PREDICTING RAINFALL IN MAINPAT USING THE BPN METHOD

Abstract

Climate change is a major concern on a global scale. The purpose of this study is to elucidate the causes of historical fluctuations in rainfall. While late monsoon and post monsoon regions will see notable fluctuations in rainfall amounts, pre-monsoon and early monsoon regions will become drier in the future. Rain prediction is a particularly difficult task for meteorologists. Numerous models have been used in recent years to evaluate and precisely predict rainfall. In this context, climate records can be quite useful. Long-term data retention can improve our ability to forecast rainfall. This study presents the modeling and rainfall prediction throughout Mainpat using statistical techniques, specifically the Modified ANN model and linear regression. The Indian Meteorological Department (IMD), India, supplied the rainfall data for the last 31 years. This Mainpat surface-based rain gauge collected rainfall data from metrological sites between 1991 and 2024. It has been established how much rain falls each month and year. To assess the accuracy of the data, the average, median, correlation coefficients, and standard deviation were computed for every station. When the quantity of rainfall predicted by the model was contrasted with the amount recorded by rain gauges at different sites, it was discovered that the model's rainfall forecast produced accurate results.