Geoinformatics of soil loss using GEE in Tigray regional state of Ethiopia.

Abstract

Abstract Soil loss and its geostatistical analysis was studied at the kebele level in Tigray. The method applied to estimate soil loss was the revised universal soil loss equation. Earth Engine's public data archive was used for data collection. The R-factor was developed from the SM2RAIN-ASCAT (2007–2021) global daily satellite rainfall data, the K-factor was developed from USDA-3A1A1A_M/v02 soil data, the C-factor was derived from MODIS/006/MOD13A2, and LS factor was derived from WWF Hydro SHEDS Hydrologically Conditioned DEM. By integrating all factor, the soil loss was obtained by the RUSLE model. Spatial Autocorrelation (Morans I) statistic was used to identify the pattern of soil loss and Ordinary Least Squares (OLS) linear regression was used to model a soil loss in terms of its relationships to R, K, LS, C, and P factors. The grouping analysis tool was used to Group kebele based on soil loss. The results indicate that the estimated average soil erosion is 82760 t ha− 1 y− 1. The pattern of soil loss at the kebele level was found highly clustered with a z-score of 23.39. The groping analysis tool divides the kebele into five categories to identify the cause of spatial variation of the soil loss in Tigray. Groups 1, 4 & 5 were found as in the outlier positions due to the high LS factor. The results deliver valuable information for decision-makers and planners to take suitable land administration measures to minimize the soil loss. It, therefore, indicates google earth engine is a significant platform to analyze the RUSLE model for evaluating and mapping soil erosion quantitatively and spatially.