Unthreaded #7

Continued from Unthreaded #6


  1. Willis Eschenbach
    Posted Mar 18, 2007 at 6:50 AM | Permalink

    fFreddy, you say:

    Dumb question : does anyone know who actually measures SO2 levels ? Where do they publish ?

    The most complete analysis I’ve found is here, but it only covers emissions, not actual atmospheric levels. Also, it ends in 2000, and so it doesn’t contain the 27% increase in Chinese SO2 emission since 2001.

    The basic message seems to be that anthropogenic SO2 emissions peaked somewhere between 1975 and 1990 (depending on who you believe), and has either stayed about constant or dropped slightly until 2000 (same caveat). Their data doesn’t include volcanic or other natural sources of sulfur.

    I’ve been looking for measurements of airborne SO2, but no joy so far.


  2. Posted Mar 18, 2007 at 7:50 AM | Permalink

    Martin Durkin was surprised that the attacks against him (after his Swindle documentary) were so feeble. 😉


    Nevertheless, critics demanded, among other things, that he would include the supermodern, state-of-the-art, most accurately verified graph of the temperatures in the last 1000 years, the so-called hockey stick graph. 😉 See also an audio interview with him.


    The page above was updated with links to videos with Sallie Baliunas and Michael Crichton.

  3. John Lang
    Posted Mar 18, 2007 at 7:59 AM | Permalink

    To Willis and Jae, I’ve found this article you might be interested in Arbiters of Energy

    The links are a little hard to follow through to get to the whole article. It is a just an article but you will find more detailed papers from the scientists who wrote the article and are using NASA’s MODIS and CERES satellites.

    The article contains charts on water vapour content of the atmosphere, IR radiation absorbed by water vapour versus CO2 and actual measurements of IR energy lost to space versus the models as in the charts below.

  4. JP
    Posted Mar 18, 2007 at 8:03 AM | Permalink

    Global Warming: How it All Began
    I found this article interesting because it was published in 1999 and it backs up some of the things said in the Swindle Documentary.

  5. John Lang
    Posted Mar 18, 2007 at 8:18 AM | Permalink

    Having some problems getting charts to show up. Here they are again.

    CO2 Absorption Versus Water Vapour

    Actual IR Radiation to Space Measured by Satellite

  6. jae
    Posted Mar 18, 2007 at 8:37 AM | Permalink

    5, John: I gotta hand it to those modelers. They show no increase in emitted radiation, when there is a substantial increase. Then they show no change in reflected radiation, when there is a clear decline. And, viola, the average is zero change. Looks to me like the Earth is automatically adjusting for the increased heat from solar radiation/less clouds.

  7. jae
    Posted Mar 18, 2007 at 8:41 AM | Permalink

    Willis: Isn’t the half-life of SO2 in the atmosphere very short? I think it forms cloud condensation nuclei and gets rained out quite quickly.

  8. S. Hales
    Posted Mar 18, 2007 at 9:48 AM | Permalink


    You might want to start here:

    This is the World Data Centre for Greenhouse Gases

    Since SO2 is washed out of the atmosphere regional differences will be large.

  9. Posted Mar 18, 2007 at 9:51 AM | Permalink


    Who remembers Mrs. Thatcher’s Incomprehensibles – The Goon Show. She did enough damage as a junior minister without going on to prove the Tories have no sense. Look at Cameron. But then they are politicians – all of them.

  10. Posted Mar 18, 2007 at 10:34 AM | Permalink

    The worldwide coverage for sulfur dioxide is impressive

  11. bernie
    Posted Mar 18, 2007 at 11:46 AM | Permalink

    I wanted to go back to the Lobell & Field article commented on in Unthreaded #6 (#319, 322, 323 and 326).

    Now my statistics are rusty but in keeping with prior discussions on CA I am a bit stunned by the way this research was done. First it seems to me that year to year chnages in t(min) and t(max) are very likely correlated and that including them in the same regression equation is extemely odd. In addition, even if the tmin and tmax were derived from correct temperature series when you use the first differnece then you essentially have to also include an interaction term in the regression equation wouldn’t you? I haven’t kept up with the econ literature for nearly 30 years, but when did it become fashionable not to include your model equations in the write up?

    Most importantly and to my mind stunning is the exclusion of any price variable in this model. It looks like the model focuses purely on yield. I would assume therefore that a poor harvest followed by higher prices that pulls more marginal land into production. Simialrly higher prices operating independently would pulllower yieldingland into production.

    Finally, I don’t get how this is a global model but the growing seasons are restricted to Northern Hemisphere growing seasons. Can someone explain?

    Overall, as someone else has already indicated this is not a very impressive article. Any commodity analysts should take a look because you may be able to help determine how meaningful are the findings in this paper. To me it smells of a conclusion in search of some data.

  12. bernie
    Posted Mar 18, 2007 at 12:26 PM | Permalink

    After my last message with its multitude of typos, is there any hope of an edit function like the one that is on Physorg Forums ?

  13. Pat Frank
    Posted Mar 18, 2007 at 1:05 PM | Permalink

    How does a high 1980’s-1990’s sulfate aerosol anomaly coexist with a steady decrease in cloudiness in both hemispheres and over both land and ocean between 1952-1997? Wouldn’t one expect an excursion away from a secular trend in cloudiness if anthropogenic SO2 was mucking up the dynamics?

    Norris, J. R. 2005 Multidecadal changes in near-global cloud cover and estimated cloud cover radiative forcing J. Geophys. Res. 110, D08206, doi:10.1029/2004JD005600.

  14. Ken Fritsch
    Posted Mar 18, 2007 at 1:07 PM | Permalink

    Re: # 322 (unthreaded #6)

    Found the cereal crop study … it’s here (1 MB pdf).

    The article concludes that the temperature effect could probably be offset by the fertilization effects of carbon dioxide. Over all an article showing the rapidly increase yields of grain crops (due to human intervention and adaptation) over time that purports to show the small negative yield effects of the AGW by increasing temperature but concludes that that same source offsets the effect with yet another effect yet (CO2 fertilization) seems, in my mind to be attempting to show a totally out of context cost due to AGW.

    On our Illinois farm, we always wanted the warmer temperatures and longer growing seasons for corn and soybeans. Higher night time (minimum) temperatures were particularly always considered a positive for corn yields. When higher temperatures were accompanied by droughty conditions, the corn and soybean yields could suffer, but over the years improved genetics mimumized this effect. It might be interesting to look into this study in more detail. Surely areas that grow warm weather crops such as corn and soybeans in less than optimally warm areas would benefit from higher temperatures.

  15. bernie
    Posted Mar 18, 2007 at 1:18 PM | Permalink

    Your practical experience confirmed my initial reactions to the article. How do prices impact your choice of acreage? How do you make decisions year over year as to how much and which land to put under wheat, maize, etc. Overall the article makes little statistical nor agricutlural seense sense, . The order effects are so small that given the trend, i.e., warmer temperatures, they have to be benign unless there are large areas of crop production that are right on the margin. There is soemthing really odd about how this model was reported – in this instance they have stressed the R2 but visual inspection of their scatter charts makes me somewhat skeptical.

  16. S. Hales
    Posted Mar 18, 2007 at 1:32 PM | Permalink

    There is poor spatial coverage at present at WCDG. But SO2 emissions historically have been concerned with the production of sulfuric acid that washes out of the atmosphere as “acid rain”

    One avenue to aproximating the atmospheric concentration at present is to look at an estimate of human and natural contributions of all sulfur oxides here

    From the above link
    Natural sources release 80 to 290 million tons of sulfur oxides into Earth’s atmosphere each year (variation presumably related to volcanic activity)
    Humans contribute another 70 to 100 tons annually

    Most of the world’s sulfur is forced production or that production required by environmental laws to capture it before or after combustion as elemental sulfur or sulfur dioxide. I don’t know whether China and India have much forced production yet, I think they don’t, so their production of SO2 should be higher and follow the sulfur content of their fossil fuel mix.

  17. S. Hales
    Posted Mar 18, 2007 at 2:39 PM | Permalink

    I have a question for any data nut out there. Is the raw station data for TMAX and TMIN for all the stations in the GHCN world network here ???

    If this is the raw data before it is massaged for V1 and V2 and gridded then it seems that it would be a simple task to inspect this daily data extract annual values and simulate the gridded data and see how it compares with the gridded cells produced by Hansen.

  18. nilram
    Posted Mar 18, 2007 at 2:52 PM | Permalink

    In the telegraph article posted in #2, Mr. Durkin states:

    Other than the discredited hockey stick, the graph used by us (and published by the Intergovernmental Panel on Climate Change) is the standard, accepted record of temperature in this period.

    Does anyone know where I can find this graph. I think it’s in the First Assessment report of the IPCC. But I can’t seem to find a copy of the report online. Also a page number would be nice


  19. bernie
    Posted Mar 18, 2007 at 3:12 PM | Permalink

    Lobell and Field reference

  20. James Erlandson
    Posted Mar 18, 2007 at 3:14 PM | Permalink

    Growing Degree Days — Corn Growth and Yield
    From the Department of Agronomy and Plant Genetics
    University of Minnesota
    July 27, 2004

    Temperature affects crop growth and development. Accumulation of heat during the growing season can be used as a predictor of plant developmental progress. Growing Degree Days (GDD’s) is a calculation to express the heat accumulation. GDD’s are calculated using the maximum and minimum daily air temperature to determine the average daily temperature. From the average temperature, the base of 50 [degrees] is subtracted to arrive at the daily GDD’s. There are temperature limits used when calculating GDD’s because little or no growth occurs when the temperature is greater than 86 [degrees] or less than 50 [degrees] F.

    When corn is planted late, tasseling and grain filling occurs later in the season when average air temperatures are lower and GDD accumulation is less. Later planted corn yields less and fewer GDD’s certainly contributes to lower grain yields.

    So if the climate warms a degree or two it will presumably extend the season allowing farmers to delay planting, let the fields dry out and reduce the risk of having young plants killed by a late frost.

    Midwest agricultural schools know exactly how temperature effects crop yields — they do real experiments.

  21. JerryB
    Posted Mar 18, 2007 at 3:20 PM | Permalink

    Re #17,

    Not all GHCN stations report daily min/max temperature data.
    Also, many of the stations at that site are not included
    among the stations for which monthly GHCN data has been
    collected. GISS uses the latter as its main source of
    monthly mean temp data.

    At http://www.john-daly.com/ghcndsum.smp
    you can find some stats on the data at the GHCN daily site.

  22. Basil Copeland
    Posted Mar 18, 2007 at 3:29 PM | Permalink

    #18 nilram…are you asking about this diagram?

    If the image doesn’t show, here’s the link (it is right here on Climate Audit).

  23. Mike Rankin
    Posted Mar 18, 2007 at 3:44 PM | Permalink

    I just ran across this reference on Debunkers:


    showing commentary by novelist/science fiction writer Orson Scott Card on the Hockey Stick and M&M’s efforts. Thought I should share with other visitors.

  24. Ken Fritsch
    Posted Mar 18, 2007 at 4:46 PM | Permalink

    Re: #11 and #20

    Bernie and Jim Erlandson make some good points here that along with my previous experiences and recollections would indicate to me that the linked article may not stand up to detailed analysis and/or a more complete model for explaining the year to year crop yield variations.

    I wanted to concentrate on better understanding the effects of an additional 3.4 W/m^2 absorption by water vapor in the troposphere on surface temperatures, but I am ready to be distracted and there appears to be plenty of food for thought coming out of these posts.

    I would guess that the paper in question would not receive much attention from either side of this issue simply because it seems to negate the AGW effect by reference to CO2 fertilization — even though it wants to illogically put a cost factor on the negative side (temperature) only.

    I would guess that if our climate became more or less permanently warmer that corn and soybean yields would improve in areas that currently grow corn and soybeans but under less than optimum temperatures (MN or ND) while the same crops grown currently in areas with marginally warm temperatures (?) might have decreased yields. Brazil has warmer conditions than Illinois but has of late been growing more and more corn and soybeans successfully in that nation — just ask a midwest farmer as some of them even farm in Brazil.

  25. Peter D. Tillman
    Posted Mar 18, 2007 at 5:39 PM | Permalink

    NY Times on Al Gore backlash

    “From a Rapt Audience, a Call to Cool the Hype”

    [free registration required and worthwhile]

    “I don’t want to pick on Al Gore,” Don J. Easterbrook, an emeritus professor of geology at Western Washington University, told hundreds of experts at the annual meeting of the Geological Society of America. “But there are a lot of inaccuracies in the statements we are seeing, and we have to temper that with real data.”


    Benny J. Peiser, a social anthropologist in Britain who runs the Cambridge-Conference Network, or CCNet, an Internet newsletter on climate change and natural disasters, challenged the claim of scientific consensus with examples of pointed disagreement.

    “Hardly a week goes by,” Dr. Peiser said, “without a new research paper that questions part or even some basics of climate change theory,” including some reports that offer alternatives to human activity for global warming.

    Geologists have documented age upon age of climate swings, and some charge Mr. Gore with ignoring such rhythms.

    “Nowhere does Mr. Gore tell his audience that all of the phenomena that he describes fall within the natural range of environmental change on our planet,” Robert M. Carter, a marine geologist at James Cook University in Australia, said in a September blog. “Nor does he present any evidence that climate during the 20th century departed discernibly from its historical pattern of constant change.”

    Pretty evenhanded report, especially considering where it was published.

    Happy reading–
    Pete Tillman

  26. Peter D. Tillman
    Posted Mar 18, 2007 at 5:44 PM | Permalink

    re #23 , commentary by novelist/science fiction writer Orson Scott Card on the Hockey Stick and M&M’s efforts.

    Card is widely considered to be a blowhard & flake in the science-fiction community, so this is a mixed blessing. I haven’t read the comment.

    Cheers — Pete Tillman

  27. bernie
    Posted Mar 18, 2007 at 6:31 PM | Permalink

    You raised an interesting issue as to the strategic and tactical value of addressing all the issues raised by the proponents of the reliability of temperature measurements and as a consequence a strong AGW effect. The strategic imperative I believe is to push hard on the validity of the temperature measures and the more general issue of insisting on appropriate processes for doing “science”. The latter is Steve and Ross’s major focus and the absence of full disclosure is the IPCC protected cancer. Tactically, some of us, i.e., me , do not have the skills, knowledge or time to address the major issues except to cheer from the sidelines. However, we still know enough to pick off the flawed arguments of those jumping on the AGW bandwagon and adding to its inertia. To some extent these secondary issues are linked to the general issue, but will not really sway the IPCC and related vested interests. We probably need more polemicists like Michael Crichton but with more scientific credentials to level the playing field. In the mean time, we should keep putting more information in the hopper and those with skills and knowledge should not be afraid to ask for the support and help they need. Steve should certainly not be afraid to exert as much editiorial direction as he sees as needed.

  28. Willis Eschenbach
    Posted Mar 18, 2007 at 7:22 PM | Permalink

    S. Hales, many thanks for your post with the link to the WDCG, very interesting.

    A quick look at a variety of the sites seems to indicate that in Europe SO2 levels stayed high until the 1990s and then dropped, while in Indonesia they have not fallen. However, the record is so scarce and spotty that no conclusions can be drawn.

    I am very suspicious of the emissions data, because it only contains anthropogenic emissions. As you point out in a following post, natural sources contribute more than human sources. In addition, it’s not clear if anyone is including the millions of tonnes of SO2 emitted from underground coal fires. The underground fires in China are estimated to produce nearly as much CO2 as the entire US car and small truck fleet … how much SO2 are they emitting?

    The main problem that I have with the “aerosols like SO2 caused the 1945-1975 cooling” theory is that the aerosols are generally short-lived, leaving the atmosphere within weeks or months. Since they are emitted mainly in the Northern Hemisphere, where the industry is located, the Northern Hemisphere should have cooled more than the Southern Hemisphere during that period. But the record shows nothing of the sort. Here’s the NH – SH difference, as reported by HadCRUT3:

    As you can see, the SH temp decoupled from the NH temp, not in 1945, but in about 1920. And far from cooling less than the Northern Hemisphere as the aerosol theory would predict, the South cooled more than the North …

    In addition, the drop was very sharp, with both the NH and particularly the SH dropping after 1945 … how could aerosols cause that change?

    All of which makes the aerosol theory very, very doubtful in my book.


  29. Peter Lloyd
    Posted Mar 18, 2007 at 7:39 PM | Permalink

    Two of UK’s climate researchers warn against exaggerated claims for AGW


    above link, then click on Listen Again, today’s top story.

    And listen carefully between the lines.

    Can I hear the sound of a pendulum hovering before the backswing?

  30. Peter Lloyd
    Posted Mar 18, 2007 at 7:43 PM | Permalink

    re: 29 –

    Sorry, ONE researcher! And one Greenpeace rep.

  31. Follow the Money
    Posted Mar 18, 2007 at 7:51 PM | Permalink

    “Can I hear the sound of a pendulum hovering before the backswing?”

    Nothing like open talk of direct “green” taxes on the British public to raise the suspicions of the British public about the underlying premises of the government’s actions.

    The reality is carbon caps and trading are costing them more, but the losses are not readily apparent – one of the profitable advantages of cap and trade, though the people may be awakening to that too via the comical trade practices of direct from the consumer carbon credits (paying extra for an airline ticket, Al Gore’s business, etc.)

  32. David Smith
    Posted Mar 18, 2007 at 8:36 PM | Permalink

    Interesting news report on Gore’s upcoming visit to testify to the US Congress ( link )

  33. absolutely
    Posted Mar 18, 2007 at 9:22 PM | Permalink

    More on sun and climate: here

  34. S. Hales
    Posted Mar 18, 2007 at 9:39 PM | Permalink


    Jerry, Thanks for your reply. I didn’t know that the daily data was not the sole source of the gridded monthly data. But surely there is enough information in the daily files. I’ll have to process them. I’ll let everyone know what I find. Thanks for the info.

    Willis, Yeah its too bad that acid rain did not become a problem sooner or we’d have some interesting data on the supposed cooling effect of aerosols in the 70’s. But I think it is possible to quantify the athropogenic portion through time on a regional basis just from total fossil fuel consumption. I haven’t seen a reference on how aerosols have been quantified for that period to estimate their cooling effects but they certainly have been blamed for a lot of cooling or at least dampening of the CO2 signal. As far as the problems with SH temps I think the SH is plagued by lack of spatial coverage and a gridded regime can lead to these large grids containing only a few stations. I don’t think you should conclude any such decoupling. The MSU data show high correlation between NH and SH.

  35. bernie
    Posted Mar 18, 2007 at 9:40 PM | Permalink

    Robert Baker has no apparent track record in this area – apparently does dynamic modelling. Obviously he could be passionate about
    this like M&M and many others. Just that I couldn’t find anything he has written prior to this press release.

  36. Willis Eschenbach
    Posted Mar 18, 2007 at 11:18 PM | Permalink

    S. Hales, thanks for your interesting comments. You say:

    Willis, Yeah its too bad that acid rain did not become a problem sooner or we’d have some interesting data on the supposed cooling effect of aerosols in the 70’s. But I think it is possible to quantify the athropogenic portion through time on a regional basis just from total fossil fuel consumption. I haven’t seen a reference on how aerosols have been quantified for that period to estimate their cooling effects but they certainly have been blamed for a lot of cooling or at least dampening of the CO2 signal.

    The real problem is not that we can’t say how much aerosols are in the air because of a lack of data. The problem is that we don’t know what the aerosols do. This is because of the complexity of their interactions. They absorb both incoming and outgoing radiation, and they also serve as cloud nuclei, with an unknown total overall outcome. Also, the term “aerosols” covers a variety of things, including sulphate, organic carbon, black carbon, nitrate and dust, which all have different effects. I have never understood why they are all lumped together under the rubric of “aerosols” … why not try to understand their separate effects and account for the separately?

    As far as the problems with SH temps I think the SH is plagued by lack of spatial coverage and a gridded regime can lead to these large grids containing only a few stations. I don’t think you should conclude any such decoupling. The MSU data show high correlation between NH and SH.

    Actually, the MSU data show very different trends for the NH and the SH. Here’s the latest UAH data (which these days is not too much different from the RSS data):

    As you can see, the two are not very similar except in a gross overall manner. The R^2 coefficient between the two is only 0.37, which is pretty low. In addition, the trend in the SH is not statistically different from zero, while the NH trend is.

    All the best,


  37. Peter D. Tillman
    Posted Mar 19, 2007 at 1:24 AM | Permalink

    More on Gore:


    “Whose Ox Is Gored?
    The media discover the former vice president’s environmental exaggerations and hypocrisy.”

    The Tenessean reported yesterday that Mr. Gore received $570,000 in royalties from the owners of zinc mines who held mineral leases on his farm. The mines, which closed in 2003 but are scheduled to reopen under a new operator later this year, “emitted thousands of pounds of toxic substances and several times, the water discharged from the mines into nearby rivers had levels of toxins above what was legal.”
    Mr. Gore requested that SRA [the new mine operator] “engage with us in a process to ensure that the mine becomes a global example of environmental best practices.” The Tennessean dryly notes that Mr. Gore wrote the letter the week after the paper posed a series of questions to him about his involvement with the zinc mines.

    Happy reading–
    Pete Tillman

  38. Barry B.
    Posted Mar 19, 2007 at 1:24 AM | Permalink

    Re: #20 & 24

    You might find this of interest:

    Also, one might surmise that another consequence to higher temps might be increased yield variability. But as this paper shows:Trends and Variability in U.S. Corn Yields Over the Twentieth Century this is clearly not the case. In fact, as temperatures have increased yield variability has gone down. I can tell you as an Agronomist here in Illinois, since the early 1990’s we have been in a period of unprecedented yield stability. Yields have been good and if (a big IF) AGW is happening it most certainly is having a positive effect on corn and soybean yields.

    For you people discussing Hurricanes. I found this while searching for the link to the above paper: New evidence for a relationship between Atlantic tropical cyclone activity and African dust outbreaks

  39. Nicholas
    Posted Mar 19, 2007 at 4:13 AM | Permalink

    Here’s a strange little article by Orson Scott Card which tells the story Steve the Canadian Businessman (wonder who he is) and his friends:

    Civilization Watch

  40. Barry B.
    Posted Mar 19, 2007 at 6:39 AM | Permalink

    Hmm, not sure why the image didn’t show. Here it is again:

    If it doesn’t show, the chart can be found here: U.S. Corn Yield Chart

  41. Barry B.
    Posted Mar 19, 2007 at 6:43 AM | Permalink

    Very frustrating. It seems there is a big difference between the preview and what is actually posted. Go here: http://www.nass.usda.gov/Charts_and_Maps/Field_Crops/cornyld.asp

  42. Ralph Baskett
    Posted Mar 19, 2007 at 8:14 AM | Permalink

    Re: 20, 24, 40, 41

    In my judgment, as a farmer in the Mississippi Delta, it would be extremely difficult to find any useful correlation between temperature and weather and crop yields. High yields depends on optimum temperature (i.e., moderate for the particular crop) and moisture throughout the growing season. It can be too wet and cold at planting and early growth; when it is too hot, plants expend most of their energy just surviving; it can be too hot for corn, cotton and other crops to pollinate; and, it can be too wet at harvest and the crop can rot in the field. I have seen one year when temperature and moisture were adequate but constant cloud cover reduced photosynthesis enough to substantially reduce yield. To generalize is to state the obvious: crops do not grow well during ice ages and generally warmer is better. If one is concerned about world food supply in the future, genetically modified crops have changed the equation by making it possible for poor countries to control weeds and pests and attain much higher yields.

  43. S. Hales
    Posted Mar 19, 2007 at 8:40 AM | Permalink

    Willis thanks for your reply. The following applies to the MSU record for Northern and Southern Hemispheres of the MSU record from 12/1978 to the present. The two arrays are NH and SH and contain the relative monthly anomaly for each. The moving correlation is a 68 month moving correlation through time. This reveals periods of very high correlation to periods like today of very low correlation. There seems to be an emerging periodicity in the data or lagged responses to different forcings of different signs.

    The simple correlation coefficient for the two arrays is .61 which means very simply that 61% of the time that the two arrays move in the same direction. That is what I meant by high correlation. The ~5 year moving correlation has values as high as .72 and as low as .17. Interestingly there is evidence of your “decoupling” occuring in 2002 and continuing until today where the correlation coefficient fell from .72 to .2 today on a ~5 year moving correlation. I would expect in the coming years a return to high correlation if there is periodicity (or lagged responses) in the behavior of the temperature profile of the two hemispheres. I hadn’t graphed this function in years, I think the last time was in 1999 or so when correlations were high and had been stable for years except for a brief divergence between 1995 and 1997. So when you spoke of decoupling I was skeptical. But there is a definite decoupling beginning in 2002 and continuing until today. This is quite surprising to me.

  44. Steve Sadlov
    Posted Mar 19, 2007 at 8:50 AM | Permalink

    RE: #1 – given the impact of the PRC and other rising industrial countries, I seriously doubt that SO2 has continued to fall. It might have fallen for a short while after 1990, but it probably started back up again later. Would love to see confirmation or denial of my hunch.

  45. Steve Sadlov
    Posted Mar 19, 2007 at 8:57 AM | Permalink

    RE: #43 – You are in a very interesting area. I never cease to be amazed at how dramatically the weather there can vary over short time frames. A truely “fire and ice” type of place. Some might say that about say, the midwest, but at least there, you don’t go from 90 deg F and humid, with an air mass from the Gulf, to freezing with an air mass from Western Canada, in a single week in November, LOL! ….

  46. Ian Blanchard
    Posted Mar 19, 2007 at 9:00 AM | Permalink

    Surely you can see this in every HADCRU and NASA-GISS graph. After all, they’ve been essentially flat for the last decade.
    Surely evidence of the cooling effect of Chinese aerosols?

    (Please don’t think this is entirely serious)

  47. bernie
    Posted Mar 19, 2007 at 9:12 AM | Permalink

    This looks like a helpful article. As an agronomist, why would a longitudinal analysis of yields not include data on either price or total acreage under the crop. Surely farmers increase or decrease their acreage based on price. I noted during a quick skim of this article that for example the % of corn grown under irrigation varies through time and has declined inrecent years. I would assume that this non-irrigated land with its lower yields would be at the margin and would be added or reduced as market conditions dictate. Its as though the economic behavior of the farmer is being ignored. Is there something I am overlooking?

  48. jae
    Posted Mar 19, 2007 at 9:14 AM | Permalink

    Oh, great, now John and Teresa Kerry are becoming champions of AGW and all the other environmental catastrophes that we are causing. LOL.

  49. bernie
    Posted Mar 19, 2007 at 9:23 AM | Permalink

    It is heartening to see the backgrounds of people interested in and contributing to this site. A huge reservoir of scientific and practical knowledge.
    My surmise is that there is probably an optimal temperature/precipitation relationship with yield and probably a pretty complex one. Statistically this would argue for some kind of interaction effect variable in the regression equation or model that Lobell and Field used. Since they never defined their model, we do not know if they included such a variable. Let me also pose the same questions I posed to Barry: What impact does price (and/or government incentives/disincentives) have on your decision to keep land under a particular crop and how do you decide which land to drop or switch?

  50. Ken Fritsch
    Posted Mar 19, 2007 at 9:40 AM | Permalink

    Re: #42

    Barry B., I have seen that graph in my browsing for background information on crop yields and I was struck also by the lessening of deviations from the trend line in recent years. Perhaps you can help direct me to some data that I want to find and analyze in the near future.

    Going back and reviewing some recent literature on the relationship between corn (and soybean) between climate factors such as temperature and moisture and yield I found that the photosynthesis of these plants is temperature dependent with the process tending to plateau at higher temperatures. The major detrimental effect I have found for higher day and night time temperatures on corn are all related to transpiration rates and the related stress that comes with higher temperatures when the higher temperatures occur with a lack of moisture and at critical process times such as pollination.

    Also a critical time that is reported for these crops is at the start of the growing season when certainly warmer weather favors higher yields. Notice that in the Lobell and Field article that the authors “search” for the months that give the best correlation with the max/min temperature and moisture to yield. I would think that if one were going to model the crop yield sensitivity to temperature and moisture one would have to account for the entire growing season and even look at interactions between moisture and temperature. For corn they selected the July/August period which from what I have gathered would put most of the sensitivity on the pollination period where high temperatures and low moisture can have very large effects on yields. I would guess that if the authors had looked at the pollination period on an area by area basis that the temperature/moisture effect would have been even more significant.

    The effect that they might be seeing, however, is not small incremental changes in temperature that they attribute to AGW as this would not make much sense in the view of what we know about corn (and soybean) development. I need to look in more detail but I would guess that they are looking at the effects of large swings in temperature and moisture that happen cyclically with or without AGW. My farm experiences have been that high temperatures combined with low moisture are what have historically significantly depressed corn and soybean yields. Other climate factors that I recollect having large detrimental yield effects were cold and wet springs, flooding, late spring frosts, and hail damage; however, these factors could be much more localized than temperature/moisture.

    I want to look particularly closely at the natural weather cycles that produce more extreme conditions such as those produced by the ENSO phenomena and see how they play into the Lobell and Field correlations. Also the less variation in current corn yields goes along with my observation that genetics have made the corn plant less susceptible to heat/moisture stress.

  51. Aileni Noyle
    Posted Mar 19, 2007 at 10:11 AM | Permalink

    I have just read that New Scientist was ‘inclined to ignore’ C4’s ‘Swindle’.
    I am having difficulty getting my head around this. Are they going crazy, is this Post Normal Science?

  52. Barry B.
    Posted Mar 19, 2007 at 10:22 AM | Permalink

    Re: #48

    bernie, increases in corn and soybean prices are largely driven by weather related events. Because these events tend to be short-lived, price spikes are also relatively short-term, thus acreage shifts are usually minimal. Also, most of the marginal land here in the U.S. is in some type of conservation program such as CRP & CREP so there is not a whole lot of “new” ground that can be put into production. Considering these two factors, I believe this is why the authors did not take acreage shifts into consideration. Historically, there has just not been enough fluctuation to have a large impact on their study.

    Re: #51

    Ken, Since your post will take a more detailed response than my current time allows, I will get back to you later this evening.

  53. S. Hales
    Posted Mar 19, 2007 at 10:32 AM | Permalink


    Aileni, Of course its post normal science. What really bugs me is that the skeptics can never put together a decent analysis. They keep trodding over the same barren landscape and commit egregious errors. It is just incompetence on their part. Hans Erren argues from the solid framework of sensitivities, CO2 lifetime and relative rates of change. These are all fertile fields for the skeptics to grow their fruit. But they continue cultivating weak arguments on CO2/Temp ice core lags as if that’s evidence of no effect. But lagged responses in the poor resolution of the ice cores is what one might expect. The 800 year lag encompases about 1/6th of the ice age cycle but the other 5/6ths of cycle length encompass a variety of lagged responses to increases in CO2 and insolation. These are the normal feedbacks with variable lags. The magnitude of the temp swings seem to indicate a diminished role for CO2 and a major role for insolation as well as ocean heat content, as the Milankovitch cycle progresses. See, the skeptics could get creative with the ice cores but they don’t. They are incompetent when it comes to the media. It really ticks me off.

  54. Jeff Norman
    Posted Mar 19, 2007 at 10:58 AM | Permalink

    Re:#303 (from Unthreaded #6)

    Pat Frank says:

    Just yesterday night, I plotted the NASA/GISS global temperature anomalies for 1913-1944 against those for 1973-2004. The slope of a linear fit to those data was 0.97. The r’-value was about 0.75 because the scatter was high, but the points were pretty symmetric about the fitted line. There doesn’t seem to be any particular increase in relative rate of warming at all in the later period. Over the same period, by the way, the comparative rate of CO2 increase was 2.5, which means warming in the later period should have been 40% faster than in the earlier. It appears that the rate of warming isn’t following CO2 at all.

    You may want to file this under “fools seldom differ” but I did almost exactly the same thing and found almost exactly the same results which lead me to almost the same conclusion.

  55. bernie
    Posted Mar 19, 2007 at 11:02 AM | Permalink

    I can quite imagine that the total farmed acreage is probably pretty near constant. And if Lobell and Field had even acknowledged the price variable and discounted it, then I would have little to complain about. Alas they (and the follow up article) simply ignored it – as if taxing hydrocarbons would have no impact on CO2 emissions.
    To your point, however, there still may be sufficient shifting at the margin to explain a significant portion of Lobell and Fields results. I just found this which suggests that the volume of land can vary fairly significantly – and, therefore, I assume possibly the yield though I have to imagine that the actual decision may be only indirectly if at all connected to yield estimates. Now this is clearly a complex issue and my naive view is that the Lobell and Field’s model is far too simple to warrant the conclusions that it reaches. The model also suggests there is a mindset that “if you only have a hammer, everything begins to look like a nail.” with AGW as the hammer and any time series data a nail.

    I guess this topic is useful if it serves to demonstrate how potentially bad data is feeding into other studies and that many of these studies are peculiarly flawed when viewed by the uninitiated skeptics. However, I do not want to take us too far from the intent of this site.

  56. Steve Sadlov
    Posted Mar 19, 2007 at 11:26 AM | Permalink

    In spite of the all but certain flip into (the dry south / SW US severe repeated drought version of) La Nina and negative PDO, and, the driest California rainy season to date in many years (since at least 1994 if not 1991, need to check on that later), a tiny sliver of hope – we’ll take it!


  57. JerryB
    Posted Mar 19, 2007 at 11:40 AM | Permalink

    Re #34,

    > I didn’t know that the daily data was not the sole source of the gridded monthly data.

    Let me express some of the differences in a different way.

    Of the 32,815 locations for which the daily GHCN site has data,
    about 16,334 of the files contain temperature data, more than
    twice the number of locations included in the monthly GHCN
    temperature data files. The station list of the daily GHCN
    files includes many towns/cities not listed in the station
    list of the monthly GHCN data. The station list of monthly
    GHCN data include many cities/towns not included in the
    station list of the daily GHCN data.

    In other words, do not let the presence of GHCN in the names
    of both collections of data cause you to infer a relationship
    that may be misleading.

  58. Steve McIntyre
    Posted Mar 19, 2007 at 11:40 AM | Permalink

    Here’s an announcement from a small Canadian company involved in the proposed revival of zinc operations in Tennessee. Some information will come out during the stock offerings.

    Here’s a story of how Al got the land:

    Al Gore Jr.’s involvement in mining can be traced to Sept. 22, 1973.

    Former U.S. Sen. Albert Gore Sr. bought about 88 acres along the Caney Fork River from Occidental Minerals, a subsidiary of Occidental Petroleum, for $160,000. Included in the deal was the subsurface area. The rights to the minerals below ground were then leased back to Occidental.

    On the same day, Gore Sr. sold the land and subsurface area to his 25-year-old son and daughter-in-law for $140,000. The mineral lease to Occidental was put in their names.

    Kreider said the terms of the 30-year agreement provided the Gores “no legal recourse” even if they had wanted to cancel it, Kreider said.

    The Gores, she said, would not comment on whether they tried to pursue legal action to void the lease. “There is a certain zone of privacy once people go into private life.”

    Gore received $20,000 a year for 27 years and $10,000 a year for three years, making a total of $570,000 in lease payments. Kreider said the Gores never considered selling the land.

    She said the lease has to be viewed in a “1973 context, not a 2007 context.”

    Viewing the lease in a 1973 context as the author urges: you could buy a house in Toronto for about $30,000 in 1973. $20,000 a year was a lot of money back then.

  59. Ralph Baskett
    Posted Mar 19, 2007 at 11:48 AM | Permalink

    Re: 50
    With regard to your question:

    “What impact does price (and/or government incentives/disincentives) have on your decision to keep land under a particular crop and how do you decide which land to drop or switch?”

    Profit should be the deciding factor but there are several qualifications: A higher price may not produce a larger profit if costs outrun prices. Landlords usually insist that a higher value crop be planted even if the price is low because it produces greater rental return, i.e. rent for cotton ground is about twice the rent for soybean ground. Farming areas such as the U. S. Midwest do not have any profitable alternative crops to grow such as cotton or rice which are presently suffering from low prices. Government farm programs in the U. S., China, the European Union and other countries have so large an effect on the price that there is really no true market. Among other consequences, these programs tend to cause crops to be grown no matter what the price is. Input costs such as equipment, chemicals, land prices or rents are relative to the price of the commodity so that eliminating the subsidies would bankrupt the farm industry’€”for example, you cannot renegotiate the price you paid for farmland and farm equipment.
    At present, the government’s ethanol requirements and subsidies have caused the corn price to almost double and this is moving other commodity prices as well. Although it is hard to get an unbiased assessment of the economic and environmental consequences of producing ethanol from corn to replace some gasoline consumption, I suspect it will be a lose-lose solution for everyone but the farmer’€”higher prices for fuel, meat, eggs, milk and many other foods with practically no environmental benefit. High corn prices pull up soybean prices and are causing many southern farmers to switch and reduce cotton and rice acreage. Over the longer term, this should cause cotton and rice prices to increase if the corn price bubble does not burst. Commodity prices are dependent upon the overall economy. They are the first to decline and the last to recover. The bottom line: rational analysis is the exception rather than the rule.

  60. Gerald Machnee
    Posted Mar 19, 2007 at 11:49 AM | Permalink

    Re #59 – They were cited for emitting excess pollution into the river. New management is taking over and pland to be compliant now.

  61. James Erlandson
    Posted Mar 19, 2007 at 11:55 AM | Permalink

    Re 59: “She said the lease has to be viewed in a “1973 context, not a 2007 context.”

    1973 was four years after the Cuyahoga River (in Cleveland, Ohio) famously caught fire. And one year after the Federal Water Pollution Amendments of 1972, better known as the Clean Water Act. The bad effects of water pollution were not unknown in 1973.

  62. Steve McIntyre
    Posted Mar 19, 2007 at 12:00 PM | Permalink

    These are very marginal zinc mines. You see marginal operations like this get promoted during a commodity boom. These mines are not being re-opened by a big player. When the price of zinc goes down again, I’d be surprised if the operations survive.

  63. DocMartyn
    Posted Mar 19, 2007 at 12:02 PM | Permalink

    Sun’s pulse ‘points to rain’
    By Matthew Warren

    March 19, 2007 12:00am
    DROUGHT-BREAKING rains across eastern Australia have been predicted in new modelling by a scientist who believes massive pulses in the sun’s magnetic field are helping to drive the Earth’s climate systems.

    If proven, the research will make the prediction of floods and droughts in Australia far more reliable and influence models projecting future climate change.

    Robert Baker, from the University of New England, claims to have found a strong relationship between the rhythmic pulsing of the sun’s magnetic field and weather systems, particularly in the southern hemisphere.

    The sun’s magnetic emissions are known to peak every 11 years, a phenomenon demonstrated by increased sunspot activity. The sun also switches poles every 11 years. It last flipped in 2001.

    Associate Professor Baker said modelling of the sun’s magnetic activity showed high rainfall during times of high activity and drought when the sun was stable.

    This suggested the fluctuations impacted on the upper atmosphere, which was in turn reflected in changes in the Southern Oscillation Index (SOI), the measure of air pressure over the Pacific Ocean used as a reliable indicator of drought and flood.

    Dr Baker said the most intense droughts in eastern Australia, including the Federation drought, tended to occur every 22 years, about a year after the southern pole of the sun flipped and became positively charged.

    In a paper, which has been submitted to the journal Solar Terrestrial Physics for peer review, he claims changes in solar magnetic fields can explain about 50 per cent of the variation in the SOI.

    The impact of solar magnetism is more noticeable in the southern hemisphere and in regions such as eastern Australia because more variable climate is driven by proximity to large oceans.

    He said this relationship between the sun and climate required further research because it might help to explain an important but little understood natural cycle influencing the Earth’s climate systems.

    “The sun drives the whole system,” he said. “There is a natural impact that the sun has in terms of weather patterns maybe over a century.”

    Dr Baker said the sun appeared to follow a longer-term magnetic cycle of about 80 years, meaning it might be possible to predict floods and droughts for the next 30 years based on historical records from the mid-1920s.

    Dr Baker said the SOI was currently following a similar pattern to that recorded after 1924 when eastern Australia enjoyed heavy falls after a period of prolonged drought.

    Dr Baker’s model puts a more scientific and transparent theory to the concepts first developed by long-range weather forecasters Lennox Walker and Inigo Jones.

    It also suggests there may be a longer 500-year solar cycle, which may help explain climate variability over the past centuries, including periods of unexplained climate variability such as the Medieval Warm Period and the Little Ice Age.

    Dr Baker said he was concerned about the welfare of rural communities amid unfounded speculation the current drought might continue for decades.


  64. Peter Lloyd
    Posted Mar 19, 2007 at 12:12 PM | Permalink

    re:46/Steve Sadlov –

    Oh, yes you do!

    I used to live just west of Cleveland. I remember coming out of a neighborhood party one winter evening (early 1980s) when the air temperature was below -40 deg.F. Only just got the car started, firing on one for the first couple of miles, not firing on all eight all the way home. Damn cold. By the following mid-afternoon, air temperature was just over 60 deg.F. One hundred degree F temp. rise in less than 24 hrs.
    Extreme example, but used to be typical winter weather for Burning River City until the 90s. As you say, due to the interface of Gulf air mass and Canadian Arctic air mass flipping just a hundred miles or so across the Great Lakes.

  65. S. Hales
    Posted Mar 19, 2007 at 12:15 PM | Permalink


    This is all rather absurd. Thanks for the warning. Here you have the daily files, which are updated daily, but are not the sum total of all the stations used to produce the v2 dataset. Another group of files “mysteriously” exist somewhere that make up part of the v2 dataset. Do you have any idea where this other data (daily I presume) might reside other than in its v2 incarnation?

  66. JerryB
    Posted Mar 19, 2007 at 1:06 PM | Permalink

    Re #66,

    S H,

    page linking PDF files describing (monthly) GHCN V2
    and how it came about.

    I have not seen comparable documentation of the daily GHCN collection.
    BTW, the daily GHCN website is only three months old; it’s a
    new kid on the block.

  67. JerryB
    Posted Mar 19, 2007 at 1:19 PM | Permalink

    From the paper announcing the release of GHCN V2 in 1997:

    “In the digital archives, there are many more station years of monthly
    data available than daily data with correspondingly much better spatial

  68. Steve Sadlov
    Posted Mar 19, 2007 at 1:33 PM | Permalink

    RE: #65 – You’ve made my point. I’d challenge you to find a single example of, specifically, 90F (or in some cases hotter) down to freezing, in November, anywhere in Ohio. Whereas, in the “hot” mid south, some of the sudden extremes on the cold side really surprise many people not familiar with the area – especially considering the latitude is in the low 30s.

  69. Follow the Money
    Posted Mar 19, 2007 at 1:39 PM | Permalink

    Aileni, #52:

    “”I have just read that New Scientist was inclined to ignore’ C4’s Swindle’.
    I am having difficulty getting my head around this. Are they going crazy, is this Post Normal Science?””

    Very funny!

    Correct, “post normal science” is not a “craziness.” More an ego defense mechanism shared (or to be shared) among the scientific community for whom selling out for money or conformity clashes with real science i.e, “skepticism”. Hijinx/reaction formation ensues…

  70. James Erlandson
    Posted Mar 19, 2007 at 2:49 PM | Permalink

    Hurricane prediction from Lloyd’s and Dow Jones:

    “Despite the complexities associated with hurricane forecasting, current conditions and climate models lend strong support to a much more active hurricane season than we experienced in 2006,” said Jim Rouiller, senior energy meteorologist with Planalytics in Philadelphia in a client’s report.

    Warmer-than-normal sea surface temperatures and a La Nina pattern, which produces light winds, are two meteorological signs that point to an open path for hurricanes to develop, Rouiller said. Both are in place right now, he said.

    Rouiller also said we are in the middle of a “tropical Atlantic multi-decadal signal,” which means enhanced frequency of hurricane development as pressure patterns and sea surface temperatures across the Atlantic are supportive of this potential.

    “We’re dead center in the period of maximum activity,” he said.

    But then he tempers his bullish prediction, The one wild card that could hamper or inhibit hurricane development in 2007 are (sic) Saharan sandstorms … “They are hurricane killers,” he said. “Hurricanes don’t like dry, dusty air and wind.”
    Time for Steve M. to put on his pointed hat, wave his magic wand, open the envelope and give us his prediction for 2007.

  71. Howastz Szat
    Posted Mar 19, 2007 at 3:27 PM | Permalink

    Interesting read in Science daily : Researchers Question Validity Of A ‘Global Temperature’

  72. Peter Lloyd
    Posted Mar 19, 2007 at 3:33 PM | Permalink

    re: 69/Steve Sadlov

    Apologies – I misunderstood your original point.

    Us Clevelanders used to laugh like crazy at the evening newsreels of Southerners sliding around on about 1/10th inch of ice when they had a cold snap…….but we certainly never came anywhere near 90 in November. Lucky to get a couple of days at 60 before winter relly set in.

  73. David Smith
    Posted Mar 19, 2007 at 3:43 PM | Permalink

    Re #71 I’m eagerly awaiting the European hurricane forecast for 2007, which is reputed to be considerably more accurate than the NOAA and other forecasts.

    Apparently the Europeans run GCM(s) for extended periods (six or eight months) and count the number/strength of storms the GCM(s) generate in that period. That is different from the pattern/analogue methodology of other forecasters, like NOAA, Gray, Lloyd’s, etc.

    With luck, the new European forecast will be made public before the 2007 season ends.

  74. bernie
    Posted Mar 19, 2007 at 3:54 PM | Permalink

    We should ask Ross to see if this is an acceptable summary of some of his paper. On the surface it looks like a clear statement of one aspect of the overall problem. I liked the notion of a universal exchange rate – it gives a sense as to the complexities behind the simplistic notion of a GMT.
    Thanks for the link.

  75. L Nettles
    Posted Mar 19, 2007 at 4:04 PM | Permalink

    re 71 Here’s Mr. Rouiller in August of 2006

    Depression Likely to Become Hurricane Next Week -Planalytics
    Date: Saturday, August 26 @ 13:22:27 EDT
    Topic: Enviromental Headlines; Climate Change

    The fifth tropical depression of the 2006 Atlantic storm season likely will become a hurricane as it moves into the U.S. Gulf of Mexico early next week, a private weather forecaster said Friday. “A strong consensus from hurricane tracking models indicate this storm will become a hurricane as it moves into the Gulf early next week and could threaten the Gulf production and rig infrastructure during the later half of next week,” Jim Rouiller, senior energy meteorologist at Planalytics, said in a note.

    Ernesto didn’t become a Hurricane and didn’t get to the Gulf

  76. Steve Sadlov
    Posted Mar 19, 2007 at 4:05 PM | Permalink

    The captains of industry have drunk the Kool Aid, see:


    I think I am going to be ill ….

  77. Posted Mar 19, 2007 at 5:14 PM | Permalink


    Will five years of really cold weather save us, or will it be too late? Should we pray for a Daulton Minimum, 25 years of very cold dry weather?

  78. Ken Fritsch
    Posted Mar 19, 2007 at 5:29 PM | Permalink

    Re: #59

    Former U.S. Sen. Albert Gore Sr. bought about 88 acres along the Caney Fork River from Occidental Minerals, a subsidiary of Occidental Petroleum, for $160,000. Included in the deal was the subsurface area. The rights to the minerals below ground were then leased back to Occidental.

    ..Gore received $20,000 a year for 27 years and $10,000 a year for three years, making a total of $570,000 in lease payments. Kreider said the Gores never considered selling the land.

    I do not know what potential minerals are buried in Gore’s 88 acres, but I do know that the mineral rights that my siblings and I recently sold from considerably more acreage than that from my grandparent’s homestead in Montana did not approach the worth of the Gore’s gold mine.

  79. John A
    Posted Mar 19, 2007 at 5:48 PM | Permalink

    Will five years of really cold weather save us, or will it be too late? Should we pray for a Daulton Minimum, 25 years of very cold dry weather?

    Be careful what you wish for. I’d rather climate modellers had five more years of smugness than a Dalton-minimum style cool down, which would kill hundreds of thousands in cold and drought.

  80. Steve Sadlov
    Posted Mar 19, 2007 at 6:29 PM | Permalink

    RE: #80 -correction, would kill millions. Then, in the war(s) to follow …. the number starts with a “b” ….

  81. David Smith
    Posted Mar 19, 2007 at 6:46 PM | Permalink

    The forecast is for weak La Nina (cool) conditions by this summer, along with a cooling Atlantic. See slide 28 of the latest outlook . The SST forecasts are the small maps on the left, each covering a three-month span. By hurricane season the tropical SST are near-normal.

    The subsurface equatorial Pacific also continues to cool, even at the deeper depths. 2007 is shaping up as an interesting year, temperature-wise.

  82. TAC
    Posted Mar 19, 2007 at 7:21 PM | Permalink

    Kerry Emanuel has a comment on Michaels et al. [2006] “Sea-surface temperatures and tropical cyclones in the Atlantic basin”. Michaels et al. reply here. Subscription (or $9 each) required.

  83. Pat Frank
    Posted Mar 19, 2007 at 7:49 PM | Permalink

    #59 — Zinc ores often contain a significant lead component. According to this Tennessee government environmental report (pdf linked), the Caney Fork mines zinc ore is apparently one such. I wonder what they’re doing with the lead.

  84. Follow the Money
    Posted Mar 19, 2007 at 8:45 PM | Permalink

    # 77

    “I think I am going to be ill ….”

    Actually, that list could be worse.

    Lots of carbon traders on the list.

    Look at the pension fund advocates – maybe someone should write them to put them “on notice” about their fiduciary obligations in investing in carbon trades re: what occured in Europe.

  85. TAC
    Posted Mar 19, 2007 at 8:47 PM | Permalink

    WRT #83, Emanuel summarizes his criticism of Michaels et al. (aka “MKD”):

    MKD’s central hypothesis is refuted: Lack of correlation of high intensity events with SST (or potential intensity) in a particular climate does not imply that temporally increasing potential intensity (SST) will have no significant effect on tropical cyclone activity; indeed observed time trends in tropical cyclone energy are highly significant and strongly correlated with SST, as observations clearly show. Nor is there any indication that the maximum wind speed achievable in hurricanes levels off at high sea surface temperature; both theories and models show a smooth, continuous increase. MKD’s conclusion that tropical cyclone intensity depends on SST only in a certain range of the latter is likewise false, arising from a confusion between spatial gradients of potential intensity, that depend mostly on the very large increase of outflow temperature from the tropics to the subtropics, with variability within the tropics, where the outflow temperature is less variable.

    In response, MKD conclude:

    We thank Emanuel [2007] for analyses that provide further detail about the complex relationship between SST and tropical cyclone intensity in the Atlantic basin. In agreement with Michaels et al. [2006], Emanuel’s modelbased findings confirm the lack of a strong, statistically significant SST signal in the average intensity of tropical cyclones encountering high SST (i.e., above the 28.25C threshold). There is, however, some indication that higher SST may lead to more intense strong storms. Currently, Figure 1. Scatterplot of maximum wind speed and the highest SST encountered prior to (or concurrent with) reaching the maximum wind speed for all Atlantic basin major (category 3, 4, 5) hurricanes. The two storms in the lower left (open circles) did not encounter the 28.25C threshold (vertical line) that is typically a prerequisite for major hurricane development. 2 of 3 observations can neither confirm nor refute that relationship. We do not contend that increasing SST has not, or will not, influence the intensity or frequency of Atlantic tropical cyclones. We have demonstrated in our initial findings [Michaels et al., 2006], and Emanuel has furthered in his comments, that a modest impact is likely. However, the overall weakness in the relationship between SST and storm intensity indicates that SST is only one of myriad determinants of tropical cyclone strength. This complicates clear attribution of observed or future changes. These results reaffirm our original conclusions that the full reason for the observed changes in tropical cyclone activity in the North Atlantic has yet to be established, and that we therefore recommend a cautious approach to assigning an underlying cause in this complex system [Michaels et al., 2006].”

    My sense is that Emanuel has raised a couple of interesting points. I will leave it to others to characterize their relevance to the findings of the original MKD paper.

  86. Barry B.
    Posted Mar 19, 2007 at 8:47 PM | Permalink

    Re: #51

    Ken, my initial reaction to Lobell and Field’s “search” for the months which show the best correlation to yield was of great concern. But after closer examination I would say they are spot-on for soybeans. The vegetative growth stages of soybeans have a very small influence on final yield as long as the initial stand is acceptable so I really see no reason why they need to take into account the entire growing season. It is not until the soybean plant reaches the reproductive stages of growth, which occurs from late June to August here in the Midwest, does the environment start to have a major impact on yield. This in my opinion gives credence to their decision to select the months of July/August for soybeans.

    Corn on the other hand is a whole different beast altogether. I can think of four critical periods in a corn plants development which can have a major impact on yield.

    First is planting date. Corn is much more sensitive to planting date than soybeans so any delays in planting beyond the optimum planting dates (early to mid-May in the Midwest) can affect yield. As you move farther away from the optimum, yields drop off precipitously so delays can get costly.

    Second is the vegetative growth stages between V6-V10 which is when the ear is being formed within the plant. At this time, ear length and number of kernels are determined so any stresses upon the plant during this period will adversely affect yield also.

    Third is the period of pollination which is probably the most critical time in a corn plants life. Excessive heat combined with moisture stress will reduce yield. The key here is heat and lack of sufficient moisture together. You can have high heat but adequate moisture and the crop will be fine. It’s the two together that’s the killer.

    Lastly, is the kernel fill period which is the end of the reproductive growth stage. Again, heat combined with moisture stress will negatively impact yield.

    As you can see, corn has a much broader range of critical growing periods so Lobell and Field’s selection of July/August is somewhat suspicious to me. But then again, correlating high temps to yield during the first two critical stages I outlined above would be almost impossible because problems during the stages of development are almost entirely moisture related (i.e. too much moisture) which I guess is why they chose the July/August time frame.

    You mention ENSO and although I’m a climate novice, I do know it has a huge impact on our crops across the United States. There’s hardly a day that goes by when we don’t hear some type of agricultural weather commentary discussing how the phenomena is/will impact our production. It’ll be interesting to see what the results are when you get your comparison to Lobell and Field’s correlations completed. If you need help finding any agricultural data let me know. I’ll be happy to help.

  87. Steve McIntyre
    Posted Mar 19, 2007 at 8:57 PM | Permalink

    I’m getting a little puzzled about Al Gore’s carbon offset purchases and his arrangements with the company. As I understand it, he did not actually purchase carbon offsets himself, but Generation Investment Management purchased carbon offsets for him as a member of senior management as part of their executive compensation package. If this was executive compensation, then should Generation Investment Management issue Al Gore what in Canada would be a T-4 slip and pay deductions to the federal government? It’s a curious arrangement when you think about it: the more profligate Al is with his energy consumption, the more carbon offsets are supposedly purchased by Gen Inv and thus the higher his personal income (assuming that the purchases on his behalf are income to him).

    In an unrelated news story, the Toronto papers are full of the Conrad Black trial in Chicago. Conrad Black, a famous Canadian, was the controlling shareholder of several public companies and is charged with using company funds as though they were personal purposes.

  88. Barry B.
    Posted Mar 19, 2007 at 8:58 PM | Permalink

    Re: # 56

    I don’t know bernie. I’m having a hard time believing that acreage shifts could have anything but a small impact on Lobell and Field’s results. Without further analysis though, it’s going to be nigh impossible to know for sure.

    However I will say that the makeup of yield is a very complex issue and I’m not sure Lobell and Field accurately account for the multitude of variables which influence it. But again, without further study it’s difficult to make an accurate judgement as to their conclusions. I wish I had the time to peruse through all the papers they referenced. Perhaps then one could shed more light on the subject.

  89. David Smith
    Posted Mar 19, 2007 at 10:01 PM | Permalink

    Re #86

    I’ve read both letters, and they actually left me feeling sad for Emanuel.

    I’d be glad to comment in detail, but I think there are limits to what can be publically extracted from for-fee papers.

  90. Ken Fritsch
    Posted Mar 19, 2007 at 10:25 PM | Permalink

    Re: #87

    Barry B., here is an update on my progress with checking Lobell and Field’s work. I will present graphs and details later as this is currently a work in progress.

    I found yield data for IL corn and soybeans going back to Lobell’s starting data of 1961 and used the data through 2006. I also found Max/Min temperatures and precipitation for areas in IL and used four well spaced locations that I estimated encompassed most of the IL grain belt. I have been able to obtain relatively high R^2 (approximately 0.4 to 0.5) for the correlations I did for corn and soybean deviations from their respective trend lines versus Max temperature and precipitation for the months July and August. I also did a product of precipitation and Max Temperature for these months versus the deviation of yields from their trend lines ((95-Max Temp) X Precipitation)and this fit was even better with R^2 for both crops near 0.6 — albeit using a log fit.

    On closer analysis one sees that the temperature and precipitation product has a plateau where most of the values result in essentially the same percentage yield increments from the trend line. At lower product values, that covers approximately 1/5 of the years, the yield increments are negative and decrease at a steep linear rate as the associated product decreases. These years with low moisture invariably are accompanied by higher than normal temperatures with the driest year having the highest temperature. The correlation between Max temperature and precipitation for this period in IL had an R^2 = 0.3.

    I did the same analysis using the Max/Min temperatures and precipitation for the months of April, May, June, July and August. My calculations to this point, that need to be rechecked, show absolutely no correlation for the soybean and corn increment % yield changes with Max temperatures or precipitation.

    From this preliminary work, I conclude that I can in essence reproduce what Lobell did and at the same time get some insights into what are the real factors operating in this process are. It would appear that IL corn and soybeans are critically affected by the weather prevailing in July and August and the effects of high Max temperatures could not be separated from low rates of precipitation as they appear together most of the time. One can see these effects of high temperatures and low moisture on corn and soybean yields by simply glancing at the tables of data. I am guessing now, but I suspect that this combination is produce cyclically. I need to look at the climate records going further back in time. I also want to look at the effects of the combination of precipitation for the months of April through August with Max temperatures for the months of July and August. I want to see what the effects of reserve soil moisture might have on this process.

  91. Ken Fritsch
    Posted Mar 19, 2007 at 10:35 PM | Permalink

    I belive I can look at the effects of putting more acreage into corn and soybeans after price increases as I have that information going back to 1961 and further for IL. There is a major debate going on amongst farmers currently about planting significantly more corn on acreage that would have been used for soybeans if the corn prices had not sky rocketed from the demand placed on it for added alcohol production.

  92. Ken Fritsch
    Posted Mar 19, 2007 at 10:41 PM | Permalink

    I’ve read both letters, and they actually left me feeling sad for Emanuel.

    I’d be glad to comment in detail, but I think there are limits to what can be publically extracted from for-fee papers.

    David Smith, I think you can, with clear conscious, at least summarize the letters for us cheapskates. We might want to shed a tear or two in sympathy for Emanuel — I think we agreed that he sounds like a very nice man.

  93. Roger Dueck
    Posted Mar 19, 2007 at 11:33 PM | Permalink

    jar #49

    “Both John and Teresa have been long-time leaders in the battle to save the Earth’s environment,” Gore stated. “Way back when it was not all fashionable, indeed when very few people in the world were even paying attention to it, both John and Teresa were providing outstanding and courageous leadership.”

    This from Al Gore. When was it not fashonable? Does he mean it now IS fashonable? Pol-i-tics!

  94. TAC
    Posted Mar 20, 2007 at 2:34 AM | Permalink

    #90, #93: I read both letters and tentatively reached similar conclusions. Emanuel confirms MKD’s findings, but points out that, in some sense, MKD employed (cherry-picked) statistics which reveal at best a weak correlation between SST and hurricane intensity. Emanuel shows, employing different (cherry-picked) statistics, that one could find a stronger relationship between storm intensity (PDI for Emanuel) and SST.

    Given how little data we have — particularly when one considers the time-series structure of the datasets and how this may reduce “degrees of freedom” — it is not surprising that different observers can find different patterns in the datasets. Torture data long enough and they confess.

    I was surprised by one thing. Emanuel criticizes MKD for [emphasis mine]:

    …plot[ting] peak storm intensity against the maximum SST that the storm encountered any time up to the time of peak intensity; given that this time lag may be many days and that the response time of tropical cyclones to changes in their environment is of order 15 hours, this is unphysical..

    Fair enough. It is always good to ensure that one’s statistical model corresponds to what we know about the physics. In fact, I think that’s what MKD tried to do.

    However, Emanuel then goes on to plot the PDI/SST relationship using a 5-year (1-3-4-3-1) filter on both variables. If 15 hours is the relevant time frame for the physics, how can one justify 5 years?

    In short, Emanuel has shown that one can get an interesting signal by employing a (1-3-4-3-1)-filtered “August’€”October sea surface temperature averaged over the region 6’€”18 N, 20’€”60 W.” This, to my ear, seems a bit arbitrary compared to MKD’s efforts to identify the SST corresponding to each storm. A skeptical reader would immediately wonder about the robustness of Emanuel’s findings.

    On the other hand, the central part of Emanuel’s conclusions are stated as a convoluted triple negative:

    Lack of correlation of high intensity events with SST (or potential intensity) in a particular climate does not imply that temporally increasing potential intensity (SST) will have no significant effect on tropical cyclone activity.

    IMHO: We’re going to have to wait for more data to settle the larger question.

  95. David Smith
    Posted Mar 20, 2007 at 5:00 AM | Permalink

    RE #93 Will do, Ken, by the weekend. TAC does a good job of capturing one of the main oddities. There are others. It’s a strange exchange.

  96. MarkW
    Posted Mar 20, 2007 at 5:15 AM | Permalink

    You also need to look at the information that is available to the farmers at the start of the growing season. I would imagine that
    this would affect their behavior. For example, if the information says that the season ahead is likely to be marginal in terms of
    weather, I would imagine that farmers would concentrate their efforts on their most productive fields. Letting the less productive
    fields go fallow for a year. (Good for the fields anyway.) Since in marginal years, it’s less likely that the yield from these
    secondary fields would, after subtracting cost of seed, fuel, fertilizer, herbicide, wear and tear on machine and land, would provide
    a profitable harvest.

    Basically, you can’t just look at the weather that did happen, but what the farmers thought the weather was going to be.

  97. bernie
    Posted Mar 20, 2007 at 5:40 AM | Permalink

    #91, #92
    Excellent. Great work. I guess the metapoint is that Lobell and Field would have helped us all out if they had specified their regression equation. The product or interaction term is clearly significant based upon your work. I can’t believe they did not use one in their model – which makes me wonder. I am still interested in what happens when you add price to your model – plus the theory of the weather that farmers carry around in their heads, e.g., dry seasons come in twos or threes, cool wet seasons are followed by hot wet – stuff like that which can influence theor behavior. Irrigation also seems to play a very significant role in adjusting yields, so that certainly should be a variable in the equation – the more so if you found an effect for hot-dry.
    As an aside, can you check the correlation between tmin and tmax? You may also want to limit the correlation with the product term by creating dummy variables for cool-wet, avg-avg, cool-dry, etc. But I am at the limits of my technical knowledge so perhaps someone else can help specify the equations that would minimize spurious correlations.

  98. bernie
    Posted Mar 20, 2007 at 6:03 AM | Permalink

    Good point Mark. That was one of the things I was trying to get at. There have to be behavioral economists out there who can pitch in on the expectations components that can be integrated into this model. I would imagine the forward price on some of these crops, dare I say it, could be a proxy for these expectations.
    But that brings me back to an earlier point of why Lobell and Field and others did not include such factors in their models. Surely, like solar variation and UHI, you need to eliminate obvious confounding variables in order to make strong statements.

  99. MarkW
    Posted Mar 20, 2007 at 6:22 AM | Permalink

    The futures market does play a big role, but the futures are an average for the whole country, perhaps the whole world since
    it is a world market.

    Each farmer would adjust the data from the futures market with his own feelings for what is likely to happen in his particular
    corner of the world. Many things affect that. From farmers almanac type historical data, like bernie mentioned, to guesses how
    the current nino/nina is going to affect rain patterns.

    All in all, this is a tough nut to crack.

  100. David Archibald
    Posted Mar 20, 2007 at 7:55 AM | Permalink

    Re 80 and 81, the Dalton Minimum was caused by sunspot cycles 5 and 6, which had maxima of 47.5 and 45.8 respectively. The preceding solar cycle, No 4, had a maximum of 130.9. Solar cycles are 10.7 years long on average. Solar cycle 4 was 13.6 years long. The longer a solar cycle, the weaker the following cycle is. Solar cycle 23, which we are still in, peaked in 2000 at 120.0. Now we haven’t seen a sunspot from solar cycle 24 yet, which, based on recent experience, means that solar minimum is still a year away. The longer we wait for a sunspot from solar cycle 24 (which will have reversed polarity to 23), the weaker solar cycle 24 will be, and the colder the earth will be. This is not a highly exact science yet, but by my calculations, each day’s delay in solar cycle 24 decreases the average temperature expected during the cycle by 0.001 degrees C. We are already a year late. It won’t take long to make something meaningful.

    Our generation has bathed in the warm glow of a benign, giving Sun, but the next generation will suffer a Sun that is less giving, and the Earth will be less fruitful.

  101. Ralph Baskett
    Posted Mar 20, 2007 at 8:17 AM | Permalink

    Re: 98, 99
    In my experience, I have never heard a farmer cite long-term weather forecasts as a factor in determining what crop to plant or how many acres to plant. I think this is because weather forecasts are usually wrong.
    In irrigated areas, high temperature and drought may not result in lower yields and can easily result in higher yields. The optimum conditions for cotton are 90-degree highs, low humidity and no rain! More marginal acres (i.e., less productive) are planted when the price is high which will almost certainly result in a lower average yield.
    Bernie, I attempted to answer your earlier question concerning how prices affect acres planted in comment 60.

  102. Steve Sadlov
    Posted Mar 20, 2007 at 10:01 AM | Permalink

    RE: #102 – “The optimum conditions for cotton are 90-degree highs, low humidity and no rain! ”

    Which is why cotton growers are attracted to the area which lies between State Rte 99, State Rte 132, and Interstate 5, here in California. Lots of bales coming out of that particular elongated triangular zone.

  103. Steve Sadlov
    Posted Mar 20, 2007 at 10:02 AM | Permalink

    RE: #77 – We completely scooped the main stream media. We got the story out several hours before they did.

  104. fFreddy
    Posted Mar 20, 2007 at 10:53 AM | Permalink

    Re #101, David Archibald

    the Dalton Minimum was caused by sunspot cycles 5 and 6, which had maxima of 47.5 and 45.8 respectively

    Sorry, what are the units of 47.5 and 45.8 ?

  105. Posted Mar 20, 2007 at 11:40 AM | Permalink

    Does anyone have a copy of

    Osborn TJ, Briffa KR, Jones PD (1997) Adjusting variance for sample-size in tree-ring chronologies and other regional-mean time-series. Dendrochronologia 15:89’€”99


  106. Steve McIntyre
    Posted Mar 20, 2007 at 11:49 AM | Permalink

    #106. I do. I can summarize the method briefly for you and will do a post on it. I’d be interested in what you think about this procedure as it is unknown to statisticians who haven’t dealt with the Team.

  107. bernie
    Posted Mar 20, 2007 at 12:07 PM | Permalink

    Many thanks for your earlier response at #60. I took from your earlier response that there is a complex interplay around pricing that involves near and non-complementary commodities. I agree. Moreover, I think it is OK if the model builders decide for simplicity sake to ignore certain classes of variables and effects. However, they then should provide a reasoned argument as to why these other variables can be ignored or they should avoid drawing significant conclusions for what are clearly partial models. The interesting thing for me, is that the Lobell and Field study is a microcosm of what seems to have happened with GCMs – too much ready, fire, aim.

  108. Posted Mar 20, 2007 at 12:19 PM | Permalink

    Steve, I’d appreciate a summary. If I understood the procedure correctly from

    Click to access Frank_2006_TRACE.pdf

    … well. I know some people who know some people who are world-class mathematicians. I’ll ask their opinion.

  109. JerryB
    Posted Mar 20, 2007 at 12:37 PM | Permalink

    Re #105,


    They are “wolf” sunspot numbers for calendar years 1804,
    and 1816, respectively. In between, 1810 had zero, and
    was the most recent calendar year to do so.

  110. Basil Copeland
    Posted Mar 20, 2007 at 12:43 PM | Permalink

    A question: does the case for AGW depend on relatively uniform trends in “global warming?” Certainly the rise in CO2 appears to be relatively steady. How does the possibility of secular shifts in temperature regime, as opposed to trends, relate to the AWG argument? Is the actual variability in global temperature (not seasonal, but say, decadal) simply considered (by the AWG hypothesis) as “noise” around a trend caused by rising CO2, or are their any physical models or hypotheses that would explain how a gradual rise in CO2 can account for all the “noise” in the actual record?

  111. cbone
    Posted Mar 20, 2007 at 1:06 PM | Permalink

    Shadow post from a submission I made to RC 2 hours ago. I guess it won’t make it through screening…

    Are there any assessments of the snowpack for the first half of the century? It seems rather ‘convenient’ to pick as your starting date for the assessment a point which coincides with the beginning of the mid-century cooling period associated with the 50’s-70’s. Wouldn’t a more accurate assesment be to compare today’s snowpack to that of the 20’s and 30’s during the peak of the early century warming?

  112. David Smith
    Posted Mar 20, 2007 at 1:08 PM | Permalink

    A convenient source for current solar conditions, as well as historical charts, is here

  113. Ken Fritsch
    Posted Mar 20, 2007 at 1:15 PM | Permalink

    To provide background for my IL crop yield analysis, I have provided below my sources for climate data (max/min temperatures and precipitation amounts and for yield data for corn and soybeans in IL.

    ISWS Climate Data for IL climate at specified IL locations:


    NASS IL Report for IL crop statistics including yields, acres planted and grain prices:


    For temperatures I used the average monthly values for the areas of Decatur, Dixon, Peoria and Jerseyville. A small amount of climate data was missing (always from only a single location) and that location was excluded from the 3 location average for that time period. The climate data were condensed into the table below for the months of July and August along with the IL annual soybean and corn yields (SoyBY and CornY) and the percentage deviations in these yields from the trend line are listed under CornTY and SoyBTY. Max and Min data are the averages of the July and August monthly averages of the daily high and low temperatures. Prec refers to precipitation in inches. Prod is the product of the maximum temperature (in degrees F) – 95 with this quantity multiplied times the precipitation or Prod = (95-Max) x Prec.

    If one scans the lowest values of Prod and looks at yields one can see a rapid fall off in yields from trend when its value drops into the low 30s and below. One can also note from viewing the table that dry periods are often accompanied with higher temperatures. I plan to show a couple of graphs with my next post and make some additional comments.

    The table set-up in the review looks like a great improvement for this board in handling tabled data– now lets see if it makes it through to the finished post.

    Year Max Min Prec SoyBY CornY Prod CornTY SoyBTY
    1961 85.0 62.6 5.0 28.5 79 50.6 -1.45% 3.88%
    1962 85.4 61.4 4.2 28.5 85 40.8 3.89% 2.36%
    1963 85.5 62.2 4.1 29.5 87 39.5 4.24% 4.42%
    1964 86.8 62.9 2.6 25 80 21.1 -6.01% -12.77%
    1965 84.3 62.2 6.1 29.5 94 65.5 8.34% 1.49%
    1966 87.5 63.1 1.9 27 82 14.7 -7.25% -8.40%
    1967 82.1 59.1 3.7 31 104 47.6 15.47% 3.74%
    1968 85.0 63.7 3.2 31.5 90 32.3 -1.87% 3.99%
    1969 86.2 64.4 3.9 34 102 34.7 9.25% 10.76%
    1970 85.8 63.6 3.3 31 74 30.7 -22.12% -0.34%
    1971 84.2 60.3 3.5 33 106 37.4 9.66% 4.72%
    1972 84.6 63.2 4.4 34.5 110 45.9 11.89% 8.08%
    1973 86.4 64.7 3.6 31.5 103 30.7 3.04% -2.57%
    1974 86.4 63.5 3.0 24.5 82 25.8 -19.30% -25.16%
    1975 85.9 63.6 4.2 36 116 38.5 12.34% 8.61%
    1976 85.9 60.8 2.6 33 107 23.3 1.99% -1.65%
    1977 85.6 63.0 5.5 38 105 51.7 -1.46% 11.89%
    1978 84.9 62.6 3.5 33.5 111 35.2 2.58% -2.53%
    1979 83.7 61.7 4.6 39 127 52.7 15.60% 12.15%
    1980 89.3 65.5 3.6 33.5 93 20.7 -16.60% -4.79%
    1981 83.2 62.9 6.7 38 126 78.5 11.35% 6.77%
    1982 83.9 62.1 4.7 38.5 131 52.4 14.10% 6.95%
    1983 92.3 66.4 2.0 29.5 79 5.5 -32.16% -18.97%
    1984 87.0 60.7 2.0 31.5 114 15.8 -3.48% -14.44%
    1985 83.1 60.6 4.6 42.5 135 54.6 12.73% 14.18%
    1986 84.1 62.3 4.2 40 135 46.2 11.20% 6.30%
    1987 86.4 65.0 4.7 38 132 40.3 7.27% -0.10%
    1988 90.4 65.1 1.8 27 73 8.1 -41.46% -29.77%
    1989 84.7 63.3 3.1 40 123 32.1 -2.65% 2.95%
    1990 82.8 63.4 4.3 39 127 53.0 -0.78% -0.67%
    1991 85.9 63.5 2.3 37.5 107 20.6 -17.47% -5.47%
    1992 80.7 60.2 5.2 43 149 75.2 13.48% 7.29%
    1993 84.5 66.5 6.1 43 130 64.3 -2.22% 6.21%
    1994 83.3 61.4 3.2 45.5 156 37.7 15.90% 11.27%
    1995 88.0 67.1 3.0 39 113 20.8 -17.07% -5.57%
    1996 82.6 61.5 3.6 40.5 136 44.2 -1.38% -2.90%
    1997 82.5 61.7 2.9 43 129 36.4 -7.56% 2.10%
    1998 84.5 66.0 3.4 44 141 35.4 -0.14% 3.47%
    1999 85.2 64.7 2.9 42 140 28.7 -2.00% -2.17%
    2000 83.8 65.0 4.0 44 151 44.9 4.50% 1.52%
    2001 86.3 65.0 3.7 45 152 32.5 4.00% 2.86%
    2002 87.2 67.0 3.0 43 135 23.6 -8.66% -2.62%
    2003 86.1 64.9 3.4 37 164 29.9 9.74% -16.97%
    2004 80.7 60.7 4.6 50 180 66.5 19.13% 11.18%
    2005 87.0 65.3 3.1 46.5 143 24.5 -6.38% 2.47%
    2006 86.2 66.2 4.5 48 163 39.2 5.57% 4.83%

  114. Posted Mar 20, 2007 at 1:53 PM | Permalink

    Has anyone ever produced a plot of average annual LOW temperature for the past several decades (100 years would be great). I suspect that you could have increased crop yields and increased glacial melting without any change at all in average temperature of both the highs and lows are moderated. If you have increased low temperatures, you would have fewer hours below freezing (increased melting) and fewer nights with frost while the average temperature might not change if daytime highs do not track with nighttime lows. Tracking average temperature would assume that highs and lows would track together while things that might cause greenhouse warming such as increased humidity might cause a moderation of both high and low leaving the average unchanged while giving more frost-free days and hours above freezing.

    Also, has anyone plotted the difference between average high and average low over time? In other words, (average high for the month) – ( average low for the month) to give a plot of any change in temperature range? I would expect to see a distinct narrowing of that range in increased greenhouse conditions.

  115. JerryB
    Posted Mar 20, 2007 at 2:13 PM | Permalink

    re #111,


    To your question: trends yes, uniform trends no, and rising
    CO2 need not account for “noise”, but something must, for
    example El Nino/La Nina, cycles. If something doesn’t,
    try aerosols, or ad hoc adjustments. 🙂

  116. Ken Fritsch
    Posted Mar 20, 2007 at 2:30 PM | Permalink

    Here are two graphs showing the relationship of percentage soybean deviations from the trend line versus Max Temperature and versus Prod (product of Max Temperature and precipitation as described in the previous post).

    The points of interest here are that if one removes the years where a Prod of 31 or less (about 1/4 of the years) occurs the remaining years give essentially no correlation of Prod to yield from trend. This occurs similarly with corn yields in this relationship and shows that both crops perform well and nearly the same yieldwise until a critical high temperature and low percipitation condition is reached in the critical July/August time period and begins to effect yeild negatively as these conditions become more extreme.

    One can perform a similar clipping operation on the Max Temperature versus yield data (as shown in the other graph below)and, while the R^2 will be significantly reduced, a value of approximately 0.2 or below will remain. As stated before there is a correlation between higher temperatures and percipitation for this time period in Illinois that makes separating the temperature effect difficult.

    I looked at the Max to Min temperature correlation for the July/August time period and calculated an R^2 of 0.3. When I did this for the period of April through August the Max to Min temperature correlation was reduced to zero.

    The r^2 correlation value for yield to Max temperature does not increase when using a log fit as was the case for the yield to Prod correlation.

  117. jae
    Posted Mar 20, 2007 at 3:01 PM | Permalink

    72: From the article you linked:

    Example 2: Take the same two glasses of water at 0 degrees and 100 degrees, respectively. Now multiply those two numbers and take the square root, and you will arrive at an average temperature of 46 degrees. This is called the geometric average. (The calculation is done in degrees Kelvin which are then converted back to degrees Celsius.)


  118. Mark T.
    Posted Mar 20, 2007 at 3:25 PM | Permalink

    The geometric mean is the nth root of the product of n values. For the example used, 0 C = 273 K, and 100 C = 373 K. The geometric mean is then sqrt(273*373) = 319 K = 46 C.


  119. Bill F
    Posted Mar 20, 2007 at 3:29 PM | Permalink

    Re: 101,

    The current predictions for Cycle 24 are in unusually bad agreement regarding the height of the max. The numbers I have seen range from around 42 to 169. (http://members.chello.be/j.janssens/SC24.html)

    An interesting thing to me is that Theo Landschiedt predicted a grand minima similar to the Maunder Minima to occur around 2025. NASA recently gave some hint that he may be right with a press release announcing that the conveyor belt that carries sunspots from the equator to the poles on the sun appears to have slowed way down. (http://science.nasa.gov/headlines/y2006/10may_longrange.htm) Since the strength of that circulation is one of the things that is used with some success to predict sunspot numbers 2 cycles from now, they are beginning to suspect that Cycle 25 may have very few sunspots. It will certainly answer the AGW question once and for all if it happens, but Lord help us if it does…

  120. Bill F
    Posted Mar 20, 2007 at 3:37 PM | Permalink

    0 Celsius = 273 Kelvin…100 Celsius = 373 Kelvin. The geometric mean is 319 Kelvin or 46 Celsius.

  121. Basil Copeland
    Posted Mar 20, 2007 at 4:32 PM | Permalink

    JerryB #116 you wrote: “trends yes, uniform trends no…”

    In case I’m not following you, all reconstructions of global temperature I’m aware of seem premised on the notion of “uniform” trends, at least over periods of years (decades). For example, IPCC AR4 refers to a 100 year (1906-2005) “trend” of 0.74C, and for the last 50 years a “trend” of 0.13C per decade. To me, these are “uniform” trends. They completely ignore, or abstract, not just from “noise,” but from data patterns that are neither “trends” nor “noise” — like shifts or step functions. (If you know econometrics, imagine a Chow test, for example.)

    But suppose all the “trend” in the last two decades is not really a trend, but a sudden upward shift of 0.26C (instead of a “trend” of 0.13C for two decades). Could this be attributed to a gradual rise in CO2, or is it more likely that such a structural shift is the result of something else, like PDO/ENSO? Is anyone saying PDO/ENSO events are caused by rising CO2 causing rising temperatures causing PDO/ENSO events? Isn’t that rather reversing what we know about causation?

  122. Barry B.
    Posted Mar 20, 2007 at 5:25 PM | Permalink

    Re: #117

    Ken, this might be of some help to you: http://www.ces.purdue.edu/extmedia/NCH/NCH-18.html

    It states: “Combined moisture-temperature stress during the reproductive period can substantially reduce final grain yield.”

    “Although separating the effects of these two stresses is difficult, most temperature stress conditions occur on high atmospheric-moisture-demand days–i.e., days when the daily mean temperature is above 77 F and the daily maximum is above 95 F, regardless of soil moisture conditions.”

  123. Steve Sadlov
    Posted Mar 20, 2007 at 5:31 PM | Permalink

    RE: #120 – We are way overdue for another period similar to ~ 450 – ~ 700 AD. The closest thing to it, since then, was the LIA, but that was pretty wimpy in terms of overall impact. See, the AGW alarmists don’t own the monopoly on pessimistic outlooks! 🙂

  124. Barry B.
    Posted Mar 20, 2007 at 5:33 PM | Permalink

    Re: #97

    MarkW, in reality, letting fields go fallow is virtually nonexistent. This is a direct effect of farm subsidies. Farmers get paid regardless of whether there is profitability in the market.

  125. Steve B
    Posted Mar 20, 2007 at 5:47 PM | Permalink

    One of the Swindle programme’s claims was that volcanoes produced a lot more CO2 than man, which various people have been shooting down by reference to a paper by TM Gerlach, EOS Trans. AGU (which seems to be more of a newsletter than a technical journal). I wanted to see how it was calculated and what estimated errors were reported. I’ve searched the AGU archive and can’t find it, but in the process of looking found another paper by RE Stoiber that quotes CO2 emission figures (Table 1) of 65 GtCO2/yr for subaerial volcanoes (i.e. not including submarine volcanoes) commenting that this stuff is hard to measure and numbers are approximate.

    Click to access 21_stoiber.pdf

    As I understand the anthropogenic emissions are estimated at 5.5GtC/yr = 20 GtCO2/yr, which is indeed smaller, I don’t understand the figure they’re quoting from Gerlach. I also find it particularly annoying that they quote without giving details a paper I can’t read, as “proof” that the claim in Swindle is wrong.

    Have I done something stupid? Does anyone know if I can get this Gerlach paper for free anywhere? Does anyone have any word on the official up-to-date figures or other sources for volcanic CO2? (The AGU seems to have any number of papers, which I plan to read further, but it would save me time if someone already knows the answer.) Who is right in this case – Swindle or the Climatologists?

    I thought this might be the best place to ask such a question. Thanks in advance for any help anyone can offer.

  126. DeWitt Payne
    Posted Mar 20, 2007 at 6:24 PM | Permalink

    For anyone interested in learning something about Physical Meteorology but who doesn’t want to spend $90 US for Curry and Webster, here’s a link I found to some on-line lecture notes. Unfortunately, the section on radiation isn’t available yet(?). Still, I learned a lot.


    Also, here’s some notes on IR transmission in gases and how to build a one dimensional model of radiative energy transmission in the atmosphere. Note that a multi-layer model is required because there can be no greenhouse effect in an isothermal atmosphere. There’s an algorithm in these notes for calculating saturated water vapor pressure at any temperature.

    Click to access IRemissionTheory.pdf

  127. David Smith
    Posted Mar 20, 2007 at 6:25 PM | Permalink

    Re #126 A webpage by the US Geological Survey ( here ) says manmade CO2 emission rates are much higher than those of volcanoes, and gives some values from Gerlach. I imagine that is correct.

    It refers to the Gerlach paper. I’ll search for a copy. Should be interesting reading.

  128. Richard Sharpe
    Posted Mar 20, 2007 at 6:39 PM | Permalink

    There seems to be two types of predictions for Solar Cycle 24.

    Some think it will be 30-50% more intense than 23. They seem to have a model that can back predict well.

    Others think it will be way down.

    What to do?

  129. John Lang
    Posted Mar 20, 2007 at 7:21 PM | Permalink

    To Steve B #126 – I’ve seen numbers that place emissions from volcanoes at 0.15 billion tonnes (carbon). Now those are averages – every now and again a really big volcanoe like Krakatoa happens and that would dwarf the human emissions but those really big eruptions do not happen that often.

    The paper you linked to is taking about subaerial (undersea) volcanoes and there are clearly way more volcanoes under the ocean than on land (especially including the mid-oceanic ridges.)

    However, do those CO2 emissions make it to the atmosphere? They do, but through the emissions from the oceans.

    The numbers on emissions for the oceans is 92 billion tonnes (Carbon) going in – and 90 billion tonnes going out. So the subaerial volcanoe emissions have to be tied into these numbers.

    I still have not seen any studies which show how much the 92 billion tonne number varies with ocean temperature which it clearly does. Think in terms of a 5C colder ocean in the ice ages which clearly absorbed alot of CO2.

  130. David Smith
    Posted Mar 20, 2007 at 7:22 PM | Permalink

    Re #126 Steve B, it may be a unit of measure problem.

    I get 20,000 megatons/yr CO2 from humans, versus 130 megatons/yr CO2 from volcanoes.

    The “65 Gt” above-water CO2 emissions from Stoiber may actually be 65 megatons. Since Gerlach estimated underwater volcanic emissions at about the same as above-water ones, so 65 + 65 = 130 megatons.

  131. Steve B
    Posted Mar 21, 2007 at 1:33 AM | Permalink

    David Smith,

    Thanks, you’re right. I’ve got the units mixed up. I’d converted Tg to Gt wrongly.

    I’d still be interested in anything anyone has on how they work it out, though. You can’t just multiply the number of volcanoes by the amount of gas each one generates, because they’re all different, and irregular. It sounds like an interesting calculation.

  132. Andrey Levin
    Posted Mar 21, 2007 at 4:29 AM | Permalink

    More elaborate article on the role of UK and Lady Margaret in AGW, from 1999 (!):


    Also testimony of Czech President Vaclav Klaus to US Congress committee, March 19:

    Click to access 3.19.07_Vaclav_Klaus_response.pdf

  133. David Smith
    Posted Mar 21, 2007 at 4:48 AM | Permalink

    There’s a nice image of a double ITCZ in the eastern Pacific today (March 21). The animation shows two lines of clouds and showers on either side of the equator.


  134. JerryB
    Posted Mar 21, 2007 at 6:15 AM | Permalink

    Re #122,


    Perhaps I use the terms trend, and unifor trend, too loosely.
    Let me try a metaphor.

    If you drew a vector from Denver to the Eisenhower tunnel,
    it would be uniform in direction. If you drive I-70 from
    Denver to the Eisenhower tunnel, your direction of travel
    would not be uniform with respect to any axis, but you
    would still reach the tunnel.

    I don’t know if that helps clarify my usages; I hope it does.

  135. Reference
    Posted Mar 21, 2007 at 7:03 AM | Permalink

    CSPAN webcast of House Energy Subcmte hearing on Global Climate Change starring Al Gore – starting 9:30 ET today

  136. Jeff Norman
    Posted Mar 21, 2007 at 7:37 AM | Permalink

    Regarding the CO2 emissions from volcanoes.

    It could be argued that even if tectonic greenhouse gas emissions were larger than anthropogenic greenhouse gas emissions, that this is just a strawman argument because the tectonic emissions are natural and therefore do not contribute to climate change, or something.

    Speaking of tectonic effects though, I recently heard (last week?) that the president (or whatever) of the WMO is claiming there will be more “tidal waves” resulting from AGW. Does this mean that AGW causes more tectonic activity which would imply volcanic emissions of CO2 could be a positive feedback of the whole AGW thingee. I know what he means. It’s just not clear if he or the people who are reporting his worlds know what he means.

  137. Michael Jankowski
    Posted Mar 21, 2007 at 7:58 AM | Permalink


    The main problem that I have with the “aerosols like SO2 caused the 1945-1975 cooling” theory…

    To add to that…if the aerosols were more persistent in the atmosphere than you and I think they are, exactly why didn’t the aerosol cooling and GHG warming go hand-in-hand and result in a flatline? It seems anthropogenic emissions of aerosols and GHGs should have pretty much matched each others trends, not alternated with GHG warming dominating pre-1945 and then aerosol cooling dominating from 1945-1975.

  138. Ian Blanchard
    Posted Mar 21, 2007 at 8:02 AM | Permalink

    The effects of a half degree temperature rise never cease to amaze me.

  139. Basil Copeland
    Posted Mar 21, 2007 at 8:03 AM | Permalink

    #135 Jerry, that helps me understand your distinction. Given it, it still seems to me that temperature changes are routinely represented as if they were vectors (to use your term), e.g. 0.13C decadal change in the last 50 years. But statistically, this isn’t so. The “vector” is more like 0.04C per decade, with a big jump of about 0.2C after 1997. So how does rising CO2 account for the jump, as opposed, say, to the trend/vector? To put it more succintly, CO2 rise has been gradual; temperature change has not. So what is the physical theory that links gradual CO2 rise to sudden shifts of temperature regime? That’s what I’m really asking about.

  140. Jean S
    Posted Mar 21, 2007 at 8:13 AM | Permalink

    Malcolm Hughes & bristlecone pines:
    Salzer, M.W. & Hughes, M. K. 2007. Bristlecone pine tree rings and volcanic eruptions over the last 5000 yr. Quaternary Research 67 (2007) 57’€”68.

  141. Ken Fritsch
    Posted Mar 21, 2007 at 9:05 AM | Permalink

    Re: #123

    “Although separating the effects of these two stresses is difficult, most temperature stress conditions occur on high atmospheric-moisture-demand days’€”i.e., days when the daily mean temperature is above 77 F and the daily maximum is above 95 F, regardless of soil moisture conditions.”

    Barry B., thanks much for that information — and now this retired guy seeking an excuse to get back to retirement mode has been given something else to analyze. I think I have the information (daily max/min temperatures) to look at the point raised by your reference and will do so along with some other unfinished items I wanted to cover.

    I am assuming that the pollination process is most susceptible to temperatures and moisture and I wanted to zero-in at that approximate time period for a correlation with yield changes from the trend line. Remember that although Lobell stated that the months used for the correlation of temperature and moisture with yield were not highly sensitive to the degree of correlation (without indicating which or how many months that was), I found that the April through August correlation was very close to zero for IL in the 1961 to 2006 time period for soybeans and corn. Lobell did say that the July/August was selected because it gave the highest correlation. Lobell’s using a wider range area for his study and thus a more varied climate conditions and growing seasons in his study may have resulted in his looking at a critical growth period for the crops that was smeared into a wider seasonal range. I am wondering whether corn and/or soybeans that I know grow in higher temperature areas such as Brazil and Florida are breed to perform better in climates with higher temperatures and percipitation.

    I also am looking at cyclical nature of the periods of high temperature and low moisture going back to 1901 with the IL temperature data.

  142. Ken Fritsch
    Posted Mar 21, 2007 at 9:31 AM | Permalink

    Re: #23

    Barry B., I just finished reading the article that you linked in post #23 and found it to be by far the best summary of temperature/moisture stress and effects on yields that I have in my growing bookmark file of links on this subject. It would appear from the article that the most critical time for the corn plant development vis-a-vis yields is approximately 80 days after planting until 90 days after planting. The problem with looking at historical time periods is that I know for a fact that when I was a kid on the farm that corn was planted mid to late in May while today corn is normally planted a month earlier. I also know that corn planted later matures faster. The more I look into these processes, the more I view Lobell’s approach of temperature sensitivity for crop yields as way too oversimplified. I think it is something that I see in other researchers’ publication that seem in a big hurry to show some detrimental effects or evidence for AGW. Maybe more digging will change my current view of these crop relationships.

  143. JerryB
    Posted Mar 21, 2007 at 9:39 AM | Permalink

    Re #140,


    My impression is that the AGWers would say something like:
    while things like ENSO, and PDO, cause sudden shifts, the
    shifts will have higher highs, and/or higher lows, as time
    passes, due to increased CO2.

  144. Steve Sadlov
    Posted Mar 21, 2007 at 9:41 AM | Permalink

    RE: #130 – And the measurement techniques used to determine 92 GT into the seas from spreading centers and other submarine sources are….. ? I am highly skeptical that anyone really knows what this value is.

  145. Bill F
    Posted Mar 21, 2007 at 10:44 AM | Permalink

    More evidence of a solar link to climate:


    Shouldn’t somebody tell these guys at NASA that the science is already settled?

  146. Sidviscous
    Posted Mar 21, 2007 at 11:29 AM | Permalink

    Other natural CO2 sources


    How much CO2 does Old Faitfull put into the atmosphere on a yearly basis?

  147. Michael Jankowski
    Posted Mar 21, 2007 at 11:58 AM | Permalink

    Don’t forget that AGW causes (or will cause) more frequent shifts to “bad” and longer durations of “bad.”

  148. Steve Sadlov
    Posted Mar 21, 2007 at 12:17 PM | Permalink

    Antarctic sea ice is well under way in Fall-Winter accretion mode. Looks like it hit its annual minimum at least 3 weeks prior to the equanox this year. Maybe a month prior …

  149. Lee
    Posted Mar 21, 2007 at 12:21 PM | Permalink

    find the major volcanic eruptions…

  150. Richard Sharpe
    Posted Mar 21, 2007 at 12:23 PM | Permalink

    It seems as if NASA are standing behind the predictions that the next solar cycle will be 30-50% stronger than this one:

    March 10, 2006: It’s official: Solar minimum has arrived. Sunspots have all but vanished. Solar flares are nonexistent. The sun is utterly quiet.

    Like the quiet before a storm.

    This week researchers announced that a storm is coming–the most intense solar maximum in fifty years. The prediction comes from a team led by Mausumi Dikpati of the National Center for Atmospheric Research (NCAR). “The next sunspot cycle will be 30% to 50% stronger than the previous one,” she says. If correct, the years ahead could produce a burst of solar activity second only to the historic Solar Max of 1958.

  151. Gerald Machnee
    Posted Mar 21, 2007 at 12:52 PM | Permalink

    Re #126 and #150 – **One of the Swindle programme’s claims was that volcanoes produced a lot more CO2 than man**.
    There is a wording that one has to check. This is not definite as I am still reading and searching. Is it “produced more in total” or is it “contributed more” to the annual increase in the amount of CO2 in the atmosphere. Also one author indicated (do not recall which one) that the historical total from volcanoes may be higher than man’s contribution. Again is it to the increase or what is in the atmosphere? Stevenson (noted in the thread on Buckets, I think)had some figures as well as Tom V. Segalstad on the contribution of volcanoes as well as anthropogenic CO2. Segalstad in http://folk.uio.no/tomvs/esef/ESEF3VO2.htm also cites a COUPLE DOZEN authors whose research indicates CO2 residence time in the atmosphere of about 5 years as opposed to the 50 to 200 years that the IPCC is preaching.
    With respect to the chart in #150 – there is checking to be done there as well. Segalstad indicated that the Mauna Loa measurements are done by IR as opposed to chemical measurements done before that. We do not know what assumptions they made. They were using monthly and seasonal averaging so that would not always show large variations. Mauna Loa was selected to be away from INDUSTRIAL CO2, but is it away from volcanic sources?
    So, we need a chemical audit as I am not drawing any conclusions on its accuracy.

  152. Steve McIntyre
    Posted Mar 21, 2007 at 1:07 PM | Permalink

    I for one do not agree that CO2 emissions from volcanoes are larger than human contributions. It’s not a topic that I want to devote much bandwidth to at this site.

  153. Posted Mar 21, 2007 at 1:25 PM | Permalink

    Svensmark and Nigel Calder were interviewed on BBC Radio 5 this afternoon.


    The interview starts at 1hr 55min.

  154. Reid
    Posted Mar 21, 2007 at 1:38 PM | Permalink

    Re # 150

    Lee, are you sure the data being plotted hasn’t been filtered to remove large scale volcanic eruptions? Seems an eruption like Pinatubo should have seen at least a minor rise. I suspect someone has “Jonesed” the data.

  155. Lee
    Posted Mar 21, 2007 at 1:43 PM | Permalink

    Gerald, the ~ 8ppm annual cycle in that record is due to seasonal changes in terrestrial vegetation – the measurements (averaged MONTHLY, not seasonally) clearly show that mixing of the atmosphere is sufficient and rapid enough to catch those monthly changes in mid-Pacific. In fact, the month-to-month variation that they resolve is significantly less than 1 ppm. In other words, they not only routinely, but ALWAYS resolve monthly variations in mixed CO2 concentrations due to global seasonal vegetation changes, and they do it to less than 1ppm.

    There is no evidence of local perturbations in that record, such as from emissions from local volcanoes. Mauna Loa is not continuously erupting, and if it has emissions that would perturb the record, one would see it in jumps coinciding with onset of eruptions and quiescent periods. There is no hint of such variations in this record. Remember that the Mauna Loa observatory is measuring predominantly tradewind-delivered mixed atmosphere, which is some 2500 miles or more downwind of urban/industrial areas.

    That record also does not show any perturbations coincident with known major volcanic eruptions. That is, volcanic eruptions during that time period do not make a measurable change in atmospheric CO2 concentrations, while the overall concentration has climbed smoothly from 315 to 375. The cause of that increases is clearly continuous, not episodic.

    Saying that you don’t believe the record because you don’t know the ‘assumptions’ behind IR-absorption measurements of CO2, while looking at a record that reliably resolves monthly variations of less than 1ppm, is simply blind dismissal.

  156. Richard Sharpe
    Posted Mar 21, 2007 at 1:47 PM | Permalink

    Randall Parker at FuturePundit says:

    Some JPL folks think they might have evidence of two separate sun cycles running at 200 and 88 years.

    He is referring to this: NASA Finds Sun-Climate Connection in Old Nile Records.

    (Is Joan Feynman related to the Feynman?)

    A key passage from the article seems to be:

    So what causes these cyclical links between solar variability and the Nile? The authors suggest that variations in the sun’s ultraviolet energy cause adjustments in a climate pattern called the Northern Annular Mode, which affects climate in the atmosphere of the Northern Hemisphere during the winter. At sea level, this mode becomes the North Atlantic Oscillation, a large-scale seesaw in atmospheric mass that affects how air circulates over the Atlantic Ocean. During periods of high solar activity, the North Atlantic Oscillation’s influence extends to the Indian Ocean. These adjustments may affect the distribution of air temperatures, which subsequently influence air circulation and rainfall at the Nile River’s sources in eastern equatorial Africa. When solar activity is high, conditions are drier, and when it is low, conditions are wetter.

  157. Steve B
    Posted Mar 21, 2007 at 2:02 PM | Permalink

    Lee, what point are you trying to make? I’m reasonably happy that CO2 emissions from volcanoes are less than the anthropogenic ones, (now that I’ve remembered that the SI units of mass are different from the base unit :-),) but its exactly this sort of nonsensical non-proof that annoyed me enough to go looking for more information. What does the lack of spikes in the graph prove? That the variable component in tectonic emissions doesn’t dominate the change in CO2. But that doesn’t say whether there is a large constant background component, it can only be interpreted in light of the smoothing that atmospheric mixing applies, if tectonic contributions had been of the same order of magnitude as anthropogenic (like one percent per year) then it wouldn’t introduce visible spikes either, and if both effects were dominated by changes in some other source, like biosphere or ocean, but with a heavily damped response due to unknown negative feedbacks, then changes in both anthropogenic and volcanic contributions wouldn’t show up against it. I’m not saying that’s the case, and I don’t think it is the case, but I’d like to know how we know.

    The Swindle programme got something wrong – fair enough. Most TV documentaries do. But in trying to answer the point the arguers only cited the lack of spikes in the CO2 rise, which as I said doesn’t mean anything of the sort, and quoted this Gerlach paper giving the accepted results. But I like ‘argument from authority’ even less than ‘argument from spurious graph’, so quoting a paper I can’t access doesn’t satisfy me any more than the graph does. I want to know how you know. It also sounds like a very interesting bit of science – how on Earth do you estimate the total CO2 emitted from all the world’s volcanoes?

    You could sample the gas mixture from a volcano at a particular point in time, but I expect it varies. Different types of volcano give different mixes, and while we probably know where all the world’s volcanoes are, to characterise the emissions of all of them sounds like a big job. And how do you measure the total volume of gas given off over a wide area where it’s too dangerous to get near? You could measure it downwind and guess how much mixing had gone on, but that sounds very inaccurate. Or is it something clever like the carbon content of the lava? And what about CO2 coming up through faults and cracks in areas where the geology isn’t obviously active? And how about underwater?

    So, how did they do it, and how accurate is it? Has nobody made any better estimates since 1991? The Swindle programme had its faults, but one thing it showed quite well was that, even if the AGWers are right, the arguments they have presented so far to the general public don’t prove it, and are missing a lot of relevant detail.

    Sorry if that comes across a little bad tempered. I guess it doesn’t matter very much. I was just hoping that since it had apparently been a matter of contention in this debate, someone here might have already looked at it. Tracking it down will give me something to do. 🙂

  158. Lee
    Posted Mar 21, 2007 at 2:27 PM | Permalink

    Steve,I am also responding to statements made in this thread. Somewhere above it was said that very large eruptions, I think citing Krakatoa, would dwarf human CO2 emissions. I was showing that even largish eruptions, like Pinatubo, do not cause detectable variations in the record. I was also responding to Gerald’s sniping at the Mauna Loa record itself.

    Overall, total ‘natural’ CO2 flux is much larger than human CO2 emissions, and background volcanic emissions are part of that natural flux, but less than human emissions. But this does not alter the compelling evidence that anthropogenic causes have driven the increase in CO2 to a level MUCH higher than any in the last 650,000 years at least, and likely in many millions of years. Many people (not necessarily you) make the ‘volcano/natural flus is larger’ point as a way of indirectly raising doubt about the anthropogenic cause of the CO2 increase – which is raising false doubt.

  159. Bill F
    Posted Mar 21, 2007 at 2:27 PM | Permalink

    Re 151:

    See 120. NCAR’s prediction is at the high end of the spread of predictions that range from 42 to 169. Pay attention to the fact that the press release you linked to is from march of last year, not this year. The article about the slowing of the conveyor belt was referring to Cycle 25, not 24, so don’t get confused between the two predictions.

    Re 157: See 146.

  160. Steve McIntyre
    Posted Mar 21, 2007 at 2:39 PM | Permalink

    Lee, I wasn’t directing my comment at you. I recognize that you were responding to points made elsewhere (not by me). I’m just saying (more for others) that there’s more bandwidth being spent on this issue than I’m interested in spending on it.

  161. Lee
    Posted Mar 21, 2007 at 2:47 PM | Permalink

    sorry – my 159 response was to “SteveB.” Apologies for not including the post number or complete handle.

  162. John A
    Posted Mar 21, 2007 at 2:52 PM | Permalink

    But this does not alter the compelling evidence that anthropogenic causes have driven the increase in CO2 to a level MUCH higher than any in the last 650,000 years at least, and likely in many millions of years. Many people (not necessarily you) make the volcano/natural flus is larger’ point as a way of indirectly raising doubt about the anthropogenic cause of the CO2 increase – which is raising false doubt.

    There are lots of “false doubts” that turned out to be true. The ice core record is a case in point: do you accept that carbon dioxide is higher than it has been for x thousand years without reservation? Without checking with other methods? Without audit of exactly how those curves were produced? As we’ve already seen, audit of these things is valuable in examining what is good science and what is not.

    If the ice core record is substantially correct it would beg the questions of why there is no carbon dioxide rise preceding temperature rise, why there is no feedback and why an “unprecedented” carbon dioxide rise has not produced “unprecedented” temperature rise.

    I don’t believe that the carbon dioxide record can be trusted until it is properly checked.

  163. Pete
    Posted Mar 21, 2007 at 2:58 PM | Permalink

    Is Joan Feynman related to the Feynman

    Joan is Richard’s sister.

  164. Bill F
    Posted Mar 21, 2007 at 3:19 PM | Permalink


    One correction I think you need to make in what you said re: Mauna Loa is this. No, the volcano isn’t erupting continuously…but it has erupted during the time of the record shown above (in 1983). I do find the lack of response from the monitoring pretty interesting, although it was a pretty small and short lived eruption for Mauna Loa.

    One question. Do you have any thoughts on why there is so little in the way of variation in the slope of the curve? It seems to me like there should be at least some variability in the slope if the rise is attributable to human activities, as those activities and the emissions they cause vary with economic activity, wars, etc. For example, the Kuwait oil fires in the early 90s actually appear to correspond to a lower rise in CO2 instead of a higher slope as would be expected if the change in CO2 is solely attributable to human activities. Am I missing something or is the mixing such a slow process that those types of changes won’t show up on an annual scale?

  165. richardT
    Posted Mar 21, 2007 at 3:34 PM | Permalink


    If the ice core record is substantially correct it would beg the questions of why there is no carbon dioxide rise preceding temperature rise

    Why should it? Terminations of the glacials were initiated by orbital forcings. Nobody, except for you, expects a CO2 forcing to precede the orbital forcing. How could it?

    why there is no feedback

    There was. Orbital forcing and albedo changes (one feedback no-one ever seems to have any problem with) are insufficient to explain the glacial-interglacial temperature difference. The CO2 feedback can account for the missing forcing.

    and why an “unprecedented” carbon dioxide rise has not produced “unprecedented” temperature rise.

    Just wait! There are substantial lags in the climate system to forcings, especially in warming the oceans and vegetation feedbacks.

  166. Steve McIntyre
    Posted Mar 21, 2007 at 3:39 PM | Permalink

    I’ve looked at IPCC 4AR about the CO2 lead-lag issue and it has virtually nothing on the topic, other than to say that temperature changes lead CO2 and the feedback isn’t understood. Maybe the question should be re-posed a little: do climate models include the CO2-temperature feedback mechanism held to be fundamental to ice age transitions? If not, why not?

  167. Richard Sharpe
    Posted Mar 21, 2007 at 3:47 PM | Permalink

    richardT says:

    Just wait! There are substantial lags in the climate system to forcings, especially in warming the oceans and vegetation feedbacks.

    Do we have to wait 800 years though? Indeed, how long do you suggest we wait?

  168. Bill F
    Posted Mar 21, 2007 at 4:11 PM | Permalink

    So if the warming of the ocean is a substantial lag in the climate system feedback to forcings, what does the big drop in oceanic heat content reported in Lyman et al 2006 tell us about the forcings?

  169. Mark T.
    Posted Mar 21, 2007 at 5:24 PM | Permalink

    I’ve looked at IPCC 4AR about the CO2 lead-lag issue and it has virtually nothing on the topic, other than to say that temperature changes lead CO2 and the feedback isn’t understood.

    It’s not understood because they can’t get a handle on causality. Feedback cannot create cause to lag effect unless it is a non-causal (or acausal) system. Nature is not non-causal.


  170. woodentop
    Posted Mar 21, 2007 at 5:38 PM | Permalink

    #166 – what were the forcings responsible for the MWP and the LIA?

  171. TAC
    Posted Mar 21, 2007 at 6:00 PM | Permalink

    #171, as I understand it, the MWP was caused by the Industrial Revolution, and the LIA will be caused by Peak Oil.

  172. Steve B
    Posted Mar 21, 2007 at 6:40 PM | Permalink

    Lee, thanks for clarifying that, and apologies to Steve McIntyre for using up his bandwidth.

    I think they use 650,000 yrs because CO2 has been falling from previously much higher levels and that’s when it last passed the current value. See fig 4. here. Given the uncertainty in such reconstructions, it could indeed be pushed back to millions of years.

    There may indeed be compelling evidence of these claims, but it’s hard for the layman to find out what that evidence actually is. All you get given are citations that you’re not supposed to examine.

    On the CO2-glacial lead/lag graphs, I would have thought the primary feedback would be water vapour rather than CO2, which would start immediately the temperature started to rise/fall. Water vapour is air temperature driven, right? And provides 95% of the greenhouse effect, I’ve been told? So if CO2 produces the results claimed for it primarily through the H2O amplifying feedback, the same applies to the solar forcing that triggers glacials. Both H2O and CO2 provide feedback to enhance the solar effect (and themselves, and each other) but H2O is going to be the larger part of that feedback.

    If there was no CO2 rise, how much less amplification would H2O provide on its own?

  173. Gerald Machnee
    Posted Mar 21, 2007 at 7:10 PM | Permalink

    Re #150 and #156 **Saying that you don’t believe the record because you don’t know the assumptions’ behind IR-absorption measurements of CO2, while looking at a record that reliably resolves monthly variations of less than 1ppm, is simply blind dismissal.**
    My exact words were — “I am not drawing any conclusions on its accuracy.”
    Lee, I stand by my comments – we do not know what they are doing in terms of averaging. I did not outright say, they filter out eruptions – I meant we do not know. As Steve M indicated, there is no point in discussing too much. I do not think any of us here today have been involved in the measurement. I mentioned Stevenson and Segalstad and nobody had referred to them. Segalstad noted that a couple of dozen or so researchers determined that the CO2 residence time in the atmosphere is about 5 years, something which the IPCC has ignored, expanding it by 10 times to 50 to 200 years and now all the AGW crowd quotes it without knowing why. Also, one of the original scientists who started the Mauna Loa site agreed that most of the increase was of natural origin. So unless anyone can get into the real documentation there is not much point in “he said, she said”.

  174. S. Hales
    Posted Mar 21, 2007 at 7:20 PM | Permalink

    Here’s a little primer on the history of solar research:

    Then read the interview with Eddy

    I thought the part about solar variability having a stronger footprint in the northern hemisphere to dovetail quite nicely with the marked difference in temperature profiles between the hemispheres as evidenced by MSU.

  175. DeWitt Payne
    Posted Mar 21, 2007 at 8:42 PM | Permalink

    Raymo, M.E. & Ruddiman, W.F., “Tectonic forcing of late Cenozoic climate” Nature, 359, 117-122 (1992) show a long term proxy record of oxygen isotope ratios in benthic foraminifera. The references for the figure are: Miller, K.G., Faiarbanks, R.G. & Mountain, G.S. ,Paleoceanography, 2 , 1-19 (1987) and Shackleton, H.J. & Kennett, J.P.,Init. Rep. Deep Sea Drilling Proj., 29 (ed. Kennett, J.P.) 801-807 (U.S.Government Printing Office, Washington, DC, 1975). The figure in Raymo and Ruddiman shows a fairly rapid increase in delta 18O in the last few million years and a general overall slower increase in the last 70 million years. The authors propose that the uplift of the Tibetan plateau, starting about 50 Myr ago when the Indo-Australian plate collided with the Asian plate, has caused a decrease in atmospheric CO2 due to an increase in chemical weathering. They also propose that this has not resulted in complete stripping of CO2 from the atmosphere because there was, among other things, a counterbalancing increase in CO2 production from increased oxidation of buried organic carbon. This would tend to alter the 12C/13C ratio of CO2 in the atmosphere because buried organic carbon is enriched in 12C. Perhaps the proposed mechanisms have been debunked in the last 15 years, but the proxy record seems to show that current temperatures (and by implication CO2 levels) are unusually low and ice volumes unusually high compared to the average of the last 70 Myr. It would be interesting to see how current oxygen isotope ratios compare to those in Raymo and Ruddiman.

  176. Barry B.
    Posted Mar 21, 2007 at 8:52 PM | Permalink

    Re: #142 & #143

    Ken, I was looking at global corn yields (which can be found here: http://nue.okstate.edu/Crop_Information/World_Wheat_Production.htm ) when it dawned on me that you are looking for a relationship between temperature and reduced yield that will probably be hard to find – at least here in the U.S. The reason is because of our soils.

    When you look at global crop production, yield is mainly tied to soil depth. With deeper soils you have a greater capacity to hold water and nutrients and both of these greatly reduce the impact of higher temperatures. It’s no secret that we have some of the most productive soils in the world and as someone who grew up in Illinois, I’m sure you can attest to our soils ability to hold enormous quantities of water. It is this ability to hold water in reserve that I believe is the reason you’re not finding any correlation.

    As you move outside of the U.S. and into areas exhibiting lower average yields I would guess that you will find a better correlation. This would be primarily a function of soil depth and water holding capacity but soil fertility would also have an effect. Yields are generally not limited by fertility here in the U.S. but it is an entirely different story throughout many parts of the world. Crops that are insufficiently supplied with nutrients such as N,P,&K are not able to withstand heat stresses as well as crops which have a sufficient supply of nutrients so this would be certainly be a factor.

    Of course Lobell and Field do not differentiate yield between regions, but I have to believe this is where most of their results are coming from. In looking at the chart from the link above, I think there is more than enough production in inherently poorer producing areas to offset the relatively stable production here in the U.S. and much of Europe. That’s not to say there are not any flaws in their methodology but this could be a plausible explanation for their results.

  177. Willis Eschenbach
    Posted Mar 22, 2007 at 2:30 AM | Permalink

    Gerald, I fear that you are conflating residence time and e-folding (or half-life) time when you say:

    I mentioned Stevenson and Segalstad and nobody had referred to them. Segalstad noted that a couple of dozen or so researchers determined that the CO2 residence time in the atmosphere is about 5 years, something which the IPCC has ignored, expanding it by 10 times to 50 to 200 years and now all the AGW crowd quotes it without knowing why.

    Residence time is the amount of time the average CO2 molecule stays in the air. It can be measured several ways, all of which come up with something around 5 years. E-folding or half-life time, on the other hand, is much more difficult to measure. It is how long it takes for a “pulse” of CO2 which has been emitted into the atmosphere to to decay to 1/e (or 1/2, in the case of a half-life value) of the initial size of the pulse.

    While the IPCC estimates the e-folding time at 50-200 years, there are good reasons to say it is shorter, on the order of 30 to 50 years.

    My best to everyone,


  178. MarkW
    Posted Mar 22, 2007 at 6:00 AM | Permalink

    Another feedback mechanism that I’ve been noodling about in my head has to do with the way that changing sea levels will affect air
    pressure at different portions of the globe.

    Let’s leave ice formation out for a moment so that I can explain what I am thinking of.
    It’s known that as you go up in elevation, temperature goes down.
    Wouldn’t a dropping sea level have the same effect as moving a particular point higher in elevation?
    IE, what’s the difference between dropping the sea level 100 feet, and climbing up a hillside 100 feet?

    That part seems straight forward enough. The problem for me is when I try to figure out how ice formation affects these air pressure
    changes. Obviously ice that forms at or near the new sea level will cancel out some of the drop in sea level.

    However, much of the ice will form in mountainous regions, well above even the old sea level.

    So thinking of an area such as the US northern plains. Areas that are currently between 500 and 1000 feet above sea level (mostly).
    Early on in the start of a ice age, the sea levels are dropping, but most of the new ice is forming at higher elevations.
    I’m thinking that this would cause a decrease in the average air pressure in such places. Which would in turn make these areas a little

  179. MarkW
    Posted Mar 22, 2007 at 6:09 AM | Permalink

    As a corollary, I was also thinking about what would be happening at the surface of the seas as water levels dropped.

    If the earth were just a ball covered by water, dropping sea levels would have very little impact on air pressure at sea level.
    There would be a small increase because dropping sea levels means that the radius of the earth has dropped a little. Thus there is
    less total surface area, which means that the atmosphere has to pile up a tiny bit higher to squeeze itself into a smaller globe.

    Now we don’t live on a water world. (though with the oceans covering 70% of the world, we’re not that far off.)
    The 30% of the world that is not covered by water obviously won’t drop in elevation as the waters shrink. There’s also the affect of
    ice forming on the land which will displace air. (Actually ice will displace a little more air than the water formerly did, since water
    expands as it freezes.) The net result is that the atmosphere gets thicker above the oceans. Which increases air pressure and causes
    the air at the new sea level to be warmer than it would have been had the atmosphere stayed constant.
    This would slow down ice formation at the new sea level, but it would also increase evaporation, which would increase the moisture available
    for precipitation. Both as rain, and as snow.

  180. Posted Mar 22, 2007 at 6:32 AM | Permalink

    Spot the Hockey Stick

    Warm winter signals climate change by Minnesota Public Radio.

    Notice how they didn’t run a story stating “Really cold February signals end to global warming” on March 1st.

    More fun from the same people: Predicting winter’s heating costs as uncertain as forecasting the weather.

  181. Gerald Machnee
    Posted Mar 22, 2007 at 8:17 AM | Permalink

    Re #178 – Thanks Willis. I do not have a full understanding of the topic, but my readings indicated that IPCC numbers were too large and not explained. So I also suggested that unless someone has the appropriate figures, back and forth chatter will not advance the topic.

  182. Steve Sadlov
    Posted Mar 22, 2007 at 9:44 AM | Permalink

    RE: #181 – There have been some fairly benign winters in Minnesota since 1994, especially during the 90s and first years of the 00s. Since that area tends to be dominated by High pressure during the Winter innately and have a summer precip maximum, I would reckon that high winter variability in both temperature and precipitation would be the norm over the long swathe of time. Furthermore, the jet stream has tended to miss the north eastern plains and favor the western plains the past several years. During that same period, California has had higher than normal rainfall and crummy springs and summers, due to being a the bottom of persistent troughs. With the onset of a PDO phase flip, it will be interesting to see the impacts on the whole of North America on the late 00s and oward.

  183. David Smith
    Posted Mar 22, 2007 at 10:19 AM | Permalink

    And now for something entirely different … a look at current ocean temperatures.

    First, the equatorial Pacific, which is the nursery of the famous El Nino and La Nina. The surface temperature is given here . Double-click on the colored map to get a blowup of March 21 surface temperatures alone the Pacific equator and check the lower map, which shows temperature anomaly.

    What is shows is what I’d call neutral conditions, with some cool and some warm. Winds are still wishy-washy, though, so a definite direction (cooler or warmer) does not appear to be set yet. However, the SOI index, which tends to forecast winds, is entering a “cooling” region, and many of the models forecast cooling.

    (One curiosity on the chart is the region of coolness that suddenly appeared (this week) at about 170E. This is like a bubbling-up of cool water. What’s odd is that this coolness is not visible on the charts of deeper-ocean temperatures. Odd.)

    Speaking of deeper ocean temperatures, a cross-section of equatorial Pacific ocean temperatures is given here . Note the darker green and bluish shades on the bottom plot – those are anomalously cool water. Much of the subsurface equatorial Pacific is cooler than normal, which provides support to a La Nina and moderation of Warm Pool temperatures (left side of chart).

    A map of global temperature anomalies 150 meters below the surface is shown here , bottom chart. There is a battle between cool and warm underway beneath the Indian Ocean surface, which illustrates (to me) the dynamic nature of currents and upwelling in that region. I continue to wonder if there are periods when one mode (cool or warm) dominates, which affects global climate. Perhaps the strength of Antarctic circumpolar winds, which affect the amount of thermohaline upwelling near Antarctica, also affect the Indian upwelling. Conjecture.

    Also of note is the tropical Atlantic, where there is little subsurface support for anomalously warm sea surface temperatures as we approach the hurricane season.

    Finally, the current global surface temperature anomaly is shown here . Yellows are warm, blues are cool. The Atlantic is warm, but not stongly so, except for the regions near and north of Europe. The Warm Pool (tropical Africa eastward to 180W, more or less) continues to cool. The Southern Hemisphere oceans remain rather anomalously cool.

  184. Steve Sadlov
    Posted Mar 22, 2007 at 10:37 AM | Permalink

    RE: #184 – I am changing my landscaping to be much more drought tolerant. Conjecture @ ~ 120W, 38N …. LOL!

  185. Ken Fritsch
    Posted Mar 22, 2007 at 12:09 PM | Permalink

    Re #177

    Here are some additional analyses that I did relative to the Lobell paper on temperature effects on crop yields. The analyses may raise more questions perhaps than they answer, although my conclusions indicate to me, anyway, that Lobell’s analysis is correct vis-à -vis soybean and corn for IL, but that the relationship of temperature to crop yield is much more complex than Lobell’s paper would indicate.

    I followed up on the suggestion that, at least, corn has a maximum temperature of 95 degrees F where yields would fall off due to stress under any moisture conditions. The best I could do data-wise was find days for July/August that exceeded 90 degrees F. Although these data gave a somewhat better correlation with corn yields (deviation from trend line) than Max temperature it remained confounded with moisture as is Max temperature.

    Max temperature and days over 90 degrees has an R^2 correlation of 0.87 and days over 90 and Max temperatures correlate with moisture with R^2 in both cases of 0.30.

    What I did discover, that was somewhat surprising, was the temperature calculations I made in attempts to estimate the crop growing climate conditions for the IL grain belt back to 1901 was the following for 1901 to 2006 temperature trends:

    The Max July/Aug temperature trended downward by 2.7 degrees F (1.5 degrees C) per century.

    The average of the Max/Min July/Aug temperature trended downward by 0.8 degrees F (0.44 degrees C) per century.

    The average Min/Max for all 12 months temperatures trended upward by 0.6 degrees F (0.33 degrees C) per century.

    I need to look further to determine whether these relationships hold up for the grid cell temperatures from the official data bases ‘€” but I intend to do it at the leisurely pace of the retired guy.

    I have included 2 graphs below that show the cyclical nature of climate favorable to corn and soybean yields. I suspect, from this information alone, or any that Lobell presented in his paper, there is no valid indication that a small incremental increase in average temperature is going to be seen in the much larger cyclical fluctuations that detrimentally effect crop (or at least IL soybean and corn) yields. That the threshold temperature/moisture conditions versus yield appear to plateau into the favorable area and fall off at a steep linear rate into the more unfavorable areas should not be surprising when considering the plant physiologies. It also reminds of the relationship that Pat Michaels indicated that he found for hurricane intensities.

    If I were advising Lobell and company I would recommend that they follow the crop yields in smaller areas where climate conditions do not get “averaged out” and confused. (I have noted that while temperatures tend to temporally correlate rather well across the IL grain belt, precipitation can be very localized at times). Once they found a reasonable model for crop yield versus climate conditions they could then proceed to apply it to what they measured as changing climate conditions and make estimates of losses or gains, albeit it would need to be applied on an area by area basis.

  186. Ken Fritsch
    Posted Mar 22, 2007 at 1:04 PM | Permalink

    Re: #172

    #171, as I understand it, the MWP was caused by the Industrial Revolution, and the LIA will be caused by Peak Oil.

    I started chuckling about this one at post #150.

  187. Bob Koss
    Posted Mar 22, 2007 at 1:21 PM | Permalink

    Here’s an interesting anomaly I haven’t seen mentioned. Link

    A MASSIVE, mysterious whirlpool of cold water has developed off the coast of Sydney, forcing the sea surface to fall almost 1m and ocean currents to change course.

    Dubbed a sea “monster” by CSIRO oceanographers, the huge body of water stretches almost 200km across and plunges 1000m towards the ocean floor. Its centre sits just 100km off the coast of Sydney and could stay there for months.

  188. Bill F
    Posted Mar 22, 2007 at 2:53 PM | Permalink


    The models predict a lot of unusual climate occurrences with global warming. I am sure this is one of the things the models predicted.


  189. bernie
    Posted Mar 22, 2007 at 4:40 PM | Permalink

    My issue with the Lobell and Field article is not that what they did was wrong – but it is great that you were able to duplicate it despite the absence of their regression equation. It is more what they did not do. You have very usefully pointed out that it is better to build the model from a specific to a general rather than jump immediately to a global data set that hides significant psssible regional variations and, most importantly, key additional variables such as soil qualities. I was intrigued by the fact that the Illinois data set included prices and acreage. Can they be factored into your model?

  190. Will J. Richardson
    Posted Mar 22, 2007 at 6:25 PM | Permalink

    To Climate Audit Community, Something humorous from a dedicated lurker:

    From James Taranto’s Best of the Web, March 22, 2007, Wall Street Journal Online

    More Cowbell!
    Several readers had interesting comments about Al Gore’s comment, which we noted yesterday: “The planet has a fever. If your baby has a fever, you go to the doctor. If the doctor says you need to intervene here, you don’t say, ‘well, I read a science fiction novel that tells me it’s not a problem.’ ” Reader Martin Shimp notes:

    Clearly Gore never cared for his children while they had a fever. Fevers can be a symptom of either a bacterial or viral issue. A doctor can attempt a solution to a bacterial issue, but not a viral one. A virus has to run its course while the body fights it–and a fever is a sign of the struggle to eliminate the virus. Let’s see a scientific consensus that the Earth’s fever is bacteria-related.

    Reader Scott Jacobson questions Gore’s premise:

    Some great news for Daddy Gore: Little baby Earth does not have a fever. It’s been awhile since Daddy Gore had a little bundle at home so maybe he has just forgotten that an infant is not considered to have a fever until her body temperature is at or above 100.4 degrees, or 1.8 degrees above normal. In the last century, little baby Earth’s temperature has only gone up “almost one degree.”

    Overreacting is common among parents. I remember one morning when my first child was still an infant. Imagine my horror when I discovered that her temperature had risen to 99.5 degrees, almost a whole degree in just under 12 hours. Naturally, I immediately sat down and built a computer model, which clearly projected that by age 30, her temperature was going to be a staggering 19,710 degrees!

    Thankfully, with the help of a patient wife and an impatient pediatrician, I came to realize that these fluctuations were normal, and that my baby daughter would not be going supernova by the time she reached her golden years.

    And Kelly Murphy puts things in perspective nicely:

    So let me see if I have this right. According to Al Gore, I would have to be an idiot to decide, after reading “a science-fiction novel” (he must be referring to “State of Fear”) that it’s OK to go about living my life in a normal fashion. Instead, I should see one science-fiction movie (his) and run screaming out of the theater prepared to change every aspect of my life to avoid certain destruction.

  191. bernie
    Posted Mar 22, 2007 at 7:24 PM | Permalink

    WSJ remains a good newspaper.

  192. geove
    Posted Mar 22, 2007 at 7:39 PM | Permalink

    With regards to the lobell paper it seems intriging to me that a pair of scientists who probably don’t know the difference between a plow and a combine can speculate on how a minor change in temperature will effect agricultural yields.

    All crops planted are affected by climate regimes this can easily be seen by any horticulturalist who is heavily into flowers. Maps are readily available defining suitable growing regions for the flowers or trees desired. Industrialized agriculture is no different. However as agriculture is not a hobby but a business the climate is an integral part of the decision making process when planting crops. Certainly in the corn states Iowa. Illinois and Ohio southwestern Ontario and western Quebec the anticipated amount of heat for each of these regions is a critical factor.

    Corn for instance is bred for different heat zones. The more anticipated heat or heat units the better the yield. However, the proper choice of hybrid corn will optimize the yield for that region. Sometimes a farmer will gamble on a higher heat unit corn if the spring is early. If there is a late spring a lower heat unit corn will be planted which will still have sufficient time to mature. In warmer climates where irrigation practices predominate different management pratices will be employed.

    In the corn states I note that you are aware that planting times have receeded from late may to late april or early may. This is because since the late sixties millions of miles of subsurface drainage tubing has been installed over millions of acres. The net effect is to remove the excess water from the soil and allow it to warm up earlier and have sufficient soil
    strength to carry the machinery necessary for cropping. The net result in the humid areas of North America is earlier cropping ie, a longer growing season. At the end of the season the drainage ensures that crops can be harvested with a minimum of difficulty.

    I think that the drainage is an interesting phenomenon since it is recognized that drained soils are at least 5 degrees warmer in the early spring. Since a large portion probaly > 60% of the corn growing areas mentioned above are drained one would think that this man made change to the enviroment would show up in the average temperature of those grids for the month or so they affect the evaporative requiremnts of these affected soils.

    Certainly one would expect a corresponding cooling effect to show up in irrigated areas.

    Certainly in the humid area I reside microclimate and macroroclimate have huge effects on crop yields rainfall wind soils fertilization are all contributing factors.

    I would also note that heat units can come any time a hot june or cold june generally if the overall heat units arrive sometime in the frost free period yields will generally be similar in humid climates if you have drainage generally you have a good crop or a better crop because you are planting for the heat units in any given area. If you don’t have drainage then you are at risk of mother nature throwing you a curve whatever the time of year between planting and harvest.

    Whatever a gross amount of data with no relevant understanding is unlikely to yield results of any significance

  193. bernie
    Posted Mar 22, 2007 at 7:58 PM | Permalink

    Another interesting point. Now I count soil quality, price, drainage, irrigation, seed to climate match,…, and on and on. As noted before, the model is underspecified and, therefore, the correlations found can be entirely spurious. It is a sophisticed version of the correlation between the number of clergymen in Boston and the consumption of rum.

  194. David Smith
    Posted Mar 22, 2007 at 8:08 PM | Permalink

    Re #184

    (One curiosity on the chart is the region of coolness that suddenly appeared (this week) at about 170E. This is like a bubbling-up of cool water. What’s odd is that this coolness is not visible on the charts of deeper-ocean temperatures. Odd.)

    This odd region was removed from the map several hours after I posted this – possibly a data quality problem.

  195. John Baltutis
    Posted Mar 22, 2007 at 11:37 PM | Permalink

    Re: #193

    With regards to the lobell paper it seems intriging to me that a pair of scientists who probably don’t know the difference between a plow and a combine can speculate on how a minor change in temperature will effect agricultural yields.

    They operate under the same rubric that allows economists to speculate on economy while never running their own business, climate modelers to speculate on future climate when they can’t get their underlying weather models to predict next week’s weather accurately, or market analysts whose prognostications historically under perform the index funds, but don’t bet with their money, but sell their wares to the unwary.

  196. Richard Hill
    Posted Mar 23, 2007 at 12:04 AM | Permalink

    re 188, 189
    The cold pool off Sydney is a clear result of AGW.
    The pools drift westward. They could hit anywhere along the coast.
    But the Newcastle-Sydney-Woolongong urban/industrial complex
    is almost certainly the greatest CO2 emitter in the South West Pacific.
    With smelters, power stations, steel mills, coal mines, gas fields
    airports and around 7 million people, many driving SUVs, it cannot be
    a coincidence that the cold pool ended up there.
    Pls dont tell AG about it.

  197. Bob Koss
    Posted Mar 23, 2007 at 7:20 AM | Permalink

    NCDC has an archive going back to 1929 of daily station data located here.

    Kind of skimpy in the early years, but becomes quite extensive.

    Actual min/max temperatures and not anomalies.

  198. Barry B.
    Posted Mar 23, 2007 at 8:08 AM | Permalink

    Re: #186

    That’s some great work Ken. I was particularly struck by the fact that max temps for July/Aug. have been trending downwards.

    I do however need to point out that Lobell & Field’s research showed heat related yield decreases have been increasing since 1981. Your results do not confirm this, which is not too much of a surprise. I still think one would need to analyze yields outside of the U.S. in order to repeat their results.

  199. Ulises
    Posted Mar 23, 2007 at 9:47 AM | Permalink

    Re # 178 : If the concept of half-life holds, i.e. a constant proportion is removed/dying/decaying/extincted per unit time interval, then what is left follows f[t] = exp(-c*t), with t for time, and c as parameter. Then you have the straightforward relationship : average_lifetime = 1/c . In the given case, 1/c = 5.

    Solving exp(-0.2*t)=0.5 for t gives a half-life of ~ 3.5 years.

    Of course, applying a different decrease function will result in another value, but I wouldn’t speak of half-life then.

  200. Posted Mar 23, 2007 at 9:50 AM | Permalink

    Re: 150

    That is a cute picture. Looking at it, it is obvious that within a year, there is a cyclical component. In addition, the rate of annual change itself seems to be changing.

    A simple linear regression model can thus be formulated has:

    y_t = \beta \mbox{year} + \sum\alpha_i\mbox{month}_i + e_t

    That is, each year contributes 12 observations (except for missing observations), the observations are stacked on top of each other, and there is a dummy for each month). If there is a linear annual trend with month-to-month variation, this model will capture it. If the assumptions of a linear trend with monthly cyclical variation and noise are not violated, then the residuals from the model will not have any patterns.

    Without further ado, here are the standardized residuals (residual = observed – predicted):

    Looking at the graph, note that residuals there is a definite overall pattern to the residuals with under-prediction at either end and over-prediction in the middle. Now, whether this pattern has anything to do with the frequency of volcanic eruptions, I have no idea but you might want to check http://www.volcano.si.edu/world/largeeruptions.cfm

    The preview is a little messed up. I hope I did the image embedding right.


  201. Ken Fritsch
    Posted Mar 23, 2007 at 10:01 AM | Permalink

    Re: #199

    Barry B., thanks for the links and comments as they and those of others in this thread have added much to my information base. My very small scale study was simply an effort on my part to attempt to confirm or deny some of my past experiences and thoughts on the subject matter.

    The major point that I derived from my spatially very restricted look at the Lobell issue was that, yes, one can find a correlation of temperature to yield, but that that correlation needs to be tempered with the recognition that temperature and moisture are both important factors when combined at more extreme values and that being the case can become confounding factors when they correlate and cannot be separated sufficiently in nature.

    I do not believe that Lobell put together a model that contained all the factors that might affect yields, but simply did individual correlations and thus was able to isolate the temperature versus yield relationship and in turn relate it to the “cost” of AGW. The latter relationship was in my opinion probably the motivation for doing the analysis and not that of learning some insights into the more complex nature that climate plays on crop yields. I can also see his paper being quoted as a cost of AGW in terms of increasing global temperatures without adding the authors own caveats on carbon dioxide fertilization which would, of course, be directly correlated in the big “A” AGW theory of warming.

    I also understand what you are saying about the remainder of the world’s (Northern Hemisphere I think otherwise Lobell growing seasons do not make sense) crop land conditions being different than IL. Crops that would be grown under normal climate conditions that put the prevailing moisture/temperature on the downward sloping part of the yield curve would make them more temperature sensitive over time, but would also call into question why they would be grown under those conditions and whether or not a better breed of crop would be more adaptable to these conditions. I suspect this would happen less often where crop produce were grown and sold under more free market conditions and without the artificial influences of trade restrictions and government price supports, i.e. obtaining the full benefits of Ricardo’s comparative advantages.

    Some other points from Lobell that were brought into contention with my limited study are:

    They restrict their published results to the months with the greatest correlation between temperature and crop yield and than condition this by stating that the sensitivity to months selected is not great. I found no correlation for IL corn and soybean when the months April through August were selected.

    They do not mention that the worst conditions for crops of higher temperatures in droughts can destroy yields to a significant extent and color the results of their relationships. Also they make no mention of even the potential for reoccurring weather patterns being the major player in crop yield reductions. Would detailed discussion of those conditions redirect the emphasis from gradually changing temperatures to the more extreme changes seen in natural weather cycles? I can also see regionally extended temperature patterns not matching precipitation amounts which tend to be, in my quick look, more localized, or at least in IL. I see a need to do these studies on a size restricted area by area basis to get more of the variables explained.

    I find some of the lack of detail with Lobell’s publication similar to what I have seen criticized at this blog as lacking in tree ring publications and while Lobell is probably as nice a guy as Rob Wilson is, I think their work can never be over critiqued or questioned.

    I would like to follow up on some of yours and others’ suggestions here ‘€” but at a more leisurely pace.

  202. Jim Hanrahan
    Posted Mar 23, 2007 at 11:04 AM | Permalink

    OK, this made me mad enough to send a letter to the editors. It’s an uncritical “profile” of Mann, where he is allowed to impugn the motives of his critics without facing any questions about whether the criticism is justified. I contrasted it to a profile of Richard Lindzen that was run in 2001, where comments from his critics appeared in almost every other paragraph. Do editors really think so little of readers these days that they think we can’t see through this kind of nonsense?

  203. Stan Palmer
    Posted Mar 23, 2007 at 11:25 AM | Permalink

    re 203

    Does anybody take the Scientific Amercian seriously? Ever since it was “dumbed down”, it is jas become very political and not scientific.

  204. Jim Hanrahan
    Posted Mar 23, 2007 at 11:44 AM | Permalink

    Re: #204

    Well, obviously I do 😉

    I’m quite sure there are many readers, like me, who respond to the name and history and still expect a certain level of honesty and integrity. This article demonstrates neither.

  205. bernie
    Posted Mar 23, 2007 at 11:48 AM | Permalink

    You wrote:

    I find some of the lack of detail with Lobell’s publication similar to what I have seen criticized at this blog as lacking in tree ring publications and while Lobell is probably as nice a guy as Rob Wilson is, I think their work can never be over critiqued or questioned.

    This I believe is the important meta-point and fits with the issues that Steve and others have worked so hard to bring to light. The recurring weakness in much of this work is a lack of specificity and transparancy, without which replication is very difficult, if not impossible. I am sure that if Lobell and Field had spelled out their model, its limitations would have been self evident. As presented, we are left with a potentially important conclusion, a ready sound bite, accompanied with a silent “trust us”. That is less the stance of scientists and more the pitch of stock promoters and politicians.

  206. Posted Mar 23, 2007 at 12:21 PM | Permalink

    Hope on the Horizon?



  207. bernie
    Posted Mar 23, 2007 at 2:58 PM | Permalink

    Cute, since the realists seemed to have won. The sad thing is that the 6th graders showed more sense than a number of our Senators – not to mention one ex-Senator.

  208. David Smith
    Posted Mar 23, 2007 at 3:58 PM | Permalink

    In the worth-a-glance category is this satellite-derived animation of clouds above the Pacific Warm Pool. The Australian coast is on the left.

    The Warm Pool is 30C+ and the atmosphere above that warmth stays unstable, producing many showers and thunderstorms. This part of the tropics is wet, warm and cloudy.

    The black spots are cloud tops that probably protrude into the tropopause. The air from the tops of the storms flows generally eastward, towards the eastern Pacific, where it cools and descends. The air was at very low temperatures at the tropopause, so it is quite dry. This dry, sinking air keeps the eastern Pacific rather dry and cloud-free (except for low clouds).

    The map currently (on 23 March) shows a lot of thunderstorms clumping together, which may organize into a tropical storm or typhoon this weekend.

  209. jae
    Posted Mar 23, 2007 at 5:04 PM | Permalink

    In the previous Unthreaded Thread (#289), I summarized my efforts to determine climate sensitivity (degrees/watt-meter-2) by looking at annual changes in Solar radiation and December ‘€” July temperature changes. I have now expanded my U.S. climate sensitivity spreadsheet to 138 locations and think I have solved the puzzle of why sensitivity west of the Sierras and Cascades is less than half of that for the rest of the Nation. I now make the following conclusions based on this data:

    1. Sensitivity varies greatly over the USA, but is remarkably consistent within general climate regimes. For example, sensitivity for all locations West of the Cascade and Sierra Nevada Ranges, including Hawaii and the Sitka Peninsula, is about 0.05 deg/watt/m-2 and is very uniform, with the exception of San Francisco, which has a sensitivity of 0.08. The sensitivity in this zone is less than half of that for the rest of the US, which I believe is attributable to the influence of the cool waters of the Pacific Ocean; as the warm air rises it has to be replaced by cool marine air in this part of the world. That is probably why the summer climate along the West Coast is more pleasant and less “muggy” than that in the East. I have a hunch that San Francisco is different because of Easterly winds in the summer (does anyone know if this is true?). Windspeed data (see spreadsheet) also show San Francisco to be much windier than all other West Coast (or inland) locations, so I think that has something to do with it’s higher sensitivity.

    2. The average sensitivity for all locations is 0.11 deg/watt/m-2, very close to that determined by Idso in his famous experiments.

    3. The sensitivity for virtually all locations East of the Rocky Mountains is about 0.10-0.14, and this is also strikingly uniform across the entire area. This is attributed to the fact that warm air that rises in this area is replaced by much warmer air than is the case on the West Coast (warmer water, much landmass). Some areas near the Great Lakes show slightly lower sensitivities (0.09), probably due to the cooling effects of air over those lakes.

    4. The sensitivities in the areas Eest of the Sierras and Cascades and West of the Rocky Mountains are linked very closely to altitude by a logarithmic relationship (R2 of 0.66 to 0.74, depending upon whether the West Coast and the Desert Southwest are included. (Figs 1 &3). The highest sensitivities, 0.20 and 0.22, we at the highest elevations, Alamosa CO and Flagstaff, AZ.

    5. July absolute humidity does not appear to be correlated to sensitivity in the West where there is significant variation (R2 = 0.003 to 0.01, Figs 1 and 2). However, when elevation x humidity is plotted against sensitivity, the correlation is a little better than that between elevation and sensitivity (R2 of 0.68 to 0.85, cf. 0.66 to 0.74, Figs. 1-3). This suggests to me that there little, if any, positive water vapor feedback mechanism.

    6. There is a strong correlation between latitude and sensitivity, but it is positive and appears to be logarithmic (R2 = 0.62, Fig. 4) for the Left Coast and negative and linear (R2 = 0.0.62, Fig. 5) for the areas between the Cascades and Rockies. Another puzzle.

    The fact that there is not much temperature difference between Sacramento, CA and the city of Dodge City, KS (24.3 vs 26.8 C),which is at approximately the same latitude, when Dodge City has twice the sensitivity, is intriguing. Where does all the extra heat go at Dodge City? The July absolute humidities are not all that much different, either.(11.8 vs. 14.0 g/m-3). It probably heats a much thicker layer of water-laden air, which releases more heat to space. Maybe there are just some very non-linear relationships going on that I don’t understand. Another puzzle.

    If these estimates mean anything, and if the 3.7 watts/m-2 for a doubling of CO2 are correct, the West Coast could warm by a whopping 0.18 degrees. The Eastern 2/3 of the US could warm by about 0.4 degrees, and Flagstaff could warm by one degree.

    As I mentioned before, it appears that the idea of a “global climate sensitivity” is about as meaningful as a “global average temperature,” for the same reasons outlined by McKitrick, et. al. in their recent paper.

    If someone sees a fatal flaw in all this, please speak up so I can quit thinking about it.

  210. Steve Sadlov
    Posted Mar 23, 2007 at 8:03 PM | Permalink

    RE: #210 – “I have a hunch that San Francisco is different because of Easterly winds in the summer ” – Not quite. Easterlies are actually quite unusual in Summer. SF only gets them in Fall through Spring. Fall and Spring they are Santa Ana like, winter they are colder. They are rarer than in SoCal and never persist more than a day or two. Dominent winds in all seasons are Westerlies. In summer, these bring, often, a thick marine layer, including “high fog” (coastal stratus). My guess as to why SF is “different” is probably it is the only city center on the West Coast, besides San Diego, with a true open ocean margin “beach climate” (LA, on the other hand, has its city center in a non beach climate, miles inland, same deal for Seattle,on the sound, not open ocean, Portland, on the river but inland, and of course places like Sacramento, Fresno, Springfield, etc). San Diego is so far south, it actually is in the northern extremes of the Baja “fog desert” regime, which tends to have higher absolute and relative humidity than coastal areas further north. I digress.

    When I lived in the Western part of SF, we had coastal stratus that would not burn off for days in mid Summer. Downtown, it would burn off by noon, then roll back in 5ish. The warmest weather in SF is actually normally in September, with late April or May being a sort of secondary warm peak.

  211. David Archibald
    Posted Mar 24, 2007 at 3:42 AM | Permalink

    Re 210, you have enough to publish on and it would be good for the common weal if you did so. The fact that you got close to Idso’s numbers, independently, suggests that your methodology is correct. It is important that more papers get published backing up Idso’s work. I get 2.86 watts for a doubling, but I suggest that you check at the MODTRANS site. Don’t be limited by the doubling thing. Choose your own levels. These are my numbers, for what it is worth:

    ppm watts/m2
    280 280.9
    300 280.6
    320 280.3
    340 280.1
    360 279.8
    380 279.6
    400 279.4
    420 279.2
    440 279.0
    460 278.8
    480 278.6
    500 278.5
    520 278.3
    540 278.1
    560 278.0
    580 277.9
    600 277.7
    620 277.6

    AGW is real but miniscule.

  212. EW
    Posted Mar 24, 2007 at 7:59 AM | Permalink

    Ad daily changes: being from Central Europe, I’m not accustomed to regular daily changes in cloud cover. Therefore I was quite surprised when I spent 6 weeks in Texcoco (near Mexico City) in September. Mornings we started to the Institute and sun greeted us with a nice view of Popo and its smoke plume. Around midday, white cumuli (?) arose over the mountain range (I was told that those are from the Atlantic). And around 3 p.m. I wandered to the Institute library equipped with a list of citations and a big umbrella. Because around 5 p.m. it started inevitably raining and rained a lot to approx. 2 a.m of the next day.

    Which was quite dangerous when we were on a field collecting trip and didn’t manage to get back to our car in time. In the course of the next hour, the rain converted a small brook we easily had crossed before into a real river, where we almost drowned our institutional VW Beetle.

    What’s that for a kind of weather?

  213. geove
    Posted Mar 24, 2007 at 8:03 AM | Permalink

    Isn’t the weather climate around San Francisco regulated by the cold oceans currents flowing down the west coast from Alaska.

  214. jae
    Posted Mar 24, 2007 at 10:51 AM | Permalink

    Aha, in looking at Figs 4 and 5 in my spreadsheet (see #210), I think I can demonstrate that water vapor has a strong NEGATIVE impact on sensitivity (Fig. 4). On the West Coast, water vapor decreases with increasing latitude, and sensitivity becomes higher. However when you go inland, water vapor decreases with increasing altitude, an sensitivity increases, also.

  215. Allan Ames
    Posted Mar 24, 2007 at 11:53 AM | Permalink

    re 210:
    I am new to the discussion, so forgive me if I have missed something.
    It strikes me that the climate sensitivity(C) and adiabatic lapse rate(L) are closely related, with C being local averages of L converted (with some difficulty with time and volume derivatives) from PV to power. If so, then the C depends sharply on total moisture content, and should then show dependence on wind direction relative to the nearest mountain chain.

  216. jae
    Posted Mar 24, 2007 at 12:56 PM | Permalink

    216. I think you are correct. I am now fairly certain that sensitivity increases with decreasing moisture in the air. This explains why the sensitivity is only about 0.13 east of the Rockies (lots of water vapor), but it is 0.2 at high elevations in the Rockies (little water), as well as explaining the variations of sensitivity with latitude.

  217. David Smith
    Posted Mar 24, 2007 at 3:02 PM | Permalink

    A question for someone statistically-gifted, and pardon me since this has likely been answered in past threads:

    Suppose I have two time series and I want to see if they are correlated to a significant extent. I do a standard correlation check and find that there is not a significant correlation.

    But then I smooth the two series (averaging adjacent values, then maybe doing a double-smoothing and triple-smoothing) so that only the broad variations of the series are left. I then do a standard correlation check and find that r has increased, such that the test indicates significant correlation.

    Have I “cheated”? It seems like the only correlation that should count is between unsmoothed time series, yet smoothing is a common practice and correlations are made between smoothed series.


  218. DeWitt Payne
    Posted Mar 24, 2007 at 3:13 PM | Permalink


    I’m pretty sure that higher water vapor means relatively more heat lost as latent heat from the surface and a lower surface temperature for a given radiative heat input. The dry adiabatic lapse rate is about -10 degrees K/km altitude. The moist adiabatic rate is less than this and decreases as the surface temperature and the associated saturated water vapor pressure increases. As the altitude increases, the moist lapse rate approaches the dry lapse rate associated with a much higher surface temperature.

  219. jae
    Posted Mar 24, 2007 at 3:30 PM | Permalink

    216, 219: Thanks for your input. I have to study lapse rates some more. Whatever the exact explanation is, it seems clear to me that water vapor exerts a NEGATIVE forcing, which is contrary to what all the GCM guys are saying.

  220. Chas
    Posted Mar 24, 2007 at 3:38 PM | Permalink

    Here is a 1955 paper reviewing Callender’s CO2 history (Slocum ” Has the amount of carbon dioxide in the atmosphere changed significantly since the begining of the twentieth century”)

    Click to access mwr-083-10-0225.pdf

    Conclusion; ” All this does not refute Callender’s thesis. The available data merely fail to confirm it”

  221. richardT
    Posted Mar 24, 2007 at 3:49 PM | Permalink

    Smoothing the data reduces the effective number of degrees of freedom, if you don’t account for this, any statistical test will be prone to type 1 errors (incorrectly rejecting the null hypothesis of no correlation).

  222. Steve McIntyre
    Posted Mar 24, 2007 at 4:12 PM | Permalink

    #222. Of course, exactly how to do the accounting for the reduced degrees of freedom is the $latex $2^{8} $ question. The handling of smoothed correlations deserves a special discussion all by itself, as the reporting of smoothed correlations in Team articles is very common.

    I’d be hard-pressed to give a statistical reference which deals with the matter in a convincing way. If I were to try to think about it from first principles, I’d be tempted to use the viewpoint of wavelet analysis, where the pyramid structure explictly reduces the number of values in each scale. So if you start with 128 years of data and go to 8-year bins, you only have 16 degrees of freedom at that scale.

  223. Ken Fritsch
    Posted Mar 24, 2007 at 4:36 PM | Permalink

    Barry B. and Bernie, just a brief update to let you know that I tried a correlation of days over 95 degrees F (DO95F) for the months of July/August for IL corn and soybeans (as suggested, through Barry, by the experts for corn) with delta yield. I used only one site (Champaign) for my DO95F, but selected it to be central to the IL grain belt. The correlation improved over the use of Max temp with corn producing an R^2 = 0.68 and for soybeans R^2 = 0.47. The DOF95 correlated with Max Temperature with R^2 = 0.52. The effects of low moisture were not separable from DO95F.

    These are strong temperature correlations here and to view how they occur over time I present the graph below showing the reoccurrences from the time period 1948 to 2006. I see something more that I would expect from cyclical natural events than a climate trend.

    I also followed up on the references from the Lobell paper to determine how difficult it would be to replicate what the authors did. The reference to the crop yields and acreages, here, was found as the authors noted in their paper and turn out to be available from 1961 to 2004 as analyzed by the authors and on a nation by nation basis.


    I would than assume that the authors used nation by nation yield and acreage data, weighted by acreage, and used the CRU grid temperatures and precipitation data on a nation wide basis. If I attempt to replicate some of the authors’ work from a different viewpoint I would initiate it by starting on a nation by nation basis. Do any of you have any preferences?

  224. richardT
    Posted Mar 24, 2007 at 5:09 PM | Permalink

    This problem is overlooked too often. I’ve had to persuade people, many times, that the increase in the correlation coefficient after smoothing is expected. There may be a simple solution, but I don’t know one.

    Your solution is too conservative, as this simulation shows
    x=filter(rnorm(n),rep(1/smoother,smoother))#does running mean
    xy=cor(x,y, use="complete")
    2 * min(p, 1 - p)

    The value should be .05, but it comes out as about 0.02 (regardless of the amount of smoothing).
    Changing df to (n-2)/(0.71*smoother) gives about the right answer.
    (This is ~ 1/sqrt(2), but I don’t know if that is relevant)

  225. Pete
    Posted Mar 24, 2007 at 5:32 PM | Permalink


    Of course, exactly how to do the accounting for the reduced degrees of freedom is the 2^{8}$ question $

    If the question is to estimate the number of effective parameters (or equivalently the number of degrees of freedom), nonparametric fitting provides the answer.

    It is the trace of the “hat” matrix.

    R documentation

    — Pete

  226. DeWitt Payne
    Posted Mar 24, 2007 at 6:17 PM | Permalink

    Steve M or John A, thanks for bringing back the Img button.

    Now to see if I can make it work in this case. If it doesn’t the page on moist adiabats and pseudoadiabats I’m linking the image from is:


    Lots of information on Physical Meteorology here.

  227. David Smith
    Posted Mar 24, 2007 at 7:31 PM | Permalink

    Re #222, #223


    My question was prompted by Figure 1 of Emanuel’s recent GRL comment on Michael et al’s paper . Figure 1 is a truncated version of his well-known 2005 plot. The Figure shows r-squared of 0.88, based on correlating two double-smoothed curves. That seems impressive.

    When I try to replicate it, I come close (r-squared = 0.82), so my data looks like a reasonable replication. When I use my data to look at the correlation of the unsmoothed values, though, I get r-squared of 0.46, which seems much less significant. So, I’m trying to make sense of all that.

    (Ken F., if you’re around, a free copy of Michael’s reply to Emanuel is here , courtesy of World Climate Report, as are the above links. )

  228. DeWitt Payne
    Posted Mar 24, 2007 at 10:15 PM | Permalink

    #220 jae

    I don’t think negative forcing is quite correct. It’s more like less positive. A highly simplified explanation of enhanced greenhouse warming is that if you increase the temperature at the top of the atmosphere, you increase the surface temperature by the same amount, although that’s backwards from what the greenhouse effect actually does, in my understanding. That doesn’t seem to be true for moist adiabats. The temperature at the surface doesn’t increase as much as at the TOA. You still have an increase in temperature with an increase in forcing. It’s just smaller. A better description of the actual physics, I think, is that a temperature increase at the surface causes a larger temperature increase at the TOA if there’s enough water around to allow the specific humidity to increase.

    In the figure in #227, the solid black lines are the moist adiabats. The solid grey lines are dry adiabats. The dotted grey lines are pseudoadiabats calculated assuming that condensed water vapor falls back to the surface. The details of how these are calculated are in the link and the pages related to that link.

  229. DaleC
    Posted Mar 25, 2007 at 3:28 AM | Permalink

    Here is an interesting one:

    “Leap Year Error Mars Global Warming”

    “…Most climate experts are predicting a sharp increase in global temperatures in the next 50 to 100 years, so the need for accurate data is more important than ever…”

    See http://www.freerepublic.com/focus/f-news/585355/posts

  230. jae
    Posted Mar 25, 2007 at 6:41 AM | Permalink

    227, 229, DeWitt: thanks again. I will study the met. stuff, because I am interested in it. At this point, I am just relaying what I see empirically. The data indicate a small forcing for a unit increase in energy and a decrease in that forcing with an increase in moisture (lower altitude or lower latitude). You state:

    A highly simplified explanation of enhanced greenhouse warming is that if you increase the temperature at the top of the atmosphere, you increase the surface temperature by the same amount, although that’s backwards from what the greenhouse effect actually does, in my understanding.

    I wonder if all the explanations of enhanced greenhouse warming are “simplified,” because nobody really understands just what is going on.

  231. paminator
    Posted Mar 25, 2007 at 7:13 AM | Permalink

    Jae, re #210

    Very clever natural experiment. Any ideas on the error bars associated with the solar insolation data?
    The link for your excel file in #210 does not seem to be functioning. Any suggestions? Has anyone else downloaded the excel sheet?

  232. Allan Ames
    Posted Mar 25, 2007 at 8:54 AM | Permalink

    Question on R^2? I presume this means the Pearson product-moment variety, but I never see distribution analysis to justify use of Pearson. I know that Pearson will give garbage with tailed or multimodal distributions. The nonparametric Spearman is nowhere near as vulnerable to artifacts arising from the distribution because it first maps the data onto a uniform scale.

  233. jae
    Posted Mar 25, 2007 at 9:32 AM | Permalink

    232. Hmm, the link works for me. Try again? I have no idea how you would determine error bars for the 30-year average solar insulation data. BTW, the data is here.

  234. Ken Fritsch
    Posted Mar 25, 2007 at 9:33 AM | Permalink

    Re: #233

    Good point. I was copying the procedures of Lobell’s paper but you are correct, of course, that the distribution analysis should be done before applying R^2. I will take a look and report back on this matter.

    I have done rankings correlations of the variables that I used in calculating R^2 with the methods and procedures reported. The results were much the same in both cases.

  235. Paul Linsay
    Posted Mar 25, 2007 at 10:09 AM | Permalink

    ##218, David,

    Smoothing is a a form of frequency filter. It removes the high frequency components and leaves the low frequency components. Do it enough on two different time series and you just have a pair of constants left, which correlate perfectly.

    If you are interested in the relation between a particular smoothing function and the frequency domain you have to know something about Fourier analysis and z-transforms, available in any standard book on signal processing. Ths short description is that smoothing of a time series is the convolution of the smoothing function, e.g. 1-2-2-1, with the time series. By the convolution theorem, in frequency space this is the product of the Fourier transform of the time series with the FT of the smoothing function. The FT of a smoothing function is typically peaked at zero frequency and falls off quickly at higher frequencies. That’s why smoothing gets rid of the wiggles in the original time series. Do the smoothing a second time and you’re squaring the FT of the smoothing function making it even more sharply peaked at zero and reducing the high frequencies even more. Do this N times and the FT of the smoothing function is raised to the Nth power, making it very sharply peaked at zero indeed.

    Smoothing is done in the analysis of climate data all the time with no apparent justification, at least none that I can discern. It throws away a lot of interesting data without adding anything except spurious correlation. These people don’t seem to understand that “noise” carries lots of information and may even be the signal itself. They seem to be stuck on a purely deterministic model of climate without understanding that a stochastic model might be more fruitful.

  236. Posted Mar 25, 2007 at 10:19 AM | Permalink


    I posted this figure some time ago. It shows what happens to sample correlation coefficient when one uses running average smoothers of different length, white noise input. Theoretical p=0, but if you choose the filter length post hoc, you’ll probably get the sample correlation coefficient you want. As the second figure shows, sign of the sample correlation can change many times.

  237. Steve McIntyre
    Posted Mar 25, 2007 at 10:25 AM | Permalink

    #236,237. People interested in this should also reflect on the Slutsky phenomenon from successive smoothing/integrations. Klemeš added the observation in a climate context that many hydrological series are natural integrals.

  238. Paul Linsay
    Posted Mar 25, 2007 at 10:27 AM | Permalink

    Addendum to #236, in response to SteveM at #223

    exactly how to do the accounting for the reduced degrees of freedom is the question

    Correlation of two time series is also a product in frequency space. If you put a threshold in frequency space such that components greater than threshold are signal and the rest noise, smoothing pushes more components down into “noise”. You could count the remaining signal components to determine the number of degrees of freedom. This requires a model of the noise to determine a cutoff, but you get the idea I’m sure.

  239. David Smith
    Posted Mar 25, 2007 at 10:48 AM | Permalink

    Re #236

    Thanks, Paul. In the case of the data I’m examining (Atlantic SST, hurricane ACE and sea level pressure time series) the use of smoothing might remove higher-frequency events (like El Nino / La Nina) and leave behind the lower-frequency oscillations.

    What’s bizarre about processing my data (SST, hurricane, SLP) using double-smoothing is that what I’m left with appears remarkably like solar signals, especially since the 1960s. I did not expect that. I’m agnostic on solar cycle / climate connections.

    It would be a hoot if Emanuel’s CO2/SST/hurricane correlation is actually a solar/SST/hurricane correlation.

  240. Posted Mar 25, 2007 at 11:00 AM | Permalink

    236, 239

    These people don’t seem to understand that “noise” carries lots of information and may even be the signal itself.

    Very true. Ritson coefficient is a good example how some people just don’t get it. There are actually good smoothing methods, which can be used when statistical properties of signal and noise are known. See e.g. A. Gelb, Applied Optimal Estimation.

    If you put a threshold in frequency space such that components greater than threshold are signal and the rest noise

    à  la Mann&Lees 1996. Using this kind of non-sense they show that there is anthropogenic signal global temperatures.. AR1 with p close to 1 means signal, unbelievable rubbish.

  241. Ken Fritsch
    Posted Mar 25, 2007 at 4:19 PM | Permalink

    Re: #233 and #235

    The distributions, that I encountered in my analysis of Lobell, fit log normals with the exception of the Days Over 95 Degrees F which obviously had too many 0s in the time series. Using rank and non-parametric Spearman correlation the results and relationships stay essentially the same. On further thought your caution on using a non-parametric statistic is well taken and something I should have done without looking at log normal distributions.

    I did report the Days over 95 F correlation with delta corn and soybean yields as a Pearson product moment and that is a mistake on my part. I did obtain a high correlation using rank correlations, but I am not certain with so many 0s if even that procedure is correct. That distribution looks like a Poisson distribution but with a variance (much) larger than the mean and as I recall a negative binomial may be the better fit in these cases. I plan to look further at it.

  242. Barry B.
    Posted Mar 25, 2007 at 5:07 PM | Permalink

    Re: #224

    Ken, one thing you might want to do before you move on to a larger sample is check yield vs. the average temperature threshold of 77 degrees F I referenced earlier. I would expect the correlation to be even stronger than you found for DO95F because of increased transpiration by the plant during nighttime hours.

    This may be of some help to you for your new endeavor: http://www.sage.wisc.edu:16080/pubs/articles/F-L/Leff/Leff2004GBC.pdf
    It should help in narrowing down the nations to focus on.

    Also, one of the more puzzling aspects of the Lobell & Field paper is this statement they made saying, “omitting the last 3-4 years of the study period had significant effects of inferred wheat and maize losses…” They then went on to explain that a previous study, which they seemed to agree with, recorded a positive yield trend for a period between 1982-1998 and then said, “these sensitivities indicate that warming since 1998 had high leverage on the estimated impact of climate trends”.

    I’m a little confused by these statements because in order for a small 3-4 year period to have such large influence on the other 17 years of data, there would have had to of been a drastic change in yield somewhere in the world. If this is correct, this would not indicate a global warming influence. Instead it would seem that natural cyclic weather events came into play.

    Maybe I’m missing something (feel free to point out where I’m in error), but if I’m not, I think further investigation is warranted and it might be something for you to pursue further – given that you’re going to be widening your focus anyway.

  243. David Smith
    Posted Mar 25, 2007 at 7:36 PM | Permalink

    And for something different, here is Barrow, Alaska’s current webcam view. The view shows a couple of town blocks with the ocean beyond. Current (25 March) temperature is -28C.

    Daylight is returning. The whiteness must be blinding at times.

  244. Ken Fritsch
    Posted Mar 25, 2007 at 7:47 PM | Permalink

    Re #243

    Barry B., I will look at the over 77 degree F minimum on delta corn and soybean yields and I will use rank correlations. Also, thanks for the linked article on crop areas of the world.

    I am attempting to resolve in my mind how Lobell went about obtaining the data for his study. The crop acreages and yields are readily available for his study through their papers reference 6 and that I linked in a prior post (fao.org). These data are available for the period that Lobell studied, i.e. 1961 – 2004. The data base is handy for filtering the data before downloading it to Excel. I do not see the yield (or acreage) data broken down further than by nation in this reference. It leads me to judge that Lobell used nation by nation yield data weighted by total nation output (or alternatively acreage planted) and than perhaps referenced a source such as the one you linked in your latest post to determine which areas in the nation to use for climate variables as derived from CRU grids. I also have to assume that he used NH data only, since they would fit his seasonal months.

    I am starting a quick download from FAO of the major crop producing nations in the NH, restricted to corn and soybeans for now, and plotting the yield trends and looking at annual deviations from the trend line. In my lazy man’s approach I want to determine whether these deviations increase gradually to the negative (as expected from a gradual trend in global warming) or whether these trends tend to be episodic as would be expected more from natural cycles such as drought (or floods or other natural disasters). If I see large deviations, I will look at the climate for that year in the nation’s crop land.

  245. bernie
    Posted Mar 26, 2007 at 7:35 AM | Permalink

    At RealClimate they are addressing (or perhaps not addressing) the issues raised in Does a global temperature exist? Is a clarification of the underlying issue needed?

  246. JerryB
    Posted Mar 26, 2007 at 7:54 AM | Permalink

    Steve M,

    In the Multiproxy PDFs page
    the link to your agu05.ppt file needs fixing

  247. Mark T.
    Posted Mar 26, 2007 at 9:11 AM | Permalink

    #236,237. People interested in this should also reflect on the Slutsky phenomenon from successive smoothing/integrations. Klemeš added the observation in a climate context that many hydrological series are natural integrals.

    This is a well known communications problem (signal processing in general). If the incoming data are not filtered with the exact transmission filter, intersymbol interference occurs. Filtering “spreads” the energy of a signal across time. In the comm context, this means that a bit at time t = 0, will have energy from future and past symbols.


  248. Steve Sadlov
    Posted Mar 26, 2007 at 9:28 AM | Permalink

    RE: #229 and upstream related posts. Pielke Sr. says, essentially, low, wet clouds are a negative feedback or forcing, whereas, high icy ones are perhaps positive. There would obviously be a continuum betweeen the extremes. Non condensed moisture may also behave differently depending on air temperature and the concentration of the moisture.

  249. Ken Fritsch
    Posted Mar 26, 2007 at 9:28 AM | Permalink

    Re #245

    Barry B., I looked at days with a minimum temperature over 77 degrees F in the July/Aug time period for the same station (Champaign-Urbana) where I found the over 95 degree maximum temperatures. In this area, over 77 F minimum temperature days are much rarer than the days with over 95 F for a maximum temperature. For the 1948-2004 period I found only 7 days with a 77 F or greater minimum temperature — 1 in 1955, 2 in 1966, 3 in 1983 and 1 in 1998. These conditions might explain why the crop yields are consistently high in this area of IL.

    For IL, minimum temperatures were not as important as maximum temperatures and the products of maximum temperature and precipitation when I included that statistic in my calculations. It was not at all important for soybean delta yields. That is not saying that in warmer climates the minimum may be more important.

  250. DeWitt Payne
    Posted Mar 26, 2007 at 12:23 PM | Permalink

    Re: #249

    Steve S., Everything I posted is about clear air atmospheric structure based on thermodynamics. Clouds are another story entirely and are a major source of uncertainty in attempts to mathematically model climate.

  251. bernie
    Posted Mar 26, 2007 at 12:55 PM | Permalink

    So you did find a significant interaction effect between temperature and precipitation? If so, then the Lobell model is under-specified, right?
    I focus on this, because I seem to see it as a recurring theme in the studies involving proxies. Bernie

  252. John Baltutis
    Posted Mar 26, 2007 at 2:33 PM | Permalink

    Re: #247

    Try http//www.climateaudit.org/pdf/agu05.ppt

  253. David Smith
    Posted Mar 26, 2007 at 3:12 PM | Permalink

    NOAA has a website that offers various maps showing recent weather history. For instance global temperature.

    The map, based on raw data, for the last 180 days is here . (The lower map shows the anomaly.)

    The map, after “reanalysis” (adjustment), is here .

    I believe that the adjusted version becomes the official record.

    The reanalysis generally warmed things, versus the raw temperature data. For some reason I’m not surprised.

  254. JerryB
    Posted Mar 26, 2007 at 5:32 PM | Permalink

    Re #253,


    Thank you. As it happens, I had tried a variant of that,
    (with a colon after the http) and downloaded the file, but
    also concluded that the bad link should be brought to the
    attention of Steve, or John A.

  255. Barry B.
    Posted Mar 26, 2007 at 6:09 PM | Permalink

    Re: #250

    Ken, I have to apologize, I sent you on a wild goose chase. I meant to suggest looking at the daily mean temperature >77 degrees F but after a quick look at some of the Urbana data, I now realize this will not provide any further insight into the heat/yield relationship. The Purdue publication which I linked to earlier defined this as a threshold where yields started to drop but it was only when mean temperature exceeded 77 degrees F on the same day that max temps exceeded 95 degrees F. So in other words, I was in error to begin with.

    Thanks for being so patient with my errant suggestion.

  256. John Baltutis
    Posted Mar 27, 2007 at 1:41 AM | Permalink

    Re: #253

    Glad you fixed the fat-fingering on my part. For everyone else, it’s http://data.climateaudit.org/pdf/agu05.ppt

  257. MarkR
    Posted Mar 27, 2007 at 5:49 AM | Permalink

    Why Bad Beliefs Don’t Die

    Because beliefs are designed to enhance our ability to survive, they are biologically designed to be strongly resistant to change. To change beliefs, skeptics must address the brain’s “survival” issues of meanings and implications in addition to discussing their data.

    The lesson for skeptics is to understand that people are generally not intending to be mean, contrary, harsh, or stupid when they are challenged. It’s a fight for survival. The only effective way to deal with this type of defensiveness is to de-escalate the fighting rather than inflame it. Becoming sarcastic or demeaning simply gives the other person’s defenses a foothold to engage in a tit-for-tat exchange that justifies their feelings of being threatened (“Of course we fight the skeptics-look what uncaring, hostile jerks they are!”) rather than a continued focus on the truth.

    Skeptics will only win the war for rational beliefs by continuing, even in the face of defensive responses from others, to use behavior that is unfailingly dignified and tactful and that communicates respect and wisdom. For the data to speak loudly, skeptics must always refrain from screaming.

    Finally, it should be comforting to all skeptics to remember that the truly amazing part of all of this is not that so few beliefs change or that people can be so irrational, but that anyone’s beliefs ever change at all. Skeptics’ ability to alter their own beliefs in response to data is a true gift; a unique, powerful, and precious ability. It is genuinely a “higher brain function” in that it goes against some of the most natural and fundamental biological urges. Skeptics must appreciate the power and, truly, the dangerousness that this ability bestows upon them. They have in their possession a skill that can be frightening, life-changing, and capable of inducing pain. In turning this ability on others it should be used carefully and wisely. Challenging beliefs must always be done with care and compassion.

    Skeptics must remember to always keep their eye on the goal. They must see the long view. They must attempt to win the war for rational beliefs, not to engage in a fight to the death over any one particular battle with any one particular individual or any one particular belief. Not only must skeptics’ methods and data be clean, direct, and unbiased, their demeanor and behavior must be as well.


  258. Ken Fritsch
    Posted Mar 27, 2007 at 7:30 AM | Permalink

    Re: #252

    So you did find a significant interaction effect between temperature and precipitation? If so, then the Lobell model is under-specified, right?
    I focus on this, because I seem to see it as a recurring theme in the studies involving proxies. Bernie

    I can briefly summarize what I was able to derive from my look (calling it an analysis would probably border on the presumptuous) at IL soybean and corn yields as affected by climate:

    1. I can explain how Lobell could derive correlations between temperature (and precipitation amounts) and crop yields from a trend line.
    2. The relationship in IL for soybeans/corn and climate can be seen, over the past 40 to 50 years, as being one where yields tend to plateau at some maximum to near maximum amount when temperature and moisture conditions are “normal” and on a decreasing, more or less linear slope of decreasing yields (from their trend lines) temperatures increase and precipitation declines in drought or near drought conditions.
    3. During the drought/drought like conditions and particularly for droughts the temperature increases went hand and hand with the low precipitation levels making the separation of these factors difficult.
    4. The higher temperature and lower moisture conditions appear in approximately 20 to 25 % of the total years and the occurrences appear to arrive in a, more or less, random pattern, more like one would expect from a natural cycle than a gradual trend in temperature as suggested by Lobell’s paper.
    5. A major problem I have with the Lobell paper is the rather open, and easily construed as arbitrary, choice of seasonal time periods for making correlations of temperature and delta crop yields. I found no correlations when I extended the soybean and corn seasons from July/August to April through August for IL soybeans and corn delta yields.

    I have initiated a look next at yield trend lines and deviations from the trend lines for the world and highest producing nations for some of the major cereal crops. I merely want and have wanted to look at these data from a different viewpoint than that expressed in the Lobell paper and as noted previously it would be presumptuous on my part to consider doing a serious analysis. The plant physiology involved here and published papers that have studied crop yields and natural climate cycles would be of interest to me also.

  259. Gerhard H. Wrodnigg
    Posted Mar 27, 2007 at 12:03 PM | Permalink

    Just a question:

    Is anyone here participating the

    European Geosciences Union
    General Assembly 2007
    Vienna, Austria, 15 ‘€” 20 April 2007

    If so, please contact me at climate.at.dagegen.org and I will organise a Climate Audit Vienna Event…

  260. David Smith
    Posted Mar 27, 2007 at 6:39 PM | Permalink

    I have an Excel chart I’d like to show here. But, I have no clue as to how to go from an Excel chart on my home computer to a post at CA. Can anyone offer guidance, in terms that even a dunderhead can understand?


  261. Hans
    Posted Mar 27, 2007 at 7:17 PM | Permalink

    TWO ITEMS, one is a reaction to #258 from MarkR, and the other is a short curious (if spurious) weather item from THE AUSTRALIAN, regarding Perth, a city in Western Australia that usually has very warm weather, even when summer is drawing to a close Down Under. The site referred to below is also a good source of up-to-date information (via RADAR precipitation mapping) on how the long drought in Australia appears to be finally breaking (let’s hope, for the sake of our brothers and sisters, and for the sake of the flora and fauna, in Australia!).

    I actually recognize MarkR’s sentiments as, at their core, being in tune with those of many of the writers on this site, foremost with Steve McIntyre’s own thinking. Steve can better speak for himself at http://www.climatesearch.com/newsDetail.cfm?nwsId=131

    Just in case MarkR has not been on this site long enough to recognize the apparent motivation of most of the contributors who visit, I have to defend them somewhat by saying that I have never found them to be primarily driven by anything resembling an “attempt to win the WAR for rational beliefs” let alone engaging “in a fight to the DEATH over any one particular battle with any one particular individual or any one particular belief.” I was not aware this was a war of beliefs, although sadly some may view it that way.

    What I have recognized here is the implicit and sometimes explicit hope and expectation of the various contributors at Climate Audit that the Convinced Climate Change Community [CCCC] exercise a little more care with respect to the CCCC’s own research methodology. To establish standards, to follow them, to work to proper peer review, to leave an audit trail and provide decent access to data archives — these would seem to be minimally acceptable standards. That is what Steve meant when he wrote “For someone used to processes where prospectuses require qualifying reports from independent geologists, the lack of independence is simply breathtaking and a recipe for problems, regardless of the reasons initially prompting this strange arrangement”.

    So I would challenge MarkR to clarify why he thinks this is primarily a site for “skeptics” while I see it primarily as a site for promoting sensible and clear thinking about the scientific methodology that must be employed for this very important endeavour. As to sarcasm, frequent visitors will acknowledge that Steve’s tolerance for that is nonexistent.

    No one here is suggesting that we ignore the possibility of AGW come what may, by sticking to our “bad beliefs” (some of which are indeed evolutionarily resistant to change). If the studies are properly done and begin to suggest that AGW is of valid concern, most contributors here would be on the forefront of the movement to take and demand decisive action. No one here wants to boil along with the proverbial frog in the barrel of water which is slowly brought up to the boiling point. That kind of “bad belief” is of little advantage to any species’ chance of long-term survival.

    “Challenging beliefs must always be done with care and compassion.” No argument there. The dismaying thing is that we have to work on belief systems at all. When what should by its nature be a scientific pursuit becomes a religion, we unfortunately all have to necessarily tread carefully.

    Cold mornings in Perth will continue
    By Rebecca Lodin
    March 27, 2007 18:06


    On Tuesday morning Perth had a minimum temperature of 6.3 degrees. This was their coldest morning this early in the year since records began 110 years ago. These cold temperatures are expected to continue on Wednesday, and may get even cooler.

    This shivery morning follows a cold front that recently passed over the southwest, bringing cooler, drier air to southwestern WA. Calm conditions with almost nonexistent winds also helped keep the temperatures down.

    Temperatures are expected to be even cooler tomorrow, as a ridge of high pressure strengthens over the southwest and dew points lower further. Although minimum temperatures will stay relatively low over the couple of days, tomorrow is expected to be the coldest.

  262. Earle Williams
    Posted Mar 27, 2007 at 7:34 PM | Permalink

    Re #261


    I have posted a few images in the past by doing the following:

    1) have my desired Excel chart visible as the top window

    2) use the Alt-PintScreen keystroke combo to capture the image

    3) paste the image into a handy image editing program. IrfanView is my preferred over the Windows defaults

    4) crop the image to the smallest necessary. resize it to 800×600 or smaller if needed

    5) save the image as PNG (not JPG as that introduces image loss that is quite noticable in charts)

    6) upload the image to any handy website. I have my own but I understand there are free ones like bitbucket. Note the final URL for the image.

    7) use that URL in your post using the [Img] button

    Hope that helps…

  263. Steve Sadlov
    Posted Mar 27, 2007 at 7:50 PM | Permalink

    RE: #262 – Similar to Fall 2005 in California. Many ski areas opened on October 20th that year.

  264. David Smith
    Posted Mar 27, 2007 at 9:27 PM | Permalink

    A question:

    I’ve been playing with some sea surface temperature time series and other climate-related data. One thing I’ve been doing is trying different smoothing methods, in particular applying double-smoothing (1-2-3-2-1). What surprised me is that some of the double-smoothed time series have distinctive 10 to 11 year oscillations.

    I asked someone about this and received this comment:

    I think you are seeing is mostly the
    result of your smoothing (filtering). A smoothed series will have a
    frequency response that corresponds with the smoothing window. Thus
    it is not surprising to see a 10 yr oscillation in a series that
    has a 5 yr smoothing window.

    My question: is the final sentence in the quote correct in those cases where there is no actual 10 year signal in the data? I realize that this has likely been answered numerous times here, but until I play with the data, the lesson doesn’t soak in.


  265. S. Hales
    Posted Mar 27, 2007 at 10:07 PM | Permalink


    The site you linked to has the following item about climate change. I guess you could call it skeptics about skeptics. Or True Skeptics and False Skeptics. The main thrust of the piece is how to better communicate urgency and breakthrough the citizen’s information overload or exhaustion. It uses a term dear to my heart “diffusion of inovations” or in this instance diffusion of ideas. The framing of the climate problem in these terms means that there are early adopters of ideas and if they are opinion leaders then there will eventually be a surge in adoption of the idea following the standard logistics curve. It is interesting that the author cites the bifurcation of opinion along party lines in the US with the democrats at near saturation of adoption and the republicans somewhere lower with no true opinion leaders in their ranks. The author frames this bifurcation in terms of level of exposure to “nuanced” treatments of the subject such as might be found in the New York Times; here is a recent nuanced treatment in the Times. Here is a quote from that article

    “A lot of people think climate change and the ecological repercussions are 50 years away,” said Thomas W. Swetnam, director of the Laboratory of Tree-Ring Research at the University of Arizona in Tucson. “But it’s happening now in the West. The data is telling us that we are in the middle of one of the first big indicators of climate change impacts in the continental United States.”

    Another quote to show another “nuance”

    The fear is that what happened to Summerhaven is a taste of things to come. As heat-stressed ecosystems provide fuel at the edges of new homes, catastrophic fires could become the new normal. Dr. Swetnam compares it to new developments in hurricane-prone areas in the Southeast.

    Since readers of this blog know that GW or AGW in particular plays little or no role in frequency or intensity of Atlantic hurricanes or TS this particular “nuance” seems to be a belief without strong scientific support (enter Chris Landsea). (As an aside it looks like western fire studies are the new hurricane attribution studies)

    For the patient reader a visit to the author’s website might prove fruitful in examing how a sceptic frames skeptics. The lead today is the Oxford Group study on the unsustainability of nuclear power. The powerful meme in this analysis is the “energy cliff” which is simply another “net energy” meme that dismisses economic efficiency and only focuses on engineering efficiency. The Oxford study fails to calculate time to depletion if reprocessing is used nor do they look at ultimate recoverable resources nor do they estimate future technological innovations. Again the nuance here seems to be if it supports your current beliefs and you are a noted sceptic with psicop sceptic credentials then it must be true.

  266. Steve Reynolds
    Posted Mar 27, 2007 at 10:27 PM | Permalink

    A Sir Stern howler:

  267. Posted Mar 28, 2007 at 3:09 AM | Permalink

    realclimate.org has censored again, when they get uncomfortable with the issues that have been raised. They apparently censor so that they can let their responses go unchallenged and appear to be accepted as conclusive. I’ve tried to contact Ray Pierrehumbert directly, twice but have received no response. The last post/exchange was #102 at:


    Here is my email to Dr. Pierrehumbert, which incorporates, my second attempt to post my response to the thread. Let noone mistake realclimate.org for an open scientific discussion.
    ====== begin my email =====
    Dr. Pierrehumbert,

    A couple of my posts in dialogue with you have not appeared on RC “Climate Reporting in Physics World”. Given the collective management of RC, and the coincidence of their failure to appear with Gavin’s arrival on this thread, I am concerned that you may not have had a chance to see them before a decision was made to avoid them.

    I am particularly intrigued by idea that the dry air deficit that you found may be a partial explanation for the endemic bias that Roesch found.

    I append only the second post after my signature below, because it incorporates the first.

    — regards,

    [my personal sig with contact info was here, this is the only deletion]
    ======== the second missing post =======
    Raypierre, in your 2005 paper “On the Relative Humidity of the Earth’s Atmosphere” pg 39, you describe a mechanism by which low resolution GCM studies such as the ERA40 analysis you used, may be deficient in dry air due to excessive mixing or leaking of moisture from saturated air. You speculate that this may be endemic to “low resolution” GCMs. It seems to me that this could partially explain the excessive snow cover and precipitation Roesch found in the AR4 models, as that moisture might have to travel farther and to colder regions to precipitate out. The extra moisture in the atmosphere would also enhance the H20 water vapor positive feedback, providing the extra warming to balance the positive albedo bias against solar. Of course, this wouldn’t be the complete explanation of the Roesch results, because he also found positive albedo biases in the tropical deserts, and positive albedo biases in the parameterizations of snow cover in forested regions. Do you have more recent information shedding more light on the extent of this issue, and thoughts about whether it might be responsible or have contributed to some of the positive surface albedo bias that Roesch found?

    Also, here I resubmit an earlier response that was apparently lost:

    Re: Raypierre’s response to #102

    The models matched the 20th century data by incorporating a major stabilizing influence in which albedo increases with temperature. Roesch’s analysis showed that by failing to reproduce the negative Snow Cover Area trend of the climate in the recent decades, the models had a positive surface albedo bias. The temperature was increasing over this period. To the extent, that the models achieve that temperature increase by replacing that reflected solar energy with increased sensitivity to GHGs, like thoroughbred horses, they are ready to burst out of the gate once no longer constrained to fit actual climate data.

    The IPCC SPD does not provide additional error bars for the projections of models, based on the errors documented in the various diagnostic studies. The IPCC assumes that the range of model sensitivities in the AR4 meta-ensemble brackets the actual climate sensitivity and represents the error in the models. A bias against solar in all of the AR4 models, shows that their whole estimate may be biased towards higher CO2 sensitivities. The SPD does not quantitatively compare the models errors against the energy imbalance quantity they must be accurate enough to represent, attribute and project.

    Correcting the models’ shortcomings, would likely result in the matching of the recent warming with less sensitivity to GHGs, and thus would result in lower future projections. At some point the temperature dependent stablizing component (positive albedo bias due to increased snow cover area and precipitation), that is in the models but not in the climate, will be overcome by further temperature increases, and the models will take the temperature excursions they project in the latter half of this century.

    Models are at the center of climate science, testing the various hypotheses, and providing insights when they do reproduce climate behaviors. While I am among those that argue that the models have little or no projective and attributive skill based on the current evidence, I am hopeful that such skill may be possible in the future.

  268. Steve Sadlov
    Posted Mar 28, 2007 at 10:08 AM | Permalink

    RE: #268 – Have those posts shown up yet or have they truly been censored? If so, then that shows that the RC managers are far uglier behaviorally than even I had previously imagined. If so, it is borderline evil.

  269. jae
    Posted Mar 28, 2007 at 10:37 AM | Permalink

    OK. C02 problem solved!

  270. Mark T.
    Posted Mar 28, 2007 at 10:52 AM | Permalink

    Subscription required, jae. Perhaps a summary from you?


  271. James Erlandson
    Posted Mar 28, 2007 at 11:27 AM | Permalink

    Re 270, 271:
    Summary here.
    CLIMATE: DOE finds sequestration capacity for 900 years of CO2 emissions

  272. Posted Mar 28, 2007 at 11:58 AM | Permalink

    Steve, Re: #268,

    The postings still haven’t appeared, and I emailed Dr. Pierrehumbert about them twice on March 1, almost 4 weeks ago. I emailed twice because I forgot to put a subject on the first email, so I thought it might have gotten caught in a spam filter.

    I don’t think we are dealing with people interested in getting the science right, but rather an application of Chomsky/”progressive” style mass action democracy (the “consensus” of a vocal insistant minority) for purposes peripheral to the science.

    I doubt you will find a scientist in the field who believes that the models are accurate to the 0.1W/m^2 globally and annually averaged, necessary to apportion among the competing hypotheses the approximately 0.8W/m^2 of net warming. Yet, the models are central not just to attribution and projection, but to gaining insight from the paleo data. In addition to the bias against solar found in the surface albedo of ALL the AR4 models by Roesch, there is probably still 40+W/m^2 of error in the models individually, if one goes through all the diagnostic information. But the modelers act as if these “larger” concerns allows them be dismissive of the 2.7 to 3.8W/m^2 of correlated error found by Roesch, and yet somehow to consider their models validated and supportive of the AGW hypothesis.

  273. jae
    Posted Mar 28, 2007 at 2:54 PM | Permalink

    273: Great. We need a modeling expert around here to audit these monsters. Maybe you can write up your thoughts and submit them to Steve M. for a special thread?

  274. David Smith
    Posted Mar 28, 2007 at 3:19 PM | Permalink

    Update: The Southern Hemisphere is approaching the end of its tropical cyclone season, with typically 85% of the season’s storms having occurred by now. It has been an uneventful year, with storm count and intensities slightly below average.
    The Southern hemisphere accounts for about one-third of the globe’s storms, which is two to three times more than the better-publicized Atlantic.

    The Western Pacific season will start soon. Sea surface temperatures are near-normal. The tropical Atlantic is slightly above normal but is expected to return to normal temperatures by the late summer.

    Apocalypse postponed, I guess.

  275. Steve Sadlov
    Posted Mar 28, 2007 at 3:31 PM | Permalink

    Shameless self promotion ….. my personal contribution to climate science:


    One of the subsystems in one of the mass storage systems described here was developed by a team I was part of about 20 years ago. 🙂

  276. Steve Sadlov
    Posted Mar 28, 2007 at 6:26 PM | Permalink

    FYI, especially John A and David Smith:


    The Western US has had a cold winter overall. In the Southwest, it’s been a deadly combo of cold and dry. Hopefully not a portent of emerging global conditions.

  277. David Smith
    Posted Mar 28, 2007 at 8:54 PM | Permalink

    WCR reviews a paper on western US drought and wildfires here . The usual suspects (ENSO, PDO, AMO) are examined with the warm-phase AMO getting a surprising amount of credit for western wildfires and drought.

    Looks like California has only a couple of chances for rain left before the Pacific shuts down for the dry season.

  278. Ken Fritsch
    Posted Mar 28, 2007 at 9:24 PM | Permalink

    I attempted to do directly what I assumed Lobell wanted to accomplish with his study of changing global temperature effects on world crop yields of the 6 leading cereal grains. I plotted the world crop yields for corn, rice, wheat, soybeans, barley and sorghum as time series for the period 1961-2004 using the Lobell referenced database — as noted previously from FAOSTAT. I established a trend line and then for each year derived a delta % yield by subtracting the actual yield from the trend line and dividing by the actual yield.

    All of the crops exhibited very high correlations with crop yields growing at an amazingly steady rate with time. The exception was sorghum which grew at a similar steady rate as the other crops up to 1981 when its yield actually decreased and the correlation with time did also. Would anyone have information on this sudden change in the trend line for sorghum? I noted that its crop acreage in the US was reduced by about ⻠from its peak in the late 1960s and early 1970s. I did not use sorghum in the remainder of my analysis.

    I also plotted the trend line for the global temperature anomaly and from that trend subtracted out the annual deviations for the period 1961-2004. I then calculated the Spearman R^2 (listed under the column heading R^2(2) in the table below) after comparing the rankings of delta crop yields and the delta temperature anomaly. All the results are listed in the table below.

    The consistency of the rate of growth of these crops with time as shown in the second column of the table under R^2(1) was a major surprise to me. The correlations of the crop yields with global temperature changes was not any where near as impressive as those derived in the Lobell paper, but it was in the same direction, i.e. yield increases with decreasing temperatures. That temperature change, however, explains little of the change in delta crop yields and might be explained by higher temperature drought effects as was noted with the analysis of IL yields as they related to temperature and moisture.

    I plan next to do multiple state climate and crop yield comparisons in order to better understand how and when temperature is affecting yields. I found a great site here that practically hands this information to me with little manipulation needed on my part.


    Item R^2(1) R^2(2)

    Corn 0.95 0.076
    Rice 0.98 0.017
    Wheat 0.98 0.086
    SoyB 0.95 0.041
    Barley 0.90 0.136
    Sorghum1 0.92 NA
    Sorghum2 0.18 NA

    Global Anom 0.69 NA

    Sorghum1 was 1961-1981 and Sorghum2 was 1982-2004

  279. Armin
    Posted Mar 29, 2007 at 2:49 AM | Permalink

    Something off-topic, but did you change today/yesterday something in the layout of this webpage? Suddenly the print-preview only shows empty pages, the type-preview is gone and the right bar is displayed partly over the right part of the article text?

    Or is my system the cause? 🙂

  280. David Archibald
    Posted Mar 29, 2007 at 4:36 AM | Permalink

    Re 279, may I suggest that you do a correlation of crop yield with CO2 level.

  281. MarkW
    Posted Mar 29, 2007 at 4:58 AM | Permalink

    There may be other motives that drive some of these people:

    Maurice Strong, senior advisor to Kofi Annan. Chaired the “UN Conference on Environment and Development”, 1992.
    Reponsible for putting together the Kyoto Protocol.

    “We may get to the point where the only way of saving the world will be for industrial civilization to collapse.”

    “quest for poverty … reduced resource consumption … and set levels of mortality control.”


    Timothy Wirth, US Undersecretary of State for Global Issues
    “We have got to ride the global warming issue. Even if the theory of global warming is wrong, we will be doing the right thing in terms of economic policy and environmental policy.”

    richard Benedick, deputy assistant secretary of state who headed policy divisions of the US State Department.
    “A global warming treaty must be implemented even if there is no scientific evidence to back the greenhouse effect.”

  282. DeWitt Payne
    Posted Mar 29, 2007 at 6:19 AM | Permalink

    If you thought that Jim Hansen was a little over-the-top before, take a look at this post on Prometheus.

  283. Ken Fritsch
    Posted Mar 29, 2007 at 10:02 AM | Permalink

    Re: #283

    The political part of the Jim Hansen split personality seems to be rapidly taking over the scientist part. I guess a skeptic in these matters might want to concede some merit to his argument from a contrarians’ point of view, but since he mostly pushes the consensus advantage when it benefits his advocacy stands, it is difficult to understand how he can plead a special contrarian case for the times (few) that he is in disagreement with the consensus (IPCC in this special case). The approach here is much more likely to be used by a politician than a scientist.

  284. Ken Fritsch
    Posted Mar 29, 2007 at 10:57 AM | Permalink

    Re: #281

    Obviously, if I merely plotted global CO2 levels versus crop yields the correlations would be very good, since both have been steadily rising. I could look at the deviations from the trend lines for crop yields and CO2 levels, however, I am not sure how local levels of CO2 vary.

    I did want to show the crop yield trend lines plus the global temperature anomaly for the 1961 ‘€” 2004 time period and will attempt to post that below.

  285. Posted Mar 29, 2007 at 2:01 PM | Permalink

    Have you seen the revolution in borehole reconstructions that curiously happened exactly when MBH98 was published?


  286. jae
    Posted Mar 29, 2007 at 4:16 PM | Permalink

    Lubos: What a coincidence!

  287. Joe Cool
    Posted Mar 29, 2007 at 10:02 PM | Permalink

    Greenhouse Gas Effect Consistent Over 420 Million Years
    (emphasis mine)

    This study confirms that in the Earth’s past 420 million years, each doubling of atmospheric CO2 translates to an average global temperature increase of about 3° Celsius, or 5° Fahrenheit

    This study used 500 data points in the geological records as “proxy data” and evaluated them in the context of the CO2 cycling models of co-author Robert Berner, professor emeritus of geology and geophysics at Yale who pioneered models of the balance of CO2 in the Earth and Earth’s atmosphere.

    So, 500 data points covering 450 million years confirms this fact. Guess you guys can all go home now.

    “Our results are consistent with estimates from shorter-term records, and indicate that climate sensitivity was almost certainly greater than 1.5, but less than 5.5 degrees Celsius over this period,” said Park. “At those extremes of CO2 sensitivity, [1.5°C or 5.5°C] the carbon-cycle would have been in a ‘perfect storm’ condition.”

    So the climate sensitivity is 0°C to 5.5°C?? Not sure this surprises me with them only using 500 points over 450 million years.

    Sorry, I’m no scientist/statistician/expert. Am I wrong for feeling they way I do when I read this “article”?

  288. Mark T.
    Posted Mar 30, 2007 at 12:35 AM | Permalink

    That’s about 1 point per million years? So any oscillation that operates with a period less than about 2 million years is unrecoverable. Hehe, there’s some “proof” for ya…

    No, you’re not wrong. Nyquist just rolled over in his grave. He was probably discussing such an obvious miss with Shannon.


  289. Mark T.
    Posted Mar 30, 2007 at 2:27 AM | Permalink

    From JMS on the protected thread…

    commentors on this blog seem to have a suspicion of deep knowledge of a subject. That does not make for good science,

    Unbelievable such arrogance is. Unbelievable. Tell that to Gavin.


  290. David Archibald
    Posted Mar 30, 2007 at 2:32 AM | Permalink

    Attention sun worshippers! Warwick Hughes has kindly posted on his blog a presention on the solar influence on climate: http://www.warwickhughes.com/blog/?p=102

    It is based on a paper in press at Energy & Environment. Imminent cooling is predicted. A further update is provided by a solar physicist who writes:

    Did you know that the number of spotless days so far during this SC-transit is much less than during previous cycles? In fact, its evolution mimicks much better the weak SC10-15 than that of the more active SC16-23 (http://members.chello.be/j.janssens/SC23web/SCweb3.pdf slide 5). SC10-15 reached their minima 66 months after the first spotless day. Because for this SC23-24-transit, the first spotless day occured in January 2004, this would mean a minimum mid-2009! Predicting such a late minimum borders to solar cycle heresy (;-)) , but as you pointed out in your presentation, SC23 would then be 13 years long, just like SC4 prior to the Dalton minimum.
    Moreover, Schatten’s poloidal field measurements continue to indicate a weak SC24. My polar faculae observations show -so far- also about 40% less PF than observed 10 years ago, and they are less bright. It’s good this overall trend continues to correspond.

  291. gb
    Posted Mar 30, 2007 at 5:25 AM | Permalink

    Re # 285. Interesting. Perhaps you should correlate world’s population or obesity with CO2 levels or temperature. I guess the correlation is quite strong.

  292. Hans Erren
    Posted Mar 30, 2007 at 6:48 AM | Permalink

    re 288:
    I think this is an attempt to refute Nir Shaviv’s value of 1.3 K/2xCO2

  293. David Smith
    Posted Mar 30, 2007 at 8:48 AM | Permalink

    RE #291 David A, I’m unable to download your paper from Hughes’ website for some unknown reason. If it’s not a problem, could you e-mail it to me at mndsmith33 at earthlink.net ? I’m quite interested in learning more about the topic.


  294. Gerald Machnee
    Posted Mar 30, 2007 at 9:11 AM | Permalink

    In today’s paper – “COPENHAGEN — A wristwatch buried in the ice at the North Pole three years ago was found by a boy some 2,900 km away after it floated ashore on the Faeroe Islands, halfway between Scotland and Iceland.
    The working watch, discovered by Niels Jakup Mortensen, 11, was accompanied by a letter from its burier, Joergen Amundsen, a descendant of Norwegian explorer Roald Amundsen.”

  295. Ken Fritsch
    Posted Mar 30, 2007 at 9:39 AM | Permalink

    Re: #292

    My guess is that you do not realize what kind of analyses I have been doing. I have been comparing the deviations from a trend line not trend lines and then only for variables that are known to be related like crop yields and climate.

  296. gb
    Posted Mar 30, 2007 at 9:53 AM | Permalink

    Re 296. In these 40 years farmers have become much more professional and got better tools. The productive increase owing to these factors are much more important than climate change. Trying to relate climate to crop yields will therefore be very difficult.

  297. Steve Sadlov
    Posted Mar 30, 2007 at 10:10 AM | Permalink

    RE: #289 – got aliasing? LOL …. Shannon is also rolling in his grave ….

  298. Steve Sadlov
    Posted Mar 30, 2007 at 10:15 AM | Permalink

    RE: #291 – “Imminent cooling is predicted.”

    Trying to avoid coming off like the boy who cried “wolf” – buuuuut …. I really do not have confidence that either most of the masses or most of the leaders really have much of a clue regarding just how bad even a modest cooling (let alone a Maunder-ish one) would be for a 6 billion plus populated world. Just the famine factor alone, setting aside for a moment the geopolitical knock on effects, would be a harsh dose. Add to that, competing demand for ethanol base crops, the current bee crisis, etc. Well, no matter what the masses do, I urge due diligence on the parts of individuals to become more self sufficient food wise. Both stocking as well as gardening wise.

  299. Steve Sadlov
    Posted Mar 30, 2007 at 10:19 AM | Permalink

    RE: #295 – this is a great teaching moment regarding the behavior of Arctic sea ice. Steve M had mentioned earlier how Arctic sea ice tends to be “manufactured” in certain locations in the far Eastern part of the basin (a bit south of the pole, North of Asia and Alaska), then slowly drifts toward the West, ending up mostly in the North Atlantic.

  300. MarkW
    Posted Mar 30, 2007 at 11:26 AM | Permalink


    Seeds have also been improving at an accelerating rate.

  301. MarkR
    Posted Mar 30, 2007 at 11:39 AM | Permalink

    #261 David Smith

    Try uploading your info to http://www.esnips.com it’s free. I think if you view your spreadsheet through internet explorer, you will be able to cut and save it for public viewing via a link.

  302. Steve Sadlov
    Posted Mar 30, 2007 at 11:48 AM | Permalink

    RE: #300 – something I was not clear on, the ice tends to make at least one circumpolar rotation prior to its exit and melting into the Atlantic. In any case, there is no such thing as truly “old ice” – The lion’s share of “old ice” is in fact only 2 to, at most, 10 years old.

  303. David Smith
    Posted Mar 30, 2007 at 12:42 PM | Permalink

    Re #302, Thanks, Mark. I’ll try that tonight.

  304. Steve Sadlov
    Posted Mar 30, 2007 at 1:27 PM | Permalink

    All stereotyping and “big oil shill” allusions aside, “W” alluded to some pending Executive Orders regarding various AGW hysteria driven reputed “countermeasures” (read arbitrary reductions in emissions, managed energy scarcity, dirigiste involvement in ethanol, etc) in some speeches he gave a while back. For US residents reading this, your tax dollars at work:


  305. James Erlandson
    Posted Mar 30, 2007 at 1:59 PM | Permalink

    And now for something completely different.
    From NewScientist.com

    Could dandruff be altering the world’s climate? Along with fur, algae, pollen, fungi, bacteria, viruses and various other “bio-aerosols” wafting around in the atmosphere, it may well be.

    Original report in Science Magazine

    Suspended atmospheric particles play a crucial role in any global climate scenario: They can both enforce and suppress radiative forcing. In developing climate modeling further, a deeper understanding of atmospheric aerosol is needed. Because of extreme local and temporal variations, proper incorporation of aerosols into models requires modeling of the aerosol itself. It turns out that cellular material and proteins compose up to 25% of the atmospheric aerosol. Consequently, the source strength of the biogenic aerosol in general must be corrected and should be estimated on the order of other major aerosol sources.

  306. Steve Sadlov
    Posted Mar 30, 2007 at 2:02 PM | Permalink

    FYI, a symposium this summer which may be of interest to some here:


    Disclosure – the organizer was a past academic advisor who helped keep me out of trouble … LOL!

  307. Ken Fritsch
    Posted Mar 30, 2007 at 2:22 PM | Permalink

    Re: #297

    My references where regarding the recently published Lobell and Field paper here.

    Did you read the paper? I think you might be getting into the tail end of the discussion.

    Click to access erl7_1_014002.pdf

  308. Ken Fritsch
    Posted Mar 30, 2007 at 3:56 PM | Permalink

    Re: #308

    That link did not work, so here’s one that should and I should have said: My references were regarding the recently published Lobell and Field paper here.

    Click to access erl7_1_014002.pdf

  309. DeWitt Payne
    Posted Mar 30, 2007 at 6:26 PM | Permalink

    Re: #299

    Combine near term global cooling with Peak Oil and things get really ugly really fast. Here’s hoping the cornucopists and warmers aren’t completely wrong.

    And on a more humorous (I hope) note, try No Impact Man .

  310. David Archibald
    Posted Mar 30, 2007 at 9:14 PM | Permalink

    For hurricane people: http://www.goingtomeet.com/conventions/details/5098

    1st International Summit on Hurricanes and Climate Change
    May 27, 2007 – Jun 01, 2007
    Aldemar Knossos Royal Village Conference Center Crete, Greece

    “The goal of this event is to frame the debate and address what research is needed to advance the science of hurricane climate. The purpose is to provide a venue for encouraging a lively, spirited exchange of ideas.”

    Speakers: Johnny C. L. Chan, Judith Curry, Jeffrey Donnelly, James B. Elsner, Kerry Emanuel, Greg Holland, and more

    It won’t be a “lively, spirited exchange of ideas” unless some Climate Audit people come along. As it is being held in Greece, I envision a rerun of “300”, except with more Powerpoint presentations.

    There has been enough original work on hurricanes published on Climate Audit for a whole session.

  311. Sprentuk
    Posted Mar 30, 2007 at 9:20 PM | Permalink


    “I should add that I am sure Phil Jones would make data available to a reputable scientists if asked. Data from NERC funded projects is archived at the British Atmopsheric Data Centre (BADC)”

    About ten posts down here – http://www.ukweatherworld.co.uk/forum/forums/thread-view.asp?tid=10946&posts=18&start=1

  312. David Smith
    Posted Mar 31, 2007 at 7:57 AM | Permalink

    Odds and ends:

    The current Equatorial Pacific sea temperatures, located here , shows an embryonic La Nina (the anomalously cool water on the right side, near South America). (Double-click on the colored image to see a closeup.)

    More interesting on that link is the “assorted plots” button. Click on that to see the cross-section of the equatorial Pacific, which shows the temperatures below the surface. The cool water on the right side of the colored image will support a La Nina. More interesting to me is that the lower waters, below 200 meters, continue to cool. That has longer-term implications.

    On a different note, the first tropical cyclone of the Western Pacific seems to be organizing today (31 March). This is a normal start to the 2007 season. A neat satellite animation of the swirling clouds (colorized) is given here . The yellows and oranges represent a huge amount of water vapor and heat being delivered to the upper troposphere.

    (Note that both links show the most-recent data and so they change over time.)

  313. David Smith
    Posted Mar 31, 2007 at 3:27 PM | Permalink


    (I’m trying to figure out how to post a spreadsheet chart, but when you’re a dunderhead, it’s not easy.)

    If anyone can successfully open this Excel spreadsheet, or not, please let me know. It may be password-protected. Thanks.

  314. jisc
    Posted Mar 31, 2007 at 3:43 PM | Permalink

    David….. the link took me to a “broken link” message on the eSnips website.

  315. Joel McDade
    Posted Mar 31, 2007 at 4:36 PM | Permalink

    This is for R newbies like me that are following this site hoping to learn something. Here is a reproduction of the distribution curve prepared by Paul Lindsay in the first of his hurricane threads:

    hur 1944)

    This is the result:

    I know the thread is months old, but I finally got off my lazy butt :) The rest of you may laugh, but simple examples for newbies are hard to come by.

    Paul's post is here (I'm having trouble linking):


  316. Joel McDade
    Posted Mar 31, 2007 at 4:55 PM | Permalink

    Crap, code was posted broken in the above. I think a url in the code screwed it up.

    Anyone interested can get it as a .txt file here:


  317. Mark T.
    Posted Mar 31, 2007 at 5:02 PM | Permalink

    From the data access thread, a reply to Armand:

    So what exactly are you suggesting *does* violate causality?

    CO2 lags temperature rise in the historic record (albeit coarse for temp). The delta is around 400-800 years or so.

    I agree a simplistic approach to all this is not necessarily valid, but we are force fed the concept of CO2 being the majority forcer in temperature rise by climatologists. If this is true, then by extension CO2 must lead temperature rise. It does not, therefore it cannot be the sole forcer, nor even the majority. There’s nothing wrong with assuming there is an impact, but that impact will show up as a temperature perturbation that lags CO2 rise.


  318. John Lang
    Posted Mar 31, 2007 at 6:35 PM | Permalink

    Speaking of the lag of CO2 versus temperature in the ice core records, a recent study by Robert Berner looks at the CO2 / temperature link over the past 420 million years. Berner is THE recognized authority on paleoclimate.


    Abstract Quote – “A firm understanding of the relationship between atmospheric carbon dioxide concentration and temperature is critical for interpreting past climate change and for predicting future climate change1. A recent synthesis2 suggests that the increase in global-mean surface temperature in response to a doubling of the atmospheric carbon dioxide concentration, termed ‘climate sensitivity’, is between 1.5 and 6.2 °C (5’€”95 per cent likelihood range), but some evidence is inconsistent with this range …”

    While the paper requires a Nature subscription, here is a link to Berner’s estimate of historical temperatures going back 540 million years.

    Here is link to Berner’s reconstructed CO2 record going back 540 million years. Berner’s estimate is the yellow orange line – the GEOCARB III line.

    Just picking the 420 million year timeline (consistent with the recent Nature article), we see:

    – CO2 – 4000 ppm (or about 4 doublings of CO2 from the 275 ppm that existed before humans started increasing it.)

    – temperature – 4C warmer than today.

    Thus, the historical sensitivity of temperature to CO2 is only 1C per doubling of CO2 (if it is not in fact the reverse relationship.)

    Thus, the estimates of 1.5C to 6.2C sensivity ranges would have produced temperatures 420 million years ago of 6C to 25C warmer than today.

    Based on everything we have seen on this website with the exageration, data selection and changing temperature records by GISS and Hadley Centre, I think we can conclude the 1C for every doubling of CO2 more accurately reflects reality.

  319. Armand MacMurray
    Posted Mar 31, 2007 at 6:44 PM | Permalink

    OK, I guess I was being too literal in my interpretation of your comments. My impression of the “dogma” is different than yours. My understanding of the claim is that while something else (Milankovic cycles?) may have initiated a multi-thousand-year warming, resulting in an initial increase in CO2 levels over a few hundred years, CO2 is plausibly responsible for the bulk of the warming past that initial stage, in a positive feedback process (at least up to a point). No causality problem there, because the CO2 is leading to later warming, with the early warming caused by “something else”.

  320. David Smith
    Posted Mar 31, 2007 at 6:48 PM | Permalink

    Re #315 Thanks. I’ll try again.

  321. John Lang
    Posted Mar 31, 2007 at 7:01 PM | Permalink

    To #320 – CO2 increases from 180 ppm to 280 pmm from the height of the ice age to the height of the interglacial.

    Temperature increases by 5C globally (10C at the poles.)

    The CO2 can only be responsible for 1C to 2C of the total warming.

    The rest of the warming 60% to 80% of it is cyclical or natural. Al Gore didn’t tell you that did he. He just said it was “complicated.”

  322. David Smith
    Posted Mar 31, 2007 at 9:01 PM | Permalink

    bender, I hope Ohio State has some mercy on those poor SEC gators

  323. DeWitt Payne
    Posted Mar 31, 2007 at 9:47 PM | Permalink


    Let me run this hypothesis up another flagpole for comment. Greenhouse warming is a logarithmic function of concentration. Each doubling produces the same delta H and thus the same delta T. My hypothesis is that latent heat transfer from water evaporation and condensation is an exponential function of temperature. That implies that as the temperature increases, delta T from a given delta H, otherwise known as climate sensitivity, decreases, possibly rapidly. At the average surface temperature of 288 K, we are well into the exponential increase with latent heat loss of about 75 W/sq.m.

  324. Mark T.
    Posted Mar 31, 2007 at 10:42 PM | Permalink

    My understanding of the claim is that while something else (Milankovic cycles?) may have initiated a multi-thousand-year warming, resulting in an initial increase in CO2 levels over a few hundred years, CO2 is plausibly responsible for the bulk of the warming past that initial stage, in a positive feedback process (at least up to a point). No causality problem there, because the CO2 is leading to later warming, with the early warming caused by “something else”.

    Yes, that’s sort of the dogma. And I have no problem with that concept if true. However, CO2 is _still_ not leading. The temperatures started rising 400 years ago or so, well before CO2 saw any change. They, those with the causality problem, have not unravelled this yet. Folks like Gaviin think that simple feedback does it. Al Gore “its complicated.” Etc…


  325. bender
    Posted Mar 31, 2007 at 11:37 PM | Permalink

    Re #323
    LOL. Ain’t seen nuthin’ yet.

  326. Willis Eschenbach
    Posted Apr 1, 2007 at 4:09 AM | Permalink

    Dewitt, I agree wholeheartedly with what you say above. My own estimate is that at present, somewhere around two thirds to three quarters of any additional change in forcing goes to parasitic losses (sensible and latent heat plus hydrometeors), but it could be more.

    In addition, the increased evaporation likely increases cloud cover, further reducing the temperature change from a given forcing change. Note that this is a true feedback, unlike the parasitic losses

    My strong suspicion is that at all times, the earth’s climate system is operating at its maximum temperature given the controlling factors (layout of the continents, resultant winds and oceanic currents, period of rotation) and that small changes in such things as CO2 forcing will have very little effect on that temperature. The earth heats up until the parasitic losses and feedbacks do not allow it to heat any further.

    In my opinion, this is why the earths temperature has been remarkably stable (it has only changed ~ ⯳%) over billions of years, during which time the sun’s output has increased by ~ 30% …

    My best to everyone,


  327. jae
    Posted Apr 1, 2007 at 6:41 AM | Permalink

    324, 327, Yup, see my spreadsheet. One can make a case that there is no water vapor feedback that causes a temperature rise by comparing Denver to Indianapolis, IN. Same latitude, almost same July avg. temp, even less solar insolation in Indianapolis. Yet July absolute humidities are twice as high in Indianapolis and the climate sensitivities are the same. There is a water bapor feedback from a temperature rise, all right, but it is simply more water vapor (and muggy nights).

  328. jae
    Posted Apr 1, 2007 at 8:44 AM | Permalink

    Disregard 328; there is some poor logic there. Back to drawing board.

  329. paminator
    Posted Apr 1, 2007 at 8:47 AM | Permalink

    Jae, #328-

    I checked out the site you used for the solar insolation data. I have not found any error estimates yet, but I will keep looking. I still think your approach is valid, and I am willing to compile some more of this data for your spreadseet to see how it works out. I am also interested in how it compares with data from more northern climates such as northern Canada.

    I still have not been able to get your spreadsheet to download. Your link takes me to the following blank page: http://esnips.com/HomeAction.ns;jsessionid=AF7DE9EA9DF469684459B5158997D68E. Any help would be much appreciated.

  330. Ken Fritsch
    Posted Apr 1, 2007 at 10:40 AM | Permalink

    Re: #323

    bender, I hope Ohio State has some mercy on those poor SEC gators

    Deja vu all over again?

  331. bender
    Posted Apr 1, 2007 at 11:20 AM | Permalink

    #331 Mmmm hmmm.

  332. DeWitt Payne
    Posted Apr 1, 2007 at 12:28 PM | Permalink

    Willis, I did some calculations using the formula 4.28E-08*EXP(0.0739*T) where T is Kelvin to calculate latent heat loss. I then add the radiated heat using the Stefan-Boltzman equation to get total forcing at a given temperature. I’m treating sensible heat loss as constant because I have no data or theory to indicate otherwise. I can then calculate a sensitivity factor in degrees/watt. At 288 K, it’s 0.089 and 0.078 at 291 K. To get from 288 K now to 291 K, the IPCC midrange prediction for 2100, requires an increase of forcing at the surface of 35 W/sq.m. or nearly ten times the 3.7 W/sq.m. from CO2 doubling.

    jae, altitude should be considered when comparing Denver to Indianapolis.

  333. bernie
    Posted Apr 1, 2007 at 3:40 PM | Permalink

    Would anyone like to comment. UHI explains 50 year rise in California Temp

    Seems when researchers look for an UHI they find it big time!!

  334. MarkW
    Posted Apr 1, 2007 at 3:50 PM | Permalink


    What I find interesting about the idea that something else started the warming, but CO2 later took over as
    the major driver, is the fact that the rate of increase in temperature is remarkably smooth. Even during
    the period of the alleged transfer from something else to CO2. This strongly implies a remarkable piece of timing. IE, the something else, whatever it may be, manages to peter out at precisely the same time that CO2 is ready to take over.

    Do you propose some mechanism whereby this handoff from one forcing to another is handled with such amazing

  335. DeWitt Payne
    Posted Apr 1, 2007 at 4:19 PM | Permalink


    CO2 never becomes the major driver of temperature change during the switches between glacial and interglacial regimes. That’s a straw man. The contribution from CO2 and water vapor is at most 30% of the total forcing over the 10,000 or so years it takes to heat or cool. Also, the increase (decrease on cooling) in CO2 is not a step function. Which is why you don’t see an obvious change of slope in the temperature curve.

  336. DeWitt Payne
    Posted Apr 1, 2007 at 9:26 PM | Permalink

    Let me also add that the insolation change from changes in the eccentricity of the Earth’s orbit and changes in the tilt and precession of the Earth’s axis (Milankovitch cycles) isn’t a step change either. Therefore deconvolution of the various contributions isn’t trivial, if it’s possible at all, and the lack of such a deconvolution is therefore not sufficient to claim that CO2 has no effect. Especially when there are good physical reason, i.e. the CO2 IR absorption spectrum, to think it should have an effect.

  337. Willis Eschenbach
    Posted Apr 1, 2007 at 9:31 PM | Permalink

    #336 But if CO2 is adding 30% to the forcing, surely the slope of the warming should get steeper, no?


  338. Mark T.
    Posted Apr 2, 2007 at 12:29 AM | Permalink

    What I find interesting about the idea that something else started the warming, but CO2 later took over as
    the major driver, is the fact that the rate of increase in temperature is remarkably smooth. Even during
    the period of the alleged transfer from something else to CO2. This strongly implies a remarkable piece of timing. IE, the something else, whatever it may be, manages to peter out at precisely the same time that CO2 is ready to take over.

    Do you propose some mechanism whereby this handoff from one forcing to another is handled with such amazing

    336 But if CO2 is adding 30% to the forcing, surely the slope of the warming should get steeper, no?

    Couple interesting nuances here.

    First, if at any time a feedback mechanism becomes greater than unity, i.e. it takes over, the system will become unstable without something to balance it. For example, the difference equation y(n) = ay(n-1) + x(n) is only marginally stable for a = 1, stable for a lt 1 and unstable for a gt 1 (exponential growth). Apparently, the concept of “global dimming” was derived to counter this since models would blow up. I don’t guarantee validity of that claim, it’s just what I heard. If so, a perfect transition to CO2 = primary cause from CO2 = effect, even with “dimming,” seems unlikely to me with out some evidence of the transition point. Nature is not kind enough to behave in discrete packets such as yearly temperature averages. If this transition happened, somebody should be able to find it.


  339. Ron Cram
    Posted Apr 2, 2007 at 8:07 AM | Permalink

    I have written the first drafts of two articles for Wikipedia recently. The first on Scientific data archiving. The second on Scientific data withholding. The best version of the data withholding article (containing the discussion of Michael Mann and Steve McIntyre) can be found here.

    William Connelly has nominated the article on data withholding for very fast deletion. It appears to me that he is too close to the subject to be objective. I made an attempt to write the article NPOV, according to wikipedia policy and all comments are well sourced. Please take
    a look to see if all the pertinent facts are available to readers. If you think the article has value, please say so on the Talk page.

  340. Ron Cram
    Posted Apr 2, 2007 at 8:27 AM | Permalink

    If you would like to comment on the discussion of whether “Scientific data withholding” should be deleted, you can find
    the deletion page here.

  341. L Nettles
    Posted Apr 2, 2007 at 9:36 AM | Permalink

    US. Supreme Court rules in CO2 Case

    Mass. v. EPA

  342. Joe B
    Posted Apr 2, 2007 at 9:48 AM | Permalink

    The Supreme Court has ruled against the Bush administration

  343. DeWitt Payne
    Posted Apr 2, 2007 at 9:57 AM | Permalink

    #336 But if CO2 is adding 30% to the forcing, surely the slope of the warming should get steeper, no?

    Here’s the transition from the last glacial to the current (Holocene) interglacial based on Petit’s 1999 deuterium analysis of Vostok ice cores and the GT4 time scale:

    Hardly a smooth curve. I fail to see how this proves anything about CO2 one way or the other.

  344. DeWitt Payne
    Posted Apr 2, 2007 at 10:00 AM | Permalink

    Sorry, I made it too big. The y axis is temperature difference from the present and ranges from -10 to +2 degrees C.

  345. S. Hales
    Posted Apr 2, 2007 at 10:37 AM | Permalink

    Bernie, In your posst about UHI the article starts out ok but devolves into a non-sequitor from the subject of the piece. The first 8 or so paragraphs are a reasonable summary of the study’s findings but beginning with the last sentence of paragraph eight the article makes a grand leap away from UHI to AGW in general listing all the ills to befall Cal residents if emissions continue to rise. The author seems not to realize that UHI and land use changes contaminates the surface record in Cal and any projections of future climate there must account for UHI and land use changes. Another possibility is the author thinks that UHI is made worse by higher atmospheric concentrations of CO2 so the leap to AGW in the last half of the article. Bizzare!

    Here is the link to the JPL study (link tags don’t seem to be working)


  346. jae
    Posted Apr 2, 2007 at 10:41 AM | Permalink

    330: I can’t figure out why you have trouble getting the spreadsheet. That link works perfectly for me. I can email it to Steve and perhaps he can send it to you? I would like to work with you on this, as I think there may be some important “messages” in all that data. So far, I’m just not smart enough to figure out what they are…

    333: I know I have to consider altitude. That’s one of the logic problems I was referring to in 329.

  347. DeWitt Payne
    Posted Apr 2, 2007 at 10:46 AM | Permalink

    Should think more before posting. Here’s a properly sized chart with CO2 added. Maybe someone else has finer time resolution on the CO2, but this is what I found. CO2 and temperature are clearly correlated, but since we don’t really know the shape of the other forcings, I don’t think you can prove anything one way or another about CO2 from this data, so I’ll still go with the physics that says CO2 must add some fraction to the warming.

  348. Steve Sadlov
    Posted Apr 2, 2007 at 10:47 AM | Permalink

    RE: #334- I’ve personally witnessed a big chunk of time within those 50 years. Places that used to get hard frost which were undergoing rural to suburban transitions 30 years ago no longer do. Whereas, places in the boonies get the same frost / freezing they always did, man o menos the usual and customary innate variations.

  349. Nicholas
    Posted Apr 2, 2007 at 10:56 AM | Permalink

    L Nettles : Well, I guess it’s official. Every time I breathe out, I’m creating “pollution”. That, and the world has gone totally insane. Is it too late to become a hermit?

  350. jae
    Posted Apr 2, 2007 at 11:01 AM | Permalink

    350: I suspect that the majority of the Supreme Court Justices are viewing “pollutant” as any substance that causes harm to humans or the environment. Heat is also considered a pollutant in water quality analyses. Semantics!

  351. beng
    Posted Apr 2, 2007 at 11:14 AM | Permalink

    RE 338: Willis_E says:

    But if CO2 is adding 30% to the forcing, surely the slope of the warming should get steeper, no?

    I’d think so, if it were 30%. Another consideration is that CO2 levels lag the temp downturns by at least several hundred yrs at the beginning of the glacial period. One would think the rate of temp downturn would be attenuated (30%?) compared to the rate at the start of the interglacial (when CO2 was lower), all else considered equal (a big assumption). But eyeballing the Vostok ice-cores doesn’t show that — the absolute rates of change look very similar.

  352. S. Hales
    Posted Apr 2, 2007 at 11:33 AM | Permalink

    Just some musings on the recent SC ruling

    The decision is just a reversal of a Court of Appeals decision. The supremes did not reach a conclusion that CO2 is or is not a pollutant. Their decision is strictly confined to give standing to a litigant who has not suffered present harm but will possibly suffer future harm. The SC placed restrictions on what the EPA can argue in defense of its decision to not regulate. This sets the stage for a new trial or settlement if the EPA feels its hands are now tied. To regulate CO2 the EPA (AFIK) must reach an endangerment finding. Then it must launch study after study to determine technological fixes and to assess cost/benefit. If the EPA does reach an endangerment finding we are probably looking at 10+ years before there is some technological advance to limit CO2 emissions at the tailpipe. Bureaucracy, you gotta love it.

  353. MarkW
    Posted Apr 2, 2007 at 12:05 PM | Permalink

    In order for the EPA to have standing to regulate CO2, the court had to find that CO2 was pollutant. They reversed the lower court
    ruling that CO2 was not a pollutant.

    I have no doubt that the litigants will claim that all the relevant studies have already been done. Specifically IPCC’s I-IV.

    And since Al Gore et. al. have been telling us that we have only a few years left before catastrophic warming becomes irreversible,
    the litigants will argue that we don’t have time for the formal process. Especially if there is a Democrat in the White House after 2008, it’s likely
    that this ruling will result in the effective implementation of Kyoto in a few years.
    Without having to go through all the bother of actually passing legislation or trying to pass anything as pesky as a treaty.

  354. MarkW
    Posted Apr 2, 2007 at 12:06 PM | Permalink

    Given the level of junkscience in the courtrooms in recent years, I have no confidence in the courts being able to resist the
    temptation to micromanage the entire economy. For our benefit of course.

  355. Steve Sadlov
    Posted Apr 2, 2007 at 12:44 PM | Permalink

    RE: #313 – It’s really freaking me out. My cistern system is likely to move from concept to plan soon.

  356. Steve Sadlov
    Posted Apr 2, 2007 at 12:52 PM | Permalink

    RE: #354 – From a standpoint of environmental law, today, we have become Ecotopian. And the Executive Branch now sit at the head of the new hierarchy. I fail to understand why some are saying this is a “loss” for the Whitehouse. It is pure, executive power, newly incremented upward.

  357. Nicholas
    Posted Apr 2, 2007 at 1:06 PM | Permalink

    jae : ANY and ALL substances can and do harm humans. I mean, oxygen and water are perhaps the most vital substances to life on this planet (as well as CO2) and yet both of them can easily be harmful. You can die from drinking too much water in certain circumstances. Likewise, 100% oxygen can be very bad for you. If the definition of a pollutant is anything that can harm a human in sufficient quantity, then what ISN’T a pollutant? I just don’t see the point of having a word which is so non-specific…

    Practically every day I am walking along and some old diesel-engined truck rumbles past, belching thick black smoke, causing me to cough and sneeze and otherwise feel ill. THAT is a pollutant. THAT is what organizations like the EPA should be regulating. Clearly they aren’t doing a very good job of it. What’s the point of declaring CO2 a pollutant when there are bigger fish for them to fry? I’m far more likely to die from lung cancer contracted from inhalation of diesel particulates than I am AGW, thank you very much.

    End of rant.

  358. rhodeymark
    Posted Apr 2, 2007 at 1:26 PM | Permalink

    If the EPA is going to now be forced to respond to other CO2 lawsuits, aren’t all data based claims of present and future harm subject to full disclosure, FOIA etc.? Seems to me that anyone citing a study for legal redress purposes exposes that study to audit, no?

  359. jae
    Posted Apr 2, 2007 at 1:33 PM | Permalink

    Here’s the type of thing that intrigues me about the data in my spreadsheet:

    Compare Phoenix and Atlanta. Both are at about the same altitude and latitude, and therefore both receive the same TOA solar radiation. Phoenix gets 100 w/m2 more radiation at the surface in July than Atlanta does, and it averages 8 degrees hotter. Atlanta has 49 percent more moisture in the air (absolute humidity), but the surface temperature is lower. Water vapor or clouds appear to be taking the extra heat in Atlanta. Water vapor/cloud feedback appears to be negative in this case. It appears that the extra insolation in Phoenix simply causes higher surface temperatures. Calculated sensitivities are the same at both locations, 0.14 deg/w/m2.

    Now compare Indianapolis with Sacramento. Again, both about same altitude and latitude and receive same TOA solar radiation. Sacramento gets 130 (!) w/m2 more surface radiation in July than Indianapolis, but in this case, the July average surface temperatures are the same. Indianapolis has 36 percent more water vapor in the air than Sacramento, but the temperatures are the same. It appears that there is no water vapor feedback in Indianapolis. Where does the extra heat go in Sacramento? Probably to cool pacific airflows, since the sensitivity is 0.05, compared to 0.11 in Indianapolis.

  360. Steve Sadlov
    Posted Apr 2, 2007 at 1:55 PM | Permalink

    RE: #360 – Maybe an overall cateloguing of differences might help. Sacramento Vs Indy during summer: Sacramento uniquely has (Vs Indy): Mountain ranges to the West and East, the eastern ones high mountains, relatively cold rivers running through / along it, a major marsh to the west and others a few miles to the south, low RH during summer, both (land modified) sea breezes and mountain breezes, many inversions, lots of dust. Indy uniquely has (Vs Sacto): an overall flatland for many miles on all sides, seasonally variable river, no major wetlands nearby, high RH during summer, mostly synoptic winds if any, rarer inversions, highly variable particulate count, mT air from the Gulf of Mexico intruding frequently, convective events at least weekly. This is just a start, the list can be added to.

  361. S. Hales
    Posted Apr 2, 2007 at 2:31 PM | Permalink

    MarkW, The supremes did not declare CO2 an air pollutant in general. They broadly defined air pollutants as per 7602 of the EPA act and simply that the EPA had the regulatory authority to regulate CO2 emissions from new vehicles. The lower court decision was not whether CO2 was not a polutant in general but a narrower interpretation of pollutant contained in the statute. All the rulings have centered around 2 issues the standing of the litigant who shows no reasonable harm in the present and the interpretation of 7602 to grant the EPA authority to regulate CO2 emissions from the tailpipes of new vehicles. No broad definition of CO2 as a pollutant outside of the statute has been claimed.

  362. DeWitt Payne
    Posted Apr 2, 2007 at 2:36 PM | Permalink

    Willis, I finally figured out why the upper atmosphere is supposed to warm faster than the surface. Look at the moist adiabats chart I posted in #227 in this thread. The surface temperatures are 15 degrees apart (-25, -10, 5, 20 and 35 degrees C). At high altitudes, especially where the moist adiabats approach the lapse rate of a dry adiabat, the temperature difference increases with temperature. The question then becomes whether the satellite MSU LT and MT readings are at an altitude high enough to see this effect. Another possibility (besides a warm bias in the surface temperature and/or a cool bias in the satellite data) is that the effect is being masked by stratospheric cooling from ozone loss and increased CO2 (which increases emissivity).

  363. Michael Jankowski
    Posted Apr 2, 2007 at 3:03 PM | Permalink

    It appears that there is no water vapor feedback in Indianapolis. Where does the extra heat go in Sacramento? Probably to cool pacific airflows, since the sensitivity is 0.05, compared to 0.11 in Indianapolis.

    What about river flows through Sacramento cooling things off? I think Indy gets pretty dried-up in the summertime (told sewage treatment plant discharge seems to be the overwhelming flow).

  364. jae
    Posted Apr 2, 2007 at 3:04 PM | Permalink

    361. Exactly, and all those things affect sensitivity to heat increases. The reason the sensitivity at Sacramento is half that at Indy is due to the reasons you listed, plus (I believe) a tremendous influence of the Pacific. This is suggested by the very uniform and low sensitivities (0.03-0.05 deg/wm-2) for all locations west of the Sierras and Cascades, as well as the Sitka Penninsula and Hawaii (except that crazy SanFran, which is 0.08).

    Now add a comparison between Boise and Madison, WI. (both same latitude, Boise 600 m higher, of course). Boise only gets 1.6 degrees hotter than Madison, despite getting 129 w/m2 more radiation. So some of the excess energy in Boise goes to surface heat, and some goes to atmosphere (some is altitude effect, for sure). Madison has almost twice as much moisture in the air, but does not get as hot. No water vapor feedback here.

    It was a surprise to me that Boise receives even more solar radiation at the surface in July than Phoenix (471 w/m2 vs. 421 w/m2), but does not get nearly as hot in July (34.2 vs 23.3 deg), again probably due to higher elevation and the influence of the Rocky Mountains. Of particular interest is the fact that the sensitivity in Boise is intermediate between the West Coast and Madison (and Phoenix) at 0.07 deg/wm-2.

  365. David Smith
    Posted Apr 2, 2007 at 3:39 PM | Permalink

    March Arctic sea ice anomaly is given here .

    Much of the anomaly was north of Europe, which is not a surprise, given the winter weather patterns.

  366. Steve Sadlov
    Posted Apr 2, 2007 at 4:08 PM | Permalink

    RE: #366 – I no longer trust the satellite data being used for ice extent. After witnessing some horrible glitches over the past couple of months, my faith has been shattered. I believe that the way these charts work, satellite glitches (all of them various “drop outs” either partical or complete of the signals) all subtract. In order to “smooth” the data, they constantly look back some amount of time so that the figures being posted are not real time areas, they are short term “averages” looking back to the time of the last point. If glitches are allowed to stand, they will lower the average. There may also be long term aging and degradition of the sensor arrays and amplifiers themselves, which also continually shave off signal strength and may also be shaving off apparent coverage percent. Like surface temperatures, these data are now highly suspect in my book.

  367. jae
    Posted Apr 2, 2007 at 4:52 PM | Permalink

    In a nutshell, here is what I’m concluding from the data in the spreadsheet:

    1. Climate sensitivity (deg/wm-2) varies according to location, due to geographical features, but it is surprisingly uniform in a given “climate regime,” which I classify into five categories: (a) Pacific Ocean dominated’€”west of the Cascades and Sierras, the Sitka Penninsula, and Hawaii (low sensitivity); (b) northern moderate elevation mountainous areas–Rocky Mountains to Cascade/Sierra crest (moderate sensitivity); (c) high altitude Rocky Mountain (medium ‘€”high sensitivity) (d) Desert Southwest (high sensitivity); and (e) East of Rocky Mountains (moderate sensitivity). Note that each of these regimes is dominated by major geological features: Pacific Ocean, mountains, elevation, desert, etc. In the case of the eastern US, the unimpeded mixing of air masses and high humidity tends to even out the sensitivities in the whole area.

    2. Sensitivity to perturbations in radiation is small, ranging from 0.03 deg/wm-2 near the Pacific Ocean to 0.22 deg/wm-2 at certain high elevation locations (Alamosa, CO, Flagstaff, AZ). The average is 0.11, which essentially matches Idso’s calculations.
    3. There does not appear to be any positive water vapor feedback, since absolute humidity varies closely with latitude east of the Rocky Mountains, while sensitivity appears unaffected. All other regimes also do not show any relationship between humidity and sensitivity. In fact, all of the most sensitive locations have low humidity (below 13 g/m3), suggesting that if there is any water vapor feedback, it is negative.

    I would like to look at Europe next. Does anyone know where I can get data on average monthly solar insolation, temperatures, altitudes, etc. for Europe (or anywhere else, for that matter)? I’ll bet there is tremendous variation in Europe, due to the presence of so many different climate regimes.

  368. trevor
    Posted Apr 2, 2007 at 5:14 PM | Permalink

    Re #340, #341: If you follow the links that Ron Cram gives to Wikipedia, it is evident that there is a very interesting discussion going on. I don’t really feel qualified to pitch in, but I am sure that many posters here could make a constructive contribution.

    At the least, it is a fascinating exposition of how Wikipedia works!

  369. David Smith
    Posted Apr 2, 2007 at 6:44 PM | Permalink

    RE #366 The Cryosphere Today chart shows an odd downward move in March. The regional charts for Barents and Greenland show instantaneous drops that look unphysical. Odd discontinuities like these do not inspire confidence in the sea ice measurement.

    Speaking of ice, here is a map of Arctic sea ice age (first year and multi-year). It is somewhat dated (November 2006) but it gives a good portrayal of how young the ice is and how dynamic its movement must be.

  370. David Smith
    Posted Apr 2, 2007 at 7:27 PM | Permalink

    Steve S, the current NOAA SST forecast shows both La Nina and a cool-phase PDO for 2007, as well as a cooling Atlantic. Your Pacific Coast rainfall may indeed suffer. The forecast maps show about an inch of rain in Southern California in about a week, then the well goes dry until next winter.

  371. Steve Sadlov
    Posted Apr 2, 2007 at 7:31 PM | Permalink

    RE: #370 – Not to “Da Vinci Code” this or anything, but there is sort of a Fibonacci Sequence describing the formation, movement and eventual dissipation of the ice. :;

  372. Willis Eschenbach
    Posted Apr 2, 2007 at 7:53 PM | Permalink

    Re the discussion on Wikipedia (see post 340 and here), this is just too good. William “Stoat” Connelly at his finest:

    This section appears to be in for attack reasons and is totally unbalanced. I’ve removed it. It needs a proper re-write before it goes back, not just a POV tag. William M. Connolley 21:47, 22 March 2007 (UTC)

    Please explain what specifically is attacking what and by whom? We should not have accusations that are vague or non-specific. — Tony of Race to the Right 17:43, 23 March 2007 (UTC)

    You just have to look at the section I deleted to see William M. Connolley 21:58, 23 March 2007 (UTC)

    In other words, you really do not want to explain the deletion? I cannot believe that is the case. What quotes are you considering to be attacks? Who is the attacker? Against whom are the attacks? Please, so we can understand what the actual reason is for the deletion, explain to those of us not as intelligent as you what justified the deletion. — Tony of Race to the Right 03:34, 25 March 2007 (UTC)

    Just though I should point out the example you as an experienced editor are setting. It is never the responsibility of the editor making changes, deletions, etc to explain why, it is up to everyone else to find and figure it out for themselves why the editor is correct beyond discussion. Is that right? I mean, that is why you are placing the burden of justifying a statement like, “appears to be in for attack reasons and is totally unbalanced” on me to “look at the section” to know what you consider to be an attack, rather than you explaining why you chose to ignore the policy about deleting text (If you feel the edit is unsatisfactory, improve it rather than simply reverting or deleting it). — Tony of Race to the Right 03:46, 25 March 2007 (UTC)

    Too good … too typical …


  373. Posted Apr 2, 2007 at 8:30 PM | Permalink

    I feel honored to be quoted among a class of people whose intelligence is far beyond mine. Thank you.

  374. Gerald Machnee
    Posted Apr 2, 2007 at 8:55 PM | Permalink

    In todays Sun newpapers, I found the following article(part of it) after the location.


    Following revelations that a high-ranking member of Wikipedia’s bureaucracy used his cloak of anonymity to lie about being a professor of religion, the free Internet encyclopedia plans to ask contributors who claim such credentials to identify themselves.

    Wikipedia founder Jimmy Wales said in interviews by phone and instant message Wednesday from Japan that contributors still would be able to remain anonymous. But he said they should only be allowed to cite some professional expertise in a subject if those credentials have been verified.

    “We always prefer to give a positive incentive rather than absolute prohibition, so that people can contribute without a lot of hassle,” Wales wrote.

    Wales suggested such a plan two years ago, but the idea suddenly gained currency after the recent discovery that a prolific Wikipedia contributor who wrote under the pen name “Essjay” and claimed to be a professor of theology turned out to be a 24-year-old college dropout, Ryan Jordan.

  375. Steve McIntyre
    Posted Apr 2, 2007 at 10:39 PM | Permalink

    bender, nice season for the Gators.

  376. bender
    Posted Apr 2, 2007 at 11:05 PM | Permalink

    Gainesville is a happy, happy place.

  377. T J Olson
    Posted Apr 3, 2007 at 4:16 AM | Permalink

    Nicholas (#358) asks:
    “What’s the point of declaring CO2 a pollutant when there are bigger fish for them to fry?” They’ve already been fried.

    Virtually all noxious air pollutants have declined each decade in the US for forty years. Regulating particulates produced from diesel engines was shown to be cancer causing during the Clinton presidency, but he delayed action. The Bush Administration simply went ahead with what was already known. Now, just today, I say a commercial on a cable TV news channel touting the benefits of low sulfur diesel fuel. This is the result of this regulation: reducing cancer-causing particulates emitted by applying a technological fix.

    The costs of air pollution prevention have followed a long path of decline with advancing and more economical technology. Take autos, for instance, Computers regulate air-flow intake and fuel mixtures that carburetors used to, primarily thanks to cheap microchips. Better, more complete fuel burning means less pollution.

    The trouble with CO2 “pollution” is that it can’t be solved with more technology, according to present-day knowledge. Generating CO2 is intrinsic to using carbon-based fuels for energy.

    Thus the purpose of going after CO2 by enviros and regulators is because the bigger fish have already been fried! Attacking fossil fuel use is an attack on industrialism itself. Bureaucrats rejoice.

  378. David Smith
    Posted Apr 3, 2007 at 4:55 AM | Permalink

    RE 377 chomp, chomp, chomp.

    bender, have mercy on jungle animals this fall – we’re endangered species, ya know

  379. Nicholas
    Posted Apr 3, 2007 at 6:14 AM | Permalink

    T. J. Olson : That may be true in the USA (although, judging by the last
    time I was in Los Angeles and the amount of pollution I encountered, I’m not so
    sure). However, here in Australia, particulate pollution is still a big
    problem. If I leave my window open, a few days later, the window sill is
    black with soot. Yet we are already regulating CO2 here (e.g. incandescent
    light bulbs are being banned, using CO2 as a justification). Yet, as I say,
    trucks and cars belching sooty smoke is a common sight on the roads. Yes, we
    have strict emissions standards. But there are lots of older vehicles which
    are grandfathered. So, I very much disagree that here at least, all the
    “bigger fish have been fried”.

    Here’s some food for thought. The more CO2 is regulated in countries like
    Australia and the USA, the more we’re going to export our CO2 production to
    countries like China and India – countries with poor emissions regulations
    and horrible pollution problems. So in effect, regulating CO2 is actually
    increasing other forms of pollution. Is that something to be glad of?

    P.S. Yes I am aware that much of the famous Los Angeles smog is not actually
    pollution. But there are still some dirty places in that city, especially
    near the airport, where there is at least one oil refinery.

  380. Richard Sharpe
    Posted Apr 3, 2007 at 7:53 AM | Permalink

    Hmmm, that should have been:


  381. Mark T.
    Posted Apr 3, 2007 at 9:39 AM | Permalink

    There may also be long term aging and degradition of the sensor arrays and amplifiers themselves, which also continually shave off signal strength and may also be shaving off apparent coverage percent.

    Satellites always have degradation in all of these areas, though the extent is somewhat dependent upon several things. First, where they are at in their respective orbits. LEOs suffer different effects than GEOs due to radiation and particle influences (single event upsets, for example, in digital logic). Also, the level of shielding in the electronics (sensors, amplifiers, even digital) has an impact, which varies depending upon orbit as well. All satellites have a lifetime, and usually there are self-correcting mechanisms to maintain some significant level of system gain over time (i.e. above 90% for the gain path over the life of the satellite). Once the limit is reached, time for an overhaul, or decommissioning.


  382. Ken Fritsch
    Posted Apr 3, 2007 at 9:50 AM | Permalink

    In further attempts to explain the relationships of temperature and precipitation on cereal grain crop yields as reported by Lobell and Field in their 2007 paper, I have been using state by state climate and crop yield data with attention to a temperature and precipitation interaction effect. I have used the product of ranked mean temperature and mean precipitation for the years 1961-2004 (the period of interest in the Lobell paper) and for the seasonal time period from June-August . Temperature was ranked from high (1) to low (44) and precipitation was ranked from low (1) to high (44). A low value of the temperature/precipitation product was assumed to be unfavorable to crop yields. The product was then plotted versus the annual deviation of corn and soybean yields (delta yield) from their trend lines for the period 1961-2004.

    What I found for all Midwest states analyzed was a plateau area of delta yields above a threshold product value where the yields were independent of temperature and precipitation. These years of independence covered most of the time period studied. As the product value decreased below the threshold value the yields fell off at a steep linear rate.

    What I did find was a couple to a few outliers for each state that always deviated significantly from the plateau and always on the negative side. I checked those years with an online search to determine whether or not some unusual circumstance not related to temperature and precipitation had occurred. The search has been slow, but to date I have found explanations for most of these outliers (actually these are only outliers after I can document an explanation) in IL and Iowa. They include an early frost, flooding, southern corn leaf blight and soybean aphids.

    I was wondering if any posters and readers, like Barry B,, here have any sources on historical events affecting crop yields ‘€” primarily the Midwestern states and corn and soybean yields.

    I have fit a log curve to the relationship above, and after excluding the outliers, I obtain R^2 values in the 0.7 range. I am not certain that it is legitimate at this point to use the log (or the R^2 value) as in effect what one sees is an intersection of a nearly 0 slope horizontal line with a steeply downward sloping line. More importantly it is the negative crop yield data points that I need to document with regards to outlier status.

    I find the temptation to cherry pick and over fit models is ever present and needs to be beaten down at all costs and by constant vigilance. My looking at crop yield does make me think about tree ring growth in a little different light.

  383. Steve Sadlov
    Posted Apr 3, 2007 at 10:12 AM | Permalink

    RE: #371 – I am going to stick my neck out and make another risky prediction (risk level even higher than my Winter 2006 – 07 Greenland – Iceland ice bridge prediction). For the next 30 years, the world will experience a sharp cooling later leveling off. There will be mega droughts the likes of which have not been seen since at least the 1970s, perhaps even longer. There will be famine.

  384. Mark T.
    Posted Apr 3, 2007 at 10:30 AM | Permalink

    That prediction almost sounds… biblical. The cooling, droughts, famine, mis-cues at the pool table, etc., will further be blamed on GW.


  385. Steve Sadlov
    Posted Apr 3, 2007 at 11:08 AM | Permalink

    RE: #386 – If they blame it on AGW, and respond accordingly, they will of course be adding fuel to the fire (or in this case, ice to the freezer). Interestingly, as Biblical as my prediction may seem, it is entirely based on secular analysis. I grabbed a bit of Paleo, threw in some Astrophysics (especially Solar Physics) and spiced it up with some Oceanography and Meteorology specific to the Pacific Ocean. In particular, David Smith and John A may find it of special interest, mostly as a “worst case scenario” thought starter. I hope my prediction is wrong.

  386. EW
    Posted Apr 3, 2007 at 11:11 AM | Permalink

    I’m depressed. It’s a neverending stream of doom and gloom in the media. There was some meeting of climatologists in Brussels and it’s official. We will all fry from CO2 effect and the culprit is The Mankind.

  387. Mark T.
    Posted Apr 3, 2007 at 11:18 AM | Permalink

    What I’m worried about, Steve S., is that they’ll examine the cooling and say “OMG!!!, we didn’t expect our reductions* to be this effective! The climate has a higher sensitivity than we originally thought! Victory!” and respond according to the new mantra, which will be even stricter controls to “guide” the climate.


    * reductions I expect will eventually be in place world-wide.

  388. David Smith
    Posted Apr 3, 2007 at 4:05 PM | Permalink

    Steve S., the preliminary global temperature report for March, 2007 shows it as the fifth or sixth warmest in the last 50 years. Despite anecdotal evidence of extraordinary warmth in some Northern Hemisphere population centers it looks like this March was not a record.

    The El Nino effect is over and soon the neutral/La Nina effect will kick in. Combine that with the anomalously cool subsurface water in areas of the Warm Pool and 2007 is not looking like a record. The Hadley (?) 2007 forecast of warmest-ever year looks off the mark.

  389. Steve Sadlov
    Posted Apr 3, 2007 at 4:18 PM | Permalink

    RE: #390 – A small anecdote. It must have been 1974 or so. February if I recall correctly. There was a most remarkable event, namely, several days of extreme Gulf of Mexico mT intrusion up the Mississippi Valley, so charged with moisture it evntually resulted in torrential flash flood producing rains as far north as Illinois, Wisconsin and Michigan. A serious warm front with so much energy there were actually thunderstorms in it. Seems that no one was talking about killer AGW at the time. LOL! Interestingly, that was the beginning of a fairly substantial drought affecting a huge chunk of the US. Right there at the end of the most recent previous negative PDO. I picture sort of a noisy square wave, with lots of harmonic oscillations, overshoots and undershoots throughout the sloppy transition periods. Harmonics fall to lower amplitudes during the flat sections of the “signal.”

  390. Steve Sadlov
    Posted Apr 3, 2007 at 4:42 PM | Permalink


  391. Barry B.
    Posted Apr 3, 2007 at 5:17 PM | Permalink

    Re: #383

    Ken, try this site:

    This is the only place I know of that provides commentary for the entire U.S. and over the time-frame you require. I don’t know if the information will be specific enough but it would be a good place to start.

    If you have certain years that remain in question, post the growing season and crop and I’ll see if I can find some info for you.

  392. Bob Weber
    Posted Apr 3, 2007 at 5:40 PM | Permalink

    Seasonal hurricane forcast for 2007 by Klotzback, Gray and Thorson (Colorado State Univ) predicting 17 named storms, 9 hurricanes with 5 being intense hurricanes. Link is http://hurricane.atmos.colostate.edu/Forecasts/2007/april2007/


  393. DeWitt Payne
    Posted Apr 3, 2007 at 7:24 PM | Permalink

    #368 jae,

    In fact, all of the most sensitive locations have low humidity (below 13 g/m3), suggesting that if there is any water vapor feedback, it is negative.

    Low specific humidity equates to lower heat capacity so higher sensitivity should be expected, I think. Desert areas, absent irrigation, also don’t have much moisture available to allow specific humidity to increase much with temperature. Areas with higher humidity and sufficient available water, either open or in the ground, should have lower sensitivity because water has a really large heat of vaporization. The greenhouse forcing from water vapor doubling, say, would only be a few watts/sq.m. and would be lost in the noise of the 100 or so w/sq.m. difference from summer to winter insolation.

  394. Steve Sadlov
    Posted Apr 3, 2007 at 7:29 PM | Permalink

    RE: #394 – Looks like they fell into the ENSO trap. ENSO is certainly important. Extenuating circumstance for 2007 is that in addition to La Nina, looks like we’re finally getting into unmistakeable Negative PDO territory. Also, what’s goin’ on with AMO and others. Atlantic may not get quite as warm SSTs in the critical places as some might be imagining. We’ll see who’s right. I made a very low bet, 9 named storms, 5 ‘canes – of these, no cat 5s, a couple cat 3/4 and the rest wimps.

  395. DeWitt Payne
    Posted Apr 3, 2007 at 7:33 PM | Permalink

    I haven’t done any programming in a long time and I’m having some problems setting up to do a finite difference calculation of moist adiabats using equation 3.74 on this page of R. Caballero’s Physical Meteorology notes . I think part of the problem may be that the equations, at least on my computer, are very difficult to read. Any help or references would be greatly appreciated.

  396. Mark T.
    Posted Apr 3, 2007 at 7:50 PM | Permalink

    T(z+delta_z) = T(z) – (g/(c_pd + (w_s+w_l)*c_l) * (1 + l_v*w_s/(R*T))/(1 + l_v^2/(c_pd*R_v*T^2)))|_z * delta_z

    The R*T and c_pd*R_v*T^2 are ambiguous in the equation, i.e. is is ({}/R)*T or {}/(R*T), I do not know. When I clicked on the equation itself, it gave me a LaTex popup, but wouldn’t let me copy it. Also, the |_z is the “evaluated at z” notation (bar with a superscript z).


  397. Ken Fritsch
    Posted Apr 3, 2007 at 7:58 PM | Permalink

    Re: #393

    Barry B. there is much information in the link you posted and I’ll need time to go through it. I most appreciate your providing it here.

    I just wanted to satisfy a personal curiosity on this subject and did not anticipate the project it has become. I know that I could be very mislead by looking at an outlier year and finding a potential cause that was not temperature or precipitation related. If I could get a statewide estimate of the yield effect I could confidently confirm the outlier, but with qualitative commentaries I would need to compare every since year, since I suspect that yield damage not related to temperature and precipitation occurs and is reported every year and it is a matter of degree.

  398. jae
    Posted Apr 3, 2007 at 9:06 PM | Permalink

    394: That is what the data show. And that is what impresses me. I believe in empirical information much more than theoretical calculations, especially when computer modeling is involved. I still don’t see anything wrong with the way I am calculating the sensitivity of pertubations in heat input. Until someone blows me out of the water, I will pursue this line of reasoning. It explains a LOT of things.

  399. george h.
    Posted Apr 3, 2007 at 9:13 PM | Permalink

    Business of Global Warming Feels a Lot Like Inquisition
    William F. Buckley’s piece here: http://www.realclearpolitics.com/articles/2007/04/business_of_global_warming_fee.html

  400. Curt
    Posted Apr 3, 2007 at 10:26 PM | Permalink

    I sure hope this is an April Fools’ joke:

    The government of Belgium’s French-speaking region of Wallonia, which has a population of about 4 million, has approved a tax on barbequing, local media reported.

    Experts said that between 50 and 100 grams of CO2, a so-called greenhouse gas, is emitted during barbequing. Beginning June 2007, residents of Wallonia will have to pay 20 euros for a grilling session.

    The local authorities plan to monitor compliance with the new tax legislation from helicopters, whose thermal sensors will detect burning grills.

    Scientists believe CO2 emissions are a major cause of global warming.


  401. DeWitt Payne
    Posted Apr 3, 2007 at 11:45 PM | Permalink

    Mark T.,

    Thanks. That confirms what I thought too. R is the gas constant and is always multiplied by temperature so it has to be R*T. Unfortunately, several of those terms are also a function of temperature like l_v the latent heat of vaporization. At least R and g are constants.

  402. T J Olson
    Posted Apr 4, 2007 at 3:19 AM | Permalink

    Nicholas (#380)-

    Thanks for the Aussie particulars.
    Just as you say, there may indeed be bigger fish (enviro concerns) down under than in the US, and thus the CO2 fetish may substitute imaginary concerns in place of substantive, needful ones.

    As an insular American, I appreciate that we have commenters from Canada, the UK, Europe, and elsewhere. In the spirit of international awareness, can I suggest to JohnM here that a flag-based visitor indicator in the CA site margin, like Lubos Motl has on his site, would help everyone broaden their own horizon consciousness?

    As for the question of advanced nations exporting CO2 emissions to developing nations? Since most GDP (and thus CO2 emission) anywhere is for domestic consumption, I think the concern is way overblown. But taking down trade barriers can help poor nations to adopt the most efficient (ie, the latest) in technological innovations. In general, the most efficient production techniques emit the least CO2.

  403. Curt
    Posted Apr 4, 2007 at 7:42 AM | Permalink

    Please tell me this is an April Fools’ joke:

    BRUSSELS, April 3 (RIA Novosti) – The government of Belgium’s French-speaking region of Wallonia, which has a population of about 4 million, has approved a tax on barbequing, local media reported.

    Experts said that between 50 and 100 grams of CO2, a so-called greenhouse gas, is emitted during barbequing. Beginning June 2007, residents of Wallonia will have to pay 20 euros for a grilling session.

    The local authorities plan to monitor compliance with the new tax legislation from helicopters, whose thermal sensors will detect burning grills.

    Scientists believe CO2 emissions are a major cause of global warming.


  404. Gerald Machnee
    Posted Apr 4, 2007 at 9:29 AM | Permalink

    Re #390 Steve S: **A small anecdote. It must have been 1974 or so**. You were not thinking about the thunderstorms in the Black Hills that caused a flash flood drowning 200 or more campers in 1972, I think. The line of thunderstorms stretched into the western Arctic. This one was in 1972 or less likely, in 1975 when I was in the High Arctic.

  405. Steve Sadlov
    Posted Apr 4, 2007 at 9:41 AM | Permalink

    RE: #404 – Nope. What I was referring to was much further to the East, in the states I mentioned, and specifically due to a warm front, not a cold front. After that system moved north up into Canada, a series of peristent, massive Highs covered most of the US for months on end. Corn was only knee high at the beginning of September across the entire mid section. A wave of farm failures ensued. As 1974 turned into 1975, the drought spread / shifted West. By early 1976, California was fully engulfed, and what had started out as a normal rainy season in fall of 1975 shut off. Winter of 1976 – 77 featured one of the lowest rainfall amounts in recorded history, at all recording stations, statewide. By 1977, Gov. Brown declared a state of emergency. It was the last blast of the most recent past Negative PDO, which gave droughts just shy of the megadrought category during the 50s and the 70s, affecting large swaths of the US. We were saved by the combination, in late 1977, by a swing to a positive PDO with an El Nino following starting in fall of 1978. A portent of things to come, as GLOBAL COOLING sets in for the next 30 or so years. The AGW fanatics have it all wrong. Global warming does not result in more droughts. Unfortunately, this coming innate cyclical swing into cooling will.

  406. Mark T.
    Posted Apr 4, 2007 at 10:32 AM | Permalink

    Curt, what’s funny about that is the use of HELICOPTERS to stop the rogue polluters. Isn’t a gas grill (propane) relatively low emission anyway? Wow.


  407. Mark T.
    Posted Apr 4, 2007 at 10:33 AM | Permalink

    DeWitt, no problem. If you had two mice, you could probably cut and paste the mouse-over text that pops up to get the LaTex. There should be an option to copy it, but I couldn’t figure out how…


  408. Greg F
    Posted Apr 4, 2007 at 11:18 AM | Permalink

    Please tell me this is an April Fools’ joke:

    It was.

  409. Michael Jankowski
    Posted Apr 4, 2007 at 11:30 AM | Permalink

    RE403, yes, April Fools’ (junkscience.com briefly mention it as such today).

  410. Richard Sharpe
    Posted Apr 4, 2007 at 11:49 AM | Permalink

    Climate Change Predicted to Reduce US Crop Yields by 25-80% by Schlenkar and Roberts.

  411. Richard Sharpe
    Posted Apr 4, 2007 at 12:38 PM | Permalink

    Note, that is not the actual title of the paper I linked to in my previous post. That title is from this discussion of it.

  412. Ken Fritsch
    Posted Apr 4, 2007 at 1:31 PM | Permalink

    Re: #410

    Thanks for the link, Richard. I have read part of that paper and from what I have garnered to this point is that they are looking at temperature effects on crop yields essentially the same way Lobell and Field did, i.e. they ignore temperature and precipitation interactions.

    They discuss precipitation effects more than Lobell and Field did in their paper and additionally mention the mitigating effects of precipitation on higher temperatures, but then say that since temperature and moisture are not correlated (inversely) they evidently can deal with temperature effects on yields without the interaction concern.

    Like Lobell and Field, they are correct as far they go, but I found that droughty conditions were the main detriment to corn and soybean yields and that those conditions, though occurring less than 20% of the time, invariably involved low precipitation and high temperatures.

    I have noted that this paper and some others that I have read on climate effects on crop yields talk about a break point in the relationship. I find that when I use the product of precipitation and temperature there is very definitely a break point. I will continue to look for papers that have studied the temperature and moisture interaction in more quantitative terms.

  413. Michael Jankowski
    Posted Apr 4, 2007 at 1:47 PM | Permalink

    Am I correct in nothing that elevating CO2 levels are not accounted for when it comes to future yields?

    Seeing that the Hadley model results were used here, my thoughts drift back to this from 2000. Both the Hadley and Canadian models seemed to generally agree on the future temp and precip changes in the US, but when they were combined to predict the resulting soil moisture content (“…soil moisture is critical for both agriculture and natural ecosystems…”), what appeared to be minor disagreements between the models become magnified, and the results are drastically different between the two models. And, “…the differences in soil moisture and drought tendencies will be significant for water supply, agriculture, forests, and lake levels…”

    Granted, the Canadian model outlook seemed worse than Hadley for the US, but that may not be true globally. But it does help raise the question that the model used here is oversimplistic.

  414. Curt
    Posted Apr 4, 2007 at 3:13 PM | Permalink

    jae #360:

    Where does the extra heat go in Sacramento? Probably to cool pacific airflows, since the sensitivity is 0.05, compared to 0.11 in Indianapolis.

    I think you’re right on this one. A key feature of Sacramento summer life is awaiting the “delta winds” that kick up almost every afternoon (and it’s miserable if they don’t). Early afternoon July temperatures often exceed 40C (104F) — which I believe is exceedingly rare in Indianapolis. But at about 4pm on a typical day the cooling winds that come through the Golden Gate and over the Sacramento River delta hit the city and rapidly cool things off, so that evening temperatures in Sacramento are well below those in humid Indianapolis, thus making the daily average temperatures quite equal.

  415. Dave B
    Posted Apr 4, 2007 at 3:22 PM | Permalink

    Martian Albedo…

    interesting article today:


    what is interesting to me, is that on earth, increased dust is assumed to cause cooling, but on mars, the same effect causes warming.

  416. Ken Fritsch
    Posted Apr 4, 2007 at 3:38 PM | Permalink

    Re: #414

    Am I correct in noting that elevating CO2 levels are not accounted for when it comes to future yields?

    Lobell and Field referred to 2-3% increase in yields due to CO2 level increases in the same period, 1961-2004, that they claimed temperature reduced yields by nearly an equal amount. I am not sure at this point in my reading what the paper referenced by Richard S says about CO2 effects on yields.

    I have seen papers that have shown much larger yield increases due to increasing CO2 and more recent ones that claim the experiments showing this positive effect were performed in greenhouses with higher moisture conditions. They argue if done in “nature” under drier conditions and higher temperatures that CO2 fertilization effects would be reduced dramatically. As always with AGW related issues there is controversy and conflicting results.

  417. welikerocks
    Posted Apr 4, 2007 at 4:05 PM | Permalink

    Hey everyone, have you seen this?
    Dust Storms Fuel Global Warming On Mars

    “Across the past two decades, the model showed the surface temperature of Mars has increased by about 0.65 degrees Celsius (1.17 degrees Fahrenheit).”

    But they come up with dust this time! It’s almost the exact same amount of warming as they say is happening on Earth.

    Um, why can’t they ever discuss the sun?

  418. John Baltutis
    Posted Apr 4, 2007 at 4:13 PM | Permalink

    Slightly OT. Selling it!

  419. Steve Sadlov
    Posted Apr 4, 2007 at 4:43 PM | Permalink

    Dynamic link, good only for a day or less:

    According to this, a new Greenland – Iceland ice bridge. Very, very late in the season for that!

  420. Barry B.
    Posted Apr 5, 2007 at 12:13 AM | Permalink

    Re: #414

    But it does help raise the question that the model used here is oversimplistic

    It is over-simplistic.

    The reason is because as evidenced in the Schlenkar and Roberts study they only look at rainfall during the growing season and do not account for moisture that has been stored in the soil prior to the growing season. Any agronomist will tell you that moisture is everything in crop production and adequate moisture will mitigate the effects of excess heat. Across the cornbelt (with the exception of the irrigated lands in the plains) our soils have the ability to hold a considerable amount of plant available water which can be utilized far into the growing season. A failure to take this into account will result in erroneous conclusions. This also comes into play in future predictions because as your link shows, the Hadley and Canadian models both show an increase in precipitation along with an increase in temperature. Why one would not account for stored moisture is beyond me.

    The CO2 link is also worth noting. There’s some research that shows a positive benefit to increased CO2 levels but the authors of these papers tend to dismiss it as the great unknown. Heaven forbid we show a positive effect to global warming.

    Probably the most egregious of these people’s errors is that they make no mention of future advancements in varietal technologies. I fully understand that it would be difficult to quantify future technological improvements – reasonable assumptions could be made – but to make no mention of such possibilities and assuming the status quo as they so diligently do in their predictions for the future, they seriously misrepresent the implications of climate change. It makes one wonder whether these so-called “studies” are in effect co-opted to produce a message of doom and gloom rather than add to the scientific database.

  421. Alan Woods
    Posted Apr 5, 2007 at 3:04 AM | Permalink

    Re: 418

    Um, why can’t they ever discuss the sun?

    Because it’s the elephant in the room.

  422. David Smith
    Posted Apr 5, 2007 at 6:39 AM | Permalink

    Steve S., I too am beginning to develop frustration with the people who measure Arctic sea ice extent. The Cryosphere Today website today is doing something or other to their records (I hope repairing that early-March discontinuity) – maybe it will become clear once they finish.

    Now, if they are changing their methods to give a more-accurate picture of ice extent, that’s fine and welcomed. But, I hope they don’t graft a new method onto historical records that were created by a different methodology.

    We shall see.

  423. Curt
    Posted Apr 5, 2007 at 8:23 AM | Permalink

    Interesting snippet I caught on the radio last week on the University of Texas McDonald Observatory’s “Stardate”:

    Quiet Neighbor
    One of our closest stellar neighbors is a near twin to the Sun. It’s about the same size, mass, and temperature. It’s a little older, though, and right now, it’s a good bit quieter.

    The star is Alpha Centauri A. It’s the brightest member of the triple-star Alpha Centauri system. The system is 4.3 light-years away — closer than any other stars.

    Right now, Alpha Centauri A isn’t quite as bright as normal. It’s putting out far less ultraviolet and X-ray energy than average — something that might have happened to the Sun about three centuries ago.

    Astronomy satellites have found that last year, the level of X-rays from the star dropped to just about one percent of what it was five years earlier. And its ultraviolet energy dropped to just four percent of what it was.

    A star’s energy output varies on a natural cycle. But astronomers from Villanova, who watch Alpha Centauri A and similar stars, say this is the biggest drop ever recorded from a Sun-like star.

    The Sun might have experienced a similar drop for several decades around 1700. We don’t know about its X-ray and ultraviolet energy, but we do know that sunspots all but vanished. These magnetic storms are an indication of the Sun’s energy output. During this period, Earth’s temperature dropped by about a degree.

    Keeping an eye on Alpha Centauri A and similar stars may help scientists figure out what causes these dropouts — and when the Sun might produce another one.

    From http://stardate.org/radio/program.php?f=detail&id=2007-03-30

  424. Jeff Norman
    Posted Apr 5, 2007 at 11:06 AM | Permalink

    Re: #420


    Did any polar bears make it across the bridge this year?

  425. Ken Fritsch
    Posted Apr 5, 2007 at 11:17 AM | Permalink

    Re: #421

    I finished my first reading of the Schlenker and Roberts paper linked below and have concluded that they are very much interested in making the case for global warming having a very detrimental and unavoidable effect on the major crop yields.

    Click to access SchlenkerRoberts.pdf

    Their study concentrates on temperatures and sees a relationship much the same as my temperature/precipitation product does with crop yields. They report an increase of yields with increasing temperatures and then a plateau and finally a sharp decrease with temperatures. I saw the plateau and sharp decline with my product indicating that low temperatures were not a limiting factor. I am sure that they have validity in pointing to yields being reduced by the time crops see temperatures over a given threshold, much as indicated in this discussion by Barry B. I do, however, suspect in their rush to show the negative effects of global warming on crop yields that their analysis does not take precipitation sufficiently into account and particularly when its deficiencies are paired with higher temperatures associated with droughts.

    They do mention CO2 fertilization, but dismiss accounting for it in their pessimistic prognosis for crop yields by noting the controversy surrounding its effects.

    They note that through their analysis they are able to determine that soybean and corn yield genetics have not been changed in the past 50 years to compensate for higher growing temperatures as derived from yields of southern grown crops. I was not able to understand how they came to this conclusion and as a result compared corn and soybean yield trend lines from southern and northern states over the past 44 years. Corn and soybean yields are lower for southern states, but the trend line for corn is somewhat more steeply increasing for southern states than northern ones indicating that temperature effects are not yet limiting. I used those southern states where acreage was increasing or stable as one could conjecture that in states with falling acreage the acreages eliminated where the inherently poorer yielding ones.

    Soybeans trend lines were not as steep for most southern states as was seen for northern ones, but what I saw was more recent yields having a significantly larger spread as though affected by diseases. Perhaps Barry B. has some inputs on that matter. For soybean reaction to higher growing temperatures I looked at Brazil and the state Mato Grosso where temperatures are higher than in our southern states. Here soybean yields have been increasing at a steady linear rate and have out-yielded US soybeans over the last 5 years. I have read articles that stated that soybeans varieties have been developed specifically for these tropical climates ‘€” something not noted in the Schlenker paper.

    What also puzzled me was that, while the maximum temperatures in the GW scenarios presented in the Schlenker paper would be detrimental to crop yields, by extending the growing season and accommodating earlier planting and maturing of crops could avoid some of the effects of these maximum temperatures. I do not recall a discussion or mention of this by the authors. Perhaps the photosynthetic effects would alter my reasoning here.

    I found a paper linked below that took into account the precipitation, temperature, soil moisture content and satellite surveillance of the amount of green the crops showed during the growing season. I would think that the amount of green would be strongly correlated with the other variables measured. These studies where designed to provide crop yield predicting tools from remote sensing during the season to estimate final yields. They discuss the breakpoints associated with predicting yields and the need for using 2 regressions.

    http://home.iitk.ac.in/~ramesh/publications_pdf under crop yield IOWA PDF

  426. Steve Sadlov
    Posted Apr 5, 2007 at 11:53 AM | Permalink

    RE: #425 – I heard they did. Icelanders in affected areas were buying Weatherbys.

  427. John Lang
    Posted Apr 5, 2007 at 1:11 PM | Permalink

    It does look like they are fixing the 2007 ice extent problems in the datasets at Cryosphere (for 2007 anyway, there are still some showing some problems that became evident late in 2006.)

    In looking around the archive area, I ran across this interesting paper from 1979 that charts the arctic ice anomalies from 1953 to 1979.

    I looks like 1960 was the lowest ice area, nearly 4 million sq kms lower than the average over the period versus the modern record low of about 1.75 million below the 1979-2000 average.

    Click to access i1520-0485-9-3-580.pdf

  428. Steve Sadlov
    Posted Apr 5, 2007 at 1:44 PM | Permalink

    RE: #423 and 428- Bill Chapman seems pretty consciencious and responsive. Past glitches I’ve emailed him on, and they got fixed real quick. At least he recognizes how difficult it is to get usable data from the satellites. Unlike certain other cryosphere oriented sites with overt agendas which shall go unnamed ….

  429. Reid
    Posted Apr 5, 2007 at 1:55 PM | Permalink

    Let’s get ready to rumble…

    “Kerry and Gingrich announce climate showdown”


    Former House Speaker and possible presidential candidate Newt Gingrich (R-Ga.) and 2004 Democratic nominee Sen. John Kerry (Mass.) are set to square off on climate change next week, their staffs announced Thursday.

    The debate, hosted by New York University’s John Brademas Center for the Study of Congress, will take place next Tuesday, April 10, at 10 a.m. in the Russell Senate Office Building.

One Trackback

  1. By Comment on Unthreaded #7 by Follow the Money on Mar 18, 2007 at 9:49 PM

    […] post by Carbon Credits News by Carbon Credit News Posted by […]

%d bloggers like this: