Exploring Machine Learning, Deep Learning, and Explainable AI Methods for Seasonal Precipitation Prediction in South America

TL;DR

This study evaluates classical machine learning and deep learning models, along with explainable AI, for seasonal precipitation prediction in South America, demonstrating the effectiveness of LSTM and XGBoost over traditional models.

Contribution

It provides a comprehensive comparison of ML, DL, and dynamic models for precipitation forecasting, highlighting the potential of DL models and explainable AI in climate prediction.

Findings

LSTM outperformed other models in accuracy, especially for heavy precipitation.

XGBoost offered a good balance between latency and accuracy.

Traditional dynamic models like BAM had the worst predictive performance.

Abstract

Forecasting meteorological variables is challenging due to the complexity of their processes, requiring advanced models for accuracy. Accurate precipitation forecasts are vital for society. Reliable predictions help communities mitigate climatic impacts. Based on the current relevance of artificial intelligence (AI), classical machine learning (ML) and deep learning (DL) techniques have been used as an alternative or complement to dynamic modeling. However, there is still a lack of broad investigations into the feasibility of purely data-driven approaches for precipitation forecasting. This study aims at addressing this issue where different classical ML and DL approaches for forecasting precipitation in South America, taking into account all 2019 seasons, are considered in a detailed investigation. The selected classical ML techniques were Random Forests and extreme gradient boosting (XGBoost), while the DL counterparts were a 1D convolutional neural network (CNN 1D), a long short-term memory (LSTM) model, and a gated recurrent unit (GRU) model. Additionally, the Brazilian Global Atmospheric Model (BAM) was used as a representative of the traditional dynamic modeling approach. We also relied on explainable artificial intelligence (XAI) to provide some explanations for the models behaviors. LSTM showed strong predictive performance while BAM, the traditional dynamic model representative, had the worst results. Despite presented the higher latency, LSTM was most accurate for heavy precipitation. If cost is a concern, XGBoost offers lower latency with slightly accuracy loss. The results of this research confirm the viability of DL models for climate forecasting, solidifying a global trend in major meteorological and climate forecasting centers.