Application of artificial neural network (ANN) for investigation of the impact of past and future land use–land cover change on streamflow in the Upper Gilgel Abay watershed, Abay Basin, Ethiopia

Abstract

Changes in land area features, such as vegetation type and soil conditions, have an impact on carbon sources and sinks and support food production; this is critical in addressing global sustainability challenges such as climate change, biodiversity loss, and food security. The study's major goal was to determine how LULC changes in the past and future might affect streamflow in the Upper Gilgel Abay watershed. The modeling was conducted using the MOLUSCE Quantum GIS plugins cellular automata simulation method and streamflow modeled using SWAT. Landsat 5 TM (1995), Landsat 7 ETM + (2007), and Landsat 8 imaging (2018) satellites were used to collect the images, which were then categorized using ERDAS 2014 software, and the kappa coefficient becomes 84.04%, 82.37%, and 85.54% for 1995, 2007, and 2018 LULC, respectively. SWAT model better performed the simulation which is R2 of 0.77 for calibration and 0.68 for validation and ENS becomes 0.71 and 0.62 for calibration and validation, respectively. The output change in streamflow due to past and future LULC maps shows an increase in LULC in cultivated areas and resulted in 39%, 46.81%, and 52.45% in each of the years 1995, 2007, and 2018, respectively. The three LULC modifications in the land cover maps from 1995, 2007, and 2018 had simulated mean monthly peak discharges of 62.20 m3/s, 66.51 m3/s, and 72.10 m3/s, respectively. The projected LULC 2027 also shows a similar increase in the study area, and dominantly cultivated land illustrates the highest change at around 53.77% but the highest change occurs on grassland during (2018–2027) land use at around 12.29%. And the highest streamflow was found around a monthly average of 1400 m3/s. The study primarily provides insight into how LULC fluctuation affects streamflow, and it is crucial for water planners and natural resource professionals whose focus is on the Upper Gilgel Abay basin.