Abstract

Historical rainfall data spanning from 1901 to 2021, sourced from the Climatic Research Unit (CRU) at the University of East Anglia (UEA), forms the foundation of this study. The study employs the Generalized Extreme Value (GEV) and Generalized Pareto distributions to model extreme rainfall events and evaluate their statistical properties, such as return levels and exceedance probabilities. A Bayesian framework is utilized to estimate the parameters of these distributions, offering a robust and flexible approach that integrates prior information and quantifies uncertainties.
Our Bayesian analysis incorporates Markov Chain Monte Carlo (MCMC) methods to derive posterior distributions of the GEV parameters. This method not only improves parameter estimation accuracy but also enables the integration of expert knowledge and historical climate data. Sensitivity analyses are performed to assess the impact of varying priors and ensure the robustness of the model.
Preliminary findings suggest notable trends in the frequency and intensity of extreme rainfall events in Southern Africa, and Lesotho over recent decades, likely influenced by climate change. These trends are corroborated by climate models that project future scenarios, indicating an increased risk of extreme rainfall events (or lack thereof) under ongoing global warming.
The implications of the findings are sure to be significant for policymakers and stakeholders in Lesotho. An increased risk of extreme rainfall necessitates revising existing infrastructure design standards, enhancing early warning systems, and implementing adaptive water management strategies. The Bayesian approach provides a comprehensive framework for continuously updating risk assessments as new data and climate projections emerge.
In summary, applying EVT combined with Bayesian inference presents a powerful methodology for understanding and managing the risks associated with extreme rainfall in Lesotho. This research highlights the critical need to incorporate climate change considerations into environmental and infrastructural planning to enhance resilience against future extreme weather events