Monday, 23 January 2017: 4:00 PM
604 (Washington State Convention Center )
A 24-year (1993-2016) retrospective analysis using the new NOAA National Water Model (NWM) configuration of the community WRF-Hydro system, driven by downscaled NLDAS2 meteorological forcing data has been generated to provide a baseline analysis of hydrologic statistics for the operational implementation of the NWM. The NOAA National NWM is a new, continental-scale, operational, hydrologic analysis and prediction system designed to provide
improved siutational awareness and forecast guidance of flood and flash flood events.
It operates at a spatial resolution designed to capture 'neighborhood' scale hydrologic impacts
related to flooding such as streamflow rate and velocity in local channel reaches as well as
in representing other flood related characteristics such as soil saturation and local inundation.
In this talk a statistical analysis of long-term flow conditions, as simulated by the NWM
are compared against observed streamflow statistics from a number of gauging networks in the U.S.
Regional syntheses of simulated and observed flow conditions are contrasted as a means of assessing
how well the NWM captures the dominant modes of streamflow variability across the U.S.
While much of the variability in streamflow is dominated by spatial and temporal variations
in NLDAS2 precipitation, it is found that hydrologic process representations can also have an impact on the
relative timing and magnitude of streamflow responses. Following this broader analysis of streamflow
variability, more focused examination of flood characterization from the NWM is provided. Evaluation
of the NWM fidelity in simulating flood events uses a statistcal distribution approach and centers
on comparing the magnitudes of 1, 5 and 10% events from both unregulated and regulated systems.
Emphasis is placed on diagnosing where and how the current representation of water management in the NWM
impacts flood simulation skill.
improved siutational awareness and forecast guidance of flood and flash flood events.
It operates at a spatial resolution designed to capture 'neighborhood' scale hydrologic impacts
related to flooding such as streamflow rate and velocity in local channel reaches as well as
in representing other flood related characteristics such as soil saturation and local inundation.
In this talk a statistical analysis of long-term flow conditions, as simulated by the NWM
are compared against observed streamflow statistics from a number of gauging networks in the U.S.
Regional syntheses of simulated and observed flow conditions are contrasted as a means of assessing
how well the NWM captures the dominant modes of streamflow variability across the U.S.
While much of the variability in streamflow is dominated by spatial and temporal variations
in NLDAS2 precipitation, it is found that hydrologic process representations can also have an impact on the
relative timing and magnitude of streamflow responses. Following this broader analysis of streamflow
variability, more focused examination of flood characterization from the NWM is provided. Evaluation
of the NWM fidelity in simulating flood events uses a statistcal distribution approach and centers
on comparing the magnitudes of 1, 5 and 10% events from both unregulated and regulated systems.
Emphasis is placed on diagnosing where and how the current representation of water management in the NWM
impacts flood simulation skill.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner