2015 Global Temperatures

[skierinvermont]

Add this to the list of studies questioning the UAH methodology.They use a different diurnal cycle correction and merge the NOAA-9 satellite differently. They find far more warming in the mid-troposphere than UAH.

 

It seems that the UAH method relies on little more than making some arbitrary corrections in favor of a pre-determined result.

 

http://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-13-00767.1

[StudentOfClimatology]

You are not reading correctly. The graph very clearly provides error bars for trend estimates. The Sherwood paper says the trend for 300hpa from 1959-2015 is +.25C/decade +/-.04C/decade.

It's funny how you managed to throw all those big words in there to try and sound smart, but end up being wrong.

The Sherwood paper is quite clear, the trend at 300hpa is +.25C/decade +/-.04C/decade.

This is in contrast with satellite derived estimates for TLT which have published error bars and/or discrepancy between research group of +/-.1C/decade.

Read the paper. You're clearly not understanding the type of analysis being done, because you're contradicting the very argument you were making last month regarding the time-dependent homogenizations used to analogize the resolution differential between the sonde and MSU datasets.

Are you saying that you didn't actually read Mears et al 2012? Why did you argue in favor of Mears' homogenization procedure last month? You're making no sense.

[StudentOfClimatology]

Add this to the list of studies questioning the UAH methodology.They use a different diurnal cycle correction and merge the NOAA-9 satellite differently. They find far more warming in the mid-troposphere than UAH.

It seems that the UAH method relies on little more than making some arbitrary corrections in favor of a pre-determined result.

http://journals.ametsoc.org/doi/abs/10.1175/JCLI-D-13-00767.1

Nice try. The Chedley et al 2013 study is an outdated TMT comparison between the MSU/AMSU merging algorithms used in the old UAH/RSS domainal derivations. The NOAA9 satellite is no longer in existence.

They argue that the approach used in UAHv5.6 is inferior to the procedure used by RSS. Obviously that is no longer a relevant issue.

[StudentOfClimatology]

Back to the original topic regarding radiosonde analysis. This is the very Mears et al 2012 cited in AR5. The same paper you were highlighting before the UAHv6.0 release.

http://onlinelibrary.wiley.com/doi/10.1029/2012JD017710/full

This was their conclusion:

Given the data limitations it is concluded that using radiosondes to validate multidecadal-scale trends in MSU data, or vice versa, or trying to use such metrics alone to pick a ‘winner’ is an ill-conditioned approach and has limited utility without one or more of additional independent measurements, or methodological, or physical analysis.

At least read the literature you cite. It'll help.

[chubbs]

SOC - a real graduate student would try to learn something from the Sherwood paper. Their technique significantly improves the signal to noise ratio in the radiosonde data.

[skierinvermont]

SOC - a real graduate student would try to learn something from the Sherwood paper. Their technique significantly improves the signal to noise ratio in the radiosonde data.

 

Yep - it significantly reduces the uncertainty level. Doesn't eliminate it but it provides a strong signal which is in agreement with climate models.

[skierinvermont]

Nice try. The Chedley et al 2013 study is an outdated TMT comparison between the MSU/AMSU merging algorithms used in the old UAH/RSS domainal derivations. The NOAA9 satellite is no longer in existence.

They argue that the approach used in UAHv5.6 is inferior to the procedure used by RSS. Obviously that is no longer a relevant issue.

 

The NOAA-9 satellite is still used in the datasets. It is extremely concerning that using alternative merging procedures and diurnal drift corrections yields completely opposing results. That concern didn't magically dissapear because UAH made a small correction that actually made the disparity between alternative methodologies even greater. UAH v6 just made the disparity even worse. The Chedley methodology still yields .25C/decade for TMT in complete disparity with UAH. STAR also yields .25C/decade. Sherwood also yields .25C/decade. Several other methods and papers also yield .25C/decade or raise other concerns with UAH.

[skierinvermont]

Back to the original topic regarding radiosonde analysis. This is the very Mears et al 2012 cited in AR5. The same paper you were highlighting before the UAHv6.0 release.

http://onlinelibrary.wiley.com/doi/10.1029/2012JD017710/full

This was their conclusion:

At least read the literature you cite. It'll help.

 

I am aware of Mears' concerns with the uncertainty in radiosonde data. The Sherwood paper reduces the uncertainty level. Mears is still quite clear the uncertainty level for MSU products is about .1C/decade -  a fact you have yet to acknowledge even though you're citing from his paper and just picking and choosing the parts you like. Sherwood is equally clear, the uncertainty level for IUKv2 is .04C/decade at 300hpa. 

[nflwxman]

The NOAA-9 satellite is still used in the datasets. It is extremely concerning that using alternative merging procedures and diurnal drift corrections yields completely opposing results. That concern didn't magically dissapear because UAH made a small correction that actually made the disparity between alternative methodologies even greater. UAH v6 just made the disparity even worse. The Chedley methodology still yields .25C/decade for TMT in complete disparity with UAH. STAR also yields .25C/decade. Sherwood also yields .25C/decade. Several other methods and papers also yield .25C/decade or raise other concerns with UAH.

 

UAH version 6: is actually measuring quite a bit into the stratosphere.  In fact, 5.6 and 6 don't even measure the same portions of the atmosphere.  Forget about the diurnal drift correction.  This is probably more significant.

 

 

Paraphrasing a post:

 

"As you can see their new weight function goes nicely up into the stratosphere, while cancelling a major part of the anyway meagre surface contributions. 

That's only one of the problems with satellite measurements, though.

Edit: So first of all - you are mixing in some parts of the atmosphere that are actually close or above the cooling line. Second, there must be a large difference between surface and 5 km altitude measurements (which is their maximum weight altitude): Many effects of global warming target the night inversions (less radiation into space at night weakens the inversion) and natural inversions, e.g. over snow layers - so the satellite data by construction miss most of the arctic amplification effects. Then you are more governed by ENSO effects and oceans (having a lower trend currently for obvious reasons) that have an increased impact in the mid-level atmosphere.  And there are very suspicious spots near mountainous regions like the Himalaya's, where the satellite data do not reproduce the surface measurements very well. Plus you are lacking the polar regions. "

[chubbs]

It would be interesting to know if the UAH6 TLT height profile varies with latitude. If not, would expect high latitudes to feel more of the stratospheric cooling.

[StudentOfClimatology]

The NOAA-9 satellite is still used in the datasets. It is extremely concerning that using alternative merging procedures and diurnal drift corrections yields completely opposing results. That concern didn't magically dissapear because UAH made a small correction that actually made the disparity between alternative methodologies even greater. UAH v6 just made the disparity even worse.

Now I'm lost. What "disparity" are you talking about? The Chedley et al 2013 paper is solely TMT based...I'm pouring over the version-6 TMT data and the thermal distribution looks significantly improved.

The Chedley methodology still yields .25C/decade for TMT in complete disparity with UAH. STAR also yields .25C/decade. Sherwood also yields .25C/decade. Several other methods and papers also yield .25C/decade or raise other concerns with UAH.

Uh, 0.25C/decade where? In the TMT layer? No dataset shows anything close to that. Not even STAR, which was never designed to measure global temperature in the first place.

[StudentOfClimatology]

UAH version 6: is actually measuring quite a bit into the stratosphere. In fact, 5.6 and 6 don't even measure the same portions of the atmosphere. Forget about the diurnal drift correction. This is probably more significant.

Paraphrasing a post:

"As you can see their new weight function goes nicely up into the stratosphere, while cancelling a major part of the anyway meagre surface contributions.

That's only one of the problems with satellite measurements, though.

Edit: So first of all - you are mixing in some parts of the atmosphere that are actually close or above the cooling line. Second, there must be a large difference between surface and 5 km altitude measurements (which is their maximum weight altitude): Many effects of global warming target the night inversions (less radiation into space at night weakens the inversion) and natural inversions, e.g. over snow layers - so the satellite data by construction miss most of the arctic amplification effects. Then you are more governed by ENSO effects and oceans (having a lower trend currently for obvious reasons) that have an increased impact in the mid-level atmosphere. And there are very suspicious spots near mountainous regions like the Himalaya's, where the satellite data do not reproduce the surface measurements very well. Plus you are lacking the polar regions. "

What blog did you pull this from?

Filtering for height is easy because the quantum oscillations governing microwave emissions slow with altitude/decompression. This is easier than correcting for orbital drift..and both are very easy to do.

If there's anything to be worried about regarding UAH, it's radiometer decay. That's hard to analyze/correct for, and every satellite will experience it to a varying degree.

[StudentOfClimatology]

I am aware of Mears' concerns with the uncertainty in radiosonde data. The Sherwood paper reduces the uncertainty level. Mears is still quite clear the uncertainty level for MSU products is about .1C/decade - a fact you have yet to acknowledge even though you're citing from his paper and just picking and choosing the parts you like. Sherwood is equally clear, the uncertainty level for IUKv2 is .04C/decade at 300hpa.

The 0.04C/decade uncertainty is in reference to the aggregation procedure used to structure the profile..nothing more. It isn't even mentioned in the abstract. The uncertainties within the necessary diurnal/time-based homogenization(s) were virtually ignored, and this is what Mears et al 2012 was discussing.

This should be basic stuff.

[nflwxman]

What blog did you pull this from?

Filtering for height is easy because the quantum oscillations governing microwave emissions slow with altitude/decompression. This is easier than correcting for orbital drift..and both are very easy to do.

If there's anything to be worried about regarding UAH, it's radiometer decay. That's hard to analyze/correct for, and every satellite will experience it to a varying degree.

 

The image is from Roy Spencer's blog.

[skierinvermont]

Now I'm lost. What "disparity" are you talking about? The Chedley et al 2013 paper is solely TMT based...I'm pouring over the version-6 TMT data and the thermal distribution looks significantly improved.

Uh, 0.25C/decade where? In the TMT layer? No dataset shows anything close to that. Not even STAR, which was never designed to measure global temperature in the first place.

 

The sole purpose of STAR is to measure global temperature. You're just making stuff up.

[StudentOfClimatology]

The sole purpose of STAR is to measure global temperature. You're just making stuff up.

You're the one making things up. Here is the STAR site:

http://www.star.nesdis.noaa.gov/star/prod_climate.php

The STAR projects are multi-disciplinary, but there is no continually-maintained global temperature associated with this network. The data is occasionally used in the literature and other projects, but there is no mainstream operational global temperature dataset in the STAR network.

[StudentOfClimatology]

The image is from Roy Spencer's blog.

I'm referring to the writing. It makes no physical sense.

[The_Global_Warmer]

Did they give a reason why?

I mean the change couldn't have been designed anymore perfect to counter the surface data sets.

Ohc rise

Sea level rise

Global mass glacial melt

Arctic sea ice loss.

Global ssta being explosive

[skierinvermont]

You're the one making things up. Here is the STAR site:

http://www.star.nesdis.noaa.gov/star/prod_climate.php

The STAR projects are multi-disciplinary, but there is no continually-maintained global temperature associated with this network. The data is occasionally used in the literature and other projects, but there is no mainstream operational global temperature dataset in the STAR network.

 

The only thing you said that is correct is that it is not continually maintained.

 

The sole purpose is still measurement of global temperature.

 

You're just making things up.

[StudentOfClimatology]

The only thing you said that is correct is that it is not continually maintained.

The sole purpose is still measurement of global temperature.

You're just making things up.

Like I thought, you have no idea what STAR is. I just provided you the link to what you're referring to, and you're literally pretending it doesn't exist.

This is the full list of operational, climate-based STAR datasets. I don't see anything on lower tropospheric global temperatures.

http://www.star.nesdis.noaa.gov/star/prod_climate.php

Again the STAR (Center for Satellite Applications and Research) is a multi-disciplinary network of hundreds of satellites used for the monitoring of global weather, climate, and environmental phenomenon. It has no "sole purpose", and the MSU/AMSU units in this satellite network are the very units used in both the UAH and RSS networks. There is no "STAR" dataset. Or, technically they're all "STAR" datasets.

[AvantHiatus]

AMSU has suffered from drift in the past. Why should we trust it now? Wasn't the defunct Aqua network based on AMSU? All this stuff needs a deep formal review before anyone has the right to use this information for science or policy making.

[skierinvermont]

Like I thought, you have no idea what STAR is. I just provided you the link to what you're referring to, and you're literally pretending it doesn't exist.

This is the full list of operational, climate-based STAR datasets. I don't see anything on lower tropospheric global temperatures.

http://www.star.nesdis.noaa.gov/star/prod_climate.php

Again the STAR (Center for Satellite Applications and Research) is a multi-disciplinary network of hundreds of satellites used for the monitoring of global weather, climate, and environmental phenomenon. It has no "sole purpose", and the MSU/AMSU units in this satellite network are the very units used in both the UAH and RSS networks. There is no "STAR" dataset. Or, technically they're all "STAR" datasets.

 

This is like saying the sole purpose of the University of Alabama Huntsville (UAH) is not tropospheric temperature monitoring.

 

Obviously, but researchers there create a product that's sole purpose is tropospheric temperature monitoring (commonly simply referred to as UAH in climate science).

 

Likewise, STAR creates a tropospheric temperature product. I guess you're not familiar with it. Keep digging, you'll find it. I'm honestly surprised, I would think a graduate student like yourself would have heard of STAR.

[StudentOfClimatology]

For the last time, "STAR" does not produce an operational LT dataset..that is done by the well known subsidiaries (UAH/RSS). The MSU/AMSU sounding units are taxpayer funded and the data can be accessed by anyone. It's the interpolation methods that are independently developed.

If you're unfamiliar with all this, just admit it. No one will think less of you for it.

[csnavywx]

SOI has really cratered over the past 10 days. 30 day average is around -14. The warm water pool has bulged out to around 150W so far as a result of the latest strong WWB. As long as we continue to see weak trades and a steady supply of WWBs every so often, this could turn out to be quite the Nino.

[skierinvermont]

For the last time, "STAR" does not produce an operational LT dataset..that is done by the well known subsidiaries (UAH/RSS). The MSU/AMSU sounding units are taxpayer funded and the data can be accessed by anyone. It's the interpolation methods that are independently developed.

If you're unfamiliar with all this, just admit it. No one will think less of you for it.

 

O rly?

 

I didn't say LT, I said tropospheric. Learn to read kid.

 

From STAR's website:

 

http://www.star.nesdis.noaa.gov/smcd/emb/mscat/

 

Bottom three traces: STAR MSU/AMSU Version 3.0, monthly global mean anomaly time series and trends for the layer temperatures of mid-troposphere (TMT), upper-troposphere (TUT), and lower-stratosphere (TLS). 

 

[StudentOfClimatology]

The STAR^3 dataset is not operational, though, and doesn't show anything close to an 0.25C/decade trend in the TMT. The trend it shows is roughly 0.14C/decade in the TMT layer, very similar to UAH^6's 0.11C/decade.

Other non-operational datasets include the ESRL, SSU, and ECM datasets

[nzucker]

In my understanding, UAH, RSS, and STAR are all really derived from the same satellites...they're just different interpolation/calibration methods which change the overall global temperature trend significantly. STAR has always had one of the highest anomalies on the scale of 0.2C warming per decade, whereas the more mainstream UAH/RSS data has shown about .1C/decade. The UAH/RSS sources have also shown more of a "plateau" since 1998 whereas STAR shows more of a continuous warming trend.

 

Interestingly, UAH v5.6 and RSS were in almost exact agreement with April 2015...I believe UAH 5.6 came in at +0.16C and RSS came in at +0.17C. This concordance definitely gives some validity to the two mainstream satellite sources. UAH v6.0 was colder at +0.07C, but there's been some argument about the latest revisions being too "cold." In any case, all the major data sources, including GISS, UK, RSS, and UAH, showed that April 2015 was cooler than March 2015. Again, the trend is similar in all the data so that gives it all a certain degree of validity. 

 

This is from my limited understanding, correct me if I'm wrong about STAR being simply a different calibration method of the same AMSU satellite systems.

[ORH_wxman]

STAR is a weird dataset...I've seen like 5 different versions of it which all show different trends....I'm not sure if it is because it has undergone multiple revisions or not.

 

Is there some site that actually keeps track of it in a homogenous manner that updates regularly?

[Typhoon Tip]

I am not sure what the point there is, though. 

 

First of all, that is NOAA's official position on matters.  Not mine, nor anyone else that reports their published findings - just to be clear.  Any contention after the fact needs to be taken to how ever it is they derive their numbers.  

 

I understand that the RSS and UAH data may or may not support their conclusions, but saying that they don't (merely) does nothing to invalidate NOAA's conclusions.  So that's where you lost me.  

 

None of which may be all that important when held up against the undeniable, black-and-white provable fact that GW has been underway really since the industrial revolution.  

 

That, in its self, may be mere coincidence ...sure. Or not.  Does it matter?  

 

None of these data sources, RSS, UAH ... great grandma spinning yarns about when she was a kid .. .they are not really in dispute: the Earth is a warming planet.  

[ORH_wxman]

I am not sure what the point there is, though. 

 

First of all, that is NOAA's official position on matters.  Not mine, nor anyone else that reports their published findings - just to be clear.  Any contention after the fact needs to be taken to how ever it is they derive their numbers.  

 

I understand that the RSS and UAH data may or may not support their conclusions, but saying that they don't (merely) does nothing to invalidate NOAA's conclusions.  So that's where you lost me.  

 

None of which may be all that important when held up against the undeniable, black-and-white provable fact that GW has been underway really since the industrial revolution.  

 

That, in its self, may be mere coincidence ...sure. Or not.  Does it matter?  

 

None of these data sources, RSS, UAH ... great grandma spinning yarns about when she was a kid .. .they are not really in dispute: the Earth is a warming planet.  

 

Nobody is disputing global warming is occurring in this thread.