The primary objective of the Techniques Development Program is to develop analysis and forecast techniques which, when implemented, will help improve forecast accuracy and service to the wide range of users of NWS products. These techniques are implemented on NOAA's computer system, when appropriate, and guidance products disseminated via AFOS, facsimile, or other NWS distribution systems. Techniques are produced for basic weather elements used in public and aviation forecasts, such as temperature and visibility. Also, special emphasis is given to marine-related forecasts and to those forecasts especially associated with mesoscale processes. For many synoptic-scale forecasts, the output of operational numerical models is used to produce forecasts of weather elements of interest to users. When dealing with forecasts of shorter time and space scales, more use is made of data sets rich in information on those scales; for example, hourly and automated surface reports and radar, satellite, and profiler data. For marine-related forecasts, numerical and statistical models relate elements of interest, such as storm surge, to atmospheric analyses and forecasts.

OBJECTIVE WEATHER PREDICTION PROJECT (P. Dallavalle)

Short-Range Weather Forecasting Task (Dallavalle):   Development of the new MOS system based on the Aviation (AVN) run of NCEP's Global Spectral Model (GSM) continues. We've now developed MOS equations to predict the maximum/minimum (max/min) temperature, 2-m temperature, and 2-m dew point during the warm season (April - September). Equations have been derived for over 1000 stations in the contiguous U.S., Alaska, Hawaii, and Puerto Rico. Equa tions are available for both the 0000 and 1200 UTC forecast cycles and for projections out to 72 hours after initial model time. Development of cloud amount and ceiling height equations for the cool season (October - March) has also been completed. These equations predict the probability of various categories of cloud amount and ceiling height valid every 3 hours from 6 to 72 hours after 0000 and 1200 UTC. Development and testing of equations to predict the probability of precipitation, precipitation amount, cloud amount, ceiling height, visibility, and obstructions to vision during the warm season are underway.

We continued our efforts to develop new MOS thunderstorm and severe local storm probability forecast equations for projections out to 60 hours from the Eta model and out to 72 hours from the AVN. Six additional months (April - September of 1999) of lightning observations and severe thunderstorm reports were collected and processed for inclusion in the MOS system. Improvements in the observation of cloud-to-ground lightning led us to refine our definition of a thunderstorm event. Based on the revised defini tion, new 5-year monthly relative frequencies for thunderstorms and severe thunderstorms were calculated. Station lists to be used in development, forecast cycles, and corresponding forecast projections were finalized, and new predictor subroutines to compute the lifted index and the product of the K-index and the monthly relative frequency were written. We're now beginning to develop and test forecast equations.

At the February meeting of the Committee for Analysis and Fore cast Techniques Implementation (CAFTI), we presented the results of testing the AVN MOS equations to predict temperature, dew point, wind speed and direction, cloud amount, ceiling height, and precipitation type. We also showed an example of the new AVN MOS alphanumeric message. CAFTI recommended that the new MOS guidance be implemented. The Eastern Region Scientific Services Division (SSD) requested that some changes be made to the alphanumeric message. We've accommodated that request and are in the process of revising the draft of the Technical Procedures Bulle tin that describes the message.

Efforts to convert the current NGM- and AVN-based MOS operations to the IBM were completed. The codes and script used to produce the MOS guidance on the Cray were moved, revised, tested, and certified as operational on the IBM. All archives of model and observational data were also moved to the IBM.

From late December 1999 through the first week of February 2000, we produced NGM MOS guidance from a revised version of the NGM that was being tested by NCEP on the IBM. A similar test was made retrospectively on NGM forecasts from July of 1999. We then verified the operational NGM MOS forecasts versus the test guidance. Forecasts of max/min temperature, dew point, probabil ity of precipitation, wind speed and direction, precipitation type, cloud amount, ceiling height, and visibility were evaluated for over 700 sites in the contiguous U.S. and Alaska. Differ ences between the operational and test guidance were small, although we did find some indication that the IBM version of the NGM MOS guidance deteriorated slightly at projections of 48 hours and greater. NCEP presented these results at the March meeting of CAFTI where the migration of the NGM to the IBM was judged to be successful.

Medium-Range Weather Forecasting Task (M. Erickson):   Development of new forecast equations to predict temperature out to 192 hours after 0000 UTC and quantitative precipitation out to 156 hours after 0000 UTC continues. These equations are based on the Medium-Range Forecast (MRF) run of the GSM. Experimental cool season equations to predict the max/min temperature, 2-m tempera tures and dew points valid every 6 hours, as well as the probability of equaling or exceeding .01, .10, .25, .50, 1.00, and 2.00 inches of liquid-equivalent precipitation over 12- and 24-h periods were developed and tested on independent data. Verifica tions were presented at the February CAFTI meeting, along with the proposed new MRF MOS message. The Eastern Region SSD re quested that some changes be made to the message, and we've addressed that concern. The CAFTI presentations and verification results were also made available to interested customers through the branch web site.

For the probability of precipitation amount during 12-h periods, the cool season verifications demonstrated improvements in the Brier Score over climate out to the 156-h projection for amounts up to .50 inch. Similarly, for precipitation amount during 24-h periods, the guidance showed skill relative to climate out to the 156-h projection for amounts up to 1.00 inch. The experimental probability of precipitation (PoP) forecasts for both 12- and 24-h periods showed approximately 5% more improvement over climate than the operational MOS PoP forecasts out to 108 hours, and smaller improvements out to 192 hours. For temperature and dew point, the experimental forecasts were approximately 1øF more accurate than direct model output at all projections; this difference amounted to a 2- to 3-day improvement in forecast accuracy due to use of the MOS approach. The MOS temperature and dew point forecasts were always more accurate than forecasts generated by simple climatic equations. Finally, the new max/min guidance met or exceeded the accuracy of the operational MOS and climatic forecasts at every projection.

At the conclusion of these experiments, cool season equations for temperature, dew point, and max/min temperatures were developed for all 1081 MOS stations by using data from the 1992-93 to 1998- 99 cool seasons. Development of final warm season equations is currently underway. Although prior experiments were conducted on warm season samples, equations to predict temperature and dew point were never developed. In addition, we will now extend the prediction of temperature and dew point from 192 to 264 hours after 0000 UTC to support the Climate Prediction Center (CPC). Likewise, development of the final PoP and precipitation amount equations for the warm season is underway.

Coordination with CPC personnel regarding TDL support for the 3-10 day threat assessment product continued. We agreed on the list of stations for which temperature, dew point, and max/min temperature will be provided. Forecasts of these elements for the 1999 warm season were generated. The forecasts and corre sponding historical observations will be supplied to CPC to use in developing an apparent temperature product.

Efforts to convert the current MRF-based MOS operations and MRF model archives from the Cray to the IBM were successfully completed.

National Verification Processing Task (V. Dagostaro):   In support of the NWS modernization, we continued the process of switching the official source of the verification data from AFOS to AWIPS for each Weather Forecast Office (WFO) in the conterminous U.S. As of March 29, approximately 73% of the WFOs had officially switched to the AWIPS-based forecast data collection software known as the AWIPS Verification Program (AVP), and most other offices were in the process of switching. During this quarter, we assisted NWS Regional Headquarters and individual offices by providing information regarding the receipt and quality of data transmitted from AWIPS and by answering questions about the AVP software. We also aided the Alaska Region in their effort to develop data collection and transmission software by answering questions regarding the encoding of the data and by checking the transmitted data for errors. Finally, the central data collec tion software was modified, tested, and implemented on the IBM mainframe.

To help the Office of Meteorology's (OM) development of a data base for public and aviation verification results, we provided a test file containing PoP scores. In addition, we compiled a master list of stations from the pre-AFOS, AFOS, and AWIPS eras for which verification data will be provided to populate OM's database. We recently began to convert the pre-AFOS data to MOS- 2000 format and to streamline the process of providing monthly verification scores for the master station list for all weather elements.

Finally, we generated verification results for the October - December period from 1995 to 1999 for max/min temperature, PoP, ceiling height, visibility, and wind speed and direction. Scores were computed for approximately 105 stations in the conterminous U.S. and were provided to OM.

LOCAL TECHNIQUES DEVELOPMENT PROJECT (S. Smith)

0-3 Hour QPF and Severe Weather Task (D. Kitzmiller):   Final preparations were made for operational twice-hourly production and dissemination of a 10-km national radar reflectivity mosaic created from Radar Coded Messages (RCM's). These mosaics receive automated quality control to remove non-precipitation echo features.

An effort was initiated to refine the quality-control procedures used in production of the 10-km radar mosaic and automated Radar Observations. It was found that the current procedures sometimes fail to remove echo features due to biological targets (insects and migrating birds) when such features appear under dense, cold clouds. A pattern recognition technique to identify these features based on their location relative to radar sites and their characteristic echo intensity distribution has been suc cessfully tested. New procedures have also been tested for determining whether echoes are actually precipitation.

Routine ingest of base reflectivity, composite reflectivity, vertically-integrated liquid (VIL), and digital precipitation radar products through a prototype File Transfer Protocol dissem ination system within the NWS Telecommunications Gateway was started in January. With this data source, real-time national reflectivity mosaics can now be produced at 2-km resolution, with 16 echo intensity levels between 0 and 75 dBZ.

Procedures to archive convective storm cell information on a national basis were initiated, utilizing data from the radar product dissemination system described above. Information including echo tops, maximum reflectivity, maximum VIL, hail probability, and mesocyclone indications are extracted from composite reflectivity products and stored for later collation with other meteorological observations and with severe local storm reports. The archive will eventually be used to refine the operational AWIPS severe weather probability nowcast and short- range QPF algorithms, which are based on older, geographically- limited samples of radar observations and which do not utilize Doppler velocity information. The archive will be used to derive climatologies of severe storm parameters like mesocyclones, tornado vortex signatures, etc.

Processing of lightning observations, Stage III rainfall esti mates, and severe local storm reports for the 1999 warm season was completed. These data will be added to the archive used in developing the 0-3 h QPF, lightning, and severe local storm forecast packages.

Thunderstorm Identification and Forecasting Task (S. Smith):   The AWIPS SCAN/Flash Flood Monitoring and Prediction (FFMP) teams delivered SCAN 2.0 and FFMP 1.0 to the PRC for systems integra tion and testing for AWIPS 5.0.

SCAN 2.0 was used successfully by forecasters in Little Rock, Arkansas, during at least two severe weather events in the on- going alpha test. Preparation for alpha testing was also made for the forecast offices in Goodland, Kansas; Dodge City, Kansas; Boulder, Colorado; Albany, New York; Western Region Headquarters; and the NEXRAD Operations Support Facility in Norman, Oklahoma.

SCAN 2.0 was demonstrated at the NWS Booth during the Annual Meeting of the AMS in Long Beach. Two papers were presented in Long Beach as well: (1) "Flash Flood Monitoring and Prediction in AWIPS Build 5 and Beyond" by Smith, Roe, Filiaggi, Glaudemans, Churma, Xin, and Erb at the 15th Conference on Hydrology and (2) "National Weather Service SAFESEAS - A New Marine/Coastal Monitoring and Forecasting Capability for AWIPS" by Hirschberg, Smith, and Mercer at the 16th International Conference on Inter active Information Processing Systems

Local AWIPS MOS Program Task (J. Ghirardelli):   LAMP code was delivered to PRC for systems integration and testing for AWIPS 5.0. Work was completed to run LAMP nationally at the national centers.

Quality control of the observations used as predictors in LAMP was added. The method is based on the climatology of meteorologically reasonable ranges for observational data, and varies both by region of the United States and season. Additional quality assurance checks were added which will cause the LAMP guidance to not be produced in the event of grossly reduced observation receipt.

Code modifications were completed to allow the LAMP system to run in the future using gridded predictor data on resolutions as fine as 20 km. The current horizontal resolution is 80 km.

In response to a request for data from the Central Weather Service Unit in Fort Worth, Texas, the full LAMP bulletins for all LAMP stations in the CONUS are now being provided via the LAMP web page and the OSO server.

Heavy Precipitation Forecasting Task (J. Charba):   Development, maintenance, and extensive testing of the LAMP QPF system contin ued for the operational AWIPS baseline as well as the national prototype system which provides experimental QPF products through the web to the general public and through NAWIPS to NWS offices.

The LAMP QPF system was delivered to the PRC for systems integra tion and testing in AWIPS 5.0.

The LAMP QPF program flow and product quality control checks were incorporated. These checks can result in the non-issuance of QPF products when the number of precipitation measurements used in the 1- and 3-h antecedent precipitation analysis (an important QPF predictor) does not meet minimum criteria.

Software was developed that interpolates the QPF products to the 948 LAMP stations over the CONUS. The interpolated QPF products are then included into the LAMP station bulletins, with the primary usage being an aid in trouble-shooting the overall operational AWIPS LAMP system.

Software and QPF-related data archives were migrated from the NCEP Cray systems to the new IBM-SP system.

MARINE ENVIRONMENTAL PREDICTION PROJECT (W. Shaffer)

Hurricane Storm Surge Forecasting Task (W. Shaffer):   We continue to update the New Orleans/Lake Pontchartrain SLOSH model's database. The bathymetry and terrain require considerable blending together along the coastline. We have found that using a GIS is quite useful for this purpose. We draw the model grid, together with terrain heights, water depths, and model features on the screen, with a background of USGS raster-scanned quadran gle maps. Unfortunately, we find that editing these features within the GIS is cumbersome and slow. As a result, we're continuing to edit model features with our own graphical editing programs.

Dr. Shaffer assisted in presenting the FEMA/NHC course "Introduc tion to hurricanes." The first course was conducted for emer gency managers along the Gulf; the second for emergency managers along the southeastern Atlantic coast; and the third was designed for emergency managers from the northeastern Atlantic coast. Each course had material and exercises tailored to the emergency managers' locations and threats. The courses included an over view of hurricanes, hurricane forecasting and inaccuracies in those forecasts, storm surges and SLOSH modeling, use of the HURREVAC and SLOSH display programs, and various topics related to hurricane evacuation studies, planning, and mitigation. Other instructors for the course included the NHC hurricane specialists and FEMA regional hurricane coordinators.

Extratropical Storm Surge Forecasting Task (J. Chen):   We converted the operational code and scripts for our extratropical storm surge model from the Cray computer to the new IBM computer. We continue to experiment with improving the model initialization by including the AVN's analyses at 0600 and 1800 UTC. Our current water level initialization produces excessively negative water levels in areas of high pressure. We are experimenting with several methods of reducing this problem, first looking at bays and estuaries, then at the influence of deep-oceanic areas. We started to examine statistics of hourly buoy winds, comparing them to AVN winds interpolated to the buoy sites in both time and space. Our reason for doing this is two-fold: to examine a possible deficiency in the time interpolation and to re-calibrate the drag coefficient based on water level verification.

Coastal Wave Forecasting Task (C.-S. Wu):   To obtain an offshore bathymetry for New Jersey, we examined several interpolation techniques and experimented with these schemes' smoothing. For coastal flooding, we recommend that the same offshore bathymetry be shared by surge forecasting and coastal wave propagation.

We worked with the University of Hawaii in using the WAM model to simulate wave run-up around the Hawaiian Islands for Hurricane Iniki. We continue to compare coastal wave models and have found that the SWAN runs very slowly since it transforms ocean wave spectra to the shore.

LOCAL PRODUCTS DEVELOPMENT PROJECT (D. Ruth)

IFP Product Development and Evolution Task (M. Peroutka):   Interactive Forecast Preparation System (IFPS) alpha software was upgraded at the WFO in Norman, Oklahoma, and it was installed for the first time at the WFO in Grand Rapids, Michigan. Developers traveled to the Grand Rapids WFO to complete the installation and train the staff. This is the first version of IFPS that can support NWS Marine Forecasts. It is planned for release nation wide in AWIPS 5.1.

IFPS developers began drafting a new version of the IFPS User Guide using Hypertext Markup Language (HTML). This work will make it possible to post the latest revisions of the IFPS User Guide on the Worldwide Web, and, eventually, distribute the User Guide with AWIPS software.

The IFPS capability for coordinating forecasts between neighbor ing WFOs was given a thorough test by using systems at Norman, Oklahoma; Tulsa, Oklahoma; and Wichita, Kansas. Several improve ments were identified, and they will be made available in future IFPS releases.

IFP Implementation and Enhancement Task (R. Meiggs):   We continued to forge ahead in implementing IFP this month. We provided a presentation and demo of current and planned IFPS capabilities to Office of Meteorology staff. We assisted with ICWF/WWA training at Eastern Region Headquarters for the New England area WFOs. We prepared several sessions for the IFPS Forecaster Working Group meeting at the NWS Training Center in Kansas City. And we participated in the second CRS Formatter Workshop hosted by the Office of Meteorology.

All IFPS software for AWIPS 5.0 was provided to PRC by February. Since then, TDL developers have continued to test and resolve problems as they arise. We coordinated special installation and test procedures for IFPS with PRC. This will enable us to deliver IFPS software to all WFOs with AWIPS release 5.0, but actually only install the software at a WFO upon completion of their IFPS training at the NWS Training Center.

IFP Interpretation and Editing Task (D. Ruth):   TDL continued to port IFPS software to LINUX PCs used at WFOs participating in the Rapid Prototype Project (RPP). We are evaluating various FORTRAN 90 compilers to support our RPP development.

We devoted considerable attention to documenting model interpre tation techniques for the IFPS User Guide. Tim Boyer traveled to the NWS Training Center in Kansas City to discuss the latest developments with instructors there.

WFO Applications Development and Support Task (D. Ruth):   TDL developers completed the design, development, and unit testing of all AWIPS applications targeted for AWIPS 5.0. We are now supporting Software Integration Testing at PRC.

We are also conducting alpha tests on several of our AWIPS 5.0 applications in the field. Alpha tests already in progress or soon-to-begin at WFOs include a TAF QC function, an updated Watch/Warning/Advisory program including a summary product for NOAA Weather Radio, and SCAN software with ranked cell-tables based on capabilities in WDSS.

Two TDL developers continued to support the AWIPS Site Support Team this quarter. A backlog of AWIPS Trouble Tickets has been eliminated and any new trouble tickets on TDL applications are now being resolved in a matter of hours rather than days. In doing this, we are documenting how to solve typical software problems. This documentation is being provided to the NCF, SST, and will also be included in updated versions of AWIPS user guides.

WFO Local Applications Development Support Task (E. Mandel):   The AWIPS Local Applications Policy, which addresses Office Director comments, was completed. The new version of the policy can be found at the AWIPS Local Applications Development web page.

The Local Applications Working Groups LAWG has completed updating the AIFM standards and guidelines and the AWIPS Local Applications Implementation Plan. The new version are available at the AWIPS Local Applications Development web page.

The LAWG, with the assistance of Jason Burks (WR), completed Beta testing of the Local Application Database (LAD). The database will allow the sites and regions to submit and register local applications, view an inventory of local applications, post and respond to user questions, report and respond to software deficiencies, and upload and download registered local applications. A LAD User Guide was prepared and made available via the Help feature of the LAD. All LAD backup procedures have been written and tested. TDL staff assumed responsibility for the maintenance of the LAD on March 31, 2000. The LAD will become operational on April 12, 2000.

Other activities the LAWG is involved with include: maintaining the Local Application Web site with the latest local application news, developing performance analysis procedures to assist in the evaluation of local applications, and participating in NWSTC training activities.