Fusion of satellite precipitation products and ground-based measurements using LightGBM with a focus on extreme quantiles

Abstract

Satellite precipitation products are not accurate in representing the actual precipitation measured by gauges. To improve their accuracy, machine learning algorithms are applied in regression settings with ground-based measurements as dependent variables and satellite precipitation data as predictor variables. Here we examine the case of light gradient-boosting machine (LightGBM) for correcting daily IMERG (Integrated Multi-satellitE Retrievals for GPM) and PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) precipitation data using daily precipitation measurements in the contiguous US. Our demonstration especially focuses on the estimation of quantiles of the conditional probability distribution of daily precipitation at given points, with emphasis on extreme values.