Long-term Rainfall Variability over Ganga River Basin, India through Clustering and Wavelet Transform Approach
Abstract Identifying spatiotemporal patterns of rainfall variability is crucial for sustainable water resource management and climate change-adaptive planning. Thus, we attempted to analyze long-term (1984–2024) seasonal and annual CHIRPS daily gridded precipitation data (0.05° × 0.05° resolution) across different agro-climatic zones in the Ganga river basin. For rainfall pattern analysis, we applied hierarchical clustering, discrete and continuous wavelet transform (CWT), and cross-correlation analyses, alongside elevation as a spatial covariate. The study captured the basin’s complex hydro-climatic behavior. Clustering delineated three coherent sub-regions of rainfall for winter, pre-monsoon, and post-monsoon seasons, and four for the monsoon season. The Mann–Kendall test and discrete wavelet analysis highlighted modest but discernible upward trends in annual and monsoon rainfall across several zones, overlaid by pronounced inter-annual variability. Notably, the study identified a significant cyclical coherence—particularly in the 2–3-years band—between pre-monsoon and monsoon rainfall, aligning with major climate oscillations such as ENSO and IOD. This relationship underscores the predictive value of pre-monsoon rainfall conditions for anticipating monsoon strength, offering practical implications for seasonal forecasting.
Authors
- Kshitij Divyansh
- vs Rathore
Institutions
- Birla Institute of Technology, Mesra (IN)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-13
- DOI
- https://doi.org/10.21203/rs.3.rs-10986681/v1
- Primary Topic
- Climate variability and models
- Type
- preprint