Did Tom Wigley Fudge SST Data to Fit a Predetermined Narrative?

In my last post I shared about a fascinating paper that was just published with improved bias corrections for the cooling bias affecting SSTs between 1900 and 1930. As I was studying up on this paper, I was reminded of one of the hacked CRU emails discussing roughly the same problem back in 2009. The email was from Tom Wigley at UCAR to Phil Jones at CRU about bias correction issues affecting SSTs, especially from the 1940s and earlier. The language indicates that there's a context between the two that is left unexplained - that is, we're entering into a conversation mid stream, and there's language between the sender and recipient that people wouldn't necessarily understand without context (like what the "blip" is). The text of the email is below:

From: Tom Wigley [University Corporation of Atmospheric Research]
To: Phil Jones [CRU]
Cc: Ben Santer [Lawrence Livermore National Laboratory]
September 27, 2009

Subject: 1940s

Phil, 

Here are some speculations on correcting SSTs to partly explain the 1940s warming blip. If you look at the attached plot you will see that the land also shows the 1940s blip (as I’m sure you know). So, if we could reduce the ocean blip by, say, 0.15 degC, then this would be significant for the global mean — but we’d still have to explain the land blip. I’ve chosen 0.15 here deliberately. This still leaves an ocean blip, and i think one needs to have some form of ocean blip to explain the land blip (via either some common forcing, or ocean forcing land, or vice versa, or all of these). When you look at other blips, the land blips are 1.5 to 2 times (roughly) the ocean blips — higher sensitivity plus thermal inertia effects. My 0.15 adjustment leaves things consistent with this, so you can see where I am coming from. Removing ENSO does not affect this. It would be good to remove at least part of the 1940s blip, but we are still left with “why the blip”. 

Let me go further. If you look at NH vs SH and the aerosol effect (qualitatively or with MAGICC) then with a reduced ocean blip we get continuous warming in the SH, and a cooling in the NH — just as one would expect with mainly NH aerosols. The other interesting thing is (as Foukal et al. note — from MAGICC) that the 1910-40 warming cannot be solar. The Sun can get at most 10% of this with Wang et al solar, less with Foukal solar. So this may well be NADW, as Sarah and I noted in 1987 (and also Schlesinger later). A reduced SST blip in the 1940s makes the 1910-40 warming larger than the SH (which it currently is not) — but not really enough. So … why was the SH so cold around 1910? Another SST problem? (SH/NH data also attached.) This stuff is in a report I am writing for EPRI, so I’d appreciate any comments you (and Ben) might have. xxx.

Conspiracy theorists eager to find some evidence of data tampering had a lot to work with here, but not because there's any indication that any of these scientists were doing anything wrong. The ambiguity generated by the lack of explanatory context allowed conspiracy theorists to fill in the gaps in their understanding of this text with their conspiracy theory. Without knowing what Wigley meant by terms like "blip," they could just make that term mean what they wanted to support the conspiracy narrative they were hoping to find. But it turns out that with some investigation it's easy to determine that there's nothing nefarious going on here.

The first thing we should notice is that Wigley is not talking about actions he's already taken. He's "speculating" about how to handle a problem that had been known for over a decade. He's clearly working through bias correction issues having to do with a warming "blip," and he's asking for advice from Phil Jones (and Ben Santer) before he finishes his report for EPRI. In other words, this email is not about how to find the best way to fudge data (as contrarians suppose); it's asking for advice regarding how to apply bias correction to SSTs prior to 1940 (and after 1945). Wigley chose a correction of +0.15 C (warming SSTs prior to 1940) knowing that would reduce but not eliminate the "ocean blip" and asked for their responses.

Background

At the time of Wigley's email, it was already well-known that changes in SST collection methods had introduced biases into the SST record. Biases could be introduced as transitions between four different SST collection methods occurred:

• Wooden Buckets were not well insulated and would allow heat loss to occur due to evaporation between the time the bucket was lifted from the water's surface to when the thermometer reading was taken.
• Canvas Buckets were less well insulated than wooden buckets, so more heat loss would occur, introducing a cooling bias into temperature readings. The transition from wooden to canvas buckets occurred during the late 19th century, since canvas buckets were much more convenient to use.
• Rubber Buckets looked much less like buckets, and they were much better insulated than wooden buckets. Switching from wooden or canvas buckets to rubber buckets would introduce a warming bias into SST trends.
• Engine Intake methods didn't use buckets at all, and this method was the "warmest" method relative to the other collection methods. The US began switching to this method in the 1920s but many fleets switched later, by 1942 in most fleets, though some fleets (like those from Japan) didn't make the transition until the 1960s.

By 2009, SST datasets like HadSST2 had already introduced bias correction for some of these transitions in the SST collection methodology, and the correction was significant.

Corrections from HadSST2

As you can see, bias correction for changes in SSTs reduces the amount of global SST warming because it's removing a cooling bias in instrumental data. But if you look at the uncorrected (blue) data, there's a warming "blip" in the 1940s - the global mean SST values jump up rapidly at a rate that is not explainable by natural variability or forcings from GHGs and aerosols. And this cooling bias affects a few years following 1945 as well. The cooling bias is also evident by comparing land and ocean temperatures. And as the email indicates, thermal inertia makes land temperatures more sensitive than SSTs, so warming blips are typically 1.5x or 2x greater for land than for SSTs. Uncorrected SSTs were out of accord with this pattern in the early 20th century.

Wigley's Email

Wigley's email was working on this problem and seeking to improve the bias correction methods that had already been applied. His proposed correction of 0.15°C would have reduced the the amount of global SST warming by increasing SSTs before 1940 (and a few years after 1945) by a small amount. Unsurprisingly, after the CRU email hack, Wigley was able to offer some clarification about what he was saying (this link downloads the pdf). Here is his clarification about this claim that he deliberately chose 0.15°C for some nefarious purpose:

This is just shorthand for ... "if, when the correction to the SSTs due to the change in instrumentation identified by Thompson et al. (Nature, 2008) is applied, and if this correction were, say, 0.15 degC ..."

This email was directed to Phil Jones only, and Phil knew exactly what I was talking about. It does not at all refer to making some arbitrary correction to existing data in order to make such data fit some preconceived ideas about global warming. The SST correction is being made on the basis of knowledge about the instrumentation change, and this alone. The a priori (and preliminary) estimate of this correction suggests that it will make land and ocean changes more consistent, providing a valuable check on the correction procedures.  

Wigley's clarification makes perfect sense both of the email and of the known biases from changes in that had affected SSTs prior to the 1940s. Thompson et al 2008[1] had also identified a cooling bias affecting temperatures in 1945. From the paper, "We argue that the abrupt temperature drop of 0.3°C in 1945 is the apparent result of uncorrected instrumental biases in the sea surface temperature record." This was due to the fact some fleets delaying their transition to engine intake until after 1945. They write, "The adjustments immediately after 1945 are expected to be as large as those made to the pre-war data (~0.3 °C; Fig. 4), and smaller adjustments are likely to be required in SSTs through at least the mid-1960s." Wigley's value appears to have cut the value proposed in Thompson et al 2008 by half - that is, he was being conservative in his proposal.

At the time that the CRU email hack had occurred, the finished product Wigley was working on had not yet been released, but above you can see the adjustments as they appeared in HadSST3. Now in 2024 we've moved passed HadSST3 to HadSST4, and the "warming blip" is still there. It has not been fully removed. But the applied corrections (which reduced but did not eliminate the blip) increased temperatures prior to 1940 (and around 1945) and therefore reduced the amount of global warming. In fact, these bias corrections were larger than corrections for warming biases affecting land temperature readings, meaning that the net effect of bias correction on global temperatures (land + ocean) was to reduce the amount of global warming. In other words, we can say with certainty that these bias corrections were not performed to make the data fit some predetermined narrative that global warming is a problem. These corrections would have slightly reduced the magnitude of the global warming problem.

Conclusion

Interestingly, in my last post I shared about a recent study evaluating cooling biases affecting SST measurements that identified a continued cooling bias affecting 1900-1930 (but not significantly affecting the 1850-1900 mean). If this study withstands scrutiny, it would show that Wigley (and others) had actually underestimated the cooling biases affecting SSTs for the early 20th century, especially due to the transition from wooden to canvas buckets. Corrections to the 1900-1930 cooling bias would further reduce the "warming blip" discussed in this email. In other words, what this paper shows is that Wigley and other scientists have been pretty careful and conservative about applying bias correction. But if they wanted to support a global warming narrative, they would have continued to use the raw SST data without bias correction, and they could have removed the "blip" by lowering SSTs during WWII.

There is simply no evidence in this email that Wigley (or others) were using bias correction to fabricate warming or fudge/manipulate SST data to conform to some predetermined narrative. One of my college professors used to say repeatedly, "The more you look, the more you see, but you only see what you're looking for." That is, we have to be careful with our presuppositions when looking for evidence. If you have already presumed something to be true, you're more likely to "find" evidence for the things you've convinced yourself about without looking for alternative explanations. This appears to be a clear case of conspiracy theorists approaching this email with the presumption that data fudging is happening, and then "finding" evidence for this by reading between the lines to insert context that isn't there and define terms in ways that would be consistent with their presumptions, regardless of what Wigley actually meant by what he said. But with more awareness of the actual context involved, there's no reason to think that this email is evidence of any data manipulation; the corrections would actually have reduced the amount of global warming.

References:

[1] Thompson, D. W. J., Kennedy, J. J., Wallace, J. M., & Jones, P. D. (2008). A large discontinuity in the mid-twentieth century in observed global-mean surface temperature. Nature, 453(7195), 646–649. doi:10.1038/nature06982