 1997-8 that the well documented increase in cloud cover from 1999 became sufficient to impose a period of cooling through to 2000. The correlation of sunspot activity with TLT is therefore strong at the initial stage of the upswing and weak when cloud cover advances. This is an illustration of the Earths thermostat working to eliminate heat gain. This relatively regular El Nino event can be traced in the behaviour of the Southern Oscillation Index at solar minimum over the period of record since 1880. It is very apparent in the data on cycles 22 and 23 that I present below.

For sunspot activity to be expressed in positive correlations with TLT temperatures other than during SH summer, the increase in sunspot activity must be relatively dramatic and it will assist if the atmosphere has been subjected to a period of cooling and drying beforehand. The four occasions marked with black rectangles in the figure below fit the bill.

The reason for strong negative correlations in mid year could be as follows. The zone of temperature measurement defined as ‘tropics’ lies between 20°N to 20°S (in this case). In NH summer the intertropical convergence moves strongly north under the influence of the forces shown in the Brightness Temperature Climatology maps below. ‘The tropics’ stays where it is and is increasingly under the influence of cooling south East trades, a growing seasonal cold water pool in the East Pacific and cooling tropical waters at all longitudes. Just as seasonal heating promotes a northward shift in the pressure belts in the atmosphere, anomalous heating events due to sunspot activity that impact on top of the seasonal event will promote a further northward shift in the cooler air from the Southern Hemisphere. Thus, increased sunspot activity can result in a fall in temperature in the tropics.

Tropical temperatures are therefore driven by
1. Seasonal factors associated with the tilt of the Earths axis and the rotation of the Earth around the sun.
2. ‘ENSO’ events due to rapidly advancing sunspot activity.
3 The solar cycle itself.

The change due to the solar cycle is well illustrated in the hovmoller diagram of surface temperatures in the 10°N to 10°S latitudes shown below. The advance to solar maximum involves a continuous warming of the Indian and Pacific Ocean east of 180° longitude and a gradual diminution in the extent of the cool waters in the Eastern Pacific. It involves an increase in precipitable water and relative humidity. There is a reduction of the pressure difference between the ‘hot pool’ in the west and the ever diminishing cold pool in the Eastern Pacific (not shown). In the decline of the cycle these changes are reversed. This pattern can be demonstrated, with obvious variations due to ENSO activity for solar cycles 21 and 22. This is a repeating, organic relationship that matches the solar cycle.

Nowhere on the surface of the Earth is it possible to derive an unambiguous relationship between sunspot activity and temperature. The atmosphere is so thin and so much conditioned by surface temperatures that the same statement applies in relation to the atmosphere at higher altitudes. However, that part of the atmosphere that responds to sunspot activity in such a way as to change the reflectivity of clouds to solar radiation most closely reflects solar activity in its temperature change. That appears to be at about 250Mb pressure or 9Km in the tropics. Heating occurs predominantly in December to May. This altitude is about the limit of where appreciable water vapour can be found over the warmest ocean areas. Here water vapour is condensed as tiny but highly reflective crystals of ice. The coming and going of these crystals affects the rate of reflection of solar radiation. This zone heats more because it has higher moisture content. It lies above the warmer portions of the oceans as is quite obvious in the hovmoller diagrams below. The episodes of heating in this zone are traced with green rectangles on the diagrams and these rectangles are transferred elsewhere so that the pattern of temperature change can be compared between the levels.

Is the 250mb warm zone due to a response to outgoing long wave radiation? No. The hovmoller diagram of altitude of 15 km (100mb pressure), where there is a small proportion of ozone, shows a strong response to northern Hemisphere seasonal warming and to the OLR that emanates from the cold zone in the Eastern Pacific, reversing the pattern of temperature distribution that occurs at lower altitude. The warm water zone exhibits cool temperatures at the 100mb level due to the relative deficiency of OLR over the hot pool. The cold water zones show warming at the 100mb pressure level due to the dominance of OLR in the cooling process of these locations. The cooling of the ‘hot pool’ is accomplished via evaporation, release of latent heat of condensation and decompression in the lower atmosphere. This is a good illustration of the importance of the spatial arrangement of land and sea and warm versus cold sea in determining the processes responsible for heat transfer. If one looks, one can see a signature of the Andes at the highest level of the atmosphere. As I remark earlier, the atmosphere is thin.

These observations have relevance in the study of heat storage by the oceans, the transfer of energy from place to place, the response in terms of hemispherical and global temperature and our thinking about how the Earth dissipates heat to space. They are crucial to the study of the relationship between the sun, terrestrial temperature, ENSO events, the PDO and other so called ‘oscillations’ of mysterious origin.

571. Posted May 26, 2008 at 5:48 AM | Permalink

571 (Erl):

At some times there is little response either way? But the pattern is far from random.

The eye [and the mind set on something] is a poor judge of a fit. There are standard ways of assessing randomness. I have already [many, many posts ago] showed how to apply these test, and at that time the pattern was random. Judging now with an open mind I come to the same conclusion. However, your scheme makes a testable ‘prediction’ [in quotes, because it is only a valid prediction if based on a mechanism – just saying ‘solar activity makes it go up’ is not a mechanism], namely that there should be a strong El Nino shortly, so we can now just wait a few years and see how it pans out.

572. Posted May 26, 2008 at 6:05 AM | Permalink

570 (Raven): no, I don’t really have any comments, except not to jump to unwarranted conclusions. On the other hand, there is little doubt that comparing data from extraterrestrial with terrestrial data will be helpful in deciding what is external and what is internal to the Earth system.

573. lgl
Posted May 26, 2008 at 11:37 AM | Permalink

The last non-sense from me (for a while)
“A New Meta-Stable Three-Body Solution, of Two Small Nearly-Resonant Objects Gravitationaly Revolving Around a Massive Primary Body”
I would love to see the graph showing the distance between the Sun and the planets’ center of mass (sun left out) over the last 300 years or so (1 yr resolution). Anybody, not having problems with numbers.

574. Posted May 26, 2008 at 4:53 PM | Permalink

574 (lgl): you are missing the point. None of the three bodies will feel any forces in their free falls, except the ever-present tidal forces that, if the distances are large enough, have negligible effect.

575. cba
Posted May 26, 2008 at 5:56 PM | Permalink

(Leif):

Hopefully, I’ve got it right for doing images here.

Below is OLR for the Earth based on the 1976 standard atmosphere as calculated from my one dimensional model of radiative transfer.

A surface temperature of 288.2k with 0.98 emissivity is used. Atmosphere is assumed to be in LTE, clear sky with chart points determined from doublings and halvings of present CO2 concentrations. The OLR value is for the altitudes indicated and includes surface and all atmospheric absorptions and emissions as the radiation travels through the 1976 std atmosphere shells.

OLR totals are for the spectrum 0.2 – 65.5 microns. Over 30 molecules from the Hitran database are included. Also, the dF values are the deltas between the power emission at the point and double the CO2 concentration from that point.

I was hoping you might have some constructive coments on the chart.

576. jae
Posted May 26, 2008 at 6:47 PM | Permalink

cba: Am I the only one who can’t figure out what all the curves represent?

577. cba
Posted May 26, 2008 at 7:06 PM | Permalink

(jae):

The top three are OLR, outgoing longwave radiation, as viewed at 3 altitudes – left y-axis. The bottom 3 are dF – delta forcings or differences in OLR for a doubling of CO2, again as viewed from the 3 different altitudes – right y-axis.

Does that help?

578. jae
Posted May 26, 2008 at 7:54 PM | Permalink

Ah, yes. Thanks!

579. Pat Keating
Posted May 26, 2008 at 8:45 PM | Permalink

576 cba

Interesting work.
1. Why such high altitudes?
2. How did you handle the multiple re-emissions of the radiation in the high-absorptivity regions of the spectrum?
3. How does your approach compare with that of Clough et al?

580. Posted May 27, 2008 at 5:00 AM | Permalink

57(lgl)

I would love to see the graph showing the distance between the Sun and the planets’ center of mass (sun left out) over the last 300 years or so (1 yr resolution).

I’ll post these graphs and their captions without further comment in deference to Leif’s wish that we don’t clutter his thread with discussion of solar oscillation.

The orbit of the centre of the Sun around the centre of
mass of the solar system (in units of 10)3 AU, astronomical
unit = 149 á 106 km). The dashed horizontal abscissa in the upper part of the figure represents the diameter of the Sun.

The orbit of the centre of the Sun around the centre of mass of
the solar system (in units of 10)3 AU) separated into two basic types,
the ordered (in a JS-trefoil) (top) and the disordered (bottom). The
area in which the Sun moves has a diameter of 0.02 AU or 4.4 rs, this
being the solar radius, or 3 á 106 km. The most disordered sections of
the intervals lying between the trefoils are plotted. The Sun enters into
the trefoils with a periodicity of 178.7 years, on the average (see the
times, years at the top of the respective figures). The value represents
the first basic cycle of solar motion. While the trefoils are nearly
identical (after a rotation), the disordered orbits differ one from the
other. The Wolf, Sporer, Maunder and Dalton prolonged minima of
solar activity coincide with the intervals of disordered solar motion.

Steve: Please take this to the Bulletin Board if you must. I dislike people presenting personal theories here. If you want to discuss peer reviewed literature, fine, but keep the personal advocacy to a minimum.

581. cba
Posted May 27, 2008 at 7:35 AM | Permalink

580 (Pat)

Interesting work.
1. Why such high altitudes?
2. How did you handle the multiple re-emissions of the radiation in the high-absorptivity regions of the spectrum?
3. How does your approach compare with that of Clough et al?

1 The information existed and I was curious about it even though it’s a fricken vacuum that may not be LTE. It’s also interesting just how much difference in OLR there is too depending on just how high you go.

Considering the variation there between the lines, it would appear that there is a substantial difference in results depending upon where you cut off the height. Also, most obviously, at the upper reaches there are other effects heavily involved in the nature of the warming due to the high T. Remember, the Sun’s photosphere is also a significant vacuum and that very low density gas is glowing bright enough to make summer days really hot down here, despite the huge distance between.

Also, it would appear that the clear sky results of any of those altitudes (and by those altitudes, it’s pretty much all clear sky) suggests an imbalance with too much outgoing radiation compared to that coming in.

My result for current CO2 levels at 70km is about 5 w/m^2 higher than a similar Modtran 3 parameter run so it’s not that far off either at 263 versus 259 of the modtran.

2 A gas of uniform temperature that is optically thick is going to emit outboud in a direction at the point where it starts to be optically thin and it’s going to absorb energy coming in about in the same fashion – as it enters through the optically thin boundary. The assumption of uniform temperature means that by what ever convective, conductive, radiative method, the energy was distributed / transfered through this black box area. The key is I assumed uniform T for my slice which is only roughly correct with just 50 slices but then I don’t know the exact T distribution that fine anyway.

Essentially, a slice of gas that is the same temperature as the rest is not going to make a difference in emission or absorbtion. If it’s cooler it absorbs, if warmer it creates emission lines.

3 I’m not sure how my efforts differ from Clough’s other than that what he does for a living and has for many years – suggesting much greater knowledge of the subject and time to get far more sophisticated. Also, it would seem his interest is in improving GCM input models with better radiative input and my efforts are basically conceptual / one dimensional modelling oriented.

582. Posted May 27, 2008 at 8:42 AM | Permalink

581(Steve’s comment)
Sorry Steve, It is from peer reviewed literature, but I omitted the reference. I will post further material on unthreaded on the BB anyway, as I have no wish to antagonise Leif or yourself.

Can the origin of the 2400-year cycle of solar activity be caused by solar inertial motion?
Journal Annales Geophysicae
Publisher Springer Berlin / Heidelberg
ISSN 0992-7689 (Print) 1432-0576 (Online)
Issue Volume 18, Number 4 / April, 2000
Category Original articles
DOI 10.1007/s00585-000-0399-x
Pages 399-405

I. Charvátová
(1) Geophysical Institute AS CR, Boccaronni II, 141 31 Praha 4, Czech Republic

Received: 30 September 1999 Revised: 14 January 2000 Accepted: 17 January 2000

583. Sam Urbinto
Posted May 27, 2008 at 12:27 PM | Permalink

Leif #569

“…the tides by the Moon on the Earth is 367 mm high [mm = millimeter = 1/25.4th of an inch].”

367? I would think 14.44 inches on a planet covered by 70% water might have more of an effect than an ant to a tire on a truck going 100 mph. What are they feeding you down there in Houston? :)

584. lgl
Posted May 27, 2008 at 1:01 PM | Permalink

581
Beautiful! Thank you Stroller.

585. Pat Keating
Posted May 27, 2008 at 4:31 PM | Permalink

582 cba

Thanks for the response. As I said, it is interesting work.

586. Posted May 27, 2008 at 8:25 PM | Permalink

584 (Sam): The ant and the truck was for Jupiter’s effect on the Sun.
Interesting math, BTW, if you work through it. There is a fascinating psychological study here: why do people believe such weird things? Michael Schermer has written a good book on that.

587. Posted May 28, 2008 at 12:41 AM | Permalink

587(Leif) Yeah, for example, some physicists are thinking maybe ‘dark matter’ is the medium which translates the frame dragging effect of the angular momentum of massive objects to other objects. Crazy.

Still I suppose someone has to risk ridicule and do some thinking outside the box when good correlations show up between phenomena whose mechanism of linkage can’t be explained by currently accepted models of reality.

Lots of mainstream scientists used to believe gravity’s action was instantaneous everywhere, because it was a consequence of life in a Newtonian universe. Then some upstart swiss clerk whose maths teacher had told him that he’d never make a scientist came along and changed all that.

Maybe the final results from gravity probe B will shed some light.

588. Posted May 28, 2008 at 4:27 AM | Permalink

588 (Stroller): First, the correlations are not good [otherwise this would have been accepted a long time ago]. Second, we don’t need another mechanism. The problem with solar activity is that we already have too many.

589. welikerocks
Posted May 28, 2008 at 5:57 AM | Permalink

re: 583

Periodicities between 6 and 16 years in surface air temperature in possible relation to solar inertial motion
Journal of Atmospheric and Solar-Terrestrial Physics
“The individual and summarised series of surface air temperature from the area of central Europe have been processed. A similarity between the sets of significant peaks (12.8, 10.4 and 7.8 years) in the spectra of surface air temperature and the solar inertial motion (SIM) in the period range 6–16 years especially when computed from the series long 179 years have been revealed. The value of 179 years represents the basic cycle of solar motion. The SIM is computable in advance. Predictive assessments of periodicity behaviour during the future decades were established: various periods with very low amplitudes could occur.”

590. kim
Posted May 28, 2008 at 6:07 AM | Permalink

590 (wlr) More Charvatova in an ’07 article in Advances in Space Research.
=====================================

591. Posted May 28, 2008 at 6:15 AM | Permalink

589(Leif)Bearing in mind the weakening correlation between sunspot number and global temperature, please would you comment on the following paper.

Abstract.
Solar activity, together with human activity, is considered a possible factor for
the global warming observed in the last century. However, in the last decades solar activity
has remained more or less constant while surface air temperature has continued to increase,
which is interpreted as an evidence that in this period human activity is the main factor for
global warming.We show that the index commonly used for quantifying long-term changes
in solar activity, the sunspot number, accounts for only one part of solar activity and using
this index leads to the underestimation of the role of solar activity in the global warming
in the recent decades. A more suitable index is the geomagnetic activity which reflects all
solar activity, and it is highly correlated to global temperature variations in the whole period
for which we have data.

From the conclusion:

The geomagnetic activity reflects the impact of
solar activity originating from both closed and
open magnetic field regions, so it is a better indicator
of solar activity than the sunspot number
which is related to only closed magnetic
field regions. It has been noted that in the last
century the correlation between sunspot number
and geomagnetic activity has been steadily
decreasing from – 0.76 in the period 1868-
972 Georgieva et al.: Global warming and solar activity
Fig. 6. Global temperature anomalies T (solid line)
and ak index of geomagnetic activity (broken line)
for the period 1856-2000; climatic normals.
1890, to 0.35 in the period 1960-1982, while
the lag has increased from 0 to 3 years (Vieira
et al. 2001). According to Echer et al. (2004),
the probable cause seems to be related to the
double peak structure of geomagnetic activity.
The second peak, related to high speed solar
wind from coronal holes, seems to have
increased relative to the first one, related to
sunspots (CMEs) but, as already mentioned,
this type of solar activity is not accounted
for by the sunspot number. In Figure 6 the
long-term variations in global temperature are
compared to the long-term variations in geomagnetic
activity as expressed by the ak-index
(Nevanlinna and Kataja 2003). The correlation
between the two quantities is 0.85 with p

592. Posted May 28, 2008 at 7:20 AM | Permalink

589(Leif)When there are too many mechanisms, it’s sometimes because the underlying one hasn’t yet been identified (or accepted). One example is the complex of cycles and epicycles used to explain the retrograde motion of planets before Galileo and Copernicus moved the sun to the centre of the solar system.

Now we know that the centre of the solar system is it’s centre of mass. Everything else wobbles round it.

As you pointed out, the sun is a very dynamic and mobile body, with texas sized chunks flying around at bullet like speeds a long way from it’s centre of gravity. If it’s observed diameter is changing slightly as noted in a post ^upthere, and it’s not due to tidal forces, the next most likely candidate is changes in it’s moments of inertia.

I think your analogy of an astronaut serenely orbiting the earth in freefall is inadequate to the varying motions of the sun induced by the ever-changing interplay of the planetary masses. The outer planets pull it a couple of sun diameters over 50 years or so and the inner planets give it 700km jerks several times every couple of years. 700km is deeper than the convective layer isn’t it?

If the change from harmonious motion periods and their concomitant shorter and more vigourous solar cycles to more chaotic motion and longer, weaker cycles noted by Charvatova, a potentially fruitful line of investigation would be to map the way the solar cycles speed up following chaotic motion periods and ‘play catch up’ with the planetary body holding 60% of the solar system’s angular momentum which oscillates only a few minutes of arc either side of the plane of invariance.

593. Basil
Posted May 28, 2008 at 7:26 AM | Permalink

#589 Leif

Second, we don’t need another mechanism. The problem with solar activity is that we already have too many.

None of which account for the anomalies enough to induce a paradigm. (Note that I didn’t say “paradigm shift.”) So people keep looking, and it isn’t particularly “scientific” to mock or discourage this. Of course, when one of these new theories comes along, if it doesn’t answer more questions than it raises, it will soon be discarded. So we don’t reject it out of hand (“we don’t need another mechanism”), but, as you said to begin with “because the correlations are not good” or some other such reason.

594. lgl
Posted May 28, 2008 at 8:30 AM | Permalink

#581

Assuming those nice motions of one narrow circle, one wide circle, one narrow … (each taking around 11 yrs btw) really are the Sun’s motion and not the barycenter motion. Doesn’t a pattern like that mean that one side of the Sun will travel a different distance than the opposite side, implying one side is accelerated relative to the other giving turbulence?

595. Posted May 28, 2008 at 9:17 AM | Permalink

595(lgl)The centre of each graph is the centre of mass of the solar system. The black curve is the path of the centre of the sun.

one side of the Sun will travel a different distance than the opposite side

That was what I was getting at when I said the arcs the sun moved in were tight compared to it’s diameter, in contrast to the planetary orbits. However, this comment was misunderstood by Leif and he took the case to a reductio ad absurdum where the centre of mass was one inch from the centre of the sun.

We need to move this discussion elsewhere before bzzzt {ATH }
:)

596. Posted May 28, 2008 at 9:23 AM | Permalink

594 (Basil): When I said we have enough theories [even too many] it meant that there are many physically viable and plausible mechanisms and we just at this point don’t know enough to select the right one [assuming that it is among the candidates. That is very different from rejecting all the other ones that are not physically possible on energy grounds. So we are not trashing around in the dark where any idea is a good as any other, just waiting for a new Einstein to show us the way. And I’m not going to continue the fruitless discussion of gravitational effects and movements. And just dredging up papers that claim large correlations ain’t helpful either; they are 13 to the dozen.

Steve: Leif has been very patient with discussion of gravitational and barycentric issues. I’ve repeatedly asked that such matters not be discussed here and Leif has been far more patient with breaches of this request than I would have been.

597. jae
Posted May 28, 2008 at 10:58 AM | Permalink

Leif: Steve Sadlov linked this over on the BB. Perhaps it would also be of interest here, where there are more eyes. Any comments?

598. lgl
Posted May 28, 2008 at 11:29 AM | Permalink

#596 Stroller

Yes they have been more than patient with us, thanks Leif and Steve.
I’m waiting for you to start a thread (or unthreaded) over at BB, or have you already?

599. Chris Knight
Posted May 28, 2008 at 12:20 PM | Permalink

569 Leif

Tides are probably more potent than you imagine. Otherwise, apart from the relativistic effect on Mercury, due to solar proximity, tidal locking would be 1:1 with its orbit. Outer planetary gravitational effects maintain the 2:3 rotation to orbital period.

Gravity, in any cosmological sense, is not an instantaneous phenomenon, its effects are demonstrated over time, and in wave-like pulses and periods, with standing harmonics and beat frequencies on the bodies they affect.

Steve:
I’ve asked that gravitational effects and things like that not be discussed here. Doubtless it’s an interesting topic, but it’s too far afield for this site.

600. Sam Urbinto
Posted May 28, 2008 at 12:33 PM | Permalink

587 Leif “The ant and the truck was for Jupiter’s effect on the Sun.” Ah, well, of course. Thanks for clarifying!

It seems the easiest thing to say is that there is an effect, but it’s too small to account for much if anything, and is background noise. Like the small amount that co2 reacts to near IR being there but in the end immaterial and simply complicating things if considered.

601. Posted May 28, 2008 at 3:44 PM | Permalink

599(lgl)

they have been more than patient with us, thanks Leif and Steve.
I’m waiting for you to start a thread (or unthreaded) over at BB, or have you already?

Agreed, and I hope we didn’t rock the apple cart too hard. :)

Can Steve give us a place to discuss barycentric nonsense or do we just mix it in with everything else on unthreaded?

602. Jim Arndt
Posted May 28, 2008 at 5:54 PM | Permalink

602, Stroller

I started one there at the bb

603. Posted May 28, 2008 at 6:20 PM | Permalink

602 (Stroller):

Agreed, and I hope we didn’t rock the apple cart too hard. :)

Its impact is like that of Jupiter’s tidal effect: nil.
So no harm, except waste of bandwidth and brainspan.

604. John Lang
Posted May 28, 2008 at 6:53 PM | Permalink

Hey Leif, I just saw you on the National Geographic channel, on the science show Naked Science which is, by far, the best science documentary show on television.

Good to put a face to a posting name.

I believe the show is about how a cooler Sun might help negate global warming for awhile (but it just started so I don’t know what the conclusions are.)

605. Posted May 28, 2008 at 7:00 PM | Permalink

605 (JohnL): I’m glad that you like the show. The various scientists appearing may not all share the same view [or that of the producers or N.G.], but by the breadth of their expertise may show the viewers how this topic spans several scientific disciplines.

606. John Lang
Posted May 28, 2008 at 7:20 PM | Permalink

I’ll post about the show later.

Actually, I’m watching the Stanley Cup Finals right now and just taping the show on my PVR (in between flipping channels and posting during the commercials.)

607. Posted May 28, 2008 at 9:51 PM | Permalink

Leif,

I never give up. Check these plots:

Both trends coincide in the last three of four decades, and it is longer than in previous decades. The Lean’s databse refers to sunspots only. Anyway, Lean’s database differs a lot from yours.

Thanks,

608. Posted May 29, 2008 at 1:23 AM | Permalink

604(Leif)

Its impact is like that of Jupiter’s tidal effect: nil.
So no harm, except waste of bandwidth and brainspan.

Fair enough, and apologies for cluttering your thread with stuff you don’t find relevant.

For anyone interested in the suns motion around the centre of the solar system and the planetary effects on it, the thread started by Jim is here:

Jupiter’s tidal effect on the sun may be negligible, but it’s gravitational effect is large enough to pull the sun around in the magnetospheric soup of it’s own making. I suspect the effects of that will eventually be recognised as non negligible, but I also recognise that this is not the place for speculative theorising, so once again, sorry for wasting bandwidth and your considerable brainspan.

I hope you won’t mind if I call in to ask for information relevant to our thought experiment from time to time.

My first one is; Over what sort of timescale do these texas sized chunks of matter emerge from the depths at bullet like speeds and reach the point of freefall back to the suns surface? What is the suns gravitational constant at the surface?

609. Posted May 29, 2008 at 2:36 AM | Permalink

609 (Stroller):

Over what sort of timescale do these texas sized chunks of matter emerge from the depths at bullet like speeds

typically 15 minutes

What is the suns gravitational constant at the surface?

27 times as big as the Earth’s, or 275 m/s2

610. Posted May 29, 2008 at 4:19 AM | Permalink

610(Leif) Awesome. And if I understood you correctly, this is going on all over the suns surface all the time?

559(Leif)

the available evidence is that solar output has not increased in any detectable way, and that it is therefore hard to ascribe warmings on three planets to a solar cause that has not been observed.

http://www.gao.spb.ru/personal/ivanov/papers/iv-mil-iaus2004.pdf
Reconstruction of open solar magnetic flux
and interplanetary magnetic field
in 19th and 20th centuries
V.G. Ivanov and E.V. Miletsky

Shows that although sunspot numbers have been declining since 1950 as you noted, the ‘High Speed Solar Winds’ from ‘open coronal holes’ has been increasing, and this affects the earths magnetosphere. Could this have a bearing on earth mars and jupiter getting warmer? Would the ‘squashing’ of the planetary magnetosperes by the stronger solar wind cause heating through some kind of planetary dynamo effect?

This is just a question, not a theory. :)

611. Posted May 29, 2008 at 5:15 AM | Permalink

611 (Stroller):

this is going on all over the suns surface all the time?

Yes:

The Russian paper you refer to was written before I had shown that the geomagnetic aa-index [used to reconstruct the Sun’s magnetic field] was wrongly calibrated before 1957. This error has now been corrected. Below I show the interplanetary magnetic field and the solar wind speed over the past ~130 years:

The magnetosphere does not influence the climate in any detectable way [although there are many claims that it does, just as there are many claims about many other things …]

612. Posted May 29, 2008 at 6:54 AM | Permalink

612(Leif) Seeing the N.America map overlaid really gives a sense of scale. A truly awe inspiring image.

If I read your second graph correctly, solar wind speed has indeed picked up 9% or so on the average since 1950. And judging by the linearity of the trend line, if it’s a cycle, it’s a long one!

So we don’t have any clue as to what effect the increase might have?

Or any educated guesses?? :)

613. Posted May 29, 2008 at 7:27 AM | Permalink

this paper discusses the effect of solar wind speed on the magnetosphere, but the numbers mean little to me. Are we in falling snowflake territory again?

Steve: We’re finished with this sort of discussion of magnetics, gravity, barycentrics. There are very important matters afoot in the SST revisions which impact the connection of solar to temperature. Stick to that.

614. Jim Arndt
Posted May 29, 2008 at 8:03 AM | Permalink

612, Leif

I think your scale is off the USA is at least 2k not 1k.

615. Posted May 29, 2008 at 9:35 AM | Permalink

613 (Stroller):
The 9% increase in solar wind speed is likely instrumental in the sense that the Earth is the instrument and the Earth’s sensitivity to the solar wind has changed by 10% due to the decrease of the Earth’s magnetic field in the last few centuries. This is still under debate and no conclusions can be drawn yet, except that in any case the influence on terrestrial climate due to this is probably undetectable [and undetected – again, despite claims of the opposite]. In trying to assess ‘solar’ impact on SST and climate in general, we also need to know what the Sun is doing. A problem here is that people advocating one or other ‘solar’ effect tend to cherry pick the version of solar data that fits their ideas the best. And just as the temperature record is flawed and uncertain, so is the solar record. So comparing one flawed record to another flawed records leaves a rather large wiggle-room.

615 (JimA): the total scale is 7k. Each white and black mark is 1k.

616. Posted May 29, 2008 at 9:47 AM | Permalink

Steve,

My posts are stuck to solar issues, but it seems my posts are not so important as to be considered seriously.

617. Posted May 29, 2008 at 10:50 AM | Permalink

617 (Nasif): That Lean’s is not the same as mine is no news. Otherwise we would not have two series :-) But Steve’s point is that the blog is for discussion of issues that bear on auditing published articles, rather than advancing private ideas [however deserving].

618. Leif
Posted May 29, 2008 at 11:27 AM | Permalink

In order to assess the impact of solar causes, we need good solar data. This is not easy, because the data spans centuries, and a reliable and stable calibration is hard to come by. My AGU talk last Tuesday on the calibration of the sunspot number is here

619. Posted May 29, 2008 at 1:34 PM | Permalink

Leif,

I still cannot understand why Lean’s amplitudes SFI in the two previous decades are higher than yours:

Temp. 79-89 = 1.9505
Lean SFI 79-89 = 1.0605
Sval. SFI 79-89 = 0.003221077

T. 89-99 = 0.336916667
Lean SFI 89-99 = 0.156
Sval. SFI 89-99 = 0.018896231

The amplitudes of SFI from your database are almost void. Is it possible in nature?

620. Posted May 29, 2008 at 1:44 PM | Permalink

# 620

Leif,

What’s more, the change of amplitude from your database for the last two decades is negative, while the change of amplitude from Lean’s work is positive:

Svalgaard’s = -0.015675154
Lean’s = 0.9045

BTW, who won the cup?

621. Leif
Posted May 29, 2008 at 5:10 PM | Permalink

620 (Nasif): I have no idea what you are getting at. This graph

shows that Lean’s and my TSI match each other closely 1940-2008

622. Erl Happ
Posted May 29, 2008 at 5:40 PM | Permalink

619 (Leif)
Great paper. Can’t pretend to understand the relationships but looks like something we can trust.

623. Posted May 29, 2008 at 10:44 PM | Permalink

# 621

Leif,

I’m sorry; I didn’t describe the process. I calculated the decadal Solar Flux Index needed for getting the normal, that is, the decadal amplitude in both reconstructions. Lean’s work coincides a lot with your reconstruction from 1940 to 2000; however, the coincidences are evident mostly on frequency and dates of events. I saw that we hadn’t information about the amplitudes of each event, so we could see something more “tangible” which would give us more light on the differences between your database and Lean’s database. I decided to take a little from which we had discussed before in this thread and considered 1366 Wm^-2 like ground zero for plotting the SFI amplitudes only. I calculated the excess of solar energy necessary to standardize the amount of TSI, the last taken from well-known databases. I found another difference, let’s say more consequential, between your reconstruction and Lean’s reconstruction. I wonder why there are differences in both reconstructions after 1939 when we know that the measurements of TSI from 1940 to date were calibrated.

Furthermore, Leif, there are not only differences in amplitudes, but the change of amplitude for the identified periods, taken from your database, is negative, while the change from Lean’s database is positive. I cannot explain this difference and, honestly, I don’t know what database is accurate and what is not, given that both reconstructions were based on the same calibrated data.

624. Posted May 29, 2008 at 11:02 PM | Permalink

# 624

Leif,

Let me tell you that your work is not my problem, but something that I’ve not found… yet.

625. See - owe to Rich
Posted May 30, 2008 at 12:47 PM | Permalink

552 (duwayne)

Sorry for the delay – I’ve been away.

Rich.

626. Posted May 30, 2008 at 2:33 PM | Permalink

I found a correlation between the asymmetry of amplitude of Lean-Svalgaard TSI and the Amplitude of ∆ Temperature:

Something wrong? ;)

627. Larry T
Posted May 31, 2008 at 10:05 AM | Permalink

I have been watching the charts for the current solar cycles and have an addition observation to the fact that cycle 24 is starting both slowly and weakly is that the residuals of data observations are also lower. Lower values and less standard deviations to me suggest a much weaker than normal cycle. I hope some of the experts will give me some educated comments.

628. See - owe to Rich
Posted May 31, 2008 at 2:55 PM | Permalink

436 (duwayne)

Hi, I’ve analyzed your model in R, and it does pretty well, but has some problems. For details see the BB.

Rich.

629. Posted May 31, 2008 at 7:24 PM | Permalink

628 (LarryT):

Lower values and less standard deviations to me suggest a much weaker than normal cycle.

Since the sunspot number is bounded below by zero, when the values are low the standard deviations necessarily have to be low as well and therefore have no predictive power. The prediction of a low cycle 24 is not based on its length. Cycle 20 was also long, yet cycle 21 was the second largest cycle ever.

630. Jon
Posted Jun 1, 2008 at 12:25 PM | Permalink

Leif,

One rebuttal is that TSI or sunspot correlations depend on a dampening parameter. Obviously this complicates the model but not significantly and allows lags between when sign changes in the trend of one and appears in the trend of temperature.

e.g., a time constant such as the one suggested by Schwartz et al 2007.

Were this to be true it would explain why the graph smoothing appeared to work in Lassen 1991. Arguably they erred in not proceeding on a physical basis e.g., assuming a simple dampening model such as
$\frac{dT}{dt} = \frac{-T}{\tau}+ \frac{S}{\tau} F(t)$
S being the climate sensitivity to the forcing and \tau being the time constant.

631. Posted Jun 1, 2008 at 2:33 PM | Permalink

631 (Jon): Rebuttal of what? It is helpful to include the comment number.

632. Jon
Posted Jun 1, 2008 at 10:52 PM | Permalink

632 (Leif): My remark applies to comment 529 among others.

633. Posted Jun 2, 2008 at 5:29 AM | Permalink

633 (Jom): why is that a rebuttal of 529? 529 was concerned with the question whether TSI and solar activity as a whole has increased over recent decades in order to explain the simultaneous warming of three planets.

634. Andrew
Posted Jun 2, 2008 at 9:10 AM | Permalink

643 (Leif): Becuase it doesn’t need to have, necessarily, as it will still be working through the full response to the low frequency rise (over all amplitude of solar cycles is higher, even if they bottom out at the same level, with no background rise at all). This is kind of counterintuitive, but its how things are supposed to work. BTW, I know you don’t agree with them by Wilson and Mordivinov do find an upward trend over the past few decades. If you find Judith Lean to be more an authority, then I will defer to your wisdom on this matter.

635. kim
Posted Jun 2, 2008 at 9:19 AM | Permalink

635 (A) I keep thinking about Pete’s graph in comment #454 of Svalgaard #2.
====================================

636. Andrew
Posted Jun 2, 2008 at 10:09 AM | Permalink

636(kim): Can’t see. Can you reproduce it here?

637. Posted Jun 2, 2008 at 10:20 AM | Permalink

# 635 # 636

Andrew, Kim…

And that’s what my graph demonstrates in # 627- Perhaps we see that there is no correlation between the longitude and frequency of lines from sunspots, but when we get the point on amplitudes and asymmetries, the correlation between the Solar Flux Index and the Change of temperature becomes manifest.

I used two databases for plotting them aside the amplitude of change of temperature, Leif’s and Lean’s database. I got first the amplitude of each database and then the asymmetry between both changes of amplitude. In the first place, the amplitudes taken individually and compared against the amplitudes of change of temperature demonstrate there is a close correlation. Then, when I got the asymmetries between Leif’s and Lean’s databases the correlation becomes almost perfect.

Taking the things so lightly, as to compare length and/or frequency with variability is quite deceptive, for that reason we must make use of something more tangible to know whether such correlation exists. This procedure demonstrates feasibly and unavoidably that the correlation exists.

638. Posted Jun 2, 2008 at 10:24 AM | Permalink

635 (Andrew):

This is kind of counterintuitive, but its how things are supposed to work.

Says who? What is the physics behind that? Go read the very first post on Svalgaard #1.
Claus Froehlich finds ftp://ftp.pmodwrc.ch/pub/Claus/AGU_SPD2008 :

During the last 30 years the cycle averaged TSI shows a decline after 1980 of about 50 mWm-2 /decade

I take the middle road between Wilson and Froehlich for the reasons set out in Svalgaard#1.

639. Andrew
Posted Jun 2, 2008 at 12:32 PM | Permalink

369 (Leif): I know your position on the whole PMOD ACRIM debate, but I just figured its worth mentioning.

I might be mangling the physics, but it is supposed to do with the fact that water takes a while to heat up, so 70% of the Earth’s surface, the ocean, is going to slow the full response to the external forcing. This “buffers” the system and keeps the system from rapidly reaching equilibrium. In models, if you double CO2 instantly, you may not get the full 3 or so degrees for a millenia.

Can someone more knowledge correct any errors I’m making here?

640. Posted Jun 2, 2008 at 1:23 PM | Permalink

640 (Andrew): what have the oceans to do with TSI not increasing for the last several cycles?

641. Andrew
Posted Jun 2, 2008 at 6:36 PM | Permalink

Nothing, but they may be important for the terrestrial climate response (recall that Scafetta and West say 10% to 30% with PMOD. The reason is because of this delayed response property.)

642. Andrew
Posted Jun 2, 2008 at 6:41 PM | Permalink

642 (Me): Gah! Sorry, slipped and forgot the reference. That should be in response to 641 (Leif).

643. Posted Jun 2, 2008 at 7:27 PM | Permalink

642 (Andrew): so you are saying that the last 30 years of heating on three planets [one of which does not have water] is a delayed effect of the minuscule increase of TSI that happened 30 years before that? Remember what the original question was: #524. Sometimes people get carried away with their ideas that they see the rest of the world around them dimly.

644. cba
Posted Jun 3, 2008 at 6:17 AM | Permalink

644 (Leif):

Now that you apparently have scuttled TSI as the culprit (I remain open to the subject as to whether you have nailed it or not), do you have any good candidates for the causes – or are the alternatives all equally lumped at the bottom of the pile? Or alternatively, do you think it’s merely a fluke and these three bodies just happen to be warming at the same time from different internal causes? Note what I’m asking is the difference between physics and math – the nonmathematical side of physics – that of the intuition and hunches that lead a physicist to a conclusion unattainable by a mathematician simply using math.

645. Posted Jun 3, 2008 at 7:16 AM | Permalink

645 (cba): My hunch is that it is just a coincidence, plus that it is not clear how big the warmings are, as some of the evidence is indirect. But there is a great potential in comparing proxies from the Moon, planets (+moons), and asteroids, once we get there. An ice core from Mars would be interesting. Temperatures in boreholes on the Moon…

646. Basil
Posted Jun 3, 2008 at 7:20 AM | Permalink

Steve: We’re finished with this sort of discussion of magnetics, gravity, barycentrics. There are very important matters afoot in the SST revisions which impact the connection of solar to temperature. Stick to that.

I didn’t know there was a connection between solar and temperature, at least to hear Leif on it. Is this “connection of solar to temperature” in one of the SST posts of late?

I’ll be interested in any major revisions to SST’s to see how that impacts my analysis of bidecadal oscillations in global temperature.

647. Andrew
Posted Jun 3, 2008 at 7:44 AM | Permalink

644 (Leif): Hmm, you are correct, that doesn’t work. However, I was really trying to explain the effect which 631 was alluding to. (Although, as I understand it, ice creates a delaying effect also, becuase it takes time to melt. That could work on Mars, which has polar icecaps.)

648. Posted Jun 3, 2008 at 9:39 AM | Permalink

I’m still awaiting for an audit to the methodology at #s 624 627.

649. Posted Jun 3, 2008 at 10:44 AM | Permalink

649 )Nasif): I don’t know what you are trying to accomplish.

650. Posted Jun 3, 2008 at 11:53 AM | Permalink

# 650

Leif,

649) Nasif): I don’t know what you are trying to accomplish.

I’m trying nothing misguiding. I only want to make sure that your hypothesis is falseable and testable. If all the methods at hand demonstrate that a correlation between the data bases for the TSI does not exist, then your hypothesis would be correct; if a fault exists, like which I supposedly have found, then it is necessary to reframe your hypothesis.

What I did was to obtain the amplitudes from the TSI Lean’s 2004 and Svalgaard’s 2007 databases, and the databases on the variability in the tropospheric temperature (VTT); next, I obtained the decadal amplitude to obtain the decadal asymmetries. Next, I obtained the asymmetries between your database and the Lean’s database 2004 for obtaining an average of both databases. Finally, I plotted the results on graphs.

651. Posted Jun 3, 2008 at 12:04 PM | Permalink

# 650

Leif,

It is normal that one be suspicious on evaluations, as the one that I am doing, because it would seem that it is an attack against one’s work; but there is no such, simply is necessary to assess a hypothesis with the methods available. I don’t want you to think that I am trying to discredit your hypothesis about no-correlation between sunspots number and change of temperature; on the contrary, I am trying to demonstrate that it is falseable and testable.

652. Posted Jun 3, 2008 at 2:05 PM | Permalink

652 (Nasif): good, please tell us in the end what your conclusions are.

653. See - owe to Rich
Posted Jun 3, 2008 at 2:42 PM | Permalink

652 (Nasif)

Are you a summer bird? I remember seeing you around a lot last summer, but not so much in the winter…

Look, you’re going to have to take things much more slowly and carefully – I can’t make out what you’re saying a lot of the time, and I don’t think I’m the only one. Intersperse English (sorry it can’t be Spanish :-)) explanations with some straightforward maths to show what you are trying to achieve.

Cheers,
Rich.

654. kim
Posted Jun 4, 2008 at 6:38 AM | Permalink

655 (Hey, You) Thanks for the lead in to my ditty:

I think I’ve never heard so loud
The quiet message in a cloud.
===================

655. jae
Posted Jun 5, 2008 at 11:36 AM | Permalink

Summary of a new solar-connections paper.

656. kim
Posted Jun 5, 2008 at 5:25 PM | Permalink

656 (jae) Tree rings. Who’d a thunk it?
========================

657. jae
Posted Jun 5, 2008 at 5:34 PM | Permalink

Kim: That study may be a “good” tree ring study, since it employs the drought index. I think tree rings can tell us a lot about droughts. I’m still quite negative about the temperature stuff, though. I need to look for a free copy and read it.

658. kim
Posted Jun 5, 2008 at 5:35 PM | Permalink

I’m just struck by the irony that bad tree ring science set of this hysteria, and good tree ring science may deflate it.
=====================================================================

659. bender
Posted Jun 5, 2008 at 5:51 PM | Permalink

#657
Mitchell (1979), that’s who.

660. bender
Posted Jun 5, 2008 at 6:08 PM | Permalink

#469
Thanks for the top panel; the bottom four are useless due to the irrelevant time-scales chosen. In the top panel, please identify for me the secondary peaks in CRF. Do you not see that those sub-peaks (1959, 1970, 1982, 1991, 2002; I can see them better now) correspond with TSI peaks? I grant you that they are tiny. That’s not my problem. I am merely pointing out the pattern. I’m an analyst, not a physicist. You tell me how amplification of second order effects to first order is not possible. It’s your argument, not mine.

Most of all, thank you for not dodging the question. I don’t know if there’s gold there, Leif. All I’m saying is we may want to dig some.

661. bender
Posted Jun 5, 2008 at 6:36 PM | Permalink

OT:
kim, ask Dano what he makes of lucia’s GMT trend analysis / invalidation. Maybe that’s why he’s got no time for the science any more? Moved on to pure advocacy.

662. Posted Jun 6, 2008 at 6:48 AM | Permalink

661 (Bender): In 60 years of peaks and valleys you find five subpeaks. This is much less than what one would expect by chance on a time scale of months. I find dozens and dozens if I just go wiggle hunting. What is so special above your five? I don’t see any pattern. There are standard statistical tools to quantify (sub)peak-hunting. The expanded time scale doesn’t seem useless to me, but shows the variations in more detail. If you can identify the peaks on the top panels, they should be there on the other panels. So, don’t dodge the issue now: print out the top panel, mark with a pencil where your peaks are, then identify and mark them with the same pencil on the lower panels, scan the result, and post it for everyone to see. I see nothing of interest or significance.

663. Posted Jun 6, 2008 at 5:11 PM | Permalink

# 654

See – owe to Rich,

Heh! Something alike a busy summer bird. Yes, my Spanish is not so good, my English is worst.

To obtain the amplitude of changes in Solar Irradiance I got a baseline from an assumed standard value, which is, from Leif’s database, 1365.81598 W/m^2. Then I applied the next formula:

A = λ/TSI average – TSI yr.

For example, from Leif’s database, the average of TSI is 1365.81598 W/m^2. Let’s take the TSI in 1700:

A = 0.0598 / 1365.81598 W/m^2 – 1365.604 W/m^2 = 0.0598 / 0.21198 W/m^2 = 0.2821 W/m^2

To obtain the oscillations of TSI, I got the median and deducted it from each value form the TSI databases.

θ = TSI median – TSI value

For example, the median from Leif’s database is 1365.927 W/m^2; then,

̀¸ = 1365.927 W/m^2 – TSI value

In 1700, the TSI from Leif’s database was 1365.604 W/m^2; consequently, θ is

θ = 1365.927 W/m^2 – 1365.604 W/m^2 = 0.323 W/m^2

The oscillation from the baseline to TSI in 1700 AD was 0.323 W/m^2

Finally, I got the Asymmetries between ∆T and the amplitude of TSI (∆T/Amp TSI)

Perhaps I’m wrong. If you make use of the algorithm, you’ll find some interesting things.

664. Posted Jun 6, 2008 at 6:09 PM | Permalink

For example this:

However, this issue must be reviewed because my calculations could be wrong.

665. JimP
Posted Jun 6, 2008 at 6:22 PM | Permalink

656 jae

A quote from your link: …These analyses indicated that the approximate 200-year cyclical variations present in the palaeoclimatic reconstructions were well correlated (R2 = 0.58-0.94) with similar variations in the Δ14C data, which obviously suggests the existence of a solar-climate connection. In addition, they say “the de Vries cycle has been found to occur not only during the last millennia but also in earlier epochs, up to hundreds of millions [of] years ago.”

A 200 year resolution millions of years ago? How did they do that?

666. Posted Jun 6, 2008 at 6:30 PM | Permalink

# 665

Myself,

It is obvious that there is an Earth’s internal mechanism that prolongs the effect of the Incident Solar irradiance smoothing the line of amplitude of ∆Temperature through some periods, for example, there are not pronounced troughs between decades 1810-1819 and 1870-1879 and 1940-1949 and 2000-2008. Leif is right when referring to terrestrial internal or intrinsic mechanisms that modify the effect; however, it seems that the amplitude of shifts from one value to another value of TSI has much to do with climate on Earth.

667. Erl Happ
Posted Jun 7, 2008 at 2:51 AM | Permalink

572 (Leif)
In ‘climate science’, it is generally accepted that the troposphere is transparent to solar radiation. But, if it were heated solely from beneath the range of variation would diminish with elevation. The opposite is the case.

The top figure shows the amplitude of variation in temperature over the entire period 1958-2008 at different elevations. Within the troposphere the widest variation is at 250hPa. The variation then diminishes, probably with water concentration as the air gets colder. Above the tropopause the increased variation may be due to changing UV radiation ozone and aerosols, together with solar wind effects. Solar wind effects will be controversial so let’s deal with that now.

At 1mPa equinoctial temperature peaks are prominent, especially at solar minimum. In Svalgaard 1 we learned that “as the Earth orbits around the Sun, southward IMF component is statistically more likely twice a year, increasing the coupling between the solar wind and magnetosphere. As a result, more storms occur during equinoctial months than during the solstitial months.” These equinoctial warmings appear to impact the Arctic. I would expect that with a weak tropical updraft (and cool SH summer) to drive the polar downdraft this polar heating effect would be more evident during La Nina events. The evidence is that, even at the equator, the temperature in the zone between 200mPa and 1 mPa exhibits a strong warming signal from diverse sources.

Examining the hovmoller diagrams in #571 one sees a strong warming signal over the warmest parts of the Indian and Pacific Oceans at 250mPa. The signal is specific to that zone. The enabling factor is the water vapour content of the air. When the air is warmed a loss of cloud drives surface warming, and most particularly so in the warmest oceans. Hence, we have the El Nino phenomenon in the Pacific. On the other hand a rise in humidity increases cloud density in the tropics, especially over the warmest oceans driving the opposite phenomenon, the La Nina. This is a major dynamic in the tropics where cloud cover is normally light because of the dominance of the Hadley Cell. The same phenomenon is evident in sea surface temperatures (and the fisheries) in the Atlantic and the Indian Oceans though less intense. This is a function of the latitude of the extreme projections of the land masses in the Southern Hemisphere.

Cirrus cloud is made up of highly reflective tiny crystals of ice.

This mechanism relies in simple physical principles. It is a lot more ‘physical’ than any proposition suggesting that temperatures in the stratosphere somehow propagate down into the troposphere.

The RSS satellite imagery shows how moist air is heated at 250mb. The warmest air is not in the tropics but at mid latitudes. There is no way that the air pockets at this latitude and altitude can sustain such a temperature difference without the absorption of solar energy? There is a very warm zone close to the Arctic at 150mPa. Correct me if I am wrong but that looks like an ionosphere/ thermosphere effect.

HadAT2 reports radiosonde temperature data dates from 1958. At the 200mPa level the negative temperature anomaly of March 2008 equalled the all time record of 1972. At 150mPa the March temperature easily set a new record anomaly. I suspect that the negative anomaly will intensify. September will be a test case because the correlation of 200mPa temperature with the aa index is better than the correlation with sunspots at solar minimum.

Tropical warming events are an inherent episodic feature of the solar cycle. It’s the microstructure of the cycle that matters, not the macrostructure. In the microstructure we are dealing with a 0.2% variation in irradiance, double that in the macrostructure. Apparently, measuring cirrus cloud in the tropics is difficult but it is easy to understand how the varying temperature dynamic changes cloud cover.

We have not had a solar minimum like the current one for 100 years. Perhaps we have to go back to the La Nina of solar minimum 1809, just short of solar cycle 6, to see its like. I predict that the current La Nina will manifest again in sea surface temperatures in September and there will be a very weak summer in both hemispheres in 2009 unless cycle 24 cranks up prior to June 2009. The earliest signal for the next El Nino will be a rise in temperatures at the 200hPa to 300hPa level over the Indian Ocean.

Finally, to make the point abundantly plain: The solar cycle signal is written in ENSO typography. Short term changes in the suns irradiance of as much as 0.2% drive the ENSO cycle in conjunction with the countervailing force of a warming tropical ocean that contributes water vapour to the atmosphere that builds cloud density over the warmest oceans. This is capable of reversing the warming trend even in the upswing of the solar cycle when irradiance continues to advance. There is no mystery in the fact that sunspots are positively correlated with temperature during tropical warming events and negatively correlated during tropical cooling events.

This tropical thermostat is described in several comments under Unthreaded 35 on this blog, #55 Bill E, #58 Jae, #61 Craig Loehle, #67 cba, #61 Craig, and of course Kim and I quote:

I think I’ve never heard so loud
The quiet message in a cloud
======================

The take home message: Changes in TSI over the entire cycle are immaterial, too small to be geo-effective in the grand scheme of things. To assess how the temperature of the Earth system might change over time we need to look at the ways in which the mechanisms that modulate the reception of solar energy change over time, i.e. cloud cover in the troposphere, aerosol and ozone in the stratosphere and how the timing of increases in irradiance can affect the amount of heat gained during warming events, for instance in January rather than July.

CO2 content in the atmosphere is a condition that affects ocean warming. It should shorten the warming cycle. However, OLR has little to do with the cooling processes over the warm oceans. Its all about evaporation, convection and de-compressive cooling. So, CO2 is also irrelevant.

668. Posted Jun 7, 2008 at 7:44 AM | Permalink

668 (Erl):

The top figure shows the amplitude of variation in temperature

calculated how?

669. bender
Posted Jun 7, 2008 at 8:57 AM | Permalink

Calculated how?

The graph label indicates the variation is calculated as a “range”. I assume it is that: a difference between maximum and minimum.

But Leif. Come on. This is all you have to say? What do you make of the idea of a “tropical thermostat” that buffers solar inputs?

670. Posted Jun 7, 2008 at 10:07 AM | Permalink

Erl Happ,

Tomatoes also respond to high temperatures; at some range of environmental temperatures, tomatoes lose flavor. You should continue your investigation.

671. Erl Happ
Posted Jun 7, 2008 at 1:36 PM | Permalink

Last link contains the data as text. It refers to the anomaly with reference to the period 1966-95. Unfortunately absolute figures are not available and neither is the ‘climatology’ of the reference period.

Data for each level was sorted from highest to lowest using the function in XL. The graph shows the extremes at each pressure level within the entire period.

The relationships shown in the graph could be cross checked with satellite data from NOAA should it become available.