Development of cool season MOS equations to predict wind speed and wind components from the 0000, 0600, 1200, and 1800 UTC forecast cycles of the
AVN has been completed. Forecasts generated from these equations will be available for over 1400 sites
in the contiguous U.S., Alaska, Hawaii, and Puerto Rico. Equations to predict the probability of
precipitation amount during the cool season have also been developed for the 0000 and 1200 UTC cycles.
Work is underway to develop MOS equations to predict temperature, dew point, sky cover, ceiling height,
thunderstorms, severe weather, and precipitation type from the 0600 and 1800 UTC cycles of the AVN.
We've now completed derivation of new MRF-based MOS maximum/minimum temperature, 2-m temperature,
and 2-m dew point forecast equations. These equations will be used to produce guidance out to 11
days in advance. The MOS guidance for projections beyond 7 days has been requested by the River
Forecast Centers and the Climate Prediction Center. MRF-based MOS equations to predict the
probability of thunderstorms out to 7 days in advance have also been developed for the October 16
- March 15 cool season.
Development of MRF-based MOS equations to forecast wind speed is underway. Equations are being
developed to predict the probability that the maximum sustained surface wind during a 12-h period
will be light (0 - 12 kts), brisk (13 - 21 kts), windy (22 - 33 kts), or strong (> 34 kts).
Operations and Software Support (M. Erickson):
New alphanumeric bulletins containing AVN and
MRF MOS guidance for 273 stations selected by the Air Force Weather Agency (AFWA) were designed and
tested. Communication headers were assigned by OOS and the message groupings were coordinated with
the AFWA. The new bulletins will be implemented in mid-September.
We've discovered a problem in the AVN- and MRF-based sky cover guidance. When we developed the sky
cover forecast equations, we used the reported sky cover as the predictand and complemented that
report at automated sites with satellite observations. However, we did not complement sites at which
the sky cover observations were supposed to be augmented by human observers. We've now found a number
of sites at which this human augmentation is sporadic or non-existent. The consequence of this has
been numerous forecasts of clear sky cover when forecasts of broken or overcast are more appropriate.
We've made a correction in our operational system and will implement the fix in mid- September.
COASTAL MARINE PREDICTION PROJECT (W. Shaffer)
Storm Surge and Wave Forecasting (W. Shaffer):
We're examining the EDAS analyses and Eta22
forecasts of wind and pressure for driving our extratropical surge model. EDAS's analyzed surface
winds for tropical storm Allison were superior to those of the AVN. Scripts have been written to run
our extratropical surge model in a test mode for the four cycles of the Eta22, using 3- hourly EDAS
analyses during the surge model's spin-up phase.
A simple model to compute the highest significant waves from a hurricane has been written, and is
being adapted for possible use by NHC. Computed wave heights compare favorably with observed waves
for cases where a wave-measuring buoy fell within the hurricane's radius of maximum wind. Six such
cases were found from hurricanes over the recent past-B hurricanes Fran, Opal, Bonnie, Floyd, Georges,
and Iniki.
We have begun to extract wave forecasts in GRIB format from NCEP. We will be modifying these wave
forecasts to give a better representation on the wave climate near the coastline. We have also
obtained detailed bathymetry for the area off Onslow Bay, North Carolina. This bathymetry will allow
us to model the near shore waves and to compute wave set-up and run-up in this area.
FORECAST EVALUATION PROJECT (W. Shaffer)
Forecast Verification Analysis (V. Dagostaro):
We implemented various changes to our MOS
verification web site based on comments provided by the Statistical Modeling Branch. This site
shows the performance (in graphical form and with statistical measures) of the AVN MOS, MRF MOS
and NGM MOS guidance products, and we expect to unveil the web site to the public shortly. Also,
we assisted WFO Honolulu and the Alaska Region in setting up the AVP software. Verification
statistics for July were provided to OCWWS for the local, NGM MOS, and AVN MOS forecasts of max/min
temperature, PoP, ceiling height, visibility, and cloud amount.
MESOSCALE PREDICTION PROJECT (D. Ruth)
National Digital Forecast Development (D. Ruth):
MDL developers have configured FSL software
to generate national web images for the NDFD prototype. A high level design of NDFD central processing
software components and data storage architecture has been completed.
STORM-SCALE PREDICTION PROJECT (S. Smith)
Hydro-Meteorological Monitoring (M. Filiaggi): The SCAN and FFMP software was successfully
converted to run on the new Linux workstations. The conversion dramatically improves the performance
of the SCAN graphical user interfaces and overlays over that available in the legacy HP workstations.
SCAN/FFMP-Linux will be available to the field in AWIPS 5.1.2.