I mentioned here that I would be presenting at a poster session at the U.S. Climate Change Workshop (Nov 14 evening, Crystal Gateway Marriott, Arlington Virginia). Poster pdf’s are supposed to be in tomorrow. Here’s a draft. I’d appreciate any comments. Obviously many of the points are ones that readers of this blog are familiar with.
Obviously there are any number of problems with all these studies. Here I’m picking a couple of the more interesting ones, which have a pretty important knock-on effect.
Climate Audit 
52 Comments
As a lurker and non-scientist I found the PDF reasonably easy to understand.
You may wish to explain the meaning of MXD and RW upon first use, if those unfamiliar with the field are likely to read it.
Bob
In paragraph 3 you say “the hockeystickness depended on two proxies known in specialist literature not to be temperature proxies…”
I think that this sentence should be followed up with one that says “If these few questionable proxy series are not included in the MBH 98-99 model, there is no hockey stick and no conclusion that recent temperatures are a millenial anomaly. Other studies often cited as backing up MBH 98-99 are based on the same flawed proxies and similarly fail in their conclusions when the bristlecones are removed.”
Looks good Steve,
I do wonder about the sentence in parenthesis in A1.
About A2, the second heavily used proxy, the Greenland d18O. The most important raw data are here:
http://www.ncdc.noaa.gov/paleo/icecore/icecore-varlist.html
I wonder how they produced that proxy. All I see is trendless noise:
Typo in last sentence of B2 (above the figure) “…an unreliable for comparing…” Either the “an” should be removed or there is a missing word between “unreliable” and “for”.
The Greenland dO18 series used here is Fisher’s stacked series described in the NATO 1996 publication, Climate Variations. It’s also used frequently.
My view of multiproxy studies (see, for example, Reply to Huybers #2, is that most proxies just function as white noise, but that there are a few cherrypicked hockey stick shaped series which impart the shape to the final reconstruction, which is rescaled once again to remove the attenuation in the first averaging. I thought that it was pretty interesting that white noise outperformed actual proxies in the analysis described in Reply to Huybers #2. (This wasn’t in our actual Reply to Huybers. It’s something that I noticed as I writing my blog entry and is only on the blog.)
Trying to look at the poster in naive, big-picture, TCO-style, my first impression is that the section headings don’t mention the key take-away points. Also, the colors on the spaghetti graph are wrongly-chosen. Finally, I’d add a bit to the first sentence in A1.
Here are my suggestions:
Figure 1: It’s important to have the viewer be able to easily pick out the few series you discuss in particular. Red, magenta and purple look too similar (at least on my monitor); the blue is easily distinguished — use that for JBO98; given your existing palette of colors, use green for MBH98-99.
Suggested title/text additions/changes:
A1: Was the Medieval Warm Period warmer than the mid-20th century? Jones et al (1998) and other “spaghetti graphs.”
Multiproxy temperature reconstructions of the past millenium are key building blocks in the argument that late 20th-century temperatures were millenially high. Although multiproxy studies…
A2: Many Supposedly “Independent” Multiproxy Studies Rely on the Same Three Important 11th Century (MWP) Proxies
A3: Are Jones et al (1998)’s proxies representative? Independent Views of the Polar Urals.
B1: Briffa’s Polar Urals Series – A Critical Quality Control Failure
Fig 6 – Is the late 20th century instrumantal or proxy? I think it is vital to clarify this point. Murray
No helpful comments from me in the near term, except that some statements seem a little premature. For example you say that the Polar Urals series was used “perhaps” in Briffa et al 2001. Should be easy to find that out? Also, you rightly (I guess – I’m no expert) complain that there weren’t enough cores in the early time series, remove some that you don’t like, and then present a figure with very high reconstructed temperatures. You seem to be saying that there aren’t enough data to support those other reconstructions but there are enough to support yours. Maybe you should weaken your own statement about the warmth early on.
In the longer term there is the important issue of validating the proxies’ abilities to record high temperature events. For the purpose of the poster, maybe there is still time to speculate intelligently on why they may not.
In http://www.climateaudit.org/?p=150 , Briffa Tronetrask Reconstruction, you went into more detail than in your poster. I would expect that at the USCC workshop you may be asked to explain the “simply coerced” statement.
Re #8: ou’d think that it would be easy to determine whether a particular series was used in Briffa et al [2001]. You also probably know enough about me to know that I’m pretty diligent about things: the sites used in Briffa et al [2001] are not listed in the article, there is no SI and coauthors Briffa and Jones have each refused to provide the information. There’s a small scale map showing dots, but insufficient as a data citation. It’s ridiculous that the sites used in this study should be unkown, but there it is.
What I’m trying to say is that the very cold (anomalously) so Polar Urals record is dependent on only 3 cores, below minium standards and additionally without a firm date. If those 3 cores are taken as undated and the Jones calculation re-done, you get a high MWP. I’m not saying that it was warm; only that the Jones statement that the 20th was warmer than the MWP cannot be stated – the usual form of one of our statements. Sort of like a proof by contradiction in math.
I guess I could throw in an upside-down U graph somewhere. The trouble with speculating is that you run the risk of going a bridge too far and there are lots of people waiting for any little mis-step so they can mau-mau anything I’ve ever done.
You are overly scared of the mau-mauing. Instead you should be interested in discovery.
Steve and Ross-
I think you have mostly been very careful in your writing to not make the claim that the 15th century temps were comparable to present temps (only that MBH’s claims re. 15th century are unsupportable), however, these two sentences in the conclusion of Ross’s paper at the Fraser Institute;
“By simply correcting two gross errors in
the way the IPCC’s poster-child chart
was created, the original conclusions are
overturned. The late twentieth century
is no longer climatically unique in comparison to recent history-a finding that refutes a fundamental claim of Kyoto supporters.”
do appear to make that claim. I don’t think your work supports this statement.
P.S. I think you are doing important work. Please keep it up.
Steve, I hope you’ll post the final .pdf version of your poster as a link along the site right margin (perhaps under “articles”?).
Thanks everyone for the comments. This is updated and is posted in the right frame as Armand suggests. I also tidied the links there to include the two recent Replies and the final version of our original GRL article (instead of the submission version).
One of the many things that is so goofy about these studies is that they are so pompous about their conclusions – claiming standard errors of 0.15 deg C – and then it turns out that it’s based on 3 scrappy pieces of core which are out of sequence and don’t date properly.
This is just one thing, there are lots of others, but, as TCO observes, it’s hard to finish anything if you try to do everything.
#12 BKC, I accept your point. It would have been better to say “…the claim that the late 20th century climate is historically unique is no longer supported, undermining one of the fundamental claims of Kyoto supporters.” Or something to that effect. Absence of evidence, not evidence of absence.
Beyond posted deadline now, but in thanks I wanted to reply to your response (#10). First point well made and well taken; that information should be freely available! Second point (and Nos. 12 & 16): I know you have other problems than the few highlighted in the poster, but in the context of the poster’s focus it seems that your main thrust is that the late 20th century should only be seen as anomalous in terms of the last 950 years instead of the last 1000 (because earlier reconstructions can’t be reliably resolved). Third point — my turn to be speculative….
Let’s argue that the proxies (I’m only going to talk tree rings [I haven’t read enough on proxies to be very intelligent here so give me some latitude — this is just an example]) follow some upsidedown U or semi-Gaussian optimality curve common in biology. For tree rings, we’re talking growth. If trees don’t grow increasingly more in the warmer conditions of the MWP this might say these two things: 1. the warm conditions were beyond the conditions to which the trees could take advantage and were therefore either infrequent or short-lived in history; 2. global warming will not magically be universally beneficial for plant growth. I admit that I would never put these arguments on a poster, but my purpose here is only to illustrate that discussing mechanism is important and far more interesting that simply claiming that relationships derived under some circumstances may not hold under other circumstances. Rather than wildly arm waving as I have done, perhaps citing spatial differences in relationships could stand in for how relationships may vary temporally.
Sorry for general nature of the criticism but I haven’t yet read enough to be more specific (or even certain that my criticism is applicable).
Good luck, Steve.
Re: #17
Steve, In science it’s all about what you can prove and what you can’t. Sometimes that’s dull, but anything else is speculation.
RE #17: Steve L. I’m not saying that it is only anomalous in the past 950 years. I’m saying that their confidence intervals aren’t worth the powder to blow them to hell. All I’m doing is a kind of reductio ad absurdum – they can’t argue millennial uniqueness without these 3 stupid cores. The actual tree line evidence from Polar Urals suggests that it was warm well past the early 11th century, but to disprove Jones, you don’t need to prove that. I still think in terms of math proofs.
The MXD series seem to max out at lower temperatures than the RW series. I posted up some updated info on the Polar Urals site, which shows very elevated ring widths in the MWP, but MXD responds much less. In modern times, this is the Briffa adjustment for Tornetrask – then why not for Polar URals in the 12th century?
Also and this is a typical Hockey Team stunt – when they got new Polar Urals data with big MWP ring widths, did they update Briffa et al [1995]? Of course not. Just the same as Hughes is sitting on the 2002 bristlecone ring widths from Sheep Mountain. Here my experience in mining promotions is a good guide: if he’d had “good” drill results, we’d have heard about it. The silence is deafening. In mining promotions, not announcing the results would be dishonest, but in the wild west world of climate science, it seems that anything goes.
Thanks Steve M (#20), those are interesting points. I was just making some general observations about the poster. I said it seemed that your poster was saying that it was warmer back in the MWP; then you replied that the poster was only trying to say that Jones shouldn’t have concluded that it was colder; then I responded that the poster then seems to agree with Jones that the climate is warm relative to the (less distant past). Now you’re saying both that it was warm back in the 11th C and also that all of the reconstruction is unreliable — both of those things may be correct, and I appreciate you telling me, but that’s not what your poster was about (I thought) and I was trying to comment on the message your poster conveyed to a climate layperson such as myself.
Response to Paul P (#19): I’m a scientist and I think that some of what I said is fairly applicable in general. That is, if one is going to speculate that a method applied in some circumstances will not work in others (something that is universally true and therefore not worth mentioning by itself), then one should further be able to speculate on a mechanism why that is particularly so in the case being investigated. IMHO.
Steve, the poster (seems like) an abortion. Tried to keep my mouth shut but have to be didactic. 90% of issues are communications, not substance. But they are quite bad those 90%. Hope your prospecti were better laid out.
(page 1-2)
Is this a poster or the publication that goes with the poster? It sure doesn’t attract the eye as a poster. As a publication, it seems rambling. Heck as a poster, it seems rambling. You sorta try to cobble a theme together around “beyond hockeysticks”, but it doesn’t quite click.
Some reactions as I read through (in order):
1. What is with the A123, B123?
2. Need an intro. You sort of have one with A1, but why not call it that if it is at a different level of hierarchy than the other “A”s?
3. I’m not so sure in a short paper (which this seems to be) if it is useful to conflate the overall spaghetti graph issues with examination of a specific reconstruction.
4. Also, if this is just a poster, might have been more useful to think about and come away with more high level CEO style briefing (with some facts). You seem to be struggling between an independent new criticism and showing general issues in the field.
5. The graphic of the “excising” REMAINS very hard to read. Sheesh, use some arrows and show the damn curve that got chopped. Maybe show a before and after. And what is that gray background for on the second figure? Sheesh. Would you put that as a slide in front of a board and expect them to understand?
6. And is this so important, to have it as on of your figure? Maybe yes, but seems like you are being tugged between Daly style critique (e.g. the appended instrumental record) and substantive analysis.
7. I don’t like the title. You seem to be trying to get a connection for your reader between your papers, but it’s confusing when we find out that there is little Mann discussion in there. I would prefer a title without hockey stick in it.
8. As I reread first para, it is sort of a cludge. Some intro stuff (but not well done, not showing overall organization and point of the paper) mixed in with a basic point: proxies not tested in the 90s. Don’t mix micro and macro points.
9. Second para is ok as part of an intro section. But ditch that micro point at end of first para then. However, it is too long as a single sentence. Revise.
10. Thinking as I criticize. An alternate intro structure (change order) might be: “we have criticized the hockey stick, famously; have been challenged on rest of the reconstructions (spaghetti), here we examine one other one. (THIS will engage the reader better as a story and will help start him with something familiar.)
11. The graphic above of the spaghetti is hard to read. Consider omitting the legend, which is unreadable as is and just use arrows to indicate the (old) Mann hockey stick graph and the (new for this paper) Jones multi-proxy study.
12. (next para, last on page 2) You have a 7 line sentence with way too many asides buried in it. Revise the heck out of the structure and consider if the asides are all needed. Also, you have a comma before a period. Did you do a grammar check on Word?
TCO – next time, I’d suggest speaking up before the submission deadline…
Page 3
13. First para: You have a good point in here about number of proxies, but perhaps something could be done to make it more clear: allude to guidance on how many proxies required or some basic statistics from sampling theory or such. Quick reader may not immediately realize what is wrong with 3 proxies. Or with amount of difference among the group of 3.
14. Second Para: Is it 3 series in the entire 11th century or the early 11th century? Watch your wording. Your panel only has 3 series and your para one indicates that 3 is number for early 11th only. Don’t play sticker and expand too far on a single argument area.
15. Content point: what years does the overall reconstruction cover and where are you kvetching about? How early does it start? Are we talking some little part?
16. The figure should be laid out differently so that the eye knows that Polar Urals is the one of concern. Perhaps having Polar Urals at top at a minimum. Or separate the two flat ones from the Polar Ural one from the summation: show the logic of (2 flat) + (very bumpy) = (bumpy).
17. “as well as in the early 15th century” is a needless parenthetical diversion. The para sentences should develop/prove/examine the topic sentence (which was nothing to do with the 15th or any time other then the 11th or the 20th). Stay on point.
18. Content Point: How does the Polar Urals have more than half of its black space as negative?
19. If the key thing is 11th century versus 20th century, I should have at least a dashed line or circles or such on the graph showing those periods. Instead the legend goes every 200 years and does not extend across as grid lines. Dashed lines extending across the panels at 1100 and 1900 (maybe also at 1000 and 2000) would be helpful to draw the eye to the key area.
20. Remove the modifier “centennial-scale”, wrt the Greenland proxy. It’s just less variable period, Steve. Don’t be wordy AND introduce parenthetical distractions (frequency of oscillation versus period) AT SAME TIME! J Besides that, it confuses the issue, since the others could be argued to have a 50 year period and the Greenland to have a 25 year one, so what the heck is “centennial scale”. Let that snake not be born…
21. Remove the whole sentence about the Briffa claim. The attack here and the theme here is on Jones (not having enough proxies, one of them carrying the load, not spectacular 20th century, etc.). Leave Briffa be. That claim was wrt his data only. Whole nother argument.
22. Content and argument point: Does Jones show the 3 proxy composite or show all together. And is there (could there) be some relevance in comparing a 20 proxy 20th with a 3 proxy 11th? For instance, if wiggle matching showed some correlation of the proxies picked up later in the years? (Sorry don’t have my thoughts clear, here.)
23. Third para is unclear. You are bringing in a different Jones paper right now. Is that paper purely a comment on the earlier work? Not stated or obvious that this is so. Also, you don’t prove out the implicit thought (that the SD is larger than stated). Also, (content-wise), I worry about “MWP” versus “11th century” versus “early 11th century”. He did not have 3 proxies over all that time.
24. Last para is extraneous to the thought structure. Excise it. This is a meta point about study to study issues. Should not be in this part of the paper (for sure). Likely should not even be in this paper, since this paper is a whack at the Jones tree.
Armand, he posted it the damn day before the deadline. And see my response. I didn’t have the heart to level with him then. So yeah, maybe I still should have. Oh,well. You can revise after submission time. (be real…you can.) Also, he still needs to learn for the future. There will be many, many more papers/posters, I hope.
20 of course should be amplitude, not period.
I need to hear that this feedback is valued before going over rest of the paper with my red pencil.
This is going to turn into a paper, so everything helps. I haven’t had the poster made up yet. I’ve gotten jammed for time, as I’msomewhat unexpectedly going to be in England for a few days.
Steve,
The norm in a poster session is to have a POSTER. It should have interesting schematic things that attract the reader to come over and chat with you directly. There you explain what is going on. You can do it in some big graphics program and have it printed on an architecture style plotter. Or you can do power point slides and print them out on 8.5 x 11 and tack them up. Or fiddle with construction paper and the like. But it’s more pictures, less words. You are there to explain things. It’s to get people over to your table, but should have enough of a story to interest people and so that they can “read it” if you are off at the drink table. But it’s not a paper tacked up on the posterboard! (and you don’t even have a good communicating paper at that!) Really, go to some meetings and see what the norm is. I’m not saying you need to be the most beautiful, but you should at least be at average level. And you are way, way below the bar.
Sometimes, they want a paper that will somehow go with the poster (but doesn’t need to be perfect one for one, really) that will be published in conference proceedings. Really this is a scam for societies to publish volumes. Make sure your submission is a mix and match of stuff that will get published in real journals. Put these papers in a stack on your table at the poster session. Interested parties can pick up the publication.
If this is going to be a paper, whether for Science or GRL or East Elbonian Tree Cutters or Albertan Wannabe Montanan’s Stripminers, it needs to have a clear logic to it. Strunk and White, baby. Minto. Rethink through your topic and make sure that you have a consistent theme. If you have extraneous information or if it takes too much space to properly develop your thoughts, break your work more in “LPUs”.
Really, I think there are posts here that you could more easily just send out the door as individual papers. For instance the misdated 3 cores piece. Sometimes, I think you are being a typical grad student and just wanting to do everything together. Sometimes I think that you are worried that individual pieces might be found out wrong, so you want to have things together so they carry each other. By the way, this is wrong from a logic standpoint. IT IS GREAT to isolate issues and analyze them. So what if a supposition is proved wrong. You should have about 30 papers in the hopper anyway. Do you really think everyone will be perfect? That noone else will be able to push the thinking forward?
I’m not going to bother more detailed line by line critique of your paper, based on lukewarm reaction to feedback. But you should be able to generalize my comments and evaluate your own work as I would when you read over it.
P.s. I love ya baby. You are working your ass off. And getting a hell of a lot of good thinking and analysis done. You have something real potential to benefit the community by making them think about things more. Please don’t let my words discourage you.
Re #29, Steve
If you’ve got some free time in London, can I buy you a beer ?
#30. The A1…B3 are poster panels. They are in Word so I can edit easier before transcribing to the poster. I’m doing the poster in Corel Draw so that it prints out. I’m doing it this a.m. before I go to England this afternoon and hope that I can get it delivered before I go to Washington.
I agree that I’m mixing messages here. I’ve got too many things that I’m trying to say and too few papers in the hopper, so, as you say, I’m interjecting bright ideas from elsewhere. The impact of these 3 stupid Polar Urals cores on multiproxy reconstructions is really quite remarkable and it took quite a bit of detailed work to locate. I had to learn how to use COFECHA and other dendro methods and work up some of my own.
In one sense, it would be nice to write something very short, showing that these 3 cores are screwed up (actually I could write 10 pages on that easily.) But this is not of interest to anyone unless it matters to something. So I have to tie it into one of the reconstructions and Jones et al 1998 is the best one to do so.
That’s the theory anyway. I’m going back at this poster to get rid of the post-1980s argument. It’s interesting, but something for another day and thin it out. I’m not sure that I even need to display a spaghetti diagram here. I’ll bring some color examples if anyone wonders.
The references and things here will nto be in the “poster” version of the poster. They were just in the cyber version. I also didn’t do a pdf in cyber version, because I find them hard to read.
Wasn’t GRL originally meant to publish short letters (hence the name) of work in progress?
I think you should do the 10 page paper on the 3 simple cores. (just publish the damn post that you did here.) I’ve thought it for a long time. I just need to do enought 2 by 4 mule hitting to get you to think so too. All you need to do is allude to its importance in other work (just reference them) and reference to other (future) paper of yours which examines those reconstructions. You don’t need to, nor should you, bother with showing what happens to the reconstructions at Jones or Mann level when those cores are out. Just do the damn paper on the Urals itself.
You can publish in a dendrochronology journal or such if you want, vice GRL. Although might take a quick shot at GRL too.
Re #20 — Do you know if Hughes submitted his Sheep Mountain data and analyses for publication and got rejected? I suspect that you’re right that he’s sitting on it, but I’m curious about how you know. Back to the poster: I think your message will be tighter if you don’t include stuff you don’t believe (i.e., that growth may be reduced in warmer climates).
What did you think of my criticism?
I’m back from England. What a lot of nonsense arguing while I was away – it sounded like sci.environment. I just started deleting flames and could hardly stop. If anyone feels hard done by in the deletes, post again. Otherwise, I apologize to readers.
I’m going to this workshop on Monday. If anyone wants to say hello in Washington, I’ll have time on Monday orTuesday so send me an email and I’ll try to arrange something.
I met in England with John A. and fFreddy, who also contacted me. I had lunch in the House of Lords on Friday, which was both interesting and fun.
Steve, I hope you had a talk with the PM. He sounds like he could use a clue. Welcome back!
Re #37, and all to cover for ‘John’…
So, any news from the workshop? Impressions, tidbits, even real science?
What he said, plus welcome back.
The primary driver of US winter climate and other seasons’ cooling events is the Arctic. A “warmer” article of faith is “Arctic warming” in the extreme. Pertinent to that are sea ice metrics. Here is something of interest to me. I often view Cryosphere Today. But of late, I’ve also been spending lots of time at the Anchorage NWS “virtual” ice desk, up until present, done “the old fashioned way” (but apparently, change may be in the wind, on that score). Anyhow, here is an example of a typical product:
http://pafc.arh.noaa.gov/ice.php?img=ice
One thing I’ve learned based on this is, for the Beufort Sea, Cryosphere Today is quite conservative and tends to depict less extent than actual.
RE: #42 – Here is Cryosphere Today’s current Arctic image, for comparison:
Notice how much less ice is shown near Alaska’s Arctic coast than what the Anchorage NWS show. Whereas the Anchorage Ice Desk has the ice edge from a point on the coast, WSW of Barrow, then more or less going West from there (w/ a few lobes) and encompassing the whole north of Wrangel Island, Cryosphere shows ice free area significantly north of the edge reported by the Ice Desk. From what I can gather, the Ice Desk use a combination of satellite vis, aircraft observations and sea observations, so in this case I believe it to be largely accurate. Bottom line is, Cryosphere shows less ice in this particular basin. How might that impact the reported anomaly figures?
A couple of other recent observations….
The Chukchi Sea is overtly showing a positive anomaly now. Might be interesting to observe how fast the Bering Sea ices up this Fall.
Also, looking at history, something which had escaped my notice until now …. right at the time last years BBC scaremongering “news” show about the poor Churchill polar bears was being shot, Hudson’s Bay was actually showing a positive anomaly! Hrumph …..
FYI – from the Anchorage Ice Desk, hot off the presses:
…SEPTEMBER MONTHLY YEAR GROUP ANALYSIS AND SEASONAL OUTLOOK…
THIS ICE SEASON REALLY DOES NOT HAVE A GOOD MATCH IN OUR HISTORICAL
RECORDS. 1995 IS CLOSE BUT THE ICE WAS FURTHER OFF SHORE FROM BARROW
TO FLAXMAN ISLAND. 1991 IS CLOSE BUT THE ICE WAS MUCH MORE SEVERE
NEAR CANADA. 1978 WAS VERY CLOSE EXCEPT THE RUSSIAN ICE WAS MORE
SEVERE. 2001 HAS BEEN THE MOST CONSISTENT YEAR GROUP MATCH THIS
SUMMER BUT 2001 WAS BOTH TOO COLD OVER RUSSIA AND CANADA AND TOO WARM
OVER ALASKA. 1986 WAS ALSO COLDER OVER RUSSIA AND CANADA BUT THE ICE
CONDITIONS NEAR ALASKA WERE ALMOST AS SEVERE AS 2006 INCLUDING THE
ANALYSIS OF MULTI YEAR ICE CLOSE TO THE ALASKA COAST.
I TURNED TO THE NOAA NATIONAL WEATHER SERVICE CLIMATE PREDICTION
CENTER TO HELP WITH MY DECISION MAKING PROCESS. THE CLIMATE CENTER
HAS A HISTORICAL RECORD OF EL NINO VS. LA NINA PACIFIC OCEAN SEA
SURFACE TEMPERATURE CONDITIONS THAT GOES BACK ON THEIR WEB SITE TO
1950. I DECIDED TO USE THESE HISTORICAL PATTERNS TO HELP IDENTIFY
WHICH OF THE ABOVE YEARS MIGHT HAVE A SIMILAR CLIMATE PATTERN AS THIS
YEAR. 2006 BEGAN WITH A MODERATE LA NINA TEMPERATURE GRADIENT IN THE
PACIFIC…SLIGHTLY COLDER THAN NORMAL. BY MID 2006 THE PACIFIC WAS IN
A NEUTRAL…OR NEAR NORMAL..STATE. THE CLIMATE CENTER/S FORECAST FOR
THE REMAINDER OF 2006 IS FOR THE OCEAN TO GRADUALLY WARM TO A WEAK EL
NINO BY THE END OF THE YEAR. OVERALL 2006 WILL BE A SLIGHT WARMING
YEAR FOR THE PACIFIC OCEAN.
HERE IS HOW THE OTHER ICE YEARS COMPARE…1995 STARTED WITH A STRONG
EL NINO AND COOLED TO A MODERATE LA NINA. 1991 STARTED WITH A WEAK EL
NINO AND WARMED TO END WITH A STRONG EL NINO. 1978 STARTED WITH A
WEAK EL NINO AND COOLED TO A WEAK LA NINA OR NEUTRAL CONDITION. 2001
STARTED WITH A WEAK LA NINA AND WARMED TO A NEUTRAL CONDITION THEN
BEGAN TO COOL AGAIN TOWARD THE END OF THE YEAR. 1986 STARTED WITH A
WEAK LA NINA AND WARMED TO A MODERATE OR STRONG EL NINO…THE SAME
TYPE OF PATTERN WE ARE EXPECTING FOR 2006.
THE YEAR GROUP FOR SEPTEMBER IS 1986…FOLLOWED BY 2001 AND 1991.
THE EL NINO VS. LA NINA SIGNATURE WAS STRONGER IN 1986 THAN IT IS
TODAY. FOR MORE INFORMATION ON EL NINO I HIGHLY RECOMMEND THE CLIMATE
PREDICTION CENTER/S WEB SITE AT (LOWER CASE ON ONE LINE) HTTP://
http://WWW.CPC.NCEP.NOAA.GOV/PRODUCTS/ANALYSIS_MONITORING/
ENSO_ADVISORY/INDEX.HTML.
1986 HAD LOW CONCENTRATIONS OF ICE REMAINING NEAR THE NORTHERN ALASKA
COAST THROUGH SEPTEMBER WITH ICE RETURNING TO THE NORTH COAST BY THE
SECOND WEEK IN OCTOBER. BY THE FIRST OF NOVEMBER ICE COVERED MUCH OF
KOTZEBUE SOUND AND MOST OF NORTON SOUND. BY THE FIRST OF DECEMBER
1986 THE ICE EDGE WAS SOUTH TO ST. LAWRENCE ISLAND…EXTENDED FROM
THE ALASKA WEST COAST TO NUNIVAK ISLAND AND REACHED AS FAR SOUTH
ALONG THE COAST AS KING SALMON. CONSIDERING THE WEAK EL NINO PATTERN
WE ARE EXPECTING FOR 2006 I EXPECT THE ICE DEVELOP ONE TO TWO WEEKS
LATER THAN IN 1986. I WILL PROVIDE A MORE DETAILED SEASONAL OUTLOOK
ON NEXT MONTH/S YEAR GROUP ANALYSIS.
THIS OUTLOOK WILL BE INCLUDED ON THE FIRST THREE SEA ICE ADVISORY
PRODUCTS OF EACH MONTH AND WILL NOT BE UPDATED BETWEEN ISSUANCES.
COLE SEPTEMBER 06
We’ve seen a flat line for Northern Hemisphere sea ice extent for weeks. The overall NH anomaly has been moving toward zero since the end of July. The current forecast progs deepening cold core Low pressure centers ringing the Arctic starting later this week. Now I’ll up the ante. I am confident we’ve now hit this year’s extent minimum in the NH.
Along the same lines I’d like to see the data on surface temperatures.
Early summer reports were out that it was the hottest years on record so far, and that if it continues hottest year on record. This during the beginning of the heat wave.
Shortly thereafter the temperature dropped of dramatically for most of August, for below normal, for North America and Europe, I have no idea for other areas. But the news reports of record worm year have certainly dropped of.
I think we’ll see the same thing for surface temperatures as for ice extent.
Not sure a yearly average is a great indicator to begin with, and this would be why.
Be interesting to see what it has done for tree rings. A drastic heat wave, following a mild to cold winter with high precipitation in the spring, then cool temperatures for a month as we enter fall.
RE: #47 – Interestingly, the NWS were progging above normal for AUG-SEP-OCT for much of N America. Same deal for SEP-OCT-NOV, etc. It would appear that the 90 day outlooks suffer from the same excessive AGW component as the GCMs. Take for example the above normal AUG – OCT 90 day. So far, below normal to normal for 33.33% of the period. September and Oct are going to have to be even more above normal than progged in order to make the 90 Day actuals equal the prog. Look at hemispheric current patterns and make an educated guess how likely that is. By the way, TS (nee Supertyphoon) Ioke is proged to head way up north, eventually getting sucked in the Aleutian Low. In the past, that has sometimes resulted in an early onset of typical North Pacific late fall – winter conditions.
Australasia record (25-30)year winter lows,the solar minima and enhanced GR neutron counts at present suggest antarctic stratospheric ozone depletion and lessening plar vortices..
A record low atmospheric pressure of 727.7 mm of mercury was registered in the Russian capital Tuesday, a Moscow meteorologist said.
“Atmospheric pressure dropped below a record low at 3 a.m. Moscow time (11 p.m. GMT Monday), and hit 727.7 mm Hg at 9 a.m. (5 a.m. GMT),” a regional metrological center’s spokesperson said. “The previous record low of 731.2 mm Hg was registered in 1936.”
Now the 90 Day is even more in doubt:
“AS NOTED IN PREVIOUS
DISCUSSIONS THE GLOBAL MODELS ARE JUST BEGINNING TO INTEGRATE THE
REMAINS OF “IOKE” AS IT TRANSLATES FROM TROPICAL TO EXTRA-TROPICAL
OUT NEAR 160 EAST AT THIS HOUR. THE ECMWF SEEMS TO TAKE THIS ENERGY
AND DEEPEN THE ALEUTIAN LOW WITH A CORRESPONDINGLY STRONG RIDGE OVER
THE EASTERN PACIFIC THROUGH NEXT WEEK. GIVEN THIS SCENARIO WOULD
EXPECT FAIRLY LARGE RUN TO RUN CHANGES IN THE EXTENDED RANGE MODELS.
AT THIS TIME NO STRONG OFFSHORE OR EARLY RAIN EVENTS IN SIGHT.”
The operative phrase here being “at this time.”
If we do get an early rain event, it could sufficiently perturb the synoptic pattern that by the time the perturbation settles, we’ll be past the point in the season where the Pacific High can be reestablished into its dry season regime. This reminds me of Fall 1983 (although the analogy is not exact, being that we were in a strong El Nino in 1983 ….).
Well we did not get the rain event yet, but the system is undoubtedly perturbed. Getting October weather here in Nocal.
Meanwhile in the Actic, the overall NH anomaly is positive and I can now safely say the minimum is past. In fact, it may well have been in August – it does not get much earlier than that. Woe to those who are troubled by a heavy ice season, that seems to be what we are going to have. Note also, an extensive outbreak of snowfall reaching quite far to the south in NE Asia. Funny the beeb were trying to hype Siberian permafrost melting JUST YESTERDAY! Just like the hype of reputed Polar Bear woes in Manitoba last year coincident with a *positive anomaly* for Hudson’s Bay!
Some oddities from the Univ of Ill. cryosphere imagery:
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.7.html
If one were to rely on the above plot, one might rightly conclude that areal sea ice extent in the Kara Sea has not been strongly expanding with the season. However:
Firstly one can clearly see the expanding freeze in Kara. Also, I’ve been monitoring this, and this growth has been ongoing for weeks. What clued me into looking at this was that the overall NH extent had reached its minium in August then had started to increase. Then all of the sudden there was an odd downward “glitch.” I drilled down and found the strongest contributor to be from the Kara Sea component. Yet, the visual image, at the time, showed no decline. I don’t trust the Kara Sea data.