Abstract:

Drought issues are becoming more frequent and persistent due to global warming, and the impact range is gradually expanding. Notably, drought monitoring, assessment, and research generally use drought indices. However, during their calculation processes, the sample, probability distribution function models, and parameters influence these indices leading to uncertainties in the drought assessment results. Therefore, this study investigated the influence of probability distribution function models and parameter estimation errors on the drought index-standardized precipitation evapotranspiration index (SPEI) and characteristics (drought intensity, peak, and duration) based on the SPEI in the Heihe River Basin in China as the study area. The results reveal that the two sources of the uncertainties, probability distribution function models, and parameter estimations, affected the drought index SPEI and assessment. Additionally, the results show that the more extreme the drought index SPEI is, the greater their effects, with both having a much greater impact on extreme and severe drought than on mild and moderate drought. The probability distribution function models resulted in more significant uncertainty in the drought intensity and peak of the extreme and severe droughts compared to the parameter estimation errors, which caused greater uncertainty in the drought duration in both drought conditions. Therefore, the study’s results can support the accurate assessment of drought and theoretically facilitate more accurate and effective decisions in drought prevention and mitigation efforts to avoid possible inadequacies in the mitigation capacity or waste of drought-resistant resources.

Key words: probability distribution model, parameter estimation, uncertainty, standardized precipitation evapotranspiration index (SPEI), Heihe River