Assessing spatio-temporal dynamics of land degradation neutrality using Goggle Earth Engine in the Alawuha Watershed of North Wello Zone, Ethiopia

Abstract

Land Degradation Neutrality (LDN) aims to maintain or enhance land resources and is assessed using three indicators: land cover change, land productivity dynamics, and soil organic carbon (SOC). This study analyzed LDN dynamics in Alawuha Watershed, Ethiopia, using remote sensing data of these indicators within Google Earth Engine (GEE) platform to identify areas requiring intervention. The study used Landsat 5 TM and 8 OLI imagery (2000–2023) for land use/land cover (LULC) change analysis with the Random Forest algorithm, MODIS Net primary productivity (NPP) data (2001–2023) for land productivity, and gridded SOC data to map SOC distribution. The spatial distribution of NPP trends was analyzed using the Theil-Sen slope estimator in GEE, and the modified Mann-Kendall (MK) trend test in RStudio was applied to analyze mean annual NPP trends. The "one out, all out" principle was used to evaluate the overall LDN status. Results revealed an increase in forest cover (+1.95 % annually) and built-up areas (+11.77 % annually) from 2000 to 2023, while farmland declined at an annual rate of −1.23 %, mainly due to urbanization. The spatial distribution of the NPP trends showed productivity gains in the western and southern regions (+35.85 gC/m²/year) but declines in the northeastern region (−10.99 gC/m²/year) and lower SOC, signaling localized degradation. The MK test confirmed a significant increase in mean annual NPP (tau = 0.3913, p = 0.00078) at (α=0.05), with a Sen’s slope of 7.51 gC/m²/year, showing overall productivity improvement. These findings support LDN targets and provide insights for sustainable land management and restoration policies.