ABSTRACT:

The multiple-try Differential Evolution Adaptive Metropolis (ZS) (MT-DREAM(ZS)) algorithm was used to quantify the uncertainty in the prediction of the Indicators of Hydrologic Alteration (IHA) and The Magnificent Seven indices from simulated streamflows. For modeling purposes, we used the Soil and Water Assessment Tool (SWAT) in an agriculture-dominated watershed in Michigan, US. We linked multi-objective calibration results using the U-NSGA-III algorithm with Bayesian parameter estimation via the prior distribution for model parameters. Here we provide the (posterior) sampled parameter values, the streamflow predictions, and the relative errors for the predicted hydrologic indices under different calibration settings. In addition, we provide the MATLAB codes for reproducing figures representing model parameter variability ranges, performance of streamflow predictions, and variability in predicted hydrologic indices.

ABSTRACT:

Two model calibration strategies using the U-NSGA-III evolutionary multi-objective optimization algorithm and the Soil and Water Assessment Tool (SWAT) were implemented to improve the representation of ecologically relevant hydrologic indices in an agriculture-dominated watershed in Michigan, US. Here we provide the U-NSGA-III optimization algorithm outcomes obtained under each strategy. These outcomes include Pareto-optimal solutions and all the evaluated objective functions, model parameters, and simulations for each algorithm generation.