FORECASTING AND MODELING RAINFALL DATA USING TREND ANALYSIS

Abstract

This study presents a comprehensive analysis and forecasting of rainfall data using three distinct trend analysis methods: linear, quadratic, and exponential. The research aims to predict future rainfall patterns to aid decision-making in the agricultural sector. Using historical rainfall data from 1901 to 2013, we developed models for each trend type and evaluated their forecasting quality using the Root Mean Square Error (RMSE). Among the models tested, the exponential trend model provided the most accurate forecasts, as indicated by its lower RMSE value compared to the linear and quadratic models. This model was then used to predict rainfall from 2014 to 2028, offering valuable insights for agricultural planning and water resource management. The results demonstrate the effectiveness of trend analysis in forecasting climatic variables, which is crucial for mitigating the impacts of climate variability on agriculture.