Analysis of drought recurrence conditions using first-order reliability method

Abstract

In this study, a probabilistic analysis using first-order reliability method is conducted to predict the reoccurrence of drought conditions. The implementation of the model involves the collection of the long-term average data during 1994–2015 from 609 study areas of Iran. The reliability criterion is defined as the changes in groundwater resource index during the years 1994–2004. Three parameters, including the beta index, the limit state function, and the probability of failure, are obtained using Kriging GeoStatistical analysis method to produce continuous layers. The results show that around 1900 selected points belong to definite drought conditions, and a major part of the other points are being exposed to a steady rise in the depth of water in aquifers with a probability of more than 50%. The spatial distribution of these changes shows that large parts of the eastern coastal aquifers, as well as major aquifers along the longitudinal Zagros range, and large parts of Fars province, are exposed to the reduction in groundwater level with a probability of more than 95%. In addition, the visual comparison of predicted drainage quality maps suggests a high correlation between the probability of occurrence and repetition of drought conditions with the specific drop in drinking water quality potential index classes. The results of this study confirm the drought reoccurrence conditions for predicting negative changes in standard drinking water quality.