Pacific Northwest Mesoscale Model Weather Forecasts: Information
For a description of the MM5-NAM forecasts click here
For a description of our older, experimental WRF forecasts click here
For a description of our older, real-time MM5 forecasts click here
General Information: The Department of Atmospheric
Sciences at the University of Washington is now running the
Advanced
Research WRF (WRF-ARW) core of the Weather Research and Forecasting Model
(WRF)
version 3.0.1.1 (WRF-ARW v3.0.1.1). Forecasts are run twice a day in
order to produce high resolution meteorological guidance for the Pacific
Northwest. This effort, supported by a consortium of local, state, and
federal agencies, has several goals:
- To provide high resolution meteorological fields for air quality
applications, weather forecasting, and other needs.
- To test the viability of local numerical weather forecasting
completed on relatively low-cost workstations.
- To provide a large number of mesoscale model forecasts to
determine model strengths, weaknesses, and biases.
- To improve mesoscale model physics and data assimilation
approaches.
- To test the potential of directly coupling an atmospheric forecast
model to air quality and hydrological numerical models.
Change Log: For the latest information on our setup, please
consult the Change Log for Pacific Northwest
WRF/MM5 Numerical Forecasts
Model Configuration: The WRF numerical forecasts currently
feature an outer grid of 36 km horizontal resolution that covers much
of western North America and the northeastern Pacific; a nested grid
of 12 km resolution that covers Washington, Oregon, Idaho, Utah, much
of Montana and Nevada, southern British Columbia, and northern
California, Nevada; and an inner 4 km resolution grid encompassing all
of Washington, Oregon, and Idaho plus the western third of Montana and
the extreme northern sections of California, Nevada, and Utah, and the
western third of Wyoming. The regions covered by each domain, showing
topography and land use categories are shown in these plots:
A land use category color key is included in the links above, you may
view it alone here.
A detailed description of the domain boundaries can be found in
the actual geogrid.exe namelist.wps file or
in the MM5 terrain.deck. Detailed land use
information for each domain was derived from the 1-km USGS digital
database, with some subjective modification using other data sources.
The model utilizes 38 vertical full-sigma levels see below, and is run in
non-hydrostatic mode in order to limit pressure gradient force errors
over the complex terrain. An upper-radiative boundary condition is used
to allow gravity waves to radiate through the model top without being
reflected. The following table describes the phyics options we are
currently using:
Phyics and Other Options:
| Physics | Comments |
| Radiation |
Until 2009012200, both CAM longwave and shortwave radiation
options used with default values. From 2009012212 onward, Dudhia
shortwave and RRTM longwave radiation schemes used. |
| Thompson Microphysics | Includes graupel. |
| Kain-Fritsch Cumulus |
Continuing what we began on 5 August 2004 (MM5 then) of using this on all domains. |
| YSU PBL | Yonsei University scheme (new MRF). Non-local-K
scheme with explicit entrainment layer and parabolic K profile in
unstable mixed layer. NOTE: the module_bl_ysu.F from version 3.0 of
the WRF was backported into the version 2.2.1 WRF that was run up to
and including the 2009010900 case. |
| Noah Land Surface Model |
Four layers with soil temperature and moisture, fractional snow
cover, and frozen soil physics. |
Here are the exact namelist.input files that are used to run the model: 36/12 km namelist.input and 4 km namelist.input (these
two links are updated with each model run.).
For more information on any of these options, see the WRF model chapter of
the
WRF User's Guide.
Initializations: Currently two model runs are performed.
One utilizes initial and lateral boundary conditions for
the MM5 that are generated by interpolation of the National Centers
for Environmental Prediction (NCEP) "NAM" (formerly the ETA)
model analysis and forecast fields. This MM5-NAM run only includes the
36 and 12 km domains and is described here.
The second initialization uses NCEP's GFS model (formerly known as AVN or
MRF) at 1-degree lat/lon resolution for nearly all fields. The
exceptions are: 1) SST is from the 1/4 degree OTIS grids, 2) surface
temperature is from the NAM/ETA 221 grids (40-km grid spacing), and 3)
subsurface soil temperature and moisture are also from NAM/ETA 221 grids.
Forecast: Forecasts are computed as follows:
- MM5-NAM: 4-processor Intex Xeon Linux box.
- WRF-GFS: Eight 8-processor Intel Xeon cluster running Debian
Linux (Note, 6 of the 8 processors are used, since very little is
gained by running with all 8. Keeping 2 available on each machine
also allows for short research runs or make-up runs to be run
concurrently with the real-time run with a small impact on that run).
Currently, the model is run for a 72 hour forecast period over all domains. The
MM5-NAM finishes in roughly 3.0 hours of wallclock time, while the
WRF-GFS finishes in roughly 5.2 hours of wallclock time.
Future Improvements: The modeling system is constantly being
improved.
Full Vertical Sigma Levels
Sigma: 1.00, 0.995*, 0.99, 0.985*, 0.98, 0.97, 0.96, 0.95*,
0.94, 0.93*, 0.92, 0.91*, 0.90,
0.88, 0.86, 0.83, 0.80, 0.77, 0.74, 0.71,
0.68, 0.64, 0.60, 0.56, 0.52, 0.48, 0.44, 0.40,
0.36, 0.32, 0.28, 0.24, 0.20, 0.16, 0.12, 0.08, 0.04, 0.
* = added 7/9/2001
Half-Sigma Levels
| K-index | Sigma | K-index | Sigma | K-index | Sigma |
| 1 | 0.02 | 2 | 0.06 | 3 | 0.10 |
| 4 | 0.14 | 5 | 0.18 | 6 | 0.22 |
| 7 | 0.26 | 8 | 0.30 | 9 | 0.34 |
| 10 | 0.38 | 11 | 0.42 | 12 | 0.46 |
| 13 | 0.50 | 14 | 0.54 | 15 | 0.58 |
| 16 | 0.62 | 17 | 0.66 | 18 | 0.695 |
| 19 | 0.725 | 20 | 0.755 | 21 | 0.785 |
| 22 | 0.815 | 23 | 0.845 | 24 | 0.87 |
| 25 | 0.89 | 26 | 0.905 | 27 | 0.915 |
| 28 | 0.925 | 29 | 0.935 | 30 | 0.945 |
| 31 | 0.955 | 32 | 0.965 | 33 | 0.975 |
| 34 | 0.9825 | 35 | 0.9875 | 36 | 0.9925 |
| 37 | 0.9975 |
Past Domain Configurations: This table provides all the
necessary information about previous domain configurations.
Questions and Comments: We're always interested
in feedback. Please direct any questions and comments you have to David
Ovens (ovens@atmos.washington.edu) or Cliff Mass (cliff@atmos.washington.edu).