Spatio-temporal assessment of rainfall erosivity in Ecuador based on RUSLE using satellite-based high frequency GPM-IMERG precipitation data

Abstract

Soil erosion caused by rainfall is one of the main soil degradation problems in Ecuador and has major implications on conservation and management of water and soil resources, disaster control, agricultural protection and other applications in the Ecuadorian territory. Extensive precipitation time series are required to estimate the rainfall erosivity as a triggering element of soil erosion, however, meteorological stations distribution in Ecuador is not homogeneous over the national territory (greater concentration in coastal basins) and present high lack of data, generating an incorrect rainfall distribution that requires resorting to other types of available information. The objective of this work was to analysis the spatio-temporal distribution of potential soil erosion through rainfall erosivity in Ecuador, through an innovative method based on the Revised Universal Soil Loss Equation (RUSLE) employing GPM-IMERG rainfall data. Our methodology considered the delimitation and evaluation of the watersheds environmental characteristics, GPM-IMERG, GPCC and ERG-data evaluation as providers of rainfall databases, and calculation of the erosive force of the rain (R-Factor) over the period 2001–2020. Basins delimitation and layers of environmental characteristics were obtained from regional literature; GPM-IMERG and GPCC from web servers; ERG-data from INAMHI-Ecuador; R-Factor (30 min) calculation was made from GPM-IMERG parameters and its was validated with RIST software using a representative subset of pixels, revealing an excellent compatibility of almost 1% average difference. New information was obtained from spatial, interannual and seasonal R-Factor distribution considering environmental characteristics, even though the satellite image products (GPM-IMERG) still underestimate the high values of R-Factor compared to Literature. New models for Ecuador have been established to estimate the R-Factor from annual rainfall, which allow it to be calculated since the beginning of the last century as well as generate projections of future soil erosion. This helps territory managers to take new measures to deal with soil loss.