1. Application of hierarchical cluster analysis to spatiotemporal variability of monthly precipitation over Khyber Pakhtunkhwa, Pakistan.
- Author
-
Tajbar, Sapna, Tajbar, Asfandiyar, Pashaie, Zahra, Khorshiddoust, Ali Mohammad, and Rafiq, Lubna
- Subjects
PRECIPITATION variability ,HIERARCHICAL clustering (Cluster analysis) ,CLUSTER analysis (Statistics) ,WATER management ,RAINFALL ,STANDARD deviations ,DROUGHTS - Abstract
Understanding the long-term spatiotemporal variability of precipitation at the regional scale is critical for developing flood and drought control strategies and water resource management. This study assessed the spatiotemporal variability of monthly precipitation over the Khyber Pakhtunkhwa province of Pakistan for 1998–2019 using hierarchical cluster analysis to cluster 156 Tropical Rainfall Measuring Mission grids. Statistical properties of clusters were calculated and the relationship of geographical features such as latitude, longitude, and altitude and statistical variables including standard deviation, maximum and minimum precipitation, and coefficient of variation (CV) with average precipitation was assessed. Findings showed that northeast parts received maximum precipitation while north and southern regions received less precipitation. Temporal analysis showed two clusters of rainy months (February, March, April, May, July, and August) and dry months (January, June, September, October, November, and December). The region was divided into two homogeneous precipitation regions. From January to April and November to December, cluster 1 occupied northern parts with maximum average precipitation while cluster 2 southern parts. From June to September, cluster 2 covered the northeast and southern parts with the highest average precipitation. During May, cluster 2 received the highest average precipitation in the northeast and southeast parts, whereas cluster 1 covered the northwest and southwest. In October, cluster 2 received maximum average precipitation covering the northeast. CV suggested higher temporal variability in cluster 2 (67.75–102.36)% than cluster 1 (65.82–99.55)%. Precipitation correlation showed that CV opposed the longitude and averages, whereas latitude and altitude demonstrated minimal correlations. These insights can assist decision-makers in devising suitable strategies to plan and control unexpected volumes of precipitation. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF