Stress on global water resources is expected to increase in the coming decades making reservoir operations critical for sustainable water management. Reservoir sedimentation is a long-term challenge for water management across the western U.S. since it reduces available storage. Sedimentation observations are limited to reservoir surveys that are costly and infrequent, with many reservoirs having two or fewer surveys spanning decades. In most cases, sediment is assumed to accumulate at a constant rate through time. It is often estimated by inaccurate instream measurements prior to reservoir construction and/or simple equations that neglect to consider the complex development from sedimentation in rivers and climate change.

This work leverages recent advances by Stewart et al. 2017 that coupled multiple sediment algorithms within the Variable Infiltration Capacity hydrologic model (VIC) to facilitate the inclusion of dynamic hydroclimate inputs into sediment calculations. The sediment algorithms include empirical, conceptual, stochastic, and processes-based approaches. This work analyzes different sediment algorithms applied to estimating reservoir sedimentation. A Sobol sensitivity analysis provides insight into the characteristics of each sediment algorithm that are most important for modeling reservoir sedimentation.