In this lesson the learner will review and interpret data regarding the early performance of the National Water Model (versions 1.0 -1.2). Verification and evaluation of the National Water Model has been occurring since it went operational in August 2016. This lesson will review some of the main issues in model performance through early 2018, including some retrospective verification extending back to 2011. You will see how model performance has been improved as a result of verification. Among the topics addressed are peak flow timing errors, model bias and correlation, the impacts of basin calibration, inclusion of reservoirs, snowpack evolution, drainage in sandy soils, Quantitative Precipitation Forecasts (QPF), and time-lagged ensembles.
Have an account?
Sign In to write a review.
- Explain the reasons for verification
- Quantify performance
- Gain confidence in how/when/where model performs well
- Identify areas for improvement
- Interpret verification statistics related to early versions of the NWM
- Characterize streamflow forecasts as measured by bias, correlation and timing errors
- Identify changes in domain-wide verification statistics from version to version
- Describe how verification of early NWM output was used to correct or reduce errors
- Describe some ongoing areas of model evaluation and development
- Identify QPF-driven error
- Describe impact of managed streams and reservoirs
- Explain the need for ensemble-based guidance
- Explain what a time-lagged ensemble is
- Explain limitations of time-lagged ensembles
National Water Model, NWM, NOAA, NWS, verification, AHPS, NWM guidance, River Forecast Center, RFC, hydrologic, streamflow, box plot, bias, correlation, lead time, peak flow, NLDAS, retrospective simulation, hydrograph, passive reservoir, reservoir operations, calibrated basins, snowpack physics, infiltration, soil water, snow water equivalent, SWE, soil saturation, quantitative precipitation forecast, QPF, HRRR, GFS, ensemble, time-lagged ensemble, probabilistic
