Sharma et al. (2025) Comparative Analysis of Machine Learning Methods for Imputing Missing Daily Rainfall Data in Complex Himalayan Terrain

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: International Journal of Climatology
• Year: 2025
• Date: 2025-09-28
• Authors: Rahul Sharma, S. Sreekesh
• DOI: 10.1002/joc.70122

Research Groups

Not explicitly mentioned in the abstract.

Short Summary

This study evaluated seven machine learning methods for imputing missing daily rainfall data across different elevations and agro-climatic zones in Himachal Pradesh, India, finding that Multilayer Perceptron (MLP) consistently demonstrated the highest accuracy and lowest estimation errors.

Objective

• To evaluate and identify the most suitable machine learning method for robustly imputing missing daily rainfall data across different elevations and agro-climatic zones in the complex Himalayan terrain.

Study Configuration

• Spatial Scale: Regional (Himachal Pradesh, India), across different elevations and three agro-climatic zones.
• Temporal Scale: Daily.

Methodology and Data

• Models used: Multiple Linear Regression (MLR), Support Vector Regression with Radial Basis Function kernel (SVR-RBF), K-Nearest Neighbours (KNN), Random Forest (RF), Multilayer Perceptron (MLP), Extreme Gradient Boosting (XGBoost), Lasso Regression.
• Data sources: Ground-based rainfall station observations.

Main Results

• Multilayer Perceptron (MLP) consistently exhibited higher accuracy and lower estimation errors across both overall and rainfall-intensity-classes (RIC) assessments.
• MLP was identified as the most suitable method for imputing missing daily rainfall data in the complex Himalayan terrain, outperforming MLR, SVR-RBF, KNN, RF, XGBoost, and Lasso regression when evaluated using the coefficient of determination (R²), correlation coefficient (r), root mean square error (RMSE), and mean absolute error (MAE).

Contributions

• This study provides a comprehensive evaluation of seven diverse machine learning methods for daily rainfall data imputation, specifically tailored for the challenging conditions of complex Himalayan terrain across varying elevations and agro-climatic zones. It uniquely employs a two-tier assessment (overall and rainfall-intensity-classes) to robustly compare methods and identifies Multilayer Perceptron (MLP) as the most accurate and reliable technique for this critical task.

Funding

Not explicitly mentioned in the abstract.

Citation

@article{Sharma2025Comparative,
  author = {Sharma, Rahul and Sreekesh, S.},
  title = {Comparative Analysis of Machine Learning Methods for Imputing Missing Daily Rainfall Data in Complex Himalayan Terrain},
  journal = {International Journal of Climatology},
  year = {2025},
  doi = {10.1002/joc.70122},
  url = {https://doi.org/10.1002/joc.70122}
}