I compiled the RC comments on how much impact the harry data has.
[Response: No-one should be against better data. It would have been nice had SM actually notified the holders of the data that there was a problem (he didn’t, preferring to play games instead). If he hadn’t left it for others to work out, he might even have got some credit ;). As for the Steig et al paper, the typo in Table S2 is just a typo, as could be seen from examining the location files on Eric’s website or noting that the figure S4b has the correct location for Harry. As for the implications of the errors in the BAS Harry file on the study, that too is visible in figure S4b – removing Harry (and a bunch of other AWS stations) doesn’t change the answer in any meaningful respect. Correction it has no impact on any of the reconstructions made with the AVHRR data. Thus the answers to the ‘questions that are being raised’ were obvious all along. We might post some more on this later. – gavin] (Id’s emphasis)
Then there was this post – very little effect.
Is harry in the full reconstruction or not?
[Response: And he gave a simple answer, which is the same answer that is very very obvious to anyone reading our paper. The answer is NO.–eric]
Again later on someone asked. No meaningful change.
What I am wondering is, with the corrected Harry data, will the calculations done in the paper still yield the same results and lead to the same conclusions?
[Response: Yes. Already done it. No meaningful change (except at Harry itself of course). Furthermore, the main figures in the paper use no AWS data in the first place. None. This is totally clear in the text, if you read it. The AWS data are only a double-check on the results from the satellite data.–eric]
and another one.
[Response: The reconstruction using the AWS stations obviously used AWS stations and the existing data is not ‘reconstructed’ – the method fills in missing data points, it doesn’t change the data that was measured. But the main reconstruction uses AVHRR data, where none of the AWS stations are used. – gavin]
Here’s another
[I already answered this question at least once above. The answer should have been obvious from the beginning to anyone that actually sat down and thought about it, but I’ve done the calculations anyway. I’ll give you a sneak preview of Gavin’s post: the difference it makes is about 8% at the offending AWS site (Harry) and less than 2% overall. Oh, and less than 1% in West Antarctica. Oh, and I forgot to mention that the reconstructed AWS now agrees better with the satellite data than before. Not suprising either, if you think about it for a minute.–eric]
How many times can they say the same thing.
[Response: The effect on the data actually used by Steig et al for the results featured in the paper (the satellite-derived ice surface temperature estimates) is precisely zero. The effect on the AWS data used as a secondary check by Steig et al could conceivably be non-zero. But for it have any noticeable impact, snow burial between successive AWS checks would have to be decidedly non-random in nature, exhibiting a continental-scale coherence, and a long-term temporal trend to boot. And somehow one would still have to get the same answer using the satellite data, you know, the data that were actually used for the results featured in Steig et al. -mike]
any comments on how this will effect the error in the overall trend estimates?
is it still the same?
cheers,[Response: I’m sure it will make no meaningful difference whatsoever. And keep in mind, as is very clear in the paper, the using the AWS is not the best method in the first place. There simply is not enough AWS data over enough time to demonstrate good validation at most places. That’s why we used the satellite data, not the AWS data, as our primary tool for estimating the spatial covariance of the temperature field.–eric]
On Eric’s webpage he posted this.
Note: Minor corrections to the AWS-based reconstruction are required, due to the discovery of errors in two of the AWS stations. See note under “Raw Data” below. A revised reconstruction will be posted here. The resulting differences in the reconstruction are too small to be discernable on the Figure S3, or in the trends for individual stations given in Table S1 in the Supplementary Information that accompanies the paper in Nature. (Corrections for all stations in the Table S1 are in the third decimal place (that is <0.01 degrees C/decade)). The mean trend for Antarctica changes by less than 0.004 degrees C/decade. The mean trend for West Antarctica changes by less than 0.02 degrees C/decade. Note also there is a typo in Table S2. The correct coordinates for station ‘Harry’ are 83.0 S 238.6 E.
So the corrections have only got meaning for the Automatic Weather Station AWS reconstructioins and that is apparently small. It’s a good thing they weren’t forced to give McIntyre credit for finding the problem.
2 February 2009 at 8:54 PM So what do the errors noted by SM do to the results of Steig’s report? The error was used in the Steig analysis, was it not?
[Response: Very little, but we’ll post something on this tomorrow. In the meantime, note that it is basically only figure S3 that uses the Harry AWS data (along with a correlation used as validation in fig. S4b and the dashed line in fig 2). – gavin]