Machine Learning Models for Climate Change Prediction on Winter Season and Impact Assessment in Tamil Nadu

Global warming is still a massive problem for areas like Tamil Nadu, especially during the season beginning with the letter W which is winter. This work employs artificial neural networks and support vector machines to forecast winter climate shifts and analyze potential consequences for agriculture, road networking, and human health. These models were trained using historical climate data and it yielded low mean absolute errors of 0.85 °C for temperature, and 5.20mm for rainfall. The extreme weather events were identified with great efficiency and their classification was determined to be 89% accurate. Scientific forecasts for winter 2024 show the crop production cut by 14%, transportation disturbances by 53%, and the threat of epidemics by 30%. The findings indicate that system integration with machine learning can help in climate adaptation interventions in Tamil Nadu. Future work will have to involve real-time adaptation and inclusion of socio-economic features for better predictions.