Statistical Analysis and Prediction via Neural Networks of water quality in the Middle Paraíba do Sul (Rio de Janeiro State, Brazil) region in the period (2012 - 2022)

Abstract This study presents a 10-year temporal assessment (2012–2022) of water quality in the Middle Paraíba do Sul hydrographic region, using the Water Quality Index (WQI) and statistical tools, with predictions via General Regression Neural Network (GRNN). The analysis, based on INEA data, highlights climatic events such as the 2014/2015 drought and differences between the WQI in rainy and dry seasons. The preservation of water quality in this region is crucial for public health, sustainability, and the economic development of the Rio de Janeiro metropolitan area, which relies on the Paraíba do Sul River. Increasing urbanization, agricultural expansion, and climate change pose challenges to water quality. Statistical tools such as Principal Component Analysis (PCA) and Analysis of Variance (ANOVA) reveal significant variations between monitoring points, and GRNN predicts WQI trends for 2023. This predictive approach is vital for informed decision-making in water resource management, particularly as environmental pressures increase.