Overall long term trend remains positive in lower troposphere.
Most often on this forum we have looked at either surface temperature data from surface observations or lower tropospheric temperature data derived from satellite sounders. Today I’d like to point out a short scale trend in global radiosonde data showing cooling in the last year, as well as examine the record back to 1958.
The HadAT2 dataset from the Hadley Climate Center takes in balloon radiosonde measurements taken twice daily from hundreds of points around the globe and compiles it. Here is how they describe it:
HadAT consists of temperature anomaly timeseries on 9 standard reporting pressure levels (850hPa to 30hPa). The data is also available as equivalent measures to the broad MSU satellite weighting functions. The gridded product is derived from 676 individual radiosonde stations with long-term records. Because of the criteria of data longevity the resulting dataset is limited to land areas and primarily Northern Hemisphere locations. Radiosondes are single launch instruments and there have been many changes in instruments and observing practices with time. HadAT has used a neighbour-based approach to attempt to adjust for these effects and produce a homogeneous product suitable for climate applications.
They also go on to add a cautionary note about data uncertainty:
It is important to note that significant uncertainty exists in radiosonde datasets reflecting the large number of choices available to researchers in their construction and the many heterogeneities in the data.
And they go on to suggest alternate data sets for “robustness”. For now, we’ll just stick to HadAT2, but if readers want to do comparisons against the other datasets I’ll post results here. Just visit the HadAT2 page for links.
Here is the plot of all the pressure altitude levels of temperature data since 1958:
Click for a larger image
The source data set in ASCII text is available here
In the graph above, the warmer (redder) colors represent lower tropospheric data closer to the surface (850mb for example) while the cooler blues (cyans) are the high altitude data (100, 50, and 30mb). You can see in the 850mb data, the familiar signature of the 1998 Super El Nino that raised temperatures globally.
You can also see the slow upward trend in temperature in the lower troposphere data since 1958, about 0.6°C.
To give laymen readers an idea of the vertical scope of the plot above, here is a graphic showing pressure versus atmospheric altitude.
Graphic Source: PhysicalGeography.net
Now what is interesting is when we zoom the data plotted above down to a five year level, as shown in the graph below.
Click for a larger image
Note that while preceding years have been relatively flat for trend, during the last 12-18 months, there has been a noticeable downward trend in all atmospheric levels except 50mb and 30mb, while 100mb appears to remain flat. The 50mb and 30mb levels don’t appear to have much of a positive trend in the last 12-18 months that differentiate it from the last 5 years.
For those who will immediately jump on the standard gripe of “cherry picking” let me say that I’m only using the zoomed 5 year time period above to better visually illustrate the change in the last 12-18 months. As I mentioned above, the overall long term trend since 1958 in the lower troposphere is still positive.
But whatever has happened globally in the last 12-18 months, the temperature downturn we see makes for interesting discussion.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Anthony,
You don’t mention the blue color peaks in the long term record. Looks like St. Helens, El Chi, & Pina are all represented.
Could the blue peaks, then declines in the early years be tied to nuc tests going underground (US & USSR last above ground tests in 1962-63)?
If the atmosphere is cooling short-term, and the seas are cooling short-term, warmists can call it short term variation, but they still need to answer the question: How and why is heat leaving the system?
It seems to me that under the “persistent forcing” view of things, the total heat in the system should be steadily increasing as CO2 levels increase.
On annual forecasting:
“Over the eight years, 2000-2007, since the Met Office has issued forecasts of annual global temperature, the mean value of the forecast error was just 0.07 °C.”
http://www.sciencedaily.com/releases/2008/01/080104091616.htm
That’s pretty darn good forecasting!
These will be the “pre-adjusted” figures then?
Doesn’t this also show quite significant stratospheric cooling? Judging from your altitude/pressure graph, I reckon the 50hPa line is at 25km altitude, in the lower stratosphere. The corresponding line on the temperature graph shows significant cooling – upwards of 2 degrees in 50 years. Isn’t this one of the models’ predictions for greenhouse warming? (I’ll grant you, of course, that the other more publicized one – the upper tropospheric “hotspot” – isn’t there.)
OT Epilogue to Polarstern thread from August 8-08:
Science Daily offers the story today, in a comprehensive translation:
http://www.sciencedaily.com/releases/2008/08/080811092458.htm
Again, the best way to monitor Arctic Sea Ice conditions day by day is the excellent overview offered here, as posted by several contributors in the past – one can enter any two dates for comparison:
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=08&fd=14&fy=2005&sm=08&sd=14&sy=2008
Right now, the obvious anomaly is a great deal of open water in the Beaufort Sea (creating sufficient fetch for wave build-up, making current oceanographic and seismic research more difficult). Three years ago, the open water was in the Chukchi and Laptev Seas – this illustrates circulation and wind patterns…
It will be interesting to observe what the next three weeks will bring.
For perspective, please check this outrageously flippant column by Thomas Friedman in the International Herald Tribune (I posted it before):
http://www.iht.com/articles/2008/08/06/opinion/edfriedman.php
his words: “Remember: Climate change means ‘global weirding,’ not just global warming.”
we’ll see – Mr. Celebrity Columnist.
If I had time, I would research Traditional Ecological Knowledge, talking to Inuit Elders about the warm spells of the 20s and 30s, among others – anybody knows of any oral history accounts?
Off topic – Hadcrut 3 is out for July +0.403
It would be interesting to see the trends for the 850mb-500mb thickness values. The last 12 months trend is interesting. It does illustrate the signifigance of ENSO on our global temperatures.
Also, what caused the huge decline in 1972-74 @ur momisugly 30mb? It doesn’t look like there’s been any change to note of since then at 30mb.
Anthony,
I would be interested in the algorithm used in the adjustment of the values. What caught my interest was the statement “neighbour-based approach” for the adjustment of error. Most of the radiosonde reporting stations in the US are hundreds of miles apart. I am very suspicious of adjustments made with such distant samples. The adjustments would have to account for significant differences in geography and current weather conditions requiring a very dynamic and analytical evaluation of each observation performed. Something I would have to have shown to me as being taken into account. Yes I am skeptical, there was once a time when that was a good quality, but when it concerns climate it is now a bad word.
REPLY: I suspect it is much like the GISS homogenization routine, perhaps even identical.
Perhaps I’m missing the point, but pray tell – what does a 12-18 month trend make for “interesting discussion?” Is the trend significantly significant given the recent data before it or something? How does a 12-18 month trend during a period which saw the evolution of ENSO-neutral to La Nina conditions as well as a possible reversal of the PDO mean anything in terms of climate change?
Two typos in that above post: what->why; significantly->statistically. This is what I get for not having a cup of coffee this morning.
One of the signals of global warming is cooling in the upper stratosphere. The theory is that additional GHGs in the lower atmosphere intercept IR photons from the surface before they reach the stratosphere.
While there has certainly been a reduction in temps in the stratosphere since 1958, as Gary in Olympia posted, it appears that the 3 major volcanic events, Agung in 1962, El Chichon in 1982 and Pinatubo in 1991 are the major drivers of these swings.
Temps increase temporarily by 1.0 to 1.5C from the volcanoes followed by a cooling of 1.25C to 1.75C within two years and the stratosphere temperatures seem to stabilize and do not change at all afterward.
There has been no real change in stratosphere temps since 1994.
[Personal opinion here – Of course, I can’t really buy the argument that global warming will lead to stratosphere cooling. There must eventually be a balance between the radiation coming to Earth with that leaving the Earth. Global warming could be thought of as a delay in the average time it takes for a photon from the Sun to escape the Earth. On everage, the delay is about 45 days (which leads to Greenhouse warming of 33C.) If global warming increases that delay by 1 day, then we will have 34C of warming. In this sense the stratosphere should not be cooling, it should be a steady temp reflecting the loss of energy required to keep in Earth in balance.]
Global warming theory has not adequately explained the impact of volcanoes on the stratosphere (and how these are the main drivers) nor why the stratosphere should continue cooling in the theory.
Sorry to be off topic, but I’ve just noted that the BBC are planning a definitive History of Climate Change programme this autumn: http://www.bbc.co.uk/pressoffice/pressreleases/stories/2008/07_july/10/bbctwo4.shtml#history refers.
I wonder if anyone reading this blog has been contacted? Thought not. Perhaps someone should try to get the hockey-stick story over to them, at least, if not the surface stations fiasco…
REPLY: They’ve already asked about the surfacestations, and I provided photos. But yes, perhaps sending over the hockey stick fiasco would be a good idea -Anthony
AustinS asks… “Doesn’t this also show quite significant stratospheric cooling? …the temperature graph shows significant cooling – upwards of 2 degrees in 50 years. Isn’t this one of the models’ predictions for greenhouse warming? (I’ll grant you, of course, that the other more publicized one – the upper tropospheric “hotspot” – isn’t there.)”
Let me also indicate interest in having this question answered in some detail. AGW does postulate that the CO2 warms the troposphere while cooling the stratosphere (the CO2 greenhouse effect traps the heat near the surface instead of leting it escape thus cooling the stratosphere), so how is this altitude cooling data not some indication that the AGW hypothethis has some validity? Enquiring minds like this non-ideological skeptic want to know!
Seems to me the increase in low altitude temps since 1958 could be related to UHI (Urban Heat Island) growth. Since these sites represent northern hemisphere land for the most part, and many are next to sources of urban heat, the increase cause due to UHI is plausable. As I eyeball the altitude chart the low 850mb altitude is a mile or so or less. Anyone who’s been in a light plane at those altitudes above a city has experienced bumps from the heat rising.
While digesting the info presented in the graphs, I found I kept having to ‘translate’ the key. As displayed, the key seems to be counterintuitive — when viewing the key and reading it from top to bottom it’s actually displaying data in reverse, from low altitude at the top and on up as working down the key. I’d suggest that the top of the key should start at 30mb and progress down to 850mb.
Everyone’s a critic…
How does the number of plots for 50MB & 30MB stack up with the lower levels? Balloon
failures are quite frequent above 100MB.
counters:
You must remember, for a decade we’ve been lectured that CO2 and other greenhouse gases overwhelm other “natural” climate forcings. Things like ENSO, solar variations, etc… are driven or at least are secondary to the build up of GHGs. Esteemed scientists like Hansen have repeatedly stated that we have less than a decade to act, that we are ever so close to the “tipping point” and the danger of endless positive feedback loops. Other than that, I agree with you; 12-18 months worth of data may be interesting but not necessairily earth shaking. But don’t blame Anthony. Dr Hansen et. als. started this train.
dreamin (03:10:35) :
‘If the atmosphere is cooling short-term, and the seas are cooling short-term, warmists can call it short term variation, but they still need to answer the question: How and why is heat leaving the system?
It seems to me that under the “persistent forcing” view of things, the total heat in the system should be steadily increasing as CO2 levels increase.’
Didn’t you get the memo? The CO2 ampification effect (2.5) went on vacation and will be back in 2015.
I made a mistake in my earlier post with respect to 1972-74. The steep dropoff at 30mb i’m questioning is in 1992-1994, which i assume would correspond to pinatubo. But then , why doesn’t it equalize like it typically did after the other eruptions? What caused it to reset to the lower value (~-1.5) and stay there?
My first thought is an instrumentation shift…
and also Micheal don’t forget the increase in lower level heat trapping clouds that we have in times of low sunspot activity…adds up to me. but at the end of the day its the whole system, from the bottom of the oceans to the edges of the atmosphere, we can plot trends at specific points, but need to understand the whole picture.
OT – some of you may be interested in reading my new article at The Register
http://www.theregister.co.uk/2008/08/15/goddard_arctic_ice_mystery/
REPLY: I’m reposting it, Anthony
Austin S says: “The corresponding line on the temperature graph shows significant cooling – upwards of 2 degrees in 50 years. Isn’t this one of the models’ predictions for greenhouse warming? (I’ll grant you, of course, that the other more publicized one – the upper tropospheric “hotspot” – isn’t there.)”
The two predictions are not really equivalent. The prediction of the upper tropospheric hotspot is a general prediction by the models (and basic theoretical considerations) of how warming in the tropics should occur independent of the mechanism causing the warming. I.e., it should be there for GHG warming but also warming due to increased solar luminosity or due to fluctuations.
The prediction of cooling in the stratosphere, by contrast, is a prediction specific to the mechanism of GHGs. An increase in solar luminosity, for example, is not expected to produce such cooling.
JP says: “You must remember, for a decade we’ve been lectured that CO2 and other greenhouse gases overwhelm other “natural” climate forcings. Things like ENSO, solar variations, etc… are driven or at least are secondary to the build up of GHGs.”
We don’t have to remember it because it simply isn’t true! You are mischaracterizing the scientific claims. The claim has been that the general climate trends over periods of decades are now dominated by the greenhouse gas forcings (although they weren’t, say, 50 years ago). However, there has never been a claim that on shorter timescales of a few years there wouldn’t be the normal ups and downs due to internal variability (such as ENSO). In fact, individual runs of climate models themselves show such variability, not a monotonic increase in global temperatures from year-to-year. See, e.g., here: http://www.realclimate.org/index.php/archives/2008/05/what-the-ipcc-models-really-say/langswitch_lang/sk
“You must remember, for a decade we’ve been lectured that CO2 and other greenhouse gases overwhelm other “natural” climate forcings over multi-decadal, long periods of time.”
That bolded part I added in is much more important than the first part. As a matter of fact, the first part is false except in the context of the bolded addition.
REPLY: Actually no I don’t think it is false. Here is why I think that. By the theory presented, the supposedly larger CO2 forcing would eventually dampen and “swallow up” smaller forcings, so that they are not as pronounced. There has been quite a bit of conjecture that this has already occurred. Hansen I believe wrote something on it a couple of years ago, though I don’t have the citation at my finger tips.
An analogy would be to think of the effect like a toy balloon. If the total forcing represents the inflation volume of the balloon, then when the balloon is empty, a small amount of air (the lesser forcings) would make a visibly noticeable change in diameter. However, when the balloon is well inflated, representing the expected larger forcing of CO2, the same addition or subtraction of air volume into the balloon would not be nearly as noticeable, because the smaller volume is “swamped” by the existing larger volume. Granted this is not a perfect analogy, but it does illustrate the concept.
Another way to look at it would be a radio receiver. A strong signal tends to “swamp” the receiver and make weaker distant signals on the same frequency disappear.
The fact that we can still observe such changes, that are of significant magnitude, such as the 0.6C drop in satellite derived temperature in the past 18 months, or the radiosonde data above, tells me that the postulated CO2 forcings are still not in primary control of the atmosphere.