Prediction and Estimation of Sediments Discharge for Reservoir Management

Abstract

Sediment transport in watersheds is associated with a wide variety of environmental and engineering issues which include erosion, loss of topsoil, high sediment loads to reservoirs, chemical transport of nutrients, silts and clays generation, and channel instability among others. The rapid reduction in the storage capacity of reservoirs due to siltation is a major sediment- related problem world-wide, which has lead to the reduction of reservoir storage volume and subsequent shortage of water supply. Irrigation activities further lead to increased sediment discharge and soil erosion. Monitoring data for sediment transport into reservoirs and productivity within reservoirs are therefore required for accurate predictions of sediment transportation to curb water loss, economic resources, and for good reservoir management. The aim of this research was to predict and estimate the sediment discharge inflows into Kangimi Dam Reservoir, Kaduna State, Nigeria with a catchment area of about 262.74 km2. Soil and water samples were collected from representative fields. The soil samples were analyzed for particle distribution using the Bouyoucos Hydrometer method. Soil organic matter (SOM) was determined using wet digestion method of Walkley and Black, while erodibility index was estimated using the Bouyoucos, Middleton, and Pieri methods. The suspended sediment concentrations (SSC) (Cs) were determined from the water sample mixtures using filtering and drying methods in the laboratory. The empirical relation between sediment concentrations as a dependent variable was expressed graphically using logarithmic transformation data with length, drainage area, drainage slope, land use, and herds’ activities as independent variables. The average soil particle size distribution, silt/clay, and sand ranged from 34.46% to 75.13%, and 33.05% to 65.54%, respectively, while the average organic matter (OM) ranged from 2.89% to 4.67%. Soil erodibility indices ranged from 1.63 to 4.08; 0.71 to 1.54 and 0.037 - 0.109, for clay ratio (CR), dispersion ratio (DR), and critical level of organic matter (CLOM), respectively. The modified clay ratio (MCR) ranged from 1.36 to 3.57. The sediment loads in Loko Koro/Ali, Loko Balbela/Bahago and Loko Gora/Dangora of sub-catchments showed downward trend with coefficient of determination, R2 = 0.46, 0.13, and 0.77, respectively, while Kangimi main/Rafin Kurmin Zaria and Jagiwa/Hadashi sub-catchments showed an upward trend with R2 = 0.26 and 0.38, respectively. The slope versus sediment load coefficient of correlation (R) for all the sub-catchments ranged from 0.92 to 0.97. For the drainage area versus sediment loads, the log exponential model was statistically significant at (α = 0.001, p = 0.00). The developed model positively predicted well with effect of cattle herds’ activities with R2 and R values equal to 0.67, and 0.71, respectively, at p=0.000, and in comparison with existing models, the own model result performed better. The study concluded that the model reasonably matched the observed data and moderately predicted within the data without deviation, and recommended the use of the model for the Kangimi Dam reservoir sediment discharge.