Estimation of precipitation in remote areas using B-SHADE model

Abstract

Meteorological stations often have biased distributions in remote areas, and sample biases typically lead to biased estimation of areal precipitation. We introduce a method termed the Biased Sentinel Hospital based Area Disease Estimation (B-SHADE) to correct for the bias in the locations of meteorological observations and to obtain a best linear unbiased estimation of the mean areal precipitation. We evaluated the performance of the technique using a precipitation dataset from 1998 to 2011 in Qilian Mountains, northeastern Tibetan Plateau. Our experiments showed that compared to arithmetic average, Thiessen polygon, and kriging, the estimations based on the B-SHADE model were better than results from these traditional methods.