Challenges and Opportunities of Implementing the National Water Model in Hawaii and Alaska

The National Water Model (NWM) simulates observed and forecast streamflow along with all of the variables that fully define the water cycle over the entire continental United States. Based on the stream reaches defined in the National Hydrography Dataset, the NWM uses mathematical representations of the various processes that track water as it moves through the atmosphere, falls to the surface of the earth as snow or rainfall, accumulates through the stream network and evaporates or transpires back into the atmosphere. The suite of computer models necessary is so complex that it runs on a high performance computer in order to complete the calculations in the time frame that decision makers require.

For various reasons, the original implementation (August of 2016) was confined to the contiguous US (CONUS), although there are portions of Canada and Mexico included in the model because of transboundary rivers. With version 2.0 of the operational NWM scheduled for implementation in early 2019, Hawaii is the first area outside of the contiguous US states to be part of the NWM domain. Work in Alaska is beginning and a pilot project for the Cook Inlet area in southcentral Alaska will begin in 2019 with operational implementation planned for a 2020 or 2021 release of the NWM. Modeling hydrologic processes in these areas outside the CONUS (OCONUS) will require a different operational configuration than that utilized for the CONUS. Both observation and forecast datasets tend to be less robust with challenges in observation density and shorter periods of record. Alaska and Hawaii have some significant orography and very little observation information at high altitudes. Developing an analysis of record for precipitation and temperature for use in model calibration is much more difficult for these areas because of the sparse gage network and lack of long-term observation stations. In addition, the high quality National Hydrography Dataset is not yet available for much of the area in Alaska. Both observed and forecast forcings for Alaska and Hawaii will be different than those utilized for the CONUS domain. Despite these challenges, implementation of the NWM in Alaska and Hawaii will provide a wealth of hydrologic forecast information and guidance. This paper will discuss the challenges and detail the opportunities available as this effort moves forward.