Analysis of Hydrological Model Uncertainty in the Response of Farmers' Decision Models: A Case Study from the Lower Mississippi River Basin Region

Ever since human have evolved on planet earth, they have been dependent on water resources for almost every aspect of their lives. From food production to industry, water has been an integral part of the ever-growing human civilization. Increasing reliance on water has resulted in development of effective and sustainable management strategies. However, water management has remained a complex task due to different human and natural stressors. Hydro-economic models are powerful tools that have been widely used to study water resource systems, management options, economic values, and infrastructures. They have been used to analyze the Water-Energy-Food nexus, climatic interactions, and resource allocation for sustainable water management. Despite their diverse applicability, these models are subject to significant uncertainties because of their dependence on hydrological and economic factors. This study investigates the effect of model variabilities on the output of a hydro-economic model. We used three different hydrological models and varied parameters of each model to analyze the uncertainty in the output of the hydro-economic model. We also studied the variation when using the response of the economic model when forced with different future climatic scenarios. Results shows that the model parameter uncertainties varied among three selected conceptual models from 4.4% to 17%. Similarly, the model selection uncertainty on the four selected basins resulted in 11%, however, the model selection varied among the basins from 7.7% to 14.4%. The total uncertainties calculated by including both parameters and model selection resulted in nearly 17% uncertainties which also varied among basins ranging from 13.7 to 24.2%. This assessment clearly indicates that uncertainties are large enough to be neglected and should be considered for studies that relies on hydrological model simulations. Even though there was significant uncertainty on hydrological model selection, the propagated effects on framers’ decision were observed to be insignificant.