Tuesday, June 14, 2016

GISS down 0.16°C in May; still hottest May in record

GISS is down from 1.09°C in April to 0.93°C in May. This is in line with the fall of 0.186° in TempLS, or as noted earlier, 0.164°C in the NCEP/NCAR index. Also similar falls in the troposphere indices. Still, it was 0.07° warmer than the next warmest May, in 2014. I'll show the map comparisons below the fold. But first, here is the comparison plot with 1998:



The GISS map is here:



The Moyhu spherical harmonics map is here:





31 comments:

  1. "Posted by Nick Stokes at 4:14 AM" - looks like you're an early bird. Thanks for the timely update. I'm noticing rapidly cooling temperature anomalies in the Antarctic in June so far in the daily CFSV2 estimates from UM CCI. If this tendency persists for the remainder of June it will be interesting to see if it shows up in the GISS estimates for June. I'm not sure how much coverage GISS has in Antarctica.

    ReplyDelete
    Replies
    1. Bryan,
      Yes, I woke up thinking GISS must be out now, and I hadn't noticed. But it was just out. Then I went back to sleep.

      NCEP/NCAR average so far is down about 0.08°C from May. Since 1998 staged a revival in Jun/Jul, it's quite possible that 2016 will drop behind for the first time in a while.

      Delete
    2. Nick,

      I think this is happen in June, not by the ENSO, but 1) in June the Anomaly in the arctic oceans becomes less, 2) antarctica is switched coold, very cold.. if NCEP/NCAR is rigth (http://www.karstenhaustein.com/reanalysis/gfs0p5/ANOM2m_mollw/ANOM2m_fcstMTH_mollw.html) then antarctica alone would bring a decline of arround 0.2K in compare to May.

      Delete
    3. Anon,
      Here are some pictures of Arctic and Antarctic, for June so far this year. They come from here. It's the time when the Arctic Ocean is dominated by areas tied fairly closely to freezing point, so anomalies are around zero (which is cool relative to recent). And Antarctic shows a mix of warm and cool, but more cool.

      Delete
    4. I'm still waiting for the MEI value, which is late this month, but with this slightly cooler than expected May temperature the forecast for the year 2016 will probably got down by 0.01°C.

      Delete
    5. The next month, June, will be more important than May for the following month and so for the year. Last year a large cooling event in Antarctica have lowered the temperature. As already noted, it seems to be the case this year again. At http://www.karstenhaustein.com/climate.php
      Antarctica is 3K below the 1981-2010 mean in June so far. If this and the June correlation with the following months hold, the GISS temperature anomaly for 2016 will more likely below +1.0°C than above.
      What may be the reason for this Antarctic cooling? The warming is really asymmetric, warm in NH cool in Antarctica.

      Delete
    6. The new model predicts 1.01+-0.07°C for the year 2016 GISS global LOTI now.
      But, as I've written above, it depends on the June value.
      The June my model predicts to 0.92+-0.14. If the central 0.92 would come true than the prediction for the year would 1.01+-0.06°C.
      But at the site linked above the June (so far) has fallen by 0.17°C. If the GISS values does the same the GISS June Temp would be only 0.76°C, even no new record. The is obvious mainly do to the very cold Antarctica, 3°C below the 1981-2010 mean. Assuming the GISS value would be 0.76°C for June (instead of 0.92°C) the prediction for the year would go down to 0.96+-0.06°C.
      Also last year had a very cool Antarctic winter, now this may happen again.

      Delete
    7. So last year the GISS anomaly bottomed in July with ONI at around 1.0 and growing. This year ONI is dropping into neutral/La Nina range. In this Civil War between the NH and the SH, the SH may fare better in 2016. This is why I'm watching the PDO and the AMO. So far they're staying high. If that continues, the NH can score victories in the fall and winter months by softening a La Nina blow.

      Delete
    8. @Uli: Just checked. The cold Antarctic June is almost certainly caused by the stunningly positive AAO. The forecast remains on the high side: CPC AAO forecast
      ... which is reflected in the GFS 2m forecast, essentially the continuation of this months extreme anomaly pattern: GFS AA anomaly forecast
      With the forecast locked, we are in for something in the order of +0.8K in June. Certainly no major record anymore, if any. Just as I anticipated earlier this year when I said that 2016 would look very different in terms of annual anomaly distribution than 1998.

      The interesting bit is that there is indeed a very strong link between AAO and AA 2m temperature. The higher the index, the lower the temperature. Not surprisingly, given that the cold is literally trapped in a very powerful freezer: Correlation AAO and AA T2m
      Usually, one would expect regression to the mean wrt AAO anomalies, but for some reason there is a hint of an upward trend lately (say since about 2008).

      As far as the remainder of the year is concerned, June might actually be the coldest month ... except AAO stays positive in July and/or August. Fall and winter will see a warmer Arctic again (and probably Antarctica too if last years patterns repeat), compensating the loss of tropical SST anomalies due to transition into weak La Nina. Of course, a major Eurasian cooling event in Nov or Dec may blow the +1.0K GISS anomaly away at the last mile, but without chances are that we just stay above the mark.

      Delete
    9. There is a paper that discusses a negative greenhouse effect above the Antarctic plateau, and recent news that the Antarctic CO2 level is now above 400 ppm.

      Delete
    10. @KarSteN: Thanks! The large Antarctic cooling surprised me, so the SH mean is colder than last year and the 1981-2010 average. I thought last year was a cold Antarctic anomaly, likely not repeating this year. But now ...
      My regression model predicts consistently a large cooling of global temperature in December 16 to a GISS value about 0.5-0.95°C, so December may be colder than June.

      An other strange finding is the large two year oscillation I found in the 1950-1980 GISS data. Odd years are warmer or warming more that even years. This decays some after that period, but needs incorporated in the regression.

      Delete
    11. @Uli: Well, April/May made the impression as if things down south wouldn't repeat this year, only to have June going nuts again. December very much depends on Eurasia (and La Nina strength of course). Any strong neg AO phase early in the winter season will send T2m diving. Your range looks perfectly sensible to me. The oscillation is probably a fluke, unless you can find a QBO-link, which, to my knowledge, hasn't been found.

      @JCH: The negative GHG effect over Antarctica is only at work in SH summer and it is very small. What we see is purely circulation driven. Regarding one of your earlier comments: AMO is mainly driven by PDO (atmospheric link between North Pacific and NAO state, which in turn controls the air temperature advection over the North Atlantic). Long-term AMO cycles are forcing-driven, superimposed by the PDO (aka PNA-NAO) signal.

      Delete
    12. My take on the current state of ENSO modeling
      https://forum.azimuthproject.org/discussion/comment/15376/#Comment_15376

      I think everyone is floundering by chasing temperatures up and down without having a basic understanding of the underlying pattern. Its like you have boat owners in a harbor who are surprised by the tide level every morning! Yes, the knowledge regarding ENSO is that poor!

      Delete
    13. That is why everybody speculates. On skeptic blogs they are speculating we are on the threshold of a prolonged, multiyear La Nina that will restore the pause. I've been guessing at a weak La Nina... maybe no La Nina at all. In terms of dynamics, I suspect we have not seen what is about to happen since the late 1970s and early 80s.

      Delete
    14. @JCH: Stop reading "skeptic" blogs. It doesn't further your understanding by any means. Only noise. No signal. WHUT has a much more interesting hypothesis to offer, and frankly, I'm pretty intrigued by his hypothesis. So thumbs up! ;)

      Delete
    15. Thanks. About reading "skeptic" blogs: interesting that the skeptic Richard Lindzen is largely responsible for how Laplace's Tidal Equations are incorporated into GCMs. Also Salby.

      example:
      http://imageshack.com/a/img921/4194/xEjxkY.png

      So its hard to get away from these guys. Whether they actually provide any useful signal is the question.

      Delete
  2. Hi Nick,

    Does anyone have access to Nino3.4 daily data in easy to view format (ie, txt)? I was curious about SSTs for the region lately owing to this post at Tamino's.

    https://tamino.wordpress.com/2016/06/14/global-temperature-update-2/#comment-95393

    Just wondering if Nino3.4 SSTs are shooting up lately per the graph at link, and if this is unexpected.

    (Late entry - found the source for the graph)

    http://www.tropicaltidbits.com/analysis/ocean/

    ReplyDelete
    Replies
    1. The BOM ENSO outlook graph also takes an oddly abrupt upward turn away from La Nina.

      Delete
    2. The mean forecast has temps climbing slightlyfrom August. All BoM models dip downward through June.

      http://www.bom.gov.au/climate/poama2.4/poama.shtml

      BoM modeling is somewhat of an outlier among international models, which forecast cooler 3.4 SSTs through the year (as you probably know, JCH).

      Delete
    3. Nick's page with the SST movies shows blobs of warmth forming off the West coast of S America over the past couple of weeks along the Nino line.

      https://moyhu.blogspot.com.au/p/sst-regional-movies-as-described-here-i.html

      Delete
    4. No, I did not really know that, but the heat boils you're referring to show up on most SST maps. The equator is blue to the surface, but it's a very thin line.

      Delete
    5. I think the hot blobs are intrusions of warm water coming in from the north, that looks warm anomaly-wise in the normally colder waters along the equator..

      Delete
    6. Its simple the weak trade winds at the moment: http://www.atmos.albany.edu/student/ventrice/real_time/timeLon/u.anom.30.5S-5N.gif the SST are very sensitiv about the wind, because this is what cause upwelling and if upwelling is less then before the SST can warm a lot. We now will see stronger trade winds, Nino3.4 and Nino 3 should go down or become negativ in the coming 5 days

      Delete
    7. That is what I think... it takes an undisturbed surface layer, not much movement and not much upwelling... then the energy coming from the sun can build up rapidly in the surface layer.

      Delete
    8. Barry,
      I don't know any source of daily data worked out for Nino3.4. I could do it from the NCEP/NCAR data, and I'll try to get time to do that today.

      Delete
    9. Barry,
      I did the daily Nino3.4 average. It has certainly flattened, but I don't see a big recovery. Graph is here, numbers in text format here.

      Delete
    10. Bounce-backs of this magnitude during a declining el Nino are not unusual.
      Here's a graph and underlying data from 1998: http://climexp.knmi.nl/data/inino34_daily_1998:1998.png
      http://climexp.knmi.nl/data/inino34_daily_1998:1998.dat
      There is a larger one in mid April-early May and a small one in mid June, for instance.

      Delete
    11. Thanks, Nick, and belatedly for clueing me weeks ago on which regression models handle autocorrelation.

      Delete
    12. I sometimes wonder if many people understand the implications of autocorrelation. If a behavior like ENSO shows oscillatory behavior that is determined by another periodic physical cause, then the time-series is completely autocorrelated. In other words, the internal correlation of any point on the waveform is correlated with points prior to that time -- it's correlated with itself, therefore AUTO-correlated.

      The following figure shows how a fitted 30-year interval of ENSO has enough information to be able to fit the other parts of the time-series.
      http://imageshack.com/a/img923/166/81PjCe.png

      This time-series is highly autocorrelated and if somebody claims that one must remove the autocorrelated part of the signal, think carefully as you may be throwing away the most important part. YMMV

      Delete
  3. A wordy narrative does not describe ENSO -- that's like stating a bird's color to explain how it flies. Instead what you do is this: Take the original Laplace Tidal Equations, simplify along the equator, and apply a seasonally aliased lunar forcing to get the dynamics. Odd that no one ever applied this approach; yet looking at the history, if there is anyone to blame it would be Richard Lindzen. He could have figured this out sooner and avoided sending everyone down a blind alley filled with heinous math for the last 40 years.

    ReplyDelete
  4. My congratulations to Nick for this and lots of other plots inserted by Anthony in a recent home post:

    https://wattsupwiththat.files.wordpress.com/2016/06/rss-global-temp-stokes.png

    That is in my opinion a real success.

    And I didn't wonder to see somewhat "below" that even WUWT commenter Javier, all but a warmist, presented another moyhu graphics:

    https://s3-us-west-1.amazonaws.com/www.moyhu.org/data/freq/latestT/mth2.png

    Magnifique!

    ReplyDelete