When a group of federal drought researchers invited me to speak at their workshop in Santa Fe back in 2006, I titled my talk “Maybe we shouldn’t call it ‘drought’” – or words to that effect. My problem with the word “drought” is twofold. The first is that we don’t have a common language word for the opposite. (The actual word is “pluvial”, but when have you heard that used?) The result is that, when things get wet, we just take it for granted, not thinking too much about the fact that we’re in the wet site of the natural range of variability. Then we it swings to dry, we freak out.
But both pluvials and droughts are, in fact, part of the normal variability in climate.
The second problem has been on display over the last few days in a series of discussions, email and by phone, with drought researchers about a paper in Nature last week that concluded that, by one measure, there has been no global increase in drought over the last 60 years. (Discussion on my other blog here and here.)
You’ll see the results of a lot of that discussion in tomorrow’s newspaper. A lot was far too deep into the technical weeds for a newspaper, but I’d like to share, with permission, some of the most interesting bits here.
The Nature paper argues that the Palmer Drought Severity Index, a standard measure of drought conditions, shows no global trend over the last 60 years. There’s been some pushback from others who believe the Nature authors have gotten some of the math wrong. I have no dog in that fight. But a lot of the more interesting discussion revolves around the Palmer index itself, and whether it’s the right tool for this job. Here’s Dave Gutzler, UNM climate guy (I learned about Palmer when taking his drought class some years ago):
Despite its many limitations PDSI is still a meaningful indicator of short term climate variability (interannual, maybe up to decadal, time scales). If you look at the uncertainties described by Sheffield et al., they are on scales of tenths of PDSI units on continental-global scales. Big short term droughts, which are primarily driven by precip anomalies, will be captured by PDSI regardless of the method used to describe evapotranspiration. For that matter, the magnitude of long-term change looking forward, associated with really big century scale temperature change, will show up as ‘long term drought’ regardless of the details of the PDSI calculation. That’s why so many different hydrologic indicators (streamflow, vapor pressure deficit, etc.) that do not incorporate similar debatable ET parameterizations all point in the same direction. And that’s why I don’t think the Sheffield paper should require us to reject the validity of the core conclusions presented in so many recent papers.
More in tomorrow’s newspaper, and I’ll have more to share here as I get permission from the people who have been sharing on a particularly useful email thread.