Are Watershed Management Plans Selected and Preferred by Stakeholders Considering Current Climate Conditions Robust against Climate Change Scenarios? A Sensitivity Study of Stakeholders Spatially-Explicit Preferences

Graduation date:2018In the light of the changing climate, the importance of designing effective watershed management plans that are likely to be implemented is becoming ever more important. This research introduces a new concept, consensus, for incorporation into stakeholder-guided interactive optimization of watershed management plans. User preferences were mathematically simulated based upon scenarios of possible stakeholder attitudes in sub-basins of an agricultural watershed in Indiana, USA, and incorporated into an existing interactive genetic algorithm (GA) framework. These simulated users along with the watershed hydrologic model were used to evaluate overall preference for and performance of hundreds of different possible distributions of wetlands throughout the Eagle Creek Watershed, weighing cost and environmental concerns on and off of their property. Solutions generated using the interactive GA with the consensus measure performed at least as well as the non-interactively generated baseline solutions, and many out-performed the baseline solutions, with higher peak flow reductions for similar total wetland areas. This result is opposite of what was expected. Previous research has characterized adding stakeholders to the optimization process as a “tradeoff” process, where users sacrifice performance for certain intangible factors. In addition to adding a consensus measure to the interactive GA as an additional objective function, this research also developed a method to select short climate model realizations that best represent extreme flow events arising from climate extremes in the projected future. When the interactively and non-interactively generated solutions were subjected to these extreme climate years, their performance was reduced, even when adjusted for the different magnitudes of expected maximum peak flows. Data issues arising from an interruption to the interactive optimization at generation 30 likely led to some irregularities in the results of this research. Nevertheless, it appears that designing watershed management plans that perform well in the present does not necessarily lead to strong performance in the projected future. Any attempts to address climate change in management plans must do so explicitly