Rainfall Extremes Analysis in Arid Regions Under Climate Change: A Structured Review of Methods and Approaches

The impact of climate change on rainfall extremes has become increasingly obvious in many climatic regions including arid regions where extreme precipitation events are thought to have augmented or at least intensified. Driven by global factors such as greenhouse gas emissions, deforestation, and industrialization, climate change has augmented hydrological variability, thus making traditional stationary models inadequate for the estimation of extreme rainfall at various return periods. Extreme value analyses, which were traditionally derived under the assumption of stationarity (i.e., constant statistical properties over time) and typically do not account for temporal variability or external climatic drivers (e.g., temperature or large-scale climate indices), may lead to inaccurate estimation of rainfall quantiles under changing climate conditions. This paper presents a structured review of applied methodologies for quantifying the influence of climate change on extreme rainfall events, with special attention to how non-stationarity is addressed in arid regions applications, which was not a major focus in previous review papers. Relevant statistical techniques, extreme value theory, machine learning models, and high-resolution climate simulations are reviewed. From an initial pool of over 340 studies, 91 were selected based on their relevance to quantify rainfall extremes induced by climate change in arid regions. Based on the reviewed studies, the analysis revealed a strong reliance on trend analysis of downscaled Global Climate Models (GCMs) and Regional Climate Models (RCMs) within a stationary framework, with limited integration of covariates, other than time, in non-stationary frequency analysis to estimate the climate change-related value. This review identifies the research gaps in the scientific literature related to climate change impact assessment on extreme rainfall in arid regions. It emphasizes the necessity for adopting more robust hybrid approaches, adopting statistical distributions more suitable to arid conditions, careful treatment of outliers, conducting regional analyses to better understand the overall climate behavior of the region, addressing the impact on short-duration rainfall, integrating key climatic drivers through the incorporation of additional climate covariates and the impact of climate change on sub-daily rainfall patterns.