STORM SCALE PREDICTION PROJECT(S. Smith)

Nowcasting Guidance Development (D. Kitzmiller):   Approximately 10 different high resolution radar mosaics are now being generated every 5 minutes on the prototype operational production system located at NCEP. The mosaics are distributed to the NCEP National Centers for viewing on their N-AWIPS systems.

Hydro-Meteorological Monitoring (T. Filiaggi):  MDL developers met with members of OCWWS, OHD, and the field for two days to receive initial feedback on the latest version of Flash Flood Monitoring and Prediction (FFMP) system which includes the high- resolution stream monitoring capability of AMBER. New requirements and priorities for FFMP were also discussed and decided upon.

A new development team for SAFESEAS began designing this new automated marine metwatch and decision assistance tool. The team is composed of government and contract employees in MDL as well as representatives from OCWWS and OST's Science Branch.

Watch/Warning Assistance (M. McInerney):  On February 20-21, NWS field forecasters from around the country attended the first WWA Application Workshop in Silver Spring. The purpose of the workshop was to obtain feedback from NWS forecasters on the current operational version of the WWA application, as well as on both short and long term plans for enhancements of the software. The WWA application is a critical component in preparing, disseminating, and monitoring NWS watches, warnings, and advisories.

SYNOPTIC-SCALE PREDICTION PROJECT (P. Dallavalle)

Statistical Forecast Development (P. Dallavalle):   As mentioned in last month's report, we've begun work to provide MOS guidance at a finer spatial resolution than is done currently. We've obtained terrain elevation data with 30 arc-second resolution and have put the terrain heights on a 5-km grid covering the contiguous United States. We are now working to get the terrain heights into the MOS-2000 system and to compute a variety of derived fields, such as slope and aspect, that can be used as predictors in the MOS system. These new variables, as well as other geo- physical variables, will be tested in the development of MOS equations that can be applied to any point in the United States.

We've obtained observational data from both the National Data Buoy Center (NDBC) and the National Climatic Data Center (NCDC). The NDBC data contain observations of wind direction and speed, temperature, and other weather elements for buoys and C-MAN sites in the Great Lakes and the coastal waters of the United States. The NCDC data contain observations of daily snowfall, maximum temperature, minimum temperature, and precipitation amount for approximately 8,000 sites in the United States. We will use the two datasets to augment the guidance available in the current MOS system. We've now designed a process to extract the data and write the observations into a file that can be packed into MOS- 2000 format. The snowfall reports will be used to develop the AVN and Eta MOS snowfall systems for the upcoming 2002-03 winter. We will be adding the buoy and C-MAN sites to the MOS system to provide additional guidance to NWS forecast offices with coastal responsibilities.

Development of spring season MOS equations to predict the probability of thunderstorms and the conditional probability of severe weather from the 0600 and 1800 UTC forecast cycles of the AVN model has been completed. Development of the summer season equations is underway. We've also completed derivation of the warm season probability of precipitation (PoP) equations required for the 0600 and 1800 UTC AVN forecast cycles. We're now working on the AVN MOS quantitative precipitation forecast (QPF) equations for the same cycles.

Development of Eta-based MOS equations to predict warm season QPF at stations in the contiguous United States has been completed. Tests of the Eta-based MOS system to predict total sky cover have also been completed. Verifications indicate that the Eta-based MOS sky cover guidance is quite comparable in skill to the AVN MOS guidance. We're now beginning development of the operational equations to predict sky cover during the warm season.

Operations and Software Support (M. Erickson): On February 26, 2002, we implemented the new Eta-based MOS system. Forecasts of wind speed and direction, probability of precipitation, and quantitative precipitation amount are now being produced twice a day for stations in the contiguous United States. At this time, the guidance is only available for internal use. Forecasts of maximum and minimum temperature and total sky cover will be added to the system prior to formal implementation and dissemination in April.

We are currently evaluating the effects on the AVN/MRF MOS guidance of changes being made to NCEP's Global Forecast System (GFS). Changes to the GFS are planned for March, April, and May 2002.

MESOSCALE PREDICTION PROJECT (D. Ruth)

Statistical Update Development (J. Ghirardelli):   Testing continued this month on the Cloud Layer Advection Model (CLAM) cloud forecasts. Various combinations of advecting winds were testing in driving the model. In addition, the cloud layer analysis was systematically tested for optimal parameters.

The saturation deficit analysis and model were tested this month. Two problems were identified which adversely affected the analysis. The software was changed to address these problems, and the analysis improved significantly.

National Digital Forecast Database Development (D. Ruth):   MDL software is now decoding and mosaicing WFO-produced GRIB bulletins received on the NDFD central server. At this time, only grids from eastern cluster NDFD WFOs are being processed. Additional grids will be incoming as new IFPS software with the MDL GRIB encoder is downloaded and installed by WFOS.

We are coordinating a National Digital Forecast Database (NDFD) meeting for March 20, 21 in Atlanta. The purpose of the meeting is to obtain specific recommendations for NDFD implementation in 2003 based on experiences during the NDFD prototype. Agenda items include digital forecast definition, digital forecast dissemination (national and local), and forecast practices that support interoffice collaboration. The group also plans to tour the Weather Channel while in Atlanta.

PRODUCTS GENERATION PROJECT (M. Peroutka) IFP Product Development and Evolution (M. Peroutka):

Alpha testing of IFPS software for AWIPS Release 5.2.1 continued at two of the three WFO clusters to support the National Digital Forecast Database (NDFD) prototype. This software was also installed at a Western Region WFO.

A new process for developing future releases of IFPS and delivering it for field testing was defined. This process, named Rapid Alpha Process (RAP), provides a mechanism for frequently updating IFPS at Alpha Test WFOS. In the future, this process can provide a path to deliver IFPS capabilities to all WFOS.

FORECAST EVALUATION PROJECT (W. Shaffer)

Forecast Verification Analysis (V. Dagostaro):   To support OCWWS' goal of increasing the number of public weather verification sites, we modified our daily data collection software to accept more data from each WFO. The software has been tested and submitted for implementation in NCEP's operational job stream. So far, eight WFOs indicated they will transmit additional data, giving us 26 new verification sites beginning March 1. We worked with several WFOs to test our procedures, checking for receipt and completeness of data, identifying possible problems in the local setup of new sites, and investigating missing data from certain sites.

We continued examining the quality of NGM- and AVN-based MOS wind forecasts, comparing them to direct model forecasts from the NGM, AVN, and Eta models. In general, we find that the wind speeds from MOS are approximately 1 to 1.5 knots better in mean absolute error (MAE) than the raw model output at the forecast locations. We also find that wind directions are improved by MOS over raw model output. We examined the effect of using various wind speed thresholds, below which we considered no data. Here, the MAE's change slightly, but the general improvement by MOS is still evident.

With offices more closely examining AVP statistics, missing data from the AVP collection are being questioned. Missing data can occur when a site's METAR observation is stripped of its remarks or when an improper product identifier is used. At OCWWS' invitation, we attended a recent MSD conference call to discuss changing the long-standing policy of not replacing missing AVP data with data from other sources. We provided to OCWWS a draft proposal stating the problem of missing data in the verification program and proposed as a solution filling in missing observations or MOS guidance data from MDL's existing archives whenever feasible. We presented the pros and cons of such a policy change. OCWWS is polling the field to see if they recommend accepting these changes.

We coordinated with NCEP to assure that new AVP data transmitted from the Alaska Region WFos will be routed properly through the NWS Telecommunications Gateway and handled properly by NCEP's data collection software.

COASTAL MARINE PREDICTION PROJECT (W. Shaffer)

Storm Surge and Wave Forecasting (W. Shaffer):    Low water levels are critical to ships traveling through navigation channels. NOS's Physical Oceanographic Real-Time System (PORTS) is being developed to forecast water levels for several critical ports. NOS runs models for the Galveston Bay area, the Chesapeake Bay area, and the New York Harbor region, all of which use the MDL extratropcial storm surge extratropicalt to drive their outer boundary specifications. Although we are generally interested in the MDL modells performance for himodelsr situations, we examined it's performance for low-water forecasts during critical winter cases in the Galveston Bay area from 1997-2001. Since water levels are at their yearly lows during the winter, large %'negative" surges (wind-depressed water levels) are critical to shipping. We find that the MDL extratropical surge model shows skill in forecasting these low-water conditions. We will conduct similar tests for PORTS areas along the East Coast.

We are working with NHC forecasters to extend our hurricane wave model to compute boundary of waves greater than 12 feet for guidance in issuing marine advisories. We examined NCEP East Coast wave model forecasts for both deep and shallow water. The forecast wave heights are consistent with local buoy observations. We are working with WFOs and NCEP to make coastal wave forecasts available within 100 n.mi. of the shore. These wave data will be fed into IFPS for an operational test for NWS forecast offices at Wakefield, Virginia, and Jacksonville, Florida.