Identifying monthly rainfall erosivity patterns using hourly rainfall data across India

Area of Science:

• Environmental Science
• Hydrology
• Soil Science

Background:

• Rainfall erosivity is a critical factor in water erosion, exhibiting significant spatial and temporal variability.
• Understanding rainfall erosivity patterns is crucial for effective soil conservation and erosion management, particularly in diverse climatic regions like India.

Purpose of the Study:

• To comprehensively analyze the spatial patterns and monthly distribution of rainfall erosivity across India.
• To examine rainfall erosivity attributes, including kinetic energy, erosive events, and peak rainfall intensity, for the first time in India.
• To develop and evaluate predictive models for monthly erosivity estimation.

Main Methods:

• Utilized data from 261 hourly and 2,525 monthly rainfall stations spanning 1969-2021.
• Developed and compared linear regression, XGBoost, and ensemble models for predicting monthly erosivity.
• Applied Geographically Weighted Regression (GWR) for high-resolution spatial interpolation of erosivity.

Main Results:

• XGBoost and ensemble models demonstrated high predictive accuracy for monthly erosivity (median R² of 0.97 and 0.96, respectively).
• GWR yielded accurate high-resolution erosivity maps (median R² of 0.90).
• Rainfall erosivity peaks during the summer monsoon (June-September), with July showing the highest values due to intense rainfall and kinetic energy. A statistically significant long-term increase in January erosivity and annual maximum 60-min rainfall intensity was detected.

Conclusions:

• The study provides a robust framework for estimating and mapping rainfall erosivity across India.
• Findings highlight the significant erosive potential during the monsoon season and a long-term increasing trend in specific non-monsoon months.
• Results support the development of targeted, location-specific soil conservation strategies to mitigate rain-driven erosion.