There are now more than 4000 stations reporting for November. TempLS mesh estimated the global temperature anomaly at 0.88°C, down from 0.922 in October. It is still a lot higher than any other month in the record (0.76°C in Jan 2007). TempLS grid was very similar. The report is here.
The result is in line with the Moyhu NCEP/NCAR index, which dropped by about 0.05°C, but again exceeded any other month by about 0.1°C.
There was warmth in Europe, N America (except western US) and Brazil. Cool around Mongolia and N Atlantic. More detail in the WebGL plot. This month, most land areas contributed equally, to the anomaly, along with a small rise in SST.
Since this month NCEP and TempLS mesh and grid seem to be in agreement, it seems likely that other indices will follow. GISS was 1.04°C in October, so whether it will exceed 1 is a near thing. I'd say likelier than not.
In other news, the weekly OiSST NINO measures seem to be just below their peak levels. The satellite indices were down slightly. However, NCEP in December has been high, and the forecast is looking very warm.
It just ain’t natural
3 hours ago
Posts
So a very warm month, but not a _record_ warm month?
ReplyDeleteOch, I can already hear the distant war cry of the MacWattite and the MacMonckton clans ringing out through the highlands, to the sound of the pipes...
NO WARMING SINCE OCTOBER 2015! FREEEEEEDOOOOOOM!
(With apologies to you, my Scottish relations and mad Mel Gibson)
I was already hit with "NO WARMING SINCE OCTOBER 2015! FREEEEEEDOOOOOOM!" at Climate Etc.
DeleteThe EL Nino has been full of surprises, and it may not be finished with delivering surprises. Could linger. Could regenerate. Could die in a week or two. People think it's 1997-98 all over again, and it may have little resemblance to that EL Nino.
Looking at Nick's reanalysis data for the first five days of December that appears to have been an awfully brief hiatus. (Back to the highlands with the clans, where they will stand ready until called.)
DeleteLate 1950’s – early 1960’s is a good analogue of what we can expect in the next decade.
ReplyDeleteYet another ambiguous what-did-she-mean? drive-by comment by Judith Curry. Is it merely a coincidence that 'wave' is 'vague' in French?
DeleteWell, she, and Tsonis, expected a flat, perhap even declining, GMST trend for another 20 years, and no, I do not think they expected JN-G's ENSO neutral surge, or the El Nino that has followed it. She clearly still thinks the 20-year thing is right and a La Nina dominate regime will return - the Kimikamikaze Wind.
DeleteCherrypickers Society present the pick of today.
ReplyDeleteQuite impressive trend.. What can we expect in the near future?
credit to Cowtan's trend calculator
I think I have a better one :)
DeleteIt's become a "skeptic" meme to quote AR4 SPM where it talks about a ~ 0.2degC/Decade trend over 'the next two decades'. Well, the SPM was published in February 2007 and taking linear trends from that date to October 2015 in GISS, NOAA, HadCRUT4 and Cowtan&Way results in a spread of 0.19-0.28degC/decade.
A quick replot of NOAA's annual anomaly data with a reasonable estimate for 2015 (well, what I thought might be reasonable based on the year-to-date came out at a stunning 0.88°C, 0.14°C above 2014) shows a 50-year trend (1965-2015) of 0.164°C/decade using basic linear regression, and little or no hint of a breakpoint visible within the noise envelope.
DeleteAnd lest I be suspected of cherry-picking 1965 as a starting point, 50 is a nice round number and the 1975-2015 trend is 0.166°C/decade and the 1985-2015 trend 0.168°C/decade.
I do it this way because it shows the warming trend increased during the 18-year period of warming. Trust me, normal people think they are being duped once they learn this. Then it shows the minor effect the pause had on the LT trend to 2014, and that the effect is dissipating.
DeleteSB - 18-year period of no warming.
DeleteSB?
DeleteSB = should be... I left out the word no in no warming. Old fart typing.
DeleteJCH, You can try this one also: http://www.woodfortrees.org/plot/uah/to:1999/last:360/trend/plot/uah/from:1969/to:2006/trend/plot/uah/from:1969/to:2014/trend/plot/uah/from:1969/trend/plot/uah/from:1969/mean:4/mean:8
ReplyDeleteThough skeptics may complain that this isn't the latest UAH revision (v6.0), UAH was alleged by them to be much more accurate than surface temperature reconstructions, even before the newest revision.
Apologies for the missing url tag. Try this instead.
DeleteSpeaking of cherries, it's still early but December is cooking so far. All but one of the latest eight NCEP/NCAR daily anomalies were above 0.7. This sustained period of warmth (however short) is unprecedented in the whole 15 months of the data set.
ReplyDeleteKarsten Hausteins month-to-date+forecast (right now reaching to Dec 16) show an anomaly that is 0.213 C above the October average
DeleteThat's pretty warm! I'm in Montreal (which was just about the only cold spot in the whole world in 2014), close to one of the forecast red spots. The current local temperatures and forecast for the next few days (MeteoMedia) indeed show minima and maxima both nearly 15°C above normal.
DeleteWhat would Nick's NCEP global anomaly have been on December 7,1941... Pearl Harbor. His December 4, 2015 is .843! There does appear to be some relief coming mid month.
DeleteSB 7th...
DeleteThe first week of gridded reanalysis data is 0.16° above the mean anomaly for October.
DeleteTurn down the damn thermostat, Nick. If the 2015 anomaly breaks the 2014 record by much more than 0.10°C my early forecast of 0.10 ± 0.01°C isn't going to look nearly as sharp as it could have been.
On the bright side the deniers will be denied the statistical escape hatched they tried to use for 2014. I look forward to Thomas Karl and Gavin Schmidt reprising their joint NOAA/NASA news conference to announce the hottest year on record, this time stating that the probability that 2015 was the warmest year is 99%. Maybe they'll politely disagree.
DeleteNASA: 98.5% likely
NOAA: 99.5% likely
Deniers: HAHAHAHA they can't even agree! And no warming since last week!! We win!!!
*hatch*
DeleteI am not really adept at this stuff, but I do not think the percentage can get that high as each year in the record, no matter how cold, gets a probability percentage. 2014 was the highest one; therefore, the warmest year. It will be interesting to see how big that number is this year. Can it exceed 50%?
DeleteThe NOAA uses a somewhat more sophisticated version of overlapping normal distributions to calculate the probability that a given year was actually the warmest on record given statistical uncertainties behind the annual anomalies and 'true' temperature, which it estimates at 0.02 to 0.03°C for recent years.
DeleteI won't try to replicate the NOAA procedure but will use overlapping normal distributions with a standard deviation of 0.0285°C for all years, which splits the difference between NASA and NOAA's 2014 probability and fits the other years tolerably well (2014: 43%, 2010: 26%, 2013: 10%, 2005: 8%, etc.)
Assuming that November comes in 0.04°C cooler than October as Nick's reanalysis suggests and that the Jan-Oct values don't change. I calculated the following cases for December: 0.04 below November, equal to November, 0.04 above November (i.e. equal to October), and 0.08 above November to obtain the following probabilities that 2015 will be the warmest year on record:
-0.04: 98.6%
0.00: 98.9%
+0.04: 99.2%
+0.08: 99.4%
I guess that is what they mean when they predict 2015 will be the warmest year "by a country mile." It's like Secretariat.
ReplyDeleteThe AMO took a dip in November.
The first week of gridded reanalysis data is 0.16° above the mean anomaly for October.
ReplyDeleteNow the Dec. average is 0.19 above October. Today's NCEP anomaly of 0.94 would seem to be a new record by a substantial margin. Nick's CSV file with the entire record of NCEP data doesn't seem to have been updated since mid October, but if the value on 6 Oct (0.865) was the previous record, this is quite a bit higher.
Prior to Oct 2015, the warmest day in the NCEP record was 0.672 (on 2 Dec 2013). Every single day so far this month has been higher than the highest single day in the entire 20+ year record from 1994 to Oct 2015.
Here's an updated version of the histogram I posted back in October. If this goes on we're going to need a bigger graph soon...
Well, that was a short-lived record. NCEP/NCAR for December 9 is 0.975°C.
DeleteThat's an interesting histogram of the 1994-date NCEP global anomalies... I wonder if the skewed shape is an artifact of the reanalysis data or its subsequent processing or if it's revealing a genuine underlying physical effect such as a change in the mean value with time. Where is the csv file containing the daily NCEP data located? I looked quickly but didn't spot it.
The zipfile is here. I'll set up a mechanism for regular updating and put the link on the data page.
DeleteThanks Nick. Much appreciated.
DeleteI have set the files to update weekly. I have added monthly averages.
DeleteI calculated monthly and annual averages myself to compare with NOAA and NASA global (land and ocean) anomalies. A plot is here http://imgur.com/4znUP9o
DeleteIt's similar to some of your own plots here, but I was interested in seeing the correlation between NCEP and NOAA and NASA anomalies in order to get a jump on the November, December and 2015 global anomaly.
One question/comment: I noticed that there are only 365 days for leap years 1996, 2000, 2004, 2008 and 2012, with no value given for December 31. I simply added the date and interpolated between Dec 30 and Jan 1.
Magma,
DeleteYes, my treatment of leap years should be improved. It's a complication, because I need daily norms to create anomalies, and it's hard to get one for Feb 29. But I could just interpolate.
As I said in the read_me, I just omit Feb 29, and use a 365 day year. I think that caused a problem in the CSV files, where I just generated the dates for printing directly, so it probably put Feb 29 back in, and shifted the rest by 1 day. I'll check. That would make a small difference to the month averages for leap years, if calculated using those dates. The ones I calculated didn't have that problem, they just omit Feb 29.
Oh right, the read_me.txt file (shuffles awkwardly, looks at feet).
ReplyDeleteI've fixed it now (and changed the read_me) so that leap years are there in full. New zip is posted. Only affects the 1994-2013 day file.
DeleteJMA November (preliminary) is out now with +0.54 C, up 0.01 from October, and a new monthly record anomaly.
ReplyDeleteThe best comparison is TempLSgrid, which is down by 0.01 from October. Small changes, they agree quite well..
Looks like November GISS is 1.05° and October has been revised up to 1.06°
ReplyDelete