A quantitative study of extreme rainfall intensity and occurrence in northern Algeria

Abstract

This paper examines the characteristics associated with the spatiotemporal evolution of extreme precipitation, assesses its recurrence frequency, and predicts future return levels over northern Algeria. The study employs extreme precipitation indices in conjunction with the application of extreme value theory to a rainfall dataset spanning from 1982 to 2022. The study focused on modeling the index that demonstrated the highest percentage of significant positive trends at the α = 0.05 significance level. This was accomplished through the utilization of the Mann-Kendall test and the generalized extreme value distribution. Subsequently, the model was validated using the Kolmogorov-Smirnov fit test. The results revealed that the northeastern region of the study area experienced a more pronounced increase in rainfall intensity compared to the southern and western regions. Significant trends in precipitation intensity were observed over time. Notably, the index of days with rainfall exceeding 20 mm demonstrated the highest percentage of positive trends, with 88% of meteorological stations exhibiting an upward trend. Furthermore, a strong correlation was identified between the index of days with rainfall exceeding 20 mm and the very wet days index, particularly in the high plateaus and western region. This finding supports the hypothesis that extreme rainfall patterns are becoming more frequent in the region.