Forum 2/2003

In This Issue

Introductory Page


Mesoscale NWP

Managing Metadata

Object Data System

GPS Forecasting

Data Distribution

Weather Products


Recent Publications

Contact the Editor
Nita Fullerton

Will von Dauster
John Osborn

Best Viewed With
Internet Explorer

Briefs Masthead

...From the Director's Office

Science On a SphereTM Unveiled at the NOAA Science Center – The "NOAA Science On a Sphere TM Illumination” took place at the NOAA Science Center in Silver Spring, Maryland, on December 2 – 4, 2002.

NOAA Science On a SphereTM allows one to view many of the Earth’s global features from a space traveler’s point of view without ever having to leave the Earth. This invention by Sandy MacDonald communicates NOAA science to the public, fosters science education, and aids scientific visualization. NOAA Science On a SphereTM is intended for museums, science centers, schools, and anywhere there is interest in science. It "is an exciting and informative way for people to see NOAA's climate, weather and ocean science," said retired Navy Vice Adm. Conrad C. Lautenbacher, Ph.D., Undersecretary of Commerce for Oceans and Atmosphere and NOAA Administrator. "For example, viewers can watch how the warm water in the Pacific that signals an El Niño travels across the ocean, or watch a hurricane form as a small storm slowly gathers strength, traveling westward from Africa, across the Atlantic Ocean toward the Gulf of Mexico." It provides a dramatic visualization of complex information in an understandable form for the public; a unique instrument for teaching science, math, and geography; and a handy scientific tool to translate numerical information into visual images.

Sandy and SOS

Sandy MacDonald discussing Science On a SphereTM at the NOAA Science Center in Silver Spring, Maryland.

The next exhibit of NOAA Science On a SphereTM will be at the 83rd Annual Meeting of the American Meteorological Society, February 9 – 13, 2003, in Long Beach, California.

Class with SOS and Sandy

Mrs. Lautenbacher's class attending the demonstration of Science On a SphereTM at the NOAA Science Center, and Dr. MacDonald at the podium.

FSL Tech Day 2002 – FSL invited businesses from the private and corporate sector, and federal, state, and local agencies to the Skaggs Research Center in Boulder last October to introduce them to recent technological advances at FSL. This exposition exceeded expectations, with good turnout, interaction, and opportunities for followup. Those in attendance came from diverse groups such as the Air Force, Colorado State University, Vaisala, Raytheon, Trimble, Johns Hopkins Applied Research Laboratory, the National Center for Atmospheric Research, Meteorologix, and Harris Corporation. Attendees expressed appreciation for the opportunity to participate and suggested that it become an annual event. FSL Tech Day provided a venue for technology exchange and gave other organizations a clearer understanding of the various projects at NOAA and FSL.

...Information and Technology Services

FSL Central Facility Data Availability – A new service is provided to external and internal users of FSL data. A new Website...

has been created to provide information about the status of FSL data availability. For example, if a data source becomes unavailable, a message will be posted indicating the name of the product, when it was last received, and what action is being taken to investigate the outage. Followup messages will indicate when the product returned and all other pertinent information regarding its status. Messages are posted on the page in reverse chronological order and can be searched for specific datasets using standard Web browser search tools. The establishment of this Internet service alleviates the need to send data availability status information to specific mailing lists.

...Forecast Research Activities

Impact of GPS-IPW Data on RUC Forecasts – A primary goal of GPS-based measurements of integrated water vapor (or integrated precipitable water, IPW) is to improve numerical weather predictions of atmospheric water vapor. Water vapor, in turn, will impact such important variables as precipitation and instability. Accurate forecasts of water vapor are generally more difficult than those for other parameters such as wind and temperature. One factor responsible for this is that atmospheric moisture is often highly variable in 3-D space and time, more so than wind or temperature. Observations of moisture are inadequate, even in relatively data-rich areas such as over the United States, to fully resolve these variations. Generally, there have been three observational sources for atmospheric moisture: rawinsondes, METARs, and satellite (not available in cloudy areas below cloud top).

Estimates of IPW from GPS signal time delays can complement these moisture observations. GPS-IPW using zenith total delay provides only a vertically integrated value, by definition, but with at least hourly resolution and in all weather conditions, including cloud and even precipitation when observations are most important for forecasts of the atmospheric moisture.

Over the years, FSL has developed a GPS-IPW network that now produces high-accuracy, hourly, near-real-time measurements at more than 130 stations in the United States.

Tracy Lorraine Smith will present the most recent results from a series of GPS-IPW data impact studies performed at FSL with the Rapid Update Cycle (RUC) data assimilation and numerical forecast system at the 2003 Annual Meeting of the American Meteorological Society in Long Beach.

Using D3D – FSL developed a visualization tool, called D3D for Display in 3D, which could potentially add 3D visualization capabilities to National Weather Service (NWS) Weather Forecast Offices (WFOs). The advent of low-cost computers using Linux and their availability at many WFOs makes it possible to reliably run D3D.

Although D3D is a very small pilot program at FSL, by working with a well-established 3-D software package (Vis5D) designed for meteorology, we have been able to focus on making changes that allow for compatibility with standard AWIPS model and analysis grids, and have made the overall package look and feel like the familiar AWIPS D2D. With Linux and economical high-powered PCs available to WFOs, we have been able to widely distribute D3D for experimental testing by operational forecasters. Although individual forecasters have been impressed with many of the D3D capabilities, despite the wide distribution of D3D, it does not yet enjoy widespread use. As the Interactive Forecast Preparation System (IFPS) becomes more routine to forecasters, they should have more time to begin seriously examining D3D. Most who have used D3D agree there is potential value to operational forecasting, in particular as an excellent way to more quickly and thoroughly examine the three-dimensional atmosphere, which could aid in alleviating the problem of potential forecaster overload in dealing with more and more numerical model output. To tap this potential we urge the operational weather agencies to begin a program to integrate D3D into operations for formal tests and risk reduction activities. Edward Szoke will present a paper on this topic at the 2003 Annual Meeting of the American Meteorological Society.

Multilevel Cloud Retrieval from the GOES Platform – Since GOES-NEXT (Series 8 – 11) has become operational, significant effort has been made to maximize information in the GOES platform, including radiances and products from the measurements. For example, cloud-top pressure derived from GOES sounder data has been used in the operational 20-km Rapid Update Cycle (RUC) at the National Centers for Environmental Prediction (NCEP) since April 2002. Also, cloud track winds derived from GOES imager data have signficantly contributed to forecast improvement. Data from each instrument have relative merit; for example, imager data have better coverage and higher spatial resolution (4 km) than sounder data (10 km). Sounder data have better measurement accuracy, and with 19 channels it is possible to retrieve vertical temperature and humidity profiles.

Drs. Dongsoo Kim and Steve Koch have carried out experiments at FSL with samples of GOES sounder and imager data collected from FSL's GVAR station. Verification of the results is performed by comparing measured radiances with computed radiances with multilevel cloud assumptions during April 2002. Cases of multilevel clouds were selected and compared to measured brightness temperatures and computed brightness temperatures using estimated cloud fraction. A paper on this topic will be presented at the 2003 Annual Meeting of the American Meteorological Society.

...Demonstration Activities

Profiler Data Supports NASA in Columbia Shuttle Investigation – NOAA's Profiler Network (NPN) Program received requests on Monday, February 3, 2003, from NASA and the National Transportation Safety Board (NTSB) for NPN upper-air wind data collected during the breakup of the Columbia space shuttle. Nine of the 35 NPN profiling clear-air radars, located in the states of New Mexico, Texas, Oklahoma, Arkansas, and Louisiana, were near or directly under the flight path of the shuttle (see map).

NPN Map with Shuttle Path

NOAA Profiler Network and the approximate flight path of the Columbia Shuttle.

Because the unmanned profilers automatically acquire wind data continuously from near the ground up to 53,000 feet, they were able to capture the times and horizontal and vertical positions of the falling fragments of Columbia.

The radar data from the profilers located at Palestine, Texas, and Winnfield, Louisiana, detected falling fragments from Columbia as it disintegrated. At Palestine, some larger fragments falling at speeds of 30 feet per second or more were detcted minutes after NASA lost communications with Columbia. At Winnfield, a fall of much smaller but far more numerous particles was detected four to six hours following the breakup. It is likely that some are fine particles of aluminium from the shuttle's skin and frame, since they were highly reflective.

FSL collected NPN data, reports from dozens of boundary-layer profilers operated by cooperative agencies, and water vapor data acquired using GPS-based technology and wrote them onto CD-ROMS. The CDs were sent by overnight express to officials at National Weather Service's (NWS) Southern Region Headquarters in Ft. Worth, Texas, who are handling all coordination with NASA, NTSB, and the Federal Emergency Management Agency (FEMA). Experts on the NPN will be available at FSL to provide technical support for the forensic analysis of the shuttle breakup. The NPN is unique in that no other observing system can provide the detail of information so critical for determining the time and space parameters associated with the Columbia breakup. Refer to the following figure.

NPN Data with Shuttle Circled

A unique view "looking upward" above the Palestine profiler site. The measured vertical velocities are displayed as color: yellow to red colors for downward motion, and blue to purple as upward motion. Vertical velocity data are measured by the Profiler once every 6 minutes. A rather consistent pattern of downward falling objects is apparent above about 45,000 feet immediately following the time of the shuttle breakup (~1400 UTC). Many of these objects are falling with a velocity of 25 – 35 feet per second, but apparently still have much larger horizontal velocities and pass quickly through the 4-degree wide radar beam. Note that objects appear to continue to fall through the radar beam for about 30 minutes.

FSL operates and maintains the NPN for NOAA. The NWS uses data from the 35 network profilers routinely in computer-generated forecasts, and its field forecasters tailor model guidance to local conditions. The NPN data are also available on the Global Telecommunication System for international users, and to everyone via the Web at Data are used by many other federal, state, and local organizations for support in weather forecasting, aviation, and monitoring climate and air quality. One of NPN's customers, Lawrence Livermore National Laboratory, uses NPN data as critical input to its dispersion model. The model supports work under contract to the Departments of Defense and Energy for Homeland Security.

NOAA Profiler Network and Other Emerging Global Profiler Networks – Wind is perhaps the most important dynamic variable that must be measured to forecast the weather. Conventional means of measuring winds above the surface utilize balloons released simultaneously around the globe. Balloon soundings are made twice daily to support weather analysis and forecasting. In the past 20 years, Doppler radar wind profilers have been developed that are capable of continuous measurement of atmospheric winds throughout the troposphere. Frequent wind measurement made possible by the profilers enables the measurement of representative winds with very good resolution. Such measurements have provided a wealth of new information on small-scale atmospheric dynamics and provided a powerful new tool for nowcasting and mesocale forecasting.

The true value of profiler data is just now being recognized. The more profiler data are used, the more their value will be understood and exploited. Moreover, new developments in numerical weather prediction with increased emphasis on mesoscale modeling, especially for air quality and quantitative precipitation forecasting (QPF), promise to increase the value of profiler observations to weather forecasting. Profilers are only one of many new technologies to be developed in the past twenty years. The challenge and opportunity for weather research is to integrate these diverse measurement systems into an efficient system capable of monitoring the atmosphere and providing the optimal data required to support operational forecasting. It is also important to recognize that profilers can do more than measure the wind. Considerable research over the past two decades has demonstrated the contribution profilers can make to the measurement of turbulence and waves as well as precipitation.

The NOAA Profiler Network (NPN), operated by FSL for more than a decade, represents the culmination of several decades of research and development of wind profiling Doppler radars. The NPN comprises 35 tropospheric wind profilers (404/449 MHz) mostly located in the central United States. The NPN infrastructure, built over the years, has the flexibility and capacity to handle many other profilers in addition to the 35 NPN systems. With recent advances in computers, networking, and communication technologies, real-time profiler data can be acquired from almost anywhere on the globe. Data from remote sites are submitted to quality control and placed onto the Global Telecommunication System. Currently, FSL receives data from about 80 sites in the continental U.S., Alaska, Canada, and along the Equator west from South America. These data are routed to operational forecast centers where they are used in a variety of numerical weather prediction models, and are also distributed to the local forecast offices to tailor model guidance to local conditions. The data are placed on the FSL Website for viewing in many graphical forms, and are also available for downloading to a user’s site in numeric format.

An article on this topic authored by Margot Ackley and Kenneth Gage was presented at a SPIE Conference on Remote Sensing of the Atmosphere, Ocean, Environment, and Space in Hangchou, China, in October 2002.

An Automatic Meteorological Data Collection System Installed at GPS Monitoring Stations – FSL is conducting a long-term experiment to test the effectiveness of using the precise geodetic position measurements made by a network of GPS monitoring stations to determine the total amount of water vapor contained in the sectional volume of the atmosphere above each station. By knowing the exact position of the GPS satellites along their orbits and the precise location of the GPS monitoring receivers on the ground, an interpretation of the location error (actual location versus receiver-derived apparent location) yields a good indication of the amount of water vapor in the atmosphere. This result occurs because the monitoring station's apparent location error is partially caused by the water vapor.

Many factors influence the propagation of the electromagnetic waves as they travel through the Earth's atmosphere from the constellation of GPS satellites (distributed along their orbits) to the GPS monitoring receivers (distributed throughout a ground surface network). One of these factors is the total amount of atmospheric water vapor. Other factors that affect the local speed of propagation of the electromagnetic waves transmitted from the GPS satellites, as the waves travel toward the GPS ground receivers, are the degree of ionization of the ionosphere and the mass density distribution of the air in the atmosphere. By subtracting the effects of the ionization and of the mass density distribution from the monitoring station's total position error, the fraction of the total error caused by atmospheric water vapor can be isolated. With this value, the quantity of water vapor in the atmosphere can be calculated.

A paper on this topic by E.D. Michelena and Seth Gutman was presented at the Oceans 2002 Conference and Exibition, Biloxi, Mississippi, in October.

...Systems Development Activities

MADIS Assists the National Weather Service with Homeland Security – In a recent announcement, NOAA’s National Weather Service (NWS) and AWS Convergence Technologies, Inc. (AWS) unveiled a major new public-private partnership that will bolster the government’s ability to respond to a Homeland Security event and protect lives and property.

In the event of a Homeland Security incident involving airborne hazardous materials, the NWS will have available real-time weather data from the AWS WeatherNet network to help them provide accurate weather forecasts to government agencies and emergency managers. AWS WeatherNet, established in 1993, is a highly dense, nationwide commercial network that includes more than 6,000 automated weather station locations based primarily at schools (see the following figure).

AWS WeatherNet

AWS WeatherNet, a nationwide commercial network of 6,000 automated weather station locations.

To assist the NWS, FSL has agreed to do all of the initial work in ingesting, quality controlling, and assessing the AWS observations. The NOAA plan for initial assessment specifies that:

  • AWS data will be sent to FSL every 15 minutes for ingest into the FSL Meteorological Assimilation Data Ingest System (MADIS),
  • FSL will quality control and gather statistics on sensor performance and provide feedback to AWS, and
  • FSL will display observations on its surface observation Webpage ( to allow for subjective assessment, and will also ingest the observations into their experimental data assimilation and forecast model systems for objective evaluation.

Through MADIS, FSL will also distribute the observations to NWS field sites for operational assessment within AWIPS, and to the NWS National Centers for Environmental Prediction (NCEP) for model development and sensitivity studies.

Work will begin at FSL by ingesting observations from nearly 100 AWS stations located in New England.

The observations will be added to more than 5000 "mesonet" observations already collected by MADIS from various state and local government agencies, universities, research organizations, and private firms.

For more information on MADIS, contact Patricia Miller at 303-497-6365 or E-mail, or see

FX-Collaborate Enhancements to Support Operations at NWS, USAF, and NASA – The FX-Collaborate (FXC) is a meteorological data display system with strong support for collaboration. Depending on its application, it is also referred to as FX-Connect or the Briefing Tool. The program is written in Java and the client software has been tested on Intel-based personal computers running Linux, various Windows operating systems, and on a Macintosh running Yellow Dog Linux. FXC is unique in that it provides an extensive set of data display functions often found only in more powerful meteorological workstations, such as AWIPS, and integrates these with collaborative functions, such as text exchange (chat room) and elaborate drawing capabilities (analogous to whiteboarding) specifically designed for meteorologists and weather forecasters.

Herb Grote will discuss recent enhancements to FXC at the 2003 Annual Meeting of the American Meteorological Society (IIPS Conference).

...Aviation Activities

FSL Winner of 2002 Excellence in Aviation Award – FSL is one of several laboratories and universities supporting the FAA’s Aviation Weather Research Program that were named winners of the 2002 Excellence in Aviation Award.

The Excellence in Aviation is a highly competitive award presented each year to individuals and/or institutions that show how their past research benefits the aviation community today. Through this award, the FAA formally recognizes significant accomplishments as a result of aviation-related research efforts. This special distinction gives the FAA an opportunity to recognize superior research efforts and to highlight the benefits of such activities. FAA's Associate Administrator for Research and Acquisitions notes: "Inclement weather is responsible for 69 percent of flight delays and approximately 30 percent of fatal accidents. The laboratories supporting our weather research program are providing critical safety enhancements by developing tools to generate more accurate and accessible weather observations, warnings, and forecasts."

...Modernization Activities

IFPS Workshop in Boulder – The National Weather Service Western Region Headquarters and FSL cohosted the first IFPS (Integrated Forecast Preparation System) Methodology Workshop at FSL on January 14 – 16, 2003. The topics covered included the lastest updates on the National Digital Forecast Database (NDFD) efforts, software plans, and some of the best methodologies.

Participants were very active in contributing and presenting methodologies used at field offices. The group included Science and Operations Officers (SOOs), IFPS Focal Points, Lead Forecasters, and Regional Program Managers. All of the Western Region Weather Forecast Offices were represented, along with other offices participating from the Pacific Region, Alaska Region, Central Region, and Southern Region.

The workshop featured a session in which each participating forecast office presented an example of its best methodology or technique for producing some portion of the digital forecast. The goal was to expose all participants to the best techniques practiced at each office. In general, each presentation was followed with helpful discussions and debates on the benefits and limitations of each methodology. All participants agreed that the workshop was a very worthwhile exchange of ideas and information. Several FSL developers also made presentations to inform forecasters of currently available and upcoming IFPS features that could aid the NWS Regions in generating digital forecasts. The agenda and many of the files and documents are available at the workshop agenda website:

Text formatting with the GFESuite – FSL is working closely with National Weather Service (NWS) field forecast offices in developing the grid-editing component of the Interactive Forecast Preparation System (IFPS). This component, the Graphical Forecast Editor Suite (GFESuite), allows forecasters to define a weather forecasting gridded digital form. Once defined, most of the NWS products are then derived from this digital forecast database. Textual forecasts are among the most widely known products of the NWS. With the advent of digital forecasting, it is possible to produce these products automatically, allowing forecasters to better focus on meteorology and provide more consistency to the products. An established set of text formatters is included in the IFPS. The FSL development team is developing an alternative set of text formatters which produce standard text products directly from the digital forecast database. The GFESuite text formatters are being developed within the Rapid Prototype Process (RPP), which delivers software in a rapid fashion and incorporates forecaster feedback to quickly improve the system.

Tracy Hansen, a GFESuite developer at FSL, will present a paper on this topic at the 2003 Annual Meeting of the American Meteorological Society.

Rapid Prototype Project – Over the last three years, FSL has been working closely with NWS field forecast offices in an effort to evolve as quickly as possible the grid-editing component of the Interactive Forecast Preparation System (IFPS). This component, called the Graphical Forecast Editor (GFESuite), allows forecasters to define a weather forecast in gridded digital form. Once defined, most of NWS products are then derived from this digital forecast database. To reduce the risk in implementing a system with such a radically different paradigm, the NWS decided to deviate from its established software development, testing, and delivery methodology and deliver software in a much more rapid fashion. This was the premise behind establishing the Rapid Prototype Project, in which forecasters' comments could be more quickly utilized, and the system could evolve in a much shorter time frame.

Thomas LeFebvre, a GFESuite developer at FSL, will present a paper on this topic at the 2003 Annual Meeting of the American Meteorological Society.

Porting AWIPS to Linux – Since AWIPS was officially commissioned in 2000, FSL has continued to play an active role in developing future releases of the AWIPS software and investigating new technologies. The FX-Linux software runs on more than one computer platform. It has been thoroughly tested on an HP computer running HP-UX 10.20 and PC/Linux hardware. The ability to run the software on another platform makes it possible to achieve AWIPS functionality and better performance on a desktop PC or a laptop computer. Greater hardware independence makes it possible to take advantage of new technological developments in the computer industry. So far, FSL has ported the AWIPS display software and basic system infrastructure, and work is underway to port remaining portions of the AWIPS system.

Darien Davis will present an article on this topic at the 2003 Annual Meeting of the American Meteorological Society.

...International Activities

AWIPS at the Korea Meteorological Administration – FSL cooperates with the Korea Meteorological Administration (KMA) in the development of the Forecaster’s Analysis System (FAS), an AWIPS-like forecaster workstation. This exclusive system can easily analyze all weather information by overlaying, combining, and animating the data in a single display. FAS support allows forecasters to focus on the weather event itself, thus reducing the overall forecast preparation time.

AWIPS has been successfully installed and is running at the KMA forecast office. FAS runs on the Linux platform and receives all observational data and operational numerical model outputs in the KMA. The system was deployed at six regional offices in July 2002, and will be deployed at 37 weather stations in 2003. The FAS system architecture consists of four to nine D2D workstations, depending on the regions, and two data servers. The Combined Meteorological Information System (COMIS), KMA’s official communication system, distributes all meteorological data in KMA headquarters through a Gigabit network and to regional offices through an ATM network. Over the next three years, nowcasting tools will be integrated into the FAS system.

Young-Sun Jung, a KMA visiting scientist at FSL, will present a paper on this topic at the 2003 Annual Meeting of the American Meteorological Society.

Wavelet-based Data Compression Technique With Precision Control – With the development of computer hardware, observation networks, and numerical weather forecast model techniques, the amount of the model data becomes increasingly large. A typical high-resolution model produces a few gigabytes of data every day. In order to effectively transmit and store the model data, we need to consider and investigate some appropriate compression techniques.

The wavelet ly unnoticeable losses, and are considered "meteorologically useful" in many applications. For model data, however, we need a more rigorous control of the losses, or errors. One common requirement is that the precision, or the magnitude of the maximum allowable round-off error, has to be within a given threshold. This is a more difficult requirement to meet for wavelet transform-based (or other orthogonal transform-based) data compression. With the same precision requirements, our compression scheme outperforms typical lossless codec (coder and decoder) by between 200 – 600%.

A paper on this topic by Drs. Ning Wang and Renate Brümmer will be presented at the 2003 Annual Meeting of the American Meteorological Society (at the IIPS Conference).

Weather Data Compression – Significant reductions in weather data size without loss of any useful information can be accomplished through the use of data compression and rounding to the least significant digit. The source of compression of weather data can be traced to the smoothness and rounding of the data (that is, rounding off the insignficant digits and extracting the smoothness results in increased symbol frequency). Common applications such as gzip and bzip2 encode frequent symbols with less bits. Hence, compression allows more data to be sent over a fixed bandwith channel. FSL has compared several compression techniques with respect to 12-kilometer weather data from the Eta model.

Weather data compression has become more important with higher resolution and ensemble model data, and the commissioning of the construction of new weather satellites using 12 bands instead of 3. The fixed bandwidth of the SBN and limited disk space are also driving the interest in data compression.

Dr. Christopher Steffen will present a paper on this topic at the 2003 Annual Meeting of the American Meteorological Society.

FSL Staff FSL in Review FSL Forum (Other Editions) Publications