Instrumental Record -vs- Satellite Record

[Skyhunter]

Last April is an interesting month.

The http://www.ncdc.noaa.gov/oa/climate/research/2009/apr/global.html#temp" ranks last April as the fifth warmest on record. The lower troposphere (<8 kilometers) however is ranked eleventh by RSS and fifteenth by UAH.

Why the large discrepancy?

Should we expect the surface temperature (< 1.5 meters) ranking to correlate somewhat with the lower troposphere?

 

[sylas]

Last April is an interesting month.

The http://www.ncdc.noaa.gov/oa/climate/research/2009/apr/global.html#temp" ranks last April as the fifth warmest on record. The lower troposphere (<8 kilometers) however is ranked eleventh by RSS and fifteenth by UAH.

Why the large discrepancy?

Should we expect the surface temperature (< 1.5 meters) ranking to correlate somewhat with the lower troposphere?

You should expect a fairly strong correlation, but with some significant differences. It is, after all, a different thing being measured; not merely different ways of estimating the same thing.

Note that the rank order of data points is of very little use for looking at correlations.

One major difference between atmosphere and surface is that air temperatures respond much faster to any transient effect. The surface record is dominated by the ocean, where there is a powerful lag from the huge heat capacity of the oceans. Put simply, the surface record looks a lot like the lower atmosphere, but with additional damping. Another more subtle source of difference is changes to lapse rate as the Earth's climate is shifting state. This changes the gap between surface temperatures and atmospheric temperatures.

Finally, you need to keep in mind the nature of the data we are looking at. It's very common to present summaries of the data and the trends for a legitimately interested public, in which there is a single global "temperature" value, changing with time.

In fact, the underlying datasets are thousands of different temperatures, from all over the planet. There are two major steps in constructing a single combined value. First, you convert from "temperature" to "anomaly". Then you average up all the anomalies to a single global value. This is quite a robust procedure, but it does mean there are small differences obtained depending on how these steps are performed. There are several groups around the world involved in producing global anomaly datasets, working independently of each other but sharing much of the same underlying data. The differences between calculated results are generally fairly small, but it is enough to alter "rankings" of individual months.

As an added complication when dealing with the atmosphere, there are all kinds of instrument difficulties which lead to much larger uncertainties than a surface measurement. The main sources of data for the atmosphere are radiosondes (weather balloons) and satellite data. Each has their own unique sources of error.

Finally, the satellite record used by RSS and UAH is the focus of a long running and still unresolved question of how to combine values from different satellites into a single record. This is something to watch for the future. I suspect we may see progress on resolving discrepancies to show up in the literature within the next year or two.

Looking at the data

The NCDC pages have links for obtaining data. Unfortunately, the surface data from NCDC is from their ftp site that I have been unable to access for a few months now. I don't know why. But I can get the GISS and HadCRUT3 surface data. There are small differences between these surface records, and the NCDC surface dataset tends to lie somewhere between them. The differences are not significant in the sense that they are less than the uncertainty bounds with each dataset. There is a clear explanation of the reasons for differences in the Frequently Asked Questions at Met Office Hadley Center (UK). It's near the bottom, as an answer to "What are the differences between HadCRUT3, GISS and NCDC global temperature analyses??

There are significant differences between surface and lower troposphere anomaly data, since this is a genuinely different quantity. The overall correlation remains strong. The atmospheric data can be obtained using links given at another NCDC page: http://www.ncdc.noaa.gov/oa/climate/research/msu.html (bottom of page).

Here's a plot of the datasets. I've shifted the two surface datasets vertically by adding 1 to GISS, and 2 to HadCRUT3. This let's you see the overall shape.

Correlations are:

$$\begin{array}{l|rrrr} \mbox{all months} & \mbox{HadCRUT3}& \mbox{UAH-LT}& \mbox{RSS-LT}\\ \hline \mbox{GISS}& 0.88& 0.82& 0.81\\ \mbox{HadCRUT3}& & 0.81& 0.87\\ \mbox{UAH-LT}& & & 0.95 \end{array}$$​

You are looking simply at April. Looking at a particular month can help get around issues of how seasonal variation is handled, but it also means you look at much smaller datasets. Anyhow, here is the data from April only.

Note that values for 2009 are still preliminary -- and that HadCRUT3 does not include April in their online dataset. Here are correlations for the April timeseries:

$$\begin{array}{l|rrrr} \mbox{April}& \mbox{HadCRUT3}& \mbox{UAH-LT}& \mbox{RSS-LT}\\ \hline \mbox{GISS}& 0.86& 0.78& 0.79\\ \mbox{HadCRUT3}& & 0.83& 0.9\\ \mbox{UAH-LT}& & & 0.97 \end{array}$$​

Here are a few observations about this data.

1. The RSS and UAH lower troposphere figures show the highest correlation. That is because they are working from the same underlying satellite data. The differences are due to methods for obtaining a lower troposphere temperature from the microwave sounding units, and especially to how they merge data from different satellites with different calibration errors. These differences are a focus of a lot of work in the various groups working with satellite measurements.
2. The biggest difference between RSS and UAH that you can see in these plots is that after about 1992, there is a jump, with UAH returning lower values than RSS. It's not the only difference, but this is the main cause of different rankings for April using these two tropospheric datasets.
3. The correlation of GISS and HadCRUT3 is strong, but not as strong as the troposphere correlations. This is mainly because of differences in how they handle parts of the globe with poor data coverage. HadCRUT3 tends to leave it out. GISS attempts an interpolation. Over long time spans this tends to even out, but it does mean some short term variation in the two datasets.
4. The strongest correlation between surface and lower troposphere is seen with HadCRUT3 and the RSS data.
5. The troposphere data does not get any input from the poles, due to the satellite orbits used. This is also the region with the lowest levels of coverage for surface temperature records. The HadCRUT3 dataset therefore tends to omit information from the poles, and this might contribute to a better correlation of troposphere temperatures. But we can't be sure until people sort out the reasons for the differences between RSS and UAH.

Cheers -- sylas

 

[Skyhunter]

Thank you Sylas.

That was an excellent and comprehensive answer.

 

[Wagmc]

As an added complication when dealing with the atmosphere, there are all kinds of instrument difficulties

When dealing with atmospheric measurement by satellite, you are dealing with only one observation point, the satellite, taking measurements over the globe. To correct for factors like orbit degradation, you only have to adjust for the single platform.

I would expect exponentially more complications when dealing with the thousands of surface measurement stations, including:

a) siting issues. It is well documented that >70% of stations surveyed do not meet established criteria for location next to parking lots, roads, buildings, air conditioners, at the sewage treatment plant, etc.

b) maintenance issues. Was the unit repainted? Is it calibrated? By whom?

c) measurement issues. In several locations, daily measurements are taken by prisoners. It was raining yesterday. Did Betty Sue really go take that measurement? Or did she "feel" like it was about the same as the day before? Are measurements taken at the same time every day? Many locations across the globe do not report daily; does this inconsistency imply a lack of attention to detail that may affect quality?

There is no possible way that GISS has taken these issues into account in their "adjustment" algorithm. Until Watts' work, no one had even documented the station siting issues in the US.

Overall, I think the surface record is contaminated by un-accounted for UHI. No significant warming in lower troposphere sat measurements should be a red flag. If the ground is warming, it should show up in the sat measures.

 

[Skyhunter]

When dealing with atmospheric measurement by satellite, you are dealing with only one observation point, the satellite, taking measurements over the globe. To correct for factors like orbit degradation, you only have to adjust for the single platform.

I would expect exponentially more complications when dealing with the thousands of surface measurement stations, including:

a) siting issues. It is well documented that >70% of stations surveyed do not meet established criteria for location next to parking lots, roads, buildings, air conditioners, at the sewage treatment plant, etc.

b) maintenance issues. Was the unit repainted? Is it calibrated? By whom?

c) measurement issues. In several locations, daily measurements are taken by prisoners. It was raining yesterday. Did Betty Sue really go take that measurement? Or did she "feel" like it was about the same as the day before? Are measurements taken at the same time every day? Many locations across the globe do not report daily; does this inconsistency imply a lack of attention to detail that may affect quality?

There is no possible way that GISS has taken these issues into account in their "adjustment" algorithm. Until Watts' work, no one had even documented the station siting issues in the US.

Overall, I think the surface record is contaminated by un-accounted for UHI. No significant warming in lower troposphere sat measurements should be a red flag. If the ground is warming, it should show up in the sat measures.

When looking at Sylas' charts it is very difficult to accept that there is a systemic error with the instrumental surface record.

And since the tropospheric trend of both UAH and RSS is < 0.15C per decade, it is a false assumption that that there is no significant warming in the lower troposphere. Instead of a red flag I see a single monthly anomaly. I am interested in learning why there is such a divergence, even if it only for a single month.

I see no reason at all to take a single month's initial analysis as evidence of a larger problem that just is not supported by comparative analysis.

 

[chriscolose]

Anthony Watts has no desire whatsoever to point out real issues with the instrumental record, he just wants to declare jihad on anything which can support human-induced climate change, as clearly evidenced by the stuff which appears on his site. Volunteer pictures with cute arrows pointing to air conditioners is not a suitable approach. Real scientists working with this data are not ignorant to real problems, and it's a good bet that people who set up these weather stations and handle the data are not incompetent. Watts is doing absolutely ntohing to help the cause and fix real issues or to suggest ways to handle them. Also, good posts by Sylas. There are some good references for looking over stuff related to the topic, two of which are

Brohan, P., J.J. Kennedy, I. Haris, S.F.B. Tett and P.D. Jones (2006). "Uncertainty estimates in regional and global observed temperature changes: a new dataset from 1850". J. Geophysical Research 111: D12106

http://www.climatescience.gov/Library/sap/sap1-1/finalreport/default.htm

 

[joelupchurch]

Dr. Roy Spencer, who does the UAH, points out a few reasons for differences between UAH and RSS numbers in his blog:

http://www.drroyspencer.com/2009/05/april-2009-global-temperature-update-009-deg-c/"

The most important I think is that UAH goes down to 84S, while RSS only goes to 70S, so there can be significant differences because of Antarctic conditions.

 

[sylas]

Dr. Roy Spencer, who does the UAH, points out a few reasons for differences between UAH and RSS numbers in his blog:

http://www.drroyspencer.com/2009/05/april-2009-global-temperature-update-009-deg-c/"

The most important I think is that UAH goes down to 84S, while RSS only goes to 70S, so there can be significant differences because of Antarctic conditions.

The largest differences, by far, are in the tropics.

Another curious issue is a strong annual (seasonal) signal which starts from about 1992, in the DIFFERENCE between RSS and UAH records. It is very marked, and a number of people (including a few amateur enthusiasts who like to play with datasets!) have identified this as an issue. The most credible reason for this, I guess, is that one of the satellites has a calibration issue that varies on an annual basis. We know they have such errors already. The issue will be getting the corrections accurate enough. Someone may have over or under corrected one of the satellites. The problem has not been identified as yet, and until it is you can't tell whether it is the RSS or the UAH dataset that is most distorted by the error!

For now, we should note that the troposphere record is very much unfinished business. Stay tuned. People are looking hard for the problem. It will probably get sorted eventually.

I've been meaning to post a longer discussion of the sources of errors in satellites and surface records. It's a huge topic with a massive associated literature. (And most definitely all the issues have been known and documented and considered long before Alan Watts ever came along.)

There's one point on this which is very interesting on a personal level...

Dr Christy is a odd case, and Dr Spencer an even odder one. For onlookers, it is all too easy to classify these guys as denialists and ignore them, or (for the other side of the debate) to hold them up as the experts who will overthrow the IPCC.

My own views on that completely aside, I have been most interested to see how closely and constructively the RSS and UAH teams work together, despite being rivals. Each group has helped in finding errors in the work of the other, and as far as I can see each group actively solicits constructive criticism from the other. I'm going to post more on the whole matter of errors in the satellite record, but in the meantime... people like us here are not going to be able to give a conclusive resolution. The scientists don't have a resolution. The differences between UAH, and RSS, and the radiosondes, is all up in the air.

As I noted previously, the errors and uncertainties are much MUCH greater with the satellite data than the surface data -- even including all the horrible surface station siting issues. Explaining that is what I am hoping to tackle in another post.

When the problems with satellites are solved to the extent that people know why RSS and UAH have different trends, especially in the tropics, I expect there will be mutual recognition... just as there has been mutual recognition of some really embarrassing errors in the past. And still each group appears to hold the other in high regard for technical competence.

It's a good example of how scientists can put aside larger differences when they are hard at work on a specific empirical question. Why are RSS and UAH numbers so different? We don't know.

Cheers -- sylas

Postscript. I looked at the blog. The point #3 is significant. One of the satellites has a substantial decay in the orbit (and the RSS guys also talk about this!). RSS uses it, with corrections. UAH doesn't use it. This would explain a seasonal effect in the differences between the datasets. If this is the cause of the seasonal signal in differences, then the question becomes... does the RSS dataset get additional overall accuracy from using corrected data?

 

[Gokul43201]

Just an FYI - here's a link to a resource that let's you plot historical data on a number of climate proxies, and do some simple analysis (running means, least square fits, etc.). I can not assert that it is bug-free, so be forewarned. But so far, I haven't found any errors. For instance, the trend I get for UAH from this is 0.13C/dec, which agrees with their published value (bottom of http://vortex.nsstc.uah.edu/public/msu/t2lt/tltglhmam_5.2 ) of 0.127C/dec. Before this, I had to plot and analyze data by myself, which can sometimes be a little tedious, especially if it's just something simple ... but this makes life a lot easier for some of that stuff.

http://www.woodfortrees.org/plot/

Some examples:

http://img36.imageshack.us/img36/6818/picture7t.png

http://img33.imageshack.us/img33/6097/picture10p.png

http://img29.imageshack.us/img29/6588/picture9y.png
I suggest this only as a tool to use for your own purposes, and not as a valid reference for supporting an assertion in PF.

 

[joelupchurch]

The largest differences, by far, are in the tropics.

There's one point on this which is very interesting on a personal level...

Dr Christy is a odd case, and Dr Spencer an even odder one. For onlookers, it is all too easy to classify these guys as denialists and ignore them, or (for the other side of the debate) to hold them up as the experts who will overthrow the IPCC.

My own views on that completely aside, I have been most interested to see how closely and constructively the RSS and UAH teams work together, despite being rivals. Each group has helped in finding errors in the work of the other, and as far as I can see each group actively solicits constructive criticism from the other. I'm going to post more on the whole matter of errors in the satellite record, but in the meantime... people like us here are not going to be able to give a conclusive resolution. The scientists don't have a resolution. The differences between UAH, and RSS, and the radiosondes, is all up in the air.

It's a good example of how scientists can put aside larger differences when they are hard at work on a specific empirical question. Why are RSS and UAH numbers so different? We don't know.

As far as UAH and RSS are concerned, it is a little depressing when people are surprised when scientists actually act like scientists instead of needing a court order to release data to skeptical researchers.

I've read a lot of what Dr. Christy has written. I think he is mostly critical of legislation that doesn't meet any rational cost-benefit analysis. Spending hundreds of billions or trillions of dollars to reduce warming by .01 degrees by 2100 is a non-starter for him.

He is hardly unique among climate scientists in this regard. Google Dr. James Hansen comments about cap and trade sometimes. Also note that they both agree about the importance of developing nuclear power to replace carbon based energy.

I think his missionary work in Africa also gives him a different perspective. Expensive energy impacts our lifestyle, but in some countries it is the difference between living and dying of starvation.

 

[joelupchurch]

I forgot to mention Dr Christy's emphasis on land use as a climate problem. Deforestation and top soil erosion are degrading the ability of the environment to sequester carbon. It seems to me that is just as important as regulating emissions and may give a much better bang for the buck.

The important thing is to buy enough time to reduce our dependence on carbon based power to a level the environment can absorb and start the transition from coal and oil power to nuclear power.

 

[Skyhunter]

Thanks for the link and the charts gokul.

It seems that the largest difference between troposphere and surface temperatures is greatest at peaks and troughs.

 

[sylas]

Thanks for the link and the charts gokul.

It seems that the largest difference between troposphere and surface temperatures is greatest at peaks and troughs.

That is the biggest difference... and also the most obvious and expected. The atmosphere responds far more quickly to any impact than the surface. Hence, when you plot graphs that are normalized to the same baseline, what stands out is the larger swings in atmospheric temperatures.

What is of major interest, over, is a comparison of the trend. Theoretically, we should expect a larger trend in the atmosphere, up until fairly high in the troposphere, where the trends reverse again. Note that this is entirely independent of greenhouse effects; it is a general question about the physics of the atmosphere for ANY prolonged temperature change.

That is... for an increase in the mean temperature at the surface by 0.1C, you expect more like 0.12 or 0.15C increase in the mean at the atmosphere. You cannot see this until you abstract AWAY from the short term variation. The UAH group records smaller trends in the atmosphere... especially in the tropics. The RSS group records a larger trend. And the radiosondes lie somewhere in between; although they have errors of their own.

Unfortunately, the woodfortrees pages don't let you calculate trends as easily as one would like, and you can't look at particular zones... the difference is greatest in the tropics. The first graph with trend lines for rss and uah does show the guts of the problem, however.

Cheers -- sylas

 

[Gokul43201]

It seems that the largest difference between troposphere and surface temperatures is greatest at peaks and troughs.

This may actually be true in this particular case (for the reason given by sylas), but keep in mind that differences always appear more distinct near regions where the slope of the curve is large with respect to the direction along which the difference is being eyeballed.

For instance, where is the difference between the two curves plotted below the largest?

http://img43.imageshack.us/img43/3193/picture13w.png

Spoiler

It's exactly the same everywhere. Those are a pair of sine waves with a constant y-offset of 0.1, but it looks like the differences are most pronounced at the peaks and troughs. If on the other hand, you had a pair of sine waves with no phase difference or offset, but different amplitudes, then the differences would indeed be maximal at peaks and troughs. This latter illustration would better represent the baseline adjusted anomaly curves for surface and LT temperatures, but the former may be more useful for illustrating the differences between UAH and RSS curves (and serves to provide a warning in general, against the inherent dangers in eyeballing differences).

 

[Skyhunter]

They appear the same.

A good point gokul however, both LT curves have a greater amplitude than the surface temps.