Development of Monthly Scale Precipitation-Forecasting Model for Indian Subcontinent using Wavelet-Based Deep Learning Approach
In one sentence. Wavelet-decomposed deep learning forecasts monthly Indian precipitation from lagged rainfall and teleconnections; beats single-scale models.
Abstract
An accurate and timely precipitation forecast is essential for water resources management in hydropower, irrigation, and reservoir control. The conventional methods are limited by their inability to capture the high precipitation variability in time and space. In the present work, a wavelet-based deep learning approach is adopted to forecast precipitation using the lagged monthly rainfall, local climate variables, and global teleconnections such as IOD, PDO, NAO, and Nino 3.4 as predictors. The method was tested and validated over the Krishna River Basin in India. Overall, the forecasting accuracy was higher using the wavelet-based hybrid models than the single-scale models. The proposed multi-scale model was then applied to the different climatic regions of the country, and it was shown that the model could forecast the rainfall at reasonable accuracy for different climate zones of the country.
Labels
- Topic
- Sub-seasonal to seasonal
- Type
- New model or method
- Method
- Other
- Time scale
- 2 weeks–2 months
- Domain
- Regional
- Data
- Other
- Status
- Research
- Identifiers
- DOI:10.3390/w15183244 OpenAlex:W4386628916
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.