Recently Jos Hagelaars published a very extensive blog post (on the blog of Bart Verheggen) about a widely discussed paper of Klotzbach et al 2009. The title of the blog post – Klotzbach revisited – is in English, however, the post itself was written in Dutch. As a fellow Dutchman I understand that writing in Dutch is easier than writing in English. However, in this case, the blog post is focussed so much on one single paper, that Jos Hagelaars, in my opinion, should have chosen for an English version, in order to give the authors of the Klotzbach papers the chance to give a reaction. I translated the article with google translator and did some minor editing. I then shared the article with a few of the coauthors. John Christy looked at some of the issues raised by Hagelaars and wrote the following reaction which I publish here as a guest blog.

Guest blog by John Christy

In a blog post entitled “Klotzbach Revisited” Jos Hagelaars updated the results of Klotzbach et al. 2009, 2010, suggesting that the main point of Klotzbach was no longer substantiated. Klotzbach et al.’s main point was that a direct comparison of the relationship of the magnitude of surface temperature trends vs. temperature trends of the troposphere revealed an inconsistency with model projections of the same quantities.  Klotzbach et al. offered suggestions for this result which included the notion that near-surface air temperatures are easily affected by factors unrelated to greenhouse gas increases, which then implies they may be poor proxies for detecting the magnitude of the greenhouse effect which has its main impact in the deep atmosphere.

It appears Hagelaars’ key point is that when the data from Klotzbach et al. are extended beyond 2008 to include data through 2012, the discrepancies, i.e. the observed difference between surface and tropospheric trends relative to what models project, are reduced somewhat.

Confusion
The reader must understand that there are two issues that have unfortunately been convoluted and misinterpreted on this issue.  The first issue deals specifically with the relationship between a surface temperature trend and the temperature trend of the corresponding tropospheric layer above (roughly surface to 10 km altitude and referred to as LT for “Lower Troposphere”).  The second issue deals with the actual magnitude of the surface and tropospheric trends.  Thus the first issue is a question of the physics of the vertical temperature structure (i.e. internal model processes) and the second issue is a question of trend magnitudes (i.e. rates of warming or climate sensitivity).  The two are, of course, related.

Here is how the confusion often happens.  As shown in many results, the observed tropospheric trend is often near (or slightly below) the magnitude of the surface trend.  Thus, someone may say “the surface and troposphere agree” as if that validates greenhouse warming theory.  However, in model results (i.e. according to theory) the surface and tropospheric trends should NOT agree because in models the troposphere warms faster than the surface.  So, if surface and tropospheric trends agree, then by implication, model output is incorrect. Below we shall look at this more closely.

Regarding the first issue, there have been many studies which have looked at the relationship between the magnitude of the surface temperature trend relative to that of the tropospheric layer as defined above (e.g. Douglass et al. 2007.)  Global climate models when forced by extra greenhouse gases on average indicate their global average troposphere warms at a rate about 1.25 times that of the surface, i.e. the trend of the troposphere is amplified by a factor of 1.25 over that of the surface.  When confined to the tropics (20°S – 20°N) the amplification is about 1.4 times that of the surface.  This model-generated tropospheric warming in the tropics is known as the “hot spot” and has been claimed to be a signature of greenhouse warming because of its prominence in models.

Amplification
When separated by land and ocean, the model amplification factor is found to be larger over oceans than land.  Klotzbach et al. 2010 calculated the ratio over global land to be 1.1, and this was confirmed by independent analysis (see http://climateaudit.org/2011/11/07/un-muddying-the-waters/).  Hagelaars follows an early calculation by Gavin Schmidt, claiming the land value should be 0.95.  As noted however, several additional calculations confirm the value of 1.1 utilized by Klotzbach et al. 2010.  The model amplification of the ocean trends is close to 1.6 as determined by the NASA-GISS model.

The second issue is the simple magnitude of global temperature trends of the surface and troposphere as depicted by models and as observed by instruments.  Since both issues can be examined by investigating the observational record, we have created the Table below to update Klotzbach et al. 2010 and address the concerns of Hagelaars.

In his last table, Hagelaars appears to be subtracting the actual observed values of LT and Sfc which produces values very similar to those shown in the upper half of our table.  It is true that these differences are a little closer to zero than shown in Klotzbach et al., but that is due to the fact that there has been no warming in the past 10 years in both types of data.  (Note too, that if the surface and tropospheric trends “agree” in absolute magnitude, that means they do not agree with model output as noted earlier – hence closer agreement of absolute trends can imply greater disagreement with model results.)

Now, a more direct, “apples to apples” comparison test for the model output is to amplify the surface trends (with model factors) for comparison with the LT trends.  We have been conservative with the amplification factors, but even so, the differences are large – and very large over land.  Thus the basic point of Klotzbach et al. 2010 is confirmed, i.e. that the average climate model warms its atmosphere, relative to its land surface, more than seen in observations. (Other studies focus on the tropical “hot spot” where it is clear models also significantly warm the troposphere relative to observations, e.g. Christy et al. 2010.) This raises at least some suspicion as to the ability of the near-surface air temperature to be used as a proxy for greenhouse detection.

Table:  1979-2012 trends (°C/decade).  No amplification factors are applied in the upper half of the table, thus they compare different quantities.  Land, Ocean and Global factors of 1.1, 1.2 and 1.4 are applied to the surface data in the lower half.  Recent results from Santer et al. 2012 indicate a global amplification factor greater than 1.3 for model LT vs. Sfc, but we use only 1.2 below.  Lower Tropospheric data are from the University of Alabama in Huntsville v5.5 (UAH) and Remote Sensing Systems v3.3 (RSS), and surface data from the National Climatic Data Center (NCDC) and the Hadley and Climate Research Unit Temperature v4 (HadCRUT4).  Artificial values of “NCDC LT” and “HadCRUT4 LT” are calculated by multiplying their actual trends by the model amplification factors.

UAH LT

RSS LT

NCDC

HadCRUT4

Actual
Land

0.175

0.182

0.266

0.272

Ocean

0.116

0.107

0.100

0.115

Globe

0.137

0.131

0.152

0.161

Difference (No Amplification)

UAH-NCDC

RSS-NCDC

UAH-HadC

RSS-HadC

Land

-0.091

-0.084

-0.097

-0.090

Ocean

+0.016

+0.007

+0.001

-0.008

Globe

-0.015

-0.021

-0.024

-0.030

 
Hypothetical (with Amplification)

NCDC LT

HadCRUT4 LT

Land (1.1xSfc)

0.293

0.300

Ocean (1.4xSfc)

0.140

0.161

Globe (1.2xSfc)

0.184

0.193

Difference LT

UAH-NCDC LT

RSS-NCDC LT

UAH-HadC LT

RSS-HadC LT

Land

-0.118

-0.111

-0.125

-0.118

Ocean

-0.024

-0.033

-0.045

-0.054

Globe

-0.047

-0.053

-0.056

-0.062

 

CMIP5 versus observations
Of equal importance here are the magnitudes of the actual trends of the surface and troposphere.  The average global surface trend for 90 model simulations for 1979-2012 (Climate Model Intercomparison Project 5 or CMIP-5 used for IPCC AR5) is +0.232 °C/decade.  The average of the observations is +0.157 °C/decade.  Therefore models, on average, depict the last 34 years as warming about 1.5 times what actually occurred.  Santer et al. 2012 (for 1979-2011 model output) noted that a subset of CMIP-5 models produce warming in LT that is 1.9 times observed, and for a deeper layer of the atmosphere (mid-troposphere, surface to about 18 km) the models warm the air 2.5 times that of observations.  These are significant differences, implying the climate sensitivity of models is too high.

Signature
All of the above addresses the two issues mentioned at the beginning.  First, global climate models on average depict a relationship between the surface and upper air that is different than that observed, i.e. models depict an amplifying factor into the upper air that is greater than observed.  Secondly, the average climate model depicts the warming rate since 1979 as much higher than observed with increasing discrepancies as the altitude increases (which is consistent with the first issue).

Since this increased warming in the upper layers is a signature of greenhouse gas forcing in models, and it is not observed, this raises questions about the ability of models to represent the true vertical heat flux processes of the atmosphere and thus to represent the climate impact of the extra greenhouses gases we are putting into the atmosphere.  It is not hard to imagine that as the atmosphere is warmed by whatever means (i.e. extra greenhouse gases) that existing processes which naturally expel heat from the Earth (i.e. negative feedbacks) can be more vigorously engaged and counteract the direct warming of the forcing. This result is related to the idea of climate sensitivity, i.e. how sensitive is the surface temperature to higher greenhouse forcing, for which several recent publications suggest models, on average, have been overly sensitive.

References:
Christy, J.R., B. Herman, R. Pielke, Sr., P. Klotzbach, R.T. McNider, J.J. Hnilo, R.W. Spencer, T. Chase and D. Douglass, 2010:  What do observational datasets say about modeled tropospheric temperature trends since 1979? Remote Sens. 2, 2138-2169. Doi:10.3390/rs2092148.

Douglass D. H., J. R. Christy, B. D. Pearson and S. F. Singer (2007) A comparison of tropical temperatures trends with model predictions.. International J of Climatology 5 Dec 2007. doi: 10.1002ijoc.1651

Klotzbach, P.J., R.A. Pielke Sr., R.A. Pielke Jr., J.R. Christy, and R.T. McNider, 2009:

An alternative explanation for differential temperature trends at the surface and in the lower troposphere. J.Geophys. Res., 114, D21102, doi:10.1029/2009JD011841.<http://pielkeclimatesci.wordpress.com/files/2009/11/r-345.pdf>

Klotzbach, P.J., R.A. Pielke Sr., R.A. Pielke Jr., J.R. Christy, and R.T. McNider, 2010:

Correction to: “An alternative explanation for differential temperature trends at the surface and in the lower troposphere. J. Geophys. Res., 114, D21102, doi:10.1029/2009JD011841”, J. Geophys. Res., 115, D1, doi:10.1029/2009JD013655.

Santer B. and 26 others (2012). Identifying human influence on atmospheric temperatures. Proceeding of the National Academy of Sciences. doi:10.1073/pnas.1210514109.