Utilising Plutonium Isotopes to Evaluate Soil Erosion in Tropical East African Agri-systems

Abstract

In the beautiful landscapes of Tropical East Africa, agriculture serves as the lifeblood of both people and ecosystems. The delicate balance between agriculture and soil preservation forms a complex web of factors that significantly impact soil and plant health in the region. The use of plutonium (Pu) isotopes is a potential strategy for unravelling the history of soil erosion in this ecologically varied and agriculturally crucial area. This thesis investigates a novel method using Pu isotopes to study small-scale soil redistribution patterns across various land use histories and management practices in the Rift Valley region of the Winam Gulf catchment of Lake Victoria in western Kenya. The study aims to develop a method for measuring Pu isotopes in African soils to provide measurements of soil redistribution patterns, to better understand the factors that drive erosion processes. Analysing tropical soils poses specific challenges, and so the first objective was the optimisation of analytical methodology to enhance sensitivity. The developed separation procedure used TEVA resin to effectively reduce UH+ interference while improving the selectivity of analysis for Pu. The application of O2 gas in ICP-MS/MS mode for the analysis of samples facilitated the mass shifting of Pu isotopes away from interfering UHn+ ions, resulting in a straightforward, robust, and cost-effective method suitable to detect ultra-trace fallout 239+240Pu in African soils with detection limits of 0.18 pg kg-1. The second objective was to understand the usability of 239+240Pu as a soil erosion tracer in western Kenya compared against conventional isotopes 210Pbex and 137Cs. The lowest coefficient of variation and greatest peak-to-detection limit ratio was found for 239+240Pu showing the greatest potential as a tropical soil erosion tracer. Additionally, 239+240Pu met the 'allowable error' criteria, establishing its applicability to large-scale studies in Western Kenya, where the selection of appropriate reference sites is a significant challenge. Consequently, 239+240Pu stands as a robust tracer for assessing soil erosion pattern estimates in western Kenya. Finally, soil erosion patterns were modelled using the MODERN model at sites with distinct land use and clearance scales to gain insights into small-scale erosion processes and the influence of differing management practices. This research highlights the importance of community-led participation in the effective management of land degradation processes and highlights the pivotal role that vegetation cover plays in limiting soil erosion. This innovative application of fallout Pu as a tracer can improve our understanding of how soil erosion processes respond to land management practices, thereby supporting the implementation of effective mitigation strategies. Additionally, Pu is presented as a robust soil erosion tracer applicable to large-scale studies in tropical Africa. Data derived from Pu activities will facilitate the validation of predictive models, paving the way for community-designed solutions to combat land degradation and inform future related policy.