Following on Daniel Collins’ discussion of the different measures and definitions of drought, here are a few useful web pages that pull together various indices.
One of the problems Daniel pointed out about my use of the Standardized Precipitation Index is that it only captures precipitation anomalies, which misses the temperature side of the equation. If it’s hotter than normal, you get more evapotranspiration, which makes drought conditions worse. SPI fails to capture that. The standard tool to get around that is the Palmer Drought Severity Index, which also considers temperature. The Palmer was developed in the 1960s as a tool to estimate soil moisture conditions using temperature and precipitation data. Its problem is that it’s such a scary black box that no one quite knows what’s going on inside it – what contributes to a particular Palmer number. It’s widely used, but you will find a lot of people in the drought community who are uncomforable with it. (In fact, that’s why SPI was developed.)
There’s a new tool that’s getting a lot of attention – the Variable Infiltration Capacity (VIC) model. It uses a much more realistic soil-weather-hydrologic system model. Dennis Lettenmaier’s group at the University of Washington is using it to produce regularly updated maps of soil moisture and other variables.
Then at Princeton, Lifeng Luo, who’s working with Lettenmaier, has a web page bringing together a whole bunch of different drought measures in one place. This may be the best approach of all. Given the slippery issue of defining drought, if you’re trying to pay attention to the issue, it’s probably best to look at many different indices and see what overall picture they are painting.