E-mail Exchange With Joshua Willis On The Diagnosis Of Global Warming From Ocean Heat Content Changes

With the permission of Joshua Willis of JPL, I have posted a set of e-mails that he and I have exchanged over the last month. I present from the earliest to the latest

Date: Fri, 27 Aug 2010 13:43:10 -0600 (MDT)

Hi Josh

 I have been invited to prepare a Letter….. after the September 7-9 2010 Exeter meeting on the surface temperature data (as a climate metric). I discussed this on my post

Comments On The Economist Article “Green View: Could Temperature Be Less Intemperate?”

which is where they saw my views.

In my Letter, I want to emphasize the point I made that

“In terms of monitoring global warming, the successful installation of an upper ocean heat monitoring system which has been in place since earlier this decade (Argo as complemented with satellite measurements of the ocean) supersedes the need to use the surface air temperature data as the primary metric for this purpose [as I summarize in my article

Pielke Sr., R.A., 2008: A broader view of the role of humans in the climate system. Physics Today, 61, Vol. 11, 54-55.

We can obtain a much more robust measure of global warming (and cooling) by monitoring the upper ocean heat changes.”

Can you provide an update of the Figure you prepared for me in the Physics Today article.

They told me I need to submit within a couple of weeks after the meeting ends on September 9th.

Please let me know if you can do this.

Best Regards


On Tue, 31 Aug 2010, Josh Willis wrote:

 Hi Roger,

I will try to get something to you in this time frame, but the next couple of weeks are very hectic for me.  Also, I have been told that the ongoing correction of pressure errors in the Argo data set is likely to be done by October or early November of this year.  It will definitely be worthwhile to recompute the ocean heat content estimates AFTER that time as many of the data in quesiton will be “let back in” to the estimate once all the pressure corrections have been made.  So anything that I do give you between now and then will necessarily be a preliminary and unpublished estimate of OHC.


Date: Wed, 1 Sep 2010 06:30:38 -0600 (MDT)

Hi Josh

 Thanks! Whatever you can provide would be valuable. Of course, the final corrected version will be the set to look forward to!

In the Nature letter, let me know how (or if) I should mention any update you send (if you are able to in the next couple of weeks), or if I should just say that a definitive set of analyses with all of the corrections will be available later this year.

Best Regards


On Sat, 11 Sep 2010, Roger A Pielke Sr wrote:

 Hi Josh

 I was entrained in a debate on the value of using the Argo network (along with satellite data) to diagnose global warming and cooling on the website posts
http: //www.skepticalscience.com/Ocean-cooling-skeptic-arguments-drowned-by data.html http: //www.skepticalscience.com/Pielke-Sr-and-scientific-equivocation-dont-beat-around-the-bush-Roger.html

 I commented also on my weblog in a post


 Here is my last comment (at least so far) on the Skeptical Scientist website. Comments on this also appear on Watts Up With That[http://wattsupwiththat.com/2010/09/11/pielke-senior-misinformation-on-the website-%e2%80%9cskeptical-science-%e2%80%93-getting-skeptical-about-global warming-skepticism%e2%80%9d/]

 “The issues have been discussed extensively [and often very  constructively] in the comments on this website and at Watts Up With That [which reposted my original post and permits comments]. I suggest waiting until new information appears (promised to us by mid Fall) on updating the Argo/satellite estimates of upper ocean heating and cooling before we continue this discussion.

 At that time we can answer the central questions

 1. Using the GISS (Jim Hansen’s value of 0.6 W/m2 for the upper ocean as the model prediction, what are the estimates of the accumulation of Joules that have accumulated in the upper ocean since 2004?

 2. What is the observation accuracy of the Argo network and associate satellite altimetry measurements since 2004?

 3. Should the upper ocean heat change observations replace (or more conservatively complement) the use of the global annual average surface temperature trend estimates as the primary metric to diagnose i) multi-year and/or decadal averaged global warming.”

 Please let me know when you can answer these questions.

 Best Regards


Date: Tue, 21 Sep 2010 12:40:16 -0700 (PDT)

Hi Roger,

Sorry I’ve been slow getting back to you.  I’ve been in the thick of proposal writing lately and that tends to take precedence over everything else.  Anyway, I didn’t get time to read all of this blog conversation very carefully, but I would offer the following general opinion (which you are welcome to quote):

I do think that ocean heat content is one of the best ways to measure the net impact of human influence on the climate.  However, at present we really need to wait for more accurate estimates of ocean heat content before we draw conclusions about the size of year-to-year, or decade-to- decade natural fluctuations in ocean heat content.  For instance, how big an influence is El Nino, or a shift in the PDO on ocean heat content?  I don’t think we can really say.  What we can say, without hesitation, is that over a time span of 2 to 5 decades, the oceans have warmed substantially.  In other words, the long term rate of ocean warming has been reasonably well observed, but the year to year fluctuations are still a bit fuzzy.  The cleanest part of the Argo record is still too short, and XBT errors are still too big to make definitive statements about short-term natural variations in ocean heat content.

In other news, I am still a bit hesitant to give you a new estimate of OHC for the Nature letter.  It may be too late anyway, but I think it would be best to let the Argo Program finish all of the data updates before putting out a new estimate.  It would really be best to also sort out some of the remaining XBT errors and look at the OHC record from the early 1990s to the present, but that is probably even farther down the road.

Hope all is well.



On Tue, 21 Sep 2010, Roger A Pielke Sr wrote:

  Hi Josh

 Thank you for the feedback. I am puzzled, however, by several of your responses. First, you placed error bars on the figure you provided me for the Physics Today paper. Are these now wrong?

 Also, there are several other questions:

 1. Have you concluded that the use of surface temperatures (land and ocean)is a superior metric to diagnose the global average radiative imbalance than the upper ocean heat content?

 2. In terms of pre-Argo, why does this matter? The heat content at any time can be compared at an earlier time to assess its change within the uncertainty of the data. Thus comparisons between 2010 and 2005 do provide an effective assessment as the claim has been made by Jim Hansen that the global radiative imbalance was 0.85 Watts per meter squared at the end of the 1990s. Are the data good enough for the period 2010 and for 2005 to address whether the observations (given their uncertainties) are are in agreement or not with the Hansen (GISS prediction) for this time period?

 3. You wrote that ocean heat content is one of the best metrics. Why is it not the optimal one? This is actually a view Jim Hansen shares.

 4. On the updated data (e.g. for the Figure in the Physics Today article), when will it become available?

 The Nature article was written and has been submitted to them. It does not require the updated ocean heat data, but I do urge a movement away from the global average surface temperature trend to upper ocean heat content as the primary global warming metric.

 Best Regards


Date: Tue, 21 Sep 2010 15:23:49 -0700 (PDT)

Hi Roger,

Of course in the most general sense, a single number can never convey all the complexities of climate change.  But I do agree that as far as a single number goes, ocean heat content is a better measure than average surface temperature (I don’t think I said otherwise in my previous email, did I?).  Personally, I think global sea level is an even better number because it includes the impact of ice melt as well.  That may mean it tells us more about the effects and less about the physical processes involved, but in terms of communicating with the public, for instance, that is not necessarily a bad thing.

Anyway, in terms of trying to understand the root causes of global warming and climate change, yes ocean heat content is the best number.  But like I said above, I think the “best” number depends on the context and the message you are trying to convey.

As far as the error bars in the pre-2005 years of the plot I sent you before, they may have been somewhat optimistic.  I think I noted at the time that those error bars represented the error due to sampling effects, but not any potential systematic errors.  In fact, correction of the Argo pressure data may result in a small but significant systematic change in the early years of that curve. However, from 2005 on, the answer will not change much.

So, yes, it is now possible to test the 5-year warming rate using Argo data. But, in comparing it with the GISS model (or any other coupled run), the presence of interannual to decadal variability in ocean heat content needs to be evaluated and understood, both in the model as well as in the data.  My point is that we may have plenty of model runs with which to evaluate interannual to decadal variability in ocean heat content in models, but until can we evaluate that type of variability in the observations, this type of comparison may be of limited use.

Along these lines, I ran across an interesting paper in GRL recently:



Date: Tue, 21 Sep 2010 20:15:39 -0600 (MDT)

Hi Josh

 Thank you for the quick feedback and the very interesting paper!

 I completely agree with you that a single number cannot communicate the complexity of climate change. This complexity (and the human role) is summarized in our short EOS paper

Pielke Sr., R., K. Beven, G. Brasseur, J. Calvert, M. Chahine, R. Dickerson, D. Entekhabi, E. Foufoula-Georgiou, H. Gupta, V. Gupta, W. Krajewski, E. Philip Krider, W. K.M. Lau, J. McDonnell,  W. Rossow,  J. Schaake, J. Smith, S. Sorooshian,  and E. Wood, 2009: Climate change: The need to consider human forcings besides greenhouse gases. Eos, Vol. 90, No. 45, 10 November 2009, 413. Copyright (2009) American Geophysical Union.

However, the global annual average radiative forcing is the single best metric for the annual global warming/cooling (as we seem to agree on). The annual average ocean heat change is the most effective way to diagnose this radiative forcing as discussed, for instance, in the Physics Today paper and my earlier BAMS paper.

The global annual average sea level change is another valuable metric. By including the non-steric component (i.e. glacial melt), however, this complicates its interpretation as you noted.

 I am glad you reconfirmed the quality of the ocean data from 2005 on.

The value of your upper ocean data in terms of global warming is that time slices can be used to estimate the global average radiative forcing between the time slices. There is no need to decadal and longer term data to assess this forcing.

 Interannual and decadal variability, of course, are important issues to explain the observations. However, the value of the heat change between time slices provides the observed global average radiative forcing over this time period (within the limitations of the uncertainty of the observations).

I agree that the time period of really accurate upper ocean heat data is relatively short. However, in my view, from 2005 forward indefinitely, this should become the primary global warming metric.

Since you have confirmed the robustness of this data from 2005 onwards, we can use this to compare with the models. Since the heating is, at most, quite muted compared with the models from 2005 to 2008, there must be enhanced warming in the coming few years to compensate. This should be a critical metric to follow.

There are two relevant papers that you may find useful:

Christy, J.R., Herman, B., Pielke, R., Sr., Klotzbach, P., McNider, R.T., Hnilo, J.J., Spencer, R.W., Chase, T., and Douglass, D. What Do Observational Datasets Say about Modeled Tropospheric Temperature Trends since 1979?. Remote Sens. 2010, 2, 2148-2169.

R. S. Knox, David H. Douglass 2010: Recent energy balance of Earth. International Journal of Geosciences, 2010, vol. 1, no. 3 (November) ā.. In press doi:10.4236/ijg2010.00000.

Any comments you have would be valuable.

Finally, what would be a very valuable product when this data is updated this Fall is the upper ocean (or total ocean) heat content changes in Joules (and diagnosed global average radiative imbalance in Watts per meter squared) from the 2005 average (this annual averaged time slice) to the 2009 average time slice (or as far up to the present as possible in 2010) along with the uncertainty estimates. This should a metric that is presented in as near real time as possible (as a running annual average anomaly)

Best Regards


Comments Off on E-mail Exchange With Joshua Willis On The Diagnosis Of Global Warming From Ocean Heat Content Changes

Filed under Climate Change Metrics

Comments are closed.