Sediment Modeling Panel (part 3): Sediment Modeling Failure Modes and Best Practices

When morphological model developers get together, we find that we have had similar experiences. We often see our models applied in unexpected ways. Some of these applications are innovative, applying the model in unanticipated but appropriate directions. But more often, we see the model applications built on misunderstandings of the model assumptions and modeler decisions that are unlikely to generate useful results.

This panel discussion makes that conversation public. The session gathers four morphological model developers to discuss the most common modeling failure modes we have seen in hundreds of sediment model applications. The panel includes 1D (HEC-RAS and SRH-1D) and 2D (ADH and HEC-RAS) sediment model developers.

This panel discussion will revolve around two primary questions: What are the most common problems you see with sediment models? What are some of the best-practices that guide your modeling projects?

We expect the conversation will touch on most of the following practices and principles: • Over-specifying result boundary conditions. • Unmanageable run times from unnecessary detail. • Ignoring bed-material load because it is not in the USGS data. • Over-calibration and equifinality traps. • Adding model complexity incrementally. • Uncertainty analysis • Analyzing processes not resolved in model