TCS Daily

Some Convergence of Global Warming Estimates

By Roy Spencer - August 11, 2005 12:00 AM

In one of a trio of new global warming papers in Science, Mears & Wentz (2005) address what they consider to be a large source of uncertainty in our (University of Alabama in Huntsville, "UAH") satellite estimate for global lower tropospheric ("LT") temperature trends since 1979. The satellite measurements come from the Microwave Sounding Units (MSUs) and Advanced Microwave Sounding Units (AMSUs) flying on NOAA's polar orbiting weather satellites. The UAH estimate of the globally averaged trend since 1979 to the present has been +0.09 deg. C/decade, considerably below the surface thermometer estimate that has been hovering around +0.20 deg. C/decade for the same period of record.

This discrepancy between the UAH satellite LT trends and the surface thermometer trends has caused some consternation, since what we understand of atmospheric physics suggests that sustained warming at the surface should be amplified with height in the troposphere, not reduced.

Mears & Wentz, who are very capable remote sensing experts from Remote Sensing Systems ("RSS", Santa Rosa, California), found that the LT trend was particularly sensitive to the UAH method for removing the drift of the satellites through the local observing time. The satellites are launched into sun-synchronous orbits that are meant to cross over the same Earth locations at approximately the same time each day. But since the satellites do not have on-board propulsion, the satellites fall slowly back to Earth, which changes their orbital characteristics. In particular, what began as early afternoon observations from the daylight side of the "afternoon satellites" orbits drift to later in the day over the several years of each satellite's lifetime. This causes a spurious cooling trend as the Earth observations are made later in the afternoon to the evening.

The UAH method for removing this drift depended upon the spacecraft roll attitude (the accuracy with which it was pointing straight down, and not sideways) being almost exactly the same during the day side of the orbit as the night side. The new research paper presents Mears & Wentz's own estimate of LT trends using diurnal cycle corrections based upon a climate model estimate of the daily (diurnal) cycle of temperature at different levels in the atmosphere, on a global basis.

Their final estimate of the global lower tropospheric trend through 2004 is +0.19 deg. C/decade, very close to the surface thermometer estimate, and this constitutes the primary news value of their report.

While their criticism of the UAH diurnal cycle adjustment method is somewhat speculative, Mears & Wentz were additionally able to demonstrate to us, privately, that there is an error that arises from our implementation of the UAH technique. This very convincing demonstration, which is based upon simple algebra and was discovered too late to make it into their published report, made it obvious to us that the UAH diurnal correction method had a bias that needed to be corrected.

Since we (UAH) had already been working on a new diurnal adjustment technique, based upon the newer and more powerful AMSUs that have been flying since 1998, we rushed our new method to completion recently, and implemented new corrections. As a result, the UAH global temperature trends for the period 1979 to the present have increased from +0.09 to +0.12 deg. C/decade -- still below the RSS estimate of +0.19 deg. C/decade.

Our new AMSU-based (observed) diurnal cycle adjustments end up being very similar to RSS's climate model (theoretical) adjustments. So why the remaining difference between the trends produced by the two groups? While this needs to be studied further, it looks like the reason is the same as that determined for the discrepancy in deep-tropospheric satellite estimates between the two groups: the way in which successive satellites in the long satellite time series are intercalibrated. There has been a continuing, honest difference of opinion between UAH and RSS about how this should best be done.

In a paper accompanying the Mears and Wentz paper, a new analysis of radiosonde (weather balloon) data by Sherwood et al. also obtains larger levels of warming than have been previously reported. No other radiosonde dataset that has attempted to adjust for the calibration artifacts discussed therein has produced warming estimates as high as those obtained in this new study. As is always the case, it will take a while for the research community to form opinions about whether the new radiosonde adjustments advocated in this work are justified. At a minimum, the new work shows that at least one method for analysis of the weather balloon data (which have traditionally supported the much smaller satellite trends from UAH) results in trends much closer to the warmer surface thermometer trends.

The third paper (Santer et al, 2005) takes a more thorough look at the theoretical expectation that surface warming should be amplified with height in the troposphere. The authors restate what had already been known: that the UAH satellite warming estimates were at odds with theoretical expectations (as had been some radiosonde measures). Now, the convergence of these newly reported satellite and radiosonde estimates toward the surface warming estimates, if taken at face value, provides better agreement with climate models' explanation of how the climate system behaves.

What will all of this mean for the global warming debate? Probably less than the media spin will make of it. At a minimum, the new reports show that it is indeed possible to analyze different temperature datasets in such a way that they agree with current global warming theory. Nevertheless, all measurements systems have errors (especially for climate trends), and researchers differ in their views of what kinds of errors exist, and how they should be corrected. As pointed out by Santer et al., it is with great difficulty that our present weather measurement systems (thermometers, weather balloons, and satellites) are forced to measure miniscule climate trends. What isn't generally recognized is that the satellite-thermometer difference that has sparked debate in recent years has largely originated over the tropical oceans -- the trends over northern hemispheric land areas, where most people live, have been almost identical.

On the positive side, at least some portion of the disagreement between satellite and thermometer estimates of global temperature trends has now been removed. This helps to further shift the global warming debate out of the realm of "is warming happening?" to "how much has it warmed, and how much will it warm in the future?". (Equally valid questions to debate are "how much of the warmth is man-made?", "is warming necessarily a bad thing?", and "what can we do about it anyway?"). And this is where the debate should be.


Mears, C.A., and F.J. Wentz, 2005: The effect of diurnal correction on satellite-derived 
lower tropospheric temperature. August 11, online at 
Santer, B.D., et al., 2005: Amplification of surface temperature trends and variability 
the tropical atmosphere. August 11, online at
Sherwood, S., J. Lanzante, and C. Meyer, 2005: Radiosonde daytime biases and late 
century warming. August 11, online at



1 Comment

Satellite cooling trend?
I am confused about this maybe somone could explain it to me. How come a cooling trend is formed by the changing satellite orbit. I understand that the satellites "orbits drift to later in the day over the several years of each satellites lifetime." I don't understand why, as the day becomes later why won't a morning reading show a warming trend that offsets the afternoon observed cooling. Do the satellites not mesure the morning temperatures?

TCS Daily Archives