Relative or Absolute?

The National Snow and Ice Data Center has detailed graphs and charts showing the extent of Arctic sea ice. They’ve recently noted that they were suffering from a short-term problem called “sensor drift” that gave false readings on Arctic ice coverage. It happens from time to time and they try to adjust for it. This seems to be a responsible way to manage scientific data collection. The problem comes with some shifting claims made later in the article.

You see there are two major methods for measuring ice coverage. There is the SSM/I sensor and the newer AMSR-E sensor. The AMSR-E is more accurate, but has only been collecting data since 2002. The SSM/I is less accurate, but has a longer track record. This leads to a scary sentence at the end of the article:

Some people might ask why we don’t simply switch to the EOS AMSR-E sensor. AMSR-E is a newer and more accurate passive microwave sensor. However, we do not use AMSR-E data in our analysis because it is not consistent with our historical data.

This can lead to cries of “they’re using the data they like and ignoring the more accurate sources” which isn’t quite true. Actually, what they claim is that they want to continue to use the less accurate data because they have a longer history and, therefore, it is easier to determine trends. The argument is that recognizing trends is more important than the actual measurements.

I’m willing to buy that, but that’s not what they were saying last May:

Taken together, an assessment of the available evidence, detailed below, points to another extreme September sea ice minimum. Could the North Pole be ice free this melt season?  Given that this region is currently covered with first-year ice, that seems quite possible.

So, here’s a claim that the NSIDC can predict actual measurements of ice coverage (namely, a prediction of zero or near-zero levels). So, despite the fact that their recent sensor drift problem was off, at times, by over 193,000 square miles and they knew that the AMSR-E data was more accurate, they were willing to make absolute measurement predictions last May. Now, they play down any concerns with their data collection by falling back to the “trends” argument.

Sure sounds like they want to have it both ways.

Leave a Reply

Your email address will not be published. Required fields are marked *