SOIL EROSION MODELLING AND SEDIMENT YIELD FOR RUFIJI RIVER BASIN IN TANZANIA

Abstract

Water Resources Management is a many-sided process that requires the entanglement of many other pieces of information to underpin various decisions in society. The land surface and water are the important resources of the universe, therefore, they have to be well maintained since they are ecologically sound. However, among the prerequisites for any proper planning for agriculture and Management of rivers are the details on erosion and sediments. Time and again, sediment distribution in rivers usually brings on dwindling of the storage area, the life cycle of reservoirs, the quality of water, river discharges leads to water shortage for other several usages, likewise, sediments induce uneconomical purification methods for water. Modeling may deliver a measurable and compactible technique to evaluate soil and sediment yield erosion based on a vast scope of circumstances. In this thesis work, the soil erosion model, Revised Universal Soil Loss Equation (𝑅𝑈𝑆𝐿𝐸) mingled with the Geographical Information System (𝐺𝐼𝑆) tool, has been employed to approximate soil loss under five different scenarios of precipitation products in the Rufiji River Basin situated in Southeastern Tanzania. Remote sensing data can offer the accessibility to quantify the 𝑅𝑈𝑆𝐿𝐸 input factors such as rainfall erosivity (𝑅), Soil erodibility (𝐾), slope length, and steepness (𝐿𝑆), Cover management (𝐶), and erosion control (𝑃) while the areas which are vulnerable to erosion were mapped by 𝐺𝐼𝑆. The erosivity factor (𝑅) for the Rufiji river basin was derived from 𝑂𝑏𝑠𝑒𝑟𝑣𝑒𝑑 precipitation and other four different satellite precipitation datasets consisting of twenty years of daily precipitation (1999−2019); The Modern-Era Retrospective analysis for Research and Applications (𝑀𝐸𝑅𝑅𝐴−2), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (𝑃𝐸𝑅𝑆𝐼𝐴𝑁𝑁− 𝐶𝐷𝑅), Tropical Rainfall Measuring Mission (𝑇𝑅𝑀𝑀−3𝐵42 𝑉7), and the Global Precipitation Climatology Project (𝐺𝑃𝐶𝑃). The observed gauged precipitation appeared to be the best dataset since it produced better soil loss and sediment yield assessment results. According to the findings, the Rufiji River basin's overall annual potential soil erosion is estimated to be 1,916,753 t/yr, 863,238 t/yr, 843,069 t/yr, 789,606 t/yr, and 690,625 t/yr based on 𝑂𝑏𝑠𝑒𝑟𝑣𝑒𝑑 precipitation, � �𝐸𝑅𝑅𝐴−2, 𝑃𝐸𝑅𝑆𝐼𝐴𝑁𝑁−𝐶𝐷𝑅, 𝑇𝑅𝑀𝑀, and 𝐺𝑃𝐶𝑃 respectively. Since the observed data for erosion was unattainable in the Rufiji river basin, the erosion simulation model (𝑅𝑈𝑆𝐿𝐸) was validated by the concept of sediment delivery ratio applications. The area under low erosion composed of 47.3% in Great Ruaha, 23.5% in Kilombero, 39.5% in Lower Rufiji, and 39.4% in Luwegu Catchment; while the area under very high erosion composed of 7%, 34.2%, 3.5%, and 13.2% in

xix

Great Ruaha, Kilombero, Lower Rufiji and Luwegu respectively. Around 76.5% of the Rufiji river basin needs abrupt consideration from soil protection experts, especially at Kilombero Catchment. The simulated sediment yield of approximately 1,648,408 t/yr is comparable with the actual sediment of 1,655,056 t/yr; since the sediment delivery ratio was found to be 0.86. The correlation between elected water quality elements, such as 𝑇𝑟𝑎𝑛𝑠𝑝𝑎𝑟𝑒𝑛𝑐𝑦, 𝑇𝑆𝑆, 𝑁𝑂3, 𝑝𝐻, 𝐷𝑂, and sedimentation along water sampling sections was investigated. It was discovered that the sediment yield and water quality parameters have a strong positive relationship. Sediment yield had a strong positive relationship with 𝑇𝑆𝑆 and 𝑡𝑢𝑟𝑏𝑖𝑑𝑖𝑡𝑦 with a correlation of about (0.97). The reliability assessment of global satellite precipitation regarding observed precipitation at the monthly timescale depicted that the 𝑀𝐸𝑅𝑅𝐴−2 and 𝑃𝐸𝑅𝑆𝐼𝐴𝑁𝑁−𝐶𝐷𝑅 satellite products perform well regarding 𝑅2, 𝑁𝑆𝐸, 𝑎𝑛𝑑 𝑅𝑀𝑆𝐸 statistics with 𝑅2, 𝑁𝑆𝐸, and 𝑅𝑀𝑆𝐸 of 0.9, 0.74 𝑎𝑛𝑑 4.2 𝑚𝑚/ 𝑚𝑜𝑛𝑡ℎ for 𝑀𝐸𝑅𝑅𝐴−2, while 0.9, 0.89 and 6.2mm/month for 𝑃𝐸𝑅𝑆𝐼𝐴𝑁𝑁−𝐶𝐷𝑅 respectively. Generally, the results prescribe that a bulky total of dissipated Soil was discharged from the Rufiji river sanctuary per year. The subsequent sediments normally scaled-down the dam capacity upstream of the basin and might threaten the ecology of the river network. On the grounds of this, it is recommended that the basin requires instant consideration from a soil protection standpoint, particularly in the Kilombero catchment.