Evaluation of Regression Models of LOADEST and Eight-Parameter Model for Nitrogen Load Estimations

In this study, the Load ESTimator (LOADEST) and eight-parameter regression models were evaluated to estimate instantaneous pollutant loads under various criteria and optimization methods. As shown in the results, LOADEST, commonly used in interpolating pollutant loads, could not necessarily provide the best results with the automatically selected regression model. The various regression models in LOADEST need to be considered to find the best solution based on the characteristics of watersheds. The recently developed eight-parameter model integrated with a genetic algorithm (GA) and the gradient descent method (GDM) was also compared with LOADEST, indicating that the eight-parameter model performed better than LOADEST; however, depending on whether the eight-parameter model was used for calibration or validation, its performance varied. The eightparameter model with GDM could reproduce the nitrogen loads properly outside the calibration period (validation). Furthermore, the accuracy and precision of model estimations were evaluated using various criteria (e.g., R-2, gradient, and constant of a linear regression line). The results showed higher precisions with the R-2 values close to 1.0 in LOADEST and better accuracy with the constants (in linear regression line) close to 0.0 in the eight-parameter model with GDM. Hence, on the basis of these findings, we recommend that users need to evaluate the regression models under various criteria and calibration methods to ensure more accurate and precise results for nitrogen load estimations.