A comprehensive investigation of event-based rainfall analysis in arid and semi-arid climates: An integration of innovative storm volume-duration-frequency (SVDF) schemes and Event-Based dimensionless hyetographs (EDH)

Abstract

Recent global shifts in climate have intensified variability in the hydrological cycle and heightened occurrences of extreme weather events. In arid regions, sporadic and intense rainfall often leads to extreme flash floods, posing serious risks to infrastructure, ecosystems, and public safety. This study investigates the frequency dynamics of rainstorm characteristics in arid and semi-arid climates, utilizing the Minimum Inter-Event Time (MIET) approach. The employed hourly precipitation datasets comprise all available pixels at a 0.1-degree resolution across the United Arab Emirates (UAE) over a 21-years. The research underscores the diversity of climatic profiles within the UAE. The appropriate MIET values, ranging from 6 to 9 h, were assigned to nine clustered climatic regions to discern storm attributes, i.e. volume, duration, intensity, and dry periods. The comprehensive frequency analysis, incorporating ten probability distributions and five statistical indices, showcases the robust performance of two distributions, Generalized Extreme Value (GEV) and the two-parameter lognormal, in capturing extreme storm features. Subsequently, the optimal models were utilized in the sensitivity analysis of the novel Storm Volume-Duration-Frequency (SVDF) scheme, revealing that longer storms do not necessarily produce the largest rainfall volumes, and extreme intensities are not restricted to the shortest storms. The innovative approach offers a more realistic interpretation of events-based rainfall patterns, challenging conventional frequency analyses. In region 5, a 7-hour storm resulted in 111 mm of rain, while the 8-hour storm yielded just 59 mm at a 100-year return period. Another critical observation was the shift in extreme storm behavior, i.e. the 5-hour storm produced the peak volume up to a 10-year return period in region 9, after which the 7-hour storm became dominant, reaching 104 mm at a 100-year return period. Overall, coastal regions are highly vulnerable to extreme events and flash flood hazards. An empirical formula was developed to describe the relationship between storm volume (V, mm), duration (D, hr), and return period (Tr, years) within the framework of SVDF curves. The derived SVDF curves were then integrated into Event-Based Dimensionless Hyetographs (EDH) to enhance rainfall intensity predictions. The EDH is delineated employing events-based historical temporal analysis. This study attempts to thoroughly describe the behavior of rainstorm patterns, deduce design storms, and thereby enhance hydrological modeling. The innovative methodology could change some hydrological concepts and contribute to more effective planning and hazard mitigation strategies.