Myth that the troposphere is not warming faster then ground temperatures

[1veedo]

I happen to have read a lot of peer-review on this subject and the idea that the troposphere isn't behaving as expected is really pretty erroneous. Several threads have had this claim come up so I figured I'd put an end to it once and for all.

 

The Earth as a whole is warming at about .2C/decade but ground temperatures are actually around .18.

 

Recording temperatures isn't all that easy. There's a lot of inherent problems with A/MSU readings and specifically they have a natural tendency to show temperatures that are lower then actual.

 

Just so you guys know what I'm talking about there's an interesting article, "Contribution of stratosphericcooling to satellite-inferred tropospheric temperature trends" published in Nature in 2004. This isn't original research but the same sort of research got the attention of Cristy et al in their serious of publications "Error estimates of version 5.0 of MSU/AMSU bulk atmospheric temperatures," the update version 5.2 of which is in line with global warming theory.

"The MSU, since 1979, and its successor, the Advanced MSU (AMSU), from 1998, provide a global measure of temperature for several atmospheric layers from NOAA polar-orbiting satellites. Although the original purpose of MSU measurements was to improve weather forecasts, a continuing data-analysis effort has been made to satisfy climate research requirements of homogeneity and calibration1,2,10,18–24. Several important non-climatic influences have been identified and removed, including diurnal temperature biases related to local sampling times of the satellite and their changes over its lifetime, errors in the MSU calibration, and biases due to decay of the satellite orbits....

 

To infer the temperature of the mid-troposphere, we use two microwave channels: MSU channels 2 and 4 (or AMSU channels 5 and 9). Their weighting functions are shown in Fig. 1a. The weighting function for MSU channel 2 (or AMSU channel 5) peaks at ,550 hPa (,4.5 km). Thus the MSU Channel 2 brightness temperatures (T 2) have often been used to represent mid-tropo- spheric temperatures1,2,3,8,15,22,24. The MSU channel 4 (or AMSU channel 9), whose weighting function peaks at ,85 hPa (,18 km), has been used to represent stratospheric temperatures1,3,8,15.

 

As ~85% of the signal for T 2 comes from the troposphere and surface, it is not a bad approximation to say that the seasonal and interannual variations of mean deep-layer temperature in the troposphere can be well represented by T 2. However, this might not be the case for trends. Figure lb shows simulated changes of T2 owing to changes of tropospheric and stratospheric temperatures, and indicates that T 2 remains constant when the changes in tropo-spheric and stratospheric temperatures have a ratio of about ~1/5. This is because the vertical integral of the weighting function from the surface to tropopause, near 200 hPa, is about 5 times the integral above the tropopause. For example, if the tropospheric temperature trend were 0.15 K per decade and the stratospheric trend were ~0.8 K per decade, the trend of T 2 would be close to zero. Intriguingly, this scenario resembles what may actually be the case in the atmosphere. The temperature trend in the lower stratosphere (15 to 23 km), as obtained from radiosondes and satellite observations, is about ~0.5 to ~0.9 K per decade for the last 20 years1,3,4,9. Therefore, because of the combined influence of stratospheric and tropospheric changes, T 2 trends are not an ideal indicator of global climate change. To derive the tropospheric temperature trends, the effects of strato-spheric cooling on T 2 must be taken into account.

 

...technical details about how they accomplish this...equation 1 T 850-300 = a0 + a2T2 + a4 T4...

 

But irrespective of the techniques used to analyse the data, T2 is subject to the effects of stratospheric cooling. Assuming a stratospheric temperature trend of ~0.5 K per decade1,3,4, the Vinnikov–Grody T 2 trend translates to a T850-300 trend of ~0.33–0.37 K per decade. This value is about twice as large as the surface warming globally. It also suggests a ratio of ~3 between tropospheric and surface tempera- ture trends for the tropical region. These ratios seem large, and suggest that the technique Vinnikov and Grody used to analyse the satellite data may require further scrutiny."

So it's not a matter of turning on a satellite, click and boom we now have temperature readings. This letter attempts to reconcile the problems with Cristy's and other publications before 2004 which indicate tropospheric warming not significantly faster, but actually slower then ground temperatures. Another study has shown tropospheric warming as expected sense 2002, and furthered in the 2005 update.

 

Schabel et al. (2002). "Stable Long-Term Retrieval of Tropospheric Temperature Time Series from the Microwave Sounding Unit" Proceedings of the International Geophysics and Remote Sensing Symposium. [updated in 2005]

 

The other study we need is specifically for ground temperatures: Jones and Moberg. (2003). "Hemispheric and large-scale surface air temperature variations: An extensive revision and an update to 2001." Journal of Climate 16: 206-223.

 

To be in line with global warming theory tropospheric warming should be 1.3 times greater then ground temperatures. Ground temperatures have been rising .187C/decade meaning tropospheric warming should be observed at .2431C/decade. From Schabel et al we happen to know that tropospheric warming is .239C/decade.

 

So here we have expected: .243C/decade and actual: .239C/decade.

 

These numbers are only .004C apart which is well within our error margin (ie it's close enough). So as you can see the troposphere is in fact behaving as it was predicted.

 

This should put an end to the whole tropospheric argument that global warming deniers constantly talk about because here is very direct proof that there's nothing really wrong. If anyone ever posts this mumbo jumbo again I'm just going to happily refer them to this thread because they're obviously getting their information from questionable global warming denier websites and not from the actual science journals. The numbers in this case clearly speak for themselves so there should be no problem understanding this.

[MolotovCocktail]

This should put an end to the whole tropospheric argument that global warming deniers constantly talk about because here is very direct proof that there's nothing really wrong. If anyone ever posts this mumbo jumbo again I'm just going to happily refer them to this thread because they're obviously getting their information from questionable global warming denier websites and not from the actual science journals. The numbers in this case clearly speak for themselves so there should be no problem understanding this.

 

Not sure if this is going to put an end to it, because the global warming deniers seem to be very, very persistent as was witnessed in the other threads. Even to the point where they would just conveniently "misinterpret" your sources :rolleyes:. Same thing happens when there is any discussion about environmental problems in general for that matter (the "A Matter of Time" thread for example). I don't know a lot about climate changes but I'll help you out when the GW deniers start posting, because I'm sick of it as well.

 

 

But anyways back on topic. From my limited knowledge of climate, this data would make sense, because the atmosphere is responsible for keeping the Earth warm in the first place, and also the level of greenhouse gases that are being dumped into the atmosphere at the present rate. How often do they update their predictions, because I'm sure that the rates will change as we put more greenhouse gases into the atmosphere, and I read somewhere that the melting ice will have a huge impact on ocean current and ocean levels, which will translate to weather and climate changes.

[bascule]

I've seen this problem addressed on RealClimate. At present the models are not accurately predicting tropospheric heating (although it's within a reasonable margin)

 

However, the models are predicting hundreds of other factors successfully across a century of instrumental record.

 

Regardless of what the reality is, I've only seen this idea used as part of an argument from incredulity.

[1veedo]

You think .004C is too large of an error? That's how far off the predicted verses actual is.

 

Just curious. I'm sure there could be specifics where certain layers are off but the troposphere as a whole (and the stratosphere at that) is warming as predicted, within .004C. Cristy et al's earlier papers had a stratospheric error built in described in the letter to nature*, and the latest (5.5 (2006?) I think, not 5.2 (2005?)) is accounting for this.

 

Again the models predict .243C/decade and the observed is .239C/decade. I'm no mathematician but these numbers are suspiciously close together on the number line.

 

 

*Which right now I'm kind of skeptical if quoting all of that is violating copyright . At the very bottom of the pdf, which is a photocopy, there's a single page with a little copy-warning on it. I missed it cause the article ends and the beginning of the next article in that publication is at the bottom.

[CPL.Luke]

as long as you site it you can quote it all that you want