Evaluation of Land Use and Land Cover Change Impacts on Streamflow and Sediment Yield, and Developing Best Management Practices: The Wollo Highland of Borkena Watershed in Ethiopia

Abstract

LULC Change Has Been Identified As A Major Research Priority Around The World, With Multi Directional Consequences For Both Human And Natural Systems. Borkena Watersheds Were Among The Most Vulnerable Natural Systems And Continue To Face A Number Of Difficult Issues In Water And Land Resource Management As A Result Of Rapid Population Growth, Urbanization, And Industrialization. The Overall Objective Of This Research Was To Assess The Impact Of LULC Change On Streamflow And Sediment Yield, As Well As To Develop Best Management Practices (Bmps), Which Are Critical For Long-Term Sustainable Water And Land Resources Management And Development In The Borkena Watershed. The LULC Change Analyses For Three Periods (1990, 2006 And 2020) Were Performed Using ERDAS Imagine 2014 With A Maximum Likelihood Classifier. The SWAT Hydrologic Model Is Used To Investigate The Impact Of LULC Change On Streamflow And Sediment Yield, As Well As To Identify And Prioritize Critical Sub-Watersheds Whose Annual Sediment Yield Limit Ranges Above The Tolerable One For Developing Watershed Best Management Practices. The SWAT Model's Performance In Simulating Catchment Hydrology And Sediment Yield Was Evaluated Using Sensitivity, Uncertainty Analysis, Calibration, And Validation With Coefficient Of Determination (R2), Nash Sutcliff Efficiency (NSE), And % Bias (PBIAS) Standard Statistical Indices. The SWAT Model Was Calibrated And Verified To Evaluate The Effect Of LULC Change Streamflow And Sediment Yield In This Watershed. Using SWAT-CUP, A SWAT Model's Streamflow And Sediment Yield Were Calibrated For The Years 1990?�?2007 And Validated For 2008?�?2015. The Statistical Analysis Results Revealed That The Observed And Simulated Values Of Streamflow And Sediment Yield Agree20d Well With R2 =0.82, NSE =0.69, PBIAS = -17.3, P-Factor = 0.71, And R-Factor = 1.64 For Streamflow Calibration And R2 =0.81, NSE =0.65, PBIAS = -14.3, P-Factor = 0.6, And R-Factor = 1.86 For Streamflow Validation. Similarly, For Sediment Yield Calibration, R2= 0.81, NSE= 0.72, PBIAS= -14.6, P-Factor = 0.61, And R-Factor = 0.71 And R2 = 0.8, NSE = 0.7, PBIAS = -17.2, P-Factor = 0.72, And R-Factor = 1.02 For Sediment Yield Validation. Cropland Increased From 25.50 To 32.30%, Settlement Increased From 1.30 To 7.80%, And Bareland Increased From 0.01to 2.90% Between 1990 And 2020, According To A Land Cover Change Analysis. While Forest Declined From 47.20 To 32.70%, Grassland Declined From 8.30 To 5.40%, And Shrubland Declined From 17.50 To 4.80%. As A Result Of LULC Changes, Changes In Streamflow And Sediment Yield Were Observed. Based On This, Simulated Model Outputs Using 1990, 2006, And 2020 LULC Showed An Increase In Average Annual Water Yield Of 35.3 Mm (4.7%) And 56.5 Mm (7.2%) Between 1990 And 2006, Respectively, While Sediment Yield Increased By 4.4 T/Ha (81.4%) Between 1990 And 2006, And By 2.8 T/Ha (28.2%) Between 2006 And 2020. Based On The Average Annual Simulated Sediment Yield Using 2020 LULC, The Six Critical Sub-Watersheds With Annual Sediment Yield Limit Ranges Above The Tolerable Limit Were Identified And Prioritized For Effective Watershed Management. As A Result, Four Bmps, Namely Filter Strips, Terracing, Contour Farming, And Reforestation, Were Developed For Critical Sub Watersheds To Reduce Sediment Yield. The Results Revealed The Terracing Management Option Reduce The Sediment Yield Up To 72.32%. Implementing Filter Strips With 5m And 3 M Wide Indicates 57.12% And 50.67% Reduction Respectively. Contour Farming Alternative Reduce The Sediment Yield Up To 63.56% Whereas, Reforestation Has Given The Least Reduction That Is 33.81%. Therefore, Applying Terrace Management Option In Critical Sub-Watershed Is The Most Effective Practice For Reducing Sediment Yield In The Wollo Highland Of Borkena Watershed In Ethiopia.