Search for a command to run...
The increased global warming has increased the likelihood of recurrent drought hazards. Potential links between the frequency of extreme weather events and global warming have been suggested by earlier research. The spatial variability of meteorological factors over short distances can cause distortions in conclusions or limit the scope of drought analysis in a particular region when extreme values predominate. Therefore, it is challenging to make trustworthy judgments regarding the spatiotemporal characteristics of regional drought. This study aims to improve the quality and accuracy of regional drought characterization and the process of continuous monitoring. The new drought indicator presented in this study is called the Support Vector Machine based drought index (SVM-DI). It is created by adding different weights to an SVM-based X-bar chart that is displayed with regional precipitation aggregate data. The SVM-DI application site is located in Pakistan's northern area. Using the Pearson correlation coefficient for pairwise comparison, the study compares the SVM-DI and the Regional Standard Precipitation Index (RSPI). Interestingly, compared to RSPI, SVM-DI shows more pronounced regional characteristics in its correlations with other meteorological stations, with a significantly lower Coefficient of Variation. These results confirm that SVM-DI is a useful tool for regional drought analysis. The SVM-DI methodology offers a unique way to reduce the impact of extreme values and outliers when aggregating regional precipitation data.