METEOROLOGICAL DEVELOPMENT LABORATORY

MONTHLY PROGRESS REPORT

May 2001

June 1, 2001

SYNOPTIC-SCALE PREDICTION PROJECT (P. Dallavalle)

Statistical Forecast Development (P. Dallavalle):   Development of MRF-based MOS equations to predict the probability of thunder storms for the summer season was completed; development of equations for the cool season is underway. The current status of the AVN/MRF MOS thunderstorm and severe weather forecast systems was documented in a paper entitled "Development of MOS Thunderstorm and Severe Thunderstorm Forecast Equations with Multiple Data Sources" by K. K. Hughes. This paper will be presented at the American Meteorological Society's 18th Conference on Weather Analysis and Forecasting to be held in Ft. Lauderdale in July.

We have begun development of an Eta-based MOS system to predict maximum/minimum (max/min) temperature, probability of precipitation, and probability of thunderstorms and severe weather over the contiguous U.S. Stations with reliable precipitation reports that can be included in the development have been determined.

In late April, several NWS forecasters alerted us to possible problems in the warm season AVN MOS visibility/obstruction to vision guidance. During the 1200 UTC forecast cycle, we found the guidance was predicting low visibility and fog at certain locations and times of the day when synoptic weather conditions indicated low relative humidity and good visibility. We deter mined that the probability thresholds required to obtain the categorical visibility and obstruction to vision forecasts during the 1200 UTC cycle were incorrect. We've redeveloped the probability equations and coincident threshold values and expect to implement them in early June. Tests on sample cases have shown a significant improvement in the guidance.

We completed our assessment of the impact of proposed MRF model changes on the MOS extended-range guidance. Results were presented at the May meeting of the Committee for Analysis and Forecast Techniques Implementation. Verifications for February 2001 as well as for a 2-week period during late April 2001 showed that: the MOS temperature guidance from the "parallel" MRF tended to be colder than the operational guidance, that the MOS dew point guidance from the parallel MRF tended to be drier than the operational guidance, and that the skill of the MOS precipitation and sky cover guidance did not vary significantly between the two model runs. Our samples were small, and we're reluctant to say that the MOS guidance will be less (or more) accurate after the AVN/MRF model is changed. However, we suspect that the temperature guidance, particularly during the warmer part of the day, will tend to be slightly cooler than last warm season. The dew point guidance will be slightly drier. We're continuing to look at some of the results, but at this point we've decided not to redevelop the warm season MOS temperature and dew point forecast equations.

Operations and Software Support (M. Erickson):   In early May, we implemented the new MOS max/min temperature guidance for the Northwest, California-Nevada, Colorado Basin, and Missouri Basin River Forecast Centers (RFC's). Initially, forecasts of the max/min were available out to 7 days in advance. Following a request from the Northwest RFC, however, we extended the guidance out to 10 days to help the RFC support water resource managers. Verification results of max/min guidance for the 8 - to 10-day projection have shown some improvement relative to a forecast based on climate.

A paper entitled "Observational Data and MOS: The Challenges in Creating High-Quality Guidance" by R. A. Allen was completed for presentation at the American Meteorological Society's 18th Conference on Weather Analysis and Forecasting. This paper de scribes some of the work required to archive and quality-control observations so that our MOS system can be effectively implemented.

MESOSCALE PREDICTION PROJECT (D. Ruth)

Statistical Update Development (J. Ghirardelli):   The LAMP team completed testing parameters with the new LAMP analysis development code. We decided to produce LAMP analyses for continuous fields like sea level pressure, temperature, and dewpoint, and winds on an 80 km grid, and analyze discontinuous fields and the saturation deficit field on a 10 km grid. These analyses will be used for initialization of the simple LAMP models.

A proposal was prepared that outlines steps necessary to establish a 10 km radar mosaic archive for use within MDL.

National Digital Forecast Development (D. Ruth):   The NDFD IWT met on May 16-17 in Silver Spring. The team made recommendations for providing hourly NDFD updates via the SBN broadcast, coordinating the first national issuance of the day 7 forecast, and handling ad hoc requests for digital data. The team proposed a schedule for phasing in forecast grids at NDFD prototype offices.

FORECAST EVALUATION PROJECT (W. Shaffer)

Forecast Verification Analysis (V. Dagostaro):   We have been working with OCWWS to expand their statistics-on-demand web site to include additional weather elements and types of forecasts. In addition to the usual monthly statistics comparing NGM MOS and local forecasts to observations, we generated the same statistics for AVN MOS. (Folks, the AVN MOS looks good!) We also provided OCWWS with cloud amount verification scores for both local forecasts and NGM MOS guidance.

We are exploring ways to graphically depict our statistics and the data used in our verifications. We've generated various types of diagrams, such as scatter plots, attributes diagrams, and discrimination plots, to show forecasters how they are doing in comparison to guidance, and how they might be able to improve their forecasts relative to guidance.

COASTAL MARINE PREDICTION PROJECT (W. Shaffer)

Storm Surge and Wave Forecasting (W. Shaffer):   A script was written to extract winds and pressures from the Eta-22, and to use them at their full resolution to drive our extratropical storm surge model. We hope to improve surge forecasts in the upper Chesapeake Bay by utilizing the Eta-22's finer wind resolution in that area.

For our wave forecasting effort, we acquired detailed bathymetry for the island of Kauai and WAM wave hindcasts, allowing us to apply our wave models to hurricane Iniki.

Dr. Shaffer made presentations at New Jersey's Hurricane Preparedness Conference and at the Long Island Hurricane Preparedness Conference.

PRODUCTS GENERATION PROJECT (R. Meiggs)

WFO Application Development and Support (E. Mandel):   Integrated Work Team for Data Encoder/Decoder Development and Maintenance.

A little over a year ago, Jack Kelly established an Integrated Work Team (IWT) led by Bob Glahn for Data Encoder/Decoder Development and Maintenance. The Team established an Encoder and Decoder Software web site (http://www.nws.noaa.gov/mdl/iwt) to provide easy access to the NWS "official" encoders and decoders. As of April 30, 2001, there were 18 decoders/encoders available from the website.

On May 16, 2001, the new GRIB2 (General Regularly-distributed Information in Binary, Edition 2) developed by MDL was added to the IWT Encoder and Decoder web site. The decoder (encoder) provides the mechanism to unpack (pack) GRIB 2 data. This new edition of GRIB can achieve greater data compression than the older version. MDL contributed much to the actual compression algorithm. The World Meteorological Organization (WMO) intends to approved GRIB2 for operational use after November 7, 2001. Until then, GRIB2 can be used experimentally to distribute products within the U.S. and will be used in the NDFD work.