Assessing Seasonal Fluctuations in Forecast Precision through Comparative Regression Modelling in Meteorology

This study provides an in-depth review of various regression models used to forecast meteorological parameters across seasons. Regression models that use traditional regression can be evaluated against advanced machine learning techniques like Random Forest and Gradient Boosting to evaluate their predictive power using metrics such as root mean square Error (RMSE), Mean Absolute Error (MAE) as well as Mean Absolute Percentage Error (MAPE) to calculate R ² and ratio between RSR/RMSE to observer Standard Deviation ratio, Kling-Gupta Efficiency (KGE). The research highlights notable performance differences over time, highlighting both the variability of weather data as well as the challenges associated with accurate forecasting. The Ridge Regression model stands out from other models with one of the most accurate error metrics (RMSE: 294.87, MAE: 232.58, MAPE 7.74 RSR = 0.81); as well as consistently producing R ² values of 0.34 and KGE values of 0.53 within its model parameters. The methods adopted in this research would help the stakeholders, civic bodies and others for attaining sustainable water resources approach to tackle the repercussions of climate change.