This group’s research work includes hydrologic modeling, coupled land-atmosphere modeling, and statistical analysis of streamflow, rainfall, and soil moisture data. The overall goal of the current research focus is to improve hydrologic forecasting of runoff in the Midwest. This research is in collaboration with the National Weather Service (NWS) North Central River Forecast Center (NCRFC) located in Chanhassen, Minnesota.
Over the past couple of years, the group's research has focused on utilizing stream gage data to assess relative performance of the National Water Model (NWM) streamflow guidance. The purpose of this baseline study is to provide insight as to how the NCRFC can best use NWM simulations in their operational forecasting. The NWM produces forecasts for 2.7 million river reaches within the continental United States. There are just over 10,000 United States Geological Service (USGS) river gages in service across the country, of which a subset was used to calibrate the NWM. Regionalization techniques were used within the NWM to apply calibration parameters across other watersheds deemed hydrologically similar. Assessing the operational forecast utility of the NWM, especially at reaches not directly calibrated with USGS stream gages, is a primary concern for forecasters.
The state of Minnesota has strategically invested in ecological and water resource monitoring equipment over the past several decades. The Minnesota Department of Natural Resources (DNR) and Minnesota Pollution Control Agency (MPCA) have installed stream gages and water quality gages on streams throughout the state to supplement the USGS network and provide localized data on rivers and streams. By using this independent stream gage data source, researchers can assess the statistical performance of the NWM streamflow at reaches not directly calibrated with USGS stream gages. Local, state, and federal stream gaging sites are being used to evaluate performance of the National Water Model retrospective streamflow. The overall goal is to improve the operational forecast utility of NWM streamflow output through either calibration or post model output bias correction.