Evaluating Polynomial, and Gaussian Approaches for Temperature Change Sub Index of Water Quality Index for Smart Environmental Management

Water temperature plays a crucial role in aquatic ecosystem health, affecting oxygen solubility, organism metabolism, and overall water quality.This study evaluates Linear Interpolation, Polynomial, and Gaussian Models for predicting the Temperature Change Sub-Index in a Water Quality Index (WQI).Using a dataset from NSF-WQI, we propose a new formula based on polynomial and Gaussian approaches, offering improved accuracy over the traditional linear interpolation method for better water quality assessment.The Gaussian Model 1 emerged as the most accurate, with a 97.54% accuracy and a mean error of 2.46, outperforming the Cubic Polynomial and Fourth-Degree Polynomial Models, which achieved accuracies of 94.60% and 96.03%, respectively.The Gaussian Model's ability to capture peak characteristics and symmetrical declines makes it particularly effective for applications requiring precise water quality predictions.The integration of Gaussian Model 1 into IoT-enabled realtime monitoring systems presents significant potential, enabling continuous, accurate predictions critical for managing sensitive aquatic environments.However, the study acknowledges limitations, including the narrow data range that may limit Model generalizability and the complexity of higher-degree polynomial models, which could reduce practical applicability.Future research should focus on validating these Models across more diverse datasets, exploring hybrid Models that combine Gaussian and polynomial strengths, and enhancing computational efficiency to support broader realtime applications.