Assessment of areal interpolation methods for spatial analysis of SPI and EDI drought indices



Drought monitoring is an essential component of drought risk management. Drought indices—functions of precipitation showing the severity of dryness during a particular time period—are often used for monitoring purposes. These indices may only be calculated originally at a limited number of sites where observations on climate variables are available. However, what is required for monitoring is to estimate the spatial distribution of drought severity over larger areas in the form of maps. In this article, several geostatistical methods including kriging, co-kriging and thin plate smoothing splines with and without secondary variables, as well as Thiessen polygons and weighted moving average were assessed for the derivation of maps of drought indices. The techniques are evaluated using 1-month Standardized Precipitation Index (SPI) and Effective Drought Index (EDI) at 43 climatic stations in the Tehran province of Iran. The results indicate that although kriging is the most accurate method, weighted moving average provides a reasonable accuracy combined with the simplicity and speed of the procedure, and thus, can be recommended for operational drought monitoring system in the region. Copyright © 2008 Royal Meteorological Society