Optimizing spatial distribution of watershed-scale hydrologic models using Gaussian Mixture Models

Common methods for spatial distribution, such as hydrologic response units, are subjective, time-consuming, and fail to capture the full range of basin attributes. Recent advances in statistical-learning techniques allow new approaches this problem. We propose use Gaussian Mixture Models (GMMs) distribution models. GMMs objectively select set modeling locations that best represent watershed features relevant cycle. demonstrate method two hydrologically distinct headwater catchments Sierra Nevada show it meets or exceeds performance traditionally distributed models multiple metrics across water balance at a fraction time cost. Finally, we univariate identify most-important drivers processes basin. The GMM allows more robust, objective, repeatable models, which critical advancing research operational decision making.