H060
From Local to Global and back again: making hydrologic modeling and prediction work across scales
From Local to Global and back again: making hydrologic modeling and prediction work across scales
Session ID#: 83817
Session Description:
Recent years have seen the rise of large-domain hydrologic prediction systems offering real-time flood forecasts everywhere and at local scales, based on coupled and uncoupled prediction approaches. Yet techniques that served well for years in the local forecast context may not translate to continental or global applications, and the performance of large-domain predictions for local applications varies widely. Our goal is to promote discussion on strategies for narrowing the gap between the local and large-domain system prediction quality. We seek insights on local validation of large-domain forecasting, including both success and failure stories. Also of interest: how to interpret or improve such forecasts though the application of techniques such as post-processing, data assimilation, model parameter estimation, ensembles, tailored forecast communication. Lastly, we encourage contributions on potential applications of machine learning techniques for local, national and global hydrologic prediction. This session supports the Hydrologic Ensemble Prediction Experiment (HEPEX; www.hepex.org).
Index Terms:
1821 Floods [HYDROLOGY]
1840 Hydrometeorology [HYDROLOGY]
1922 Forecasting [INFORMATICS]
1942 Machine learning [INFORMATICS]
Primary Convener: Andrew W Wood, National Center for Atmospheric Research, Boulder, CO, United States
Conveners: Nathalie Voisin, PNNL, Seattle, WA, United States, Yu Zhang, University of Texas at Arlington, Department of Civil Engineering, Arlington, TX, United States and Albrecht Weerts, Deltares, Operational Water Management, Delft, Netherlands
Primary Liaison: Nathalie Voisin, PNNL, Seattle, WA, United States
See more of: Hydrology