Simulation of nutrient load and identification of critical areas for better watershed management using SWAT model : case study : River Manafwa Catchment

Abstract

Nutrient deposition in water bodies is a major cause of pollution. Therefore, the need to do modelling of nutrient load at a watershed level is very vital. SWAT 2012 (Soil and Water Assessment Tool) was selected for the simulation of the nutrient load in River Manafwa Catchment in Uganda. The main objective of the study was to estimate nutrient load and identify critical areas for better watershed management. To setup the model for simulation, a DEM (Digital Elevation Model), land use map, soil map and weather data (temperature, solar radiation, wind speed, precipitation and relative humidity) was used. The stream flow data was available from 2000 to 2013.The model was calibrated using SUFI algorithm in SWAT-CUP by using the flow data from 2003 to 2007 and validated for 2008 and 2013.The model was not validated for nutrient load because nutrients data was not available. The performance of the model was evaluated by using a time series plots of observed and simulated value and the statistical measures of coefficient of determination(R2) and the Nash Sutcliffe Efficiency(NSE). The statistical analysis of calibration results showed agreement between observed and simulated daily values with an R2 value of 0.74 and NSE of 0.61. The R2 and NS value for flow validation period was 0.61 and 0.53 respectively. The model identified sub basins 4,7,11,24 and 25 as the sub basins that generate the highest amount of nutrient load in the watershed and 2,6,18,19,20,22 and 23 as sub basins that contribute the least nutrient load to the in the water shed. The simulation results run with BMPs indicate that BMPs can greatly reduce on the amount of nutrient load generated from the watershed into the river. The simulation results run with BMPs indicated that grassed water way with an overall reduction efficiency of 59.45% for TN and 67.30% for TP is the most efficient BMP to adopt for the entire watershed. Taking into consideration the influence of Land slope, contouring with an overall reduction efficiency of 50.35% for TN and 60.06% for TP can be considered for implementation in the mountainous and hilly regions of the watershed. Terracing with an overall reduction efficiency of 51.95% for TN and 61.89% for TP can be Adopted for implementation in the low land and relatively flat regions of the watershed. Generally, the model was capable of simulating nutrient load in River Manafwa Catchment