Multidimensional Precipitation Index Prediction Based on CNN-LSTM Hybrid Framework
Yuchen Wang, Pengfei Jia, Zhitao Shu, Keyan Liu, Abdul Rashid Mohamed Shariff
Abstract
Amid accelerating global climate change, accurate prediction of meteorological indicators is crucial for disaster prevention, agriculture, and transportation. Precipitation, a key factor, significantly impacts water resource management, agriculture, and urban flood control. This study proposes a CNN-LSTM hybrid model for multidimensional precipitation index prediction to enhance forecasting accuracy. Using monthly mean precipitation data from Pune, India (1972–2002), the model captures both local features and long-term dependencies in the time series. Experimental results demonstrate that the model achieves an RMSE of 6.752, outperforming traditional time series prediction methods. Although effective, the model demands high computational resources and its ability to predict multidimensional precipitation data remains limited. Future work should focus on optimizing computational efficiency and expanding the model’s capacity for multidimensional prediction, contributing to more accurate and efficient meteorological forecasting technologies.