Global surface temperatures have essentially remained static since 1998 – a record el Niño year. The hiatus in land surface warming is real, unexpected, and puzzling. Recent anomaly data are shown below.
Comparison of CRUTEM4, CRUTEM3 and GHCN V3C. Error bars are those quoted for CRUTEM3.
Until 2010 CRUTEM3 was the IPCC reference land temperature data, and was used for the IPCC 4th assessment report in 2007. It is still updated and shows 1998 as the warmest year with no warming trend since then. GHCN V3C is in agreement with that conclusion. CRUTEM4 was released in 2010, and the main difference to CRUTEM3 was the addition of 628 stations near the Arctic where warming has been strongest. GISS also added a significant number of new Arctic stations. This sampling effect alone has moved the land temperature anomaly to slightly warmer values post 2000.
The blue dots show locations of the 176 stations that were removed in the transition from CRUTEM3 to CRUTEM4. The red dots shows the location of the 628 stations that were added
CRUTEM4 now shows 2010 as the hottest year and a small 0.05C/decade warming trend since 1998. In addition there also seems to have been some recent corrections made to station data enhancing this trend since I first made the comparison in 2010 as shown below.
Fig 2: Detailed comparison of temperature anomaly results from CRUTEM4 and CRUTEM3 as of 2010
Cowtan and Way(2013) also claimed that adding yet more arctic coverage would confirm some continued warming. As I understand it they essentially fill all empty Arctic grid cells with values interpolated from satellite data. However, do they do they also do the same in the Antarctic or the Sahara? However, since then this claim has essentially evaporated as shown below in their updated version V2. The figure below shows their latest result based on CRUTEM4 and compares to the other main temperature series.
Other main tempertaure indexes compared to V3C. Cowton and Way Version 2
The growing evidence of an unexpected hiatus in surface warming, now lasting 16 years clearly caused concern in the climate science community. AR5 skirted round the issue by essentially deciding that such 15 year pauses would happen occasionally due to natural variability. Of course a cynic might argue that natural variability has been tuned in the models exactly for that purpose. Phenomena like ENSO are not fully understood and cannot be predicted.
Fig 2: Comparison of CMIP5 models and observed temperature trends.
However even with natural variability included only 5% of CMIP5 model ensembles can reproduce such a pause. Furthermore the statistics behind the all important AR5 attribution statement depends on natural variability being essentially random. If not, then a proportion of the observed warming since 1950 would then be due to natural cycles such as AMO/PDO. In this case model predictions of anthropogenic warming would be too high to explain observed anthropogenic warming and the attribution statement would be wrong.
Already most of the CMIP5 model predictions for future warming are on the high side when compared to the exisiting combined land/ocean surface temperatures (HadCrut4).
CMIP5 model ensemble compared to obeservations (Hadcrut4)
A new proposal to explain the hiatus is that much of the excess heat from TOA radiation imbalance has instead been stored in the deep ocean and will reappear later to warm the surface. However this proposal implies that there are indeed natural ocean driven climate cycles. If so then how much of the rapid warming from 1970-2000 was really due to an upturn in these cycles? The next few years will be interesting.
29/5: updated to use C&W V2 anomalies.
Clive, is it clear how the new polar stations for Crutem4 were standardized to produce a baseline? How many years prior recordings did they use, for example? Do you feel their method was congruent to their already established methods, or do you see possible bias?
Each station is normalised to its 12 monthly averages between 1961-1990. So all stations must have some minimum coverage in that period. The monthly anomaly is the measured value minus the ‘normal’ value. Several of the new stations only started in the 1940’s to 1960’s so any bias would be by loading recent years with stations that are known to be warming. I wonder also if the UHI could be much greater in very cold regions. Buildings at 20C when the outside is -30C would give a larger effect.
Here is a good example of how bias may occur.
Data only available post 1960. You can view more stations on this page
Interesting major plinian eruptions are overcooled in CMIP: Krakatoa 1883,Agung 1963, Pinatubo 1991, proof that aerosol cooling is overestimated.
The cynic in me suspects that the aerosol forcing is tuned to the temperature data, and not vice versa. It has even been proposed that a series of recent small eruptions are the cause of the current hiatus.
How does the CMIP cloud compare with hadcrut if the deep volcanic cooling runs are taken out?
On this blog comment Greg Goodman shows two groups of CMIP5 models detrended against HADCRUT4 (the flat line) with markers for the three significant 20th century volcanoes.
Confirms my expectations, thanks!
Hans, I was reading a post on CA from 2007 re Cowtan and Way and saw your comment re skeptics still being in the storming phase of group dynamics. That struck me as I agree on that analysis. As a Boy Scout adult I am wondering if it is possible to get to the forming stage, and I decided to try a nudge here. I would appreciate your input as a climate science veteran.
“CRUTEM4 now shows 2010 as the hottest year and a small 0.05C/decade warming trend since 1998.”
I don’t think that’s right. My temp calculator is here. “CRUTEM” is CRUTEM4. I don’t see a trend less than 0.1 C/dec.
I find useful here to plot trend from variable start year (x-axis) to present. Gadget here. Here is CRUTEM 4 and NOAA
CRUTEM is a lot lower. But it isn’t until 2004-5 that it drops much below 0.05 C/Decade (and below 0), and that mainly reflects that on land, 2005 and 2007 were very warm. Here are some land/ocean trends:
Cowtan and Way still show higher trends before about 2004.
I just checked the values for CRUTEM4. In fact 2007 anomaly was 0.924C and 2010 was 0.918C ! So even though it is a meaningless record because the error is ~ 0.05, 2007 actually equals the 2010 record. Sorry – My trend was based on Hadcrut4 not CRUTEM4 – so you are right !
This post is fundamentally flawed in a number of ways – but i’m just going to start with a nitpick – it’s Cowtan and Way not Cowton and Way.
“Cowton and Way(2013) also claimed that adding yet more arctic coverage would confirm some continued warming. However, since then this claim has essentially evaporated as shown below in their updated version V2. This figure also compares to the other main temperature series.”
I don’t know what you’ve done wrong but I would re-look at the how you prepared your comparison because your Figure comparing our record with the others is off by a fair margin. The entire premise that our record is lower than the others at present is just plain wrong. Sorry to be blunt but you’re incorrect in your assessment and you’ve clearly made a mistake somewhere.
More importantly the first part of your thesis pertaining to Arctic warming is also incorrect. We have direct evidence of rapid Arctic warming in regions with reduced coverage including from satellite LST data (Comiso and Hall, 2014), AIRS data (see our updates) and atmospheric reanalysis (Simmons and Poli, 2015).
I would read through our update papers on Kevin’s website before making statements that do not reflect the preponderance of evidence…
Thanks for replying.
Sorry about the Cowtan spelling – fixed now.
I found a V2 series via Skeptical Science. One of the dangers of blogging is that you might pick up the wrong data. If so I apologise if I did so – and will correct. Looking again on the York site I find yet another version which is again slightly different. This is based on GHCNV3.
What is your normalisation? You seem to normalise to 1990 ? Is that correct ? The base figures I used were these ones but then renormalised to ~1975
1996,-0.075,0.018,0.018,0.005
1997,0.071,0.019,0.018,0.005
1998,0.217,0.019,0.018,0.007
1999,-0.008,0.019,0.018,0.007
2000,0.000,0.020,0.018,0.007
2001,0.138,0.021,0.019,0.007
2002,0.229,0.021,0.020,0.008
2003,0.217,0.021,0.019,0.008
2004,0.134,0.020,0.019,0.007
2005,0.292,0.021,0.019,0.008
2006,0.237,0.021,0.019,0.008
2007,0.257,0.021,0.019,0.008
2008,0.122,0.021,0.019,0.008
2009,0.266,0.021,0.019,0.008
2010,0.323,0.021,0.019,0.008
2011,0.208,0.021,0.019,0.008
2012,0.228,0.021,0.020,0.009
2013,0.237,0.021,0.019,0.009
2014,0.314,0.021,0.019,0.009
P.S. No problem about being blunt !
cheers
Clive
Sorry for mistake: I found your CRUTEM4 based anomaly set on the York site and updated the post accordingly. They are higher.
Question: Did you also in-fill other areas of poor coverage in Africa and Antarctica ?
Yes we infill completely – the paper (and updates) are freely available and it provides the information.
Pingback: Horror science fiction van de TU Delft | Klimaathype
That last one from Delft was nearly all double Dutch to me. I don’t see the point in posting it on an English language blog. Maybe the point could be summarised in English ?
From my perspective, all this is excellent science. But, as a byproduct of the discussion, it underscores the importance of authors and researchers providing for the interested public a set of open source computing codes, well documented, which will reproduce their results.
The funding agencies and their employers, presumably agencies of one sort or another, need to agree to pay for the time of these engaged individuals, even if the estimates for task activity prove to be understatements.
Hi: There have been huge forest fires in Australia, California, Canada and North India, Can we estimate what increase of CO2 ppmv they might be contributing to the current levels? Will they not worsen the speed of global warming?
By the way I have been recently enjoying your blogs on CO2 warming on Earth and Venus. I have just translated your Perl codes into Python scripts and they are working. I am now adding PyLab graphing commands to plot graphs directly. When they too work I will send you the scripts.
Cheers and Thank you,
Mandyam Anandaram
Thanks sounds great.