B
Case Studies of Transitions from Research to Operations

INFRARED SOUNDERS

Infrared sounders provide information on the vertical structure of temperature and water vapor. These soundings have become a routine and essential part of day-to-day numerical atmospheric modeling around the world. This case study outlines the unplanned and long pathway of infrared sounders from research to operations.

Research Origin/Heritage

Temperature soundings have been obtained from weather satellites since the late 1960s when the NASA environmental research satellite Nimbus, the Defense Meteorological Satellite Program (DMSP) satellite, and the Improved TIROS Operational Satellite (ITOS) flew early research versions of infrared spectrometers and microwave radiometers. They were flown by NASA, the U.S. Air Force, and NOAA because the climate and weather research community wanted to find a way to observe temperature and humidity profiles over data-sparse regions. The missions were flown as research and development (R&D) efforts, but with the objective of having their data used in operational weather forecasting. In the case of the DMSP, the Air Force used operational systems to fly R&D sensors such as the sounder. In the case of the civil satellites, NASA flew the R&D sensors on its satellites (Nimbus) and made their research sensor data available to the ESSA (Environmental Science Services Administration)—the predecessor to NOAA—for operational use. The



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 101
B Case Studies of Transitions from Research to Operations INFRARED SOUNDERS Infrared sounders provide information on the vertical structure of temperature and water vapor. These soundings have become a routine and essential part of day-to-day numerical atmospheric modeling around the world. This case study outlines the unplanned and long pathway of infrared sounders from research to operations. Research Origin/Heritage Temperature soundings have been obtained from weather satellites since the late 1960s when the NASA environmental research satellite Nimbus, the Defense Meteorological Satellite Program (DMSP) satellite, and the Improved TIROS Operational Satellite (ITOS) flew early research versions of infrared spectrometers and microwave radiometers. They were flown by NASA, the U.S. Air Force, and NOAA because the climate and weather research community wanted to find a way to observe temperature and humidity profiles over data-sparse regions. The missions were flown as research and development (R&D) efforts, but with the objective of having their data used in operational weather forecasting. In the case of the DMSP, the Air Force used operational systems to fly R&D sensors such as the sounder. In the case of the civil satellites, NASA flew the R&D sensors on its satellites (Nimbus) and made their research sensor data available to the ESSA (Environmental Science Services Administration)—the predecessor to NOAA—for operational use. The

OCR for page 101
TABLE B.1 Chronology of Early Satellite Sounders Flown on NASA and NOAA Satellites Instrument Satellite Primary Period of Operation SIRS-A Nimbus-3 1969-1971 SIRS-B Nimbus-4 1970-1972 ITPR, NEMS, SCR Nimbus-5 1972-1975 VTPR ITOS series 1972-1979 HIRS, SCAMS, PMR Nimbus-6 1975-1979 HIRS-2, MSU, PMR NOAA series 1978-1998 VAS GOES series 1980-1996 HIRS-3, AMSU NOAA series 1998-present NOTE: The acronyms are spelled out in Appendix E. SOURCE: Kalnay et al. (1996). willingness to use the data in operational forecasts, however, involved negotiations between the data providers (NOAA) and the data users (the National Meteorological Center [NMC]). In 1969, the NMC director told a NOAA scientist, referring to satellite soundings, “If you can make them look like radiosonde data we can use them.”1 It would be many years involving much study before the satellite-derived soundings would be fully employed in operational numerical weather prediction. It is worth noting that NMC began to experiment with the operational sounding data from the very first sounder, the SIRS (Solar Infrared Radiation Station)-A, flown on the Nimbus-3 satellite, less than 2 months after its launch. Data used for the National Center for Atmospheric Research/National Centers for Environmental Prediction (NCAR/NCEP) Reanalysis Project reflect sensors flown on NASA and NOAA satellites as indicated in Table B.1 (Kalnay et al., 1996). The first infrared High-resolution Infrared Radiation Sounder (HIRS-2) and Pressure-Modulator Radiometer (PMR) and Microwave Sounding Unit (MSU) sounding system flown together on an operational satellite was the Television Infrared Observational Satellite (TIROS) Operational Vertical Sounder (TOVS), which began flying on TIROS-N in October 1978. TOVS data were made available to the global numerical modeling community in 1979, with the launch of NOAA-6. One of two different types of algorithms was used to transform the sounder radiance observations into temperature and moisture values at given levels. One of these algorithms was based on statistical regression relations between the temperature and moisture values at specific vertical levels of the atmosphere and the radiances observed within all the spectral channels of the sounding radiometers. This was the method used with SIRS-A, Nimbus-6, and the early NOAA TOVS data. Alternatively, a 1   Personal communication from Ronald McPherson, Executive Director, American Meteorological Society, to committee member George L. Frederick.

OCR for page 101
nonstatistical matrix inverse method (i.e., a one-dimensional variational analysis method), which used the forecast as a first guess in the retrieval, was used throughout various periods of the satellite data processing, beginning with SIRS-B in 1970. In either case, these “retrievals,” as they were called, of temperature and moisture values were used in numerical model runs without consistent positive impact, except in the data-sparse Southern Hemisphere, until the mid-1990s, at which time researchers discovered a better way to use the observations. Rather than trying to assimilate the temperature retrievals derived from the radiances, they assimilated the radiances themselves using a new three-dimensional variational analysis technique. The resulting improvement was dramatic, as reflected in Figures B.1 and B.2. Figure B.2 shows root-mean-square (RMS) observational increments (differences between FIGURE B.1 Impact of the direct assimilation of radiances on the 5-day forecast 500-hPa anomaly correlations in the Northern Hemisphere, June through August. The improvement in 1995 was due to the assimilation of radiances from infrared sounders. SOURCES: Steve Lilly, National Centers for Environmental Prediction; Kalnay et al. (1998). Reprinted with permission.

OCR for page 101
FIGURE B.2 RMS difference (M) between 6-hour forecasts of 500-hPa heights and radiosonde observations. SOURCES: Steve Lilly, National Centers for Environmental Prediction; Kalnay et al. (1998). Reprinted with permission. 6-hour forecasts and rawinsonde observations) for 500-hPa heights. The large improvements in 1995 are associated with the direct assimilation of radiances. The reason why the satellite soundings had little impact on the forecast was that the satellite retrievals were treated as poor-quality radiosondes (i.e., point measurements) rather than as high-quality volume measurements (i.e., what the radiances represent). By assimilating the radiances rather than the retrievals, the proper spatial resolution of the data was necessarily represented in the model, thereby avoiding the prior aliasing of smaller-scale features with low spatial resolution data (i.e., the satellite soundings). It was not until the data user fully understood the characteristics of the satellite observations that a technique was devised for the proper assimilation of these data in their analysis/forecast operation. Transition Process There was no clear starting point for the transition from research on infrared sounders to their use in operational service and numerical weather prediction models. Early attempts by NMC to use the retrieved soundings occurred immedi-

OCR for page 101
ately after the first observations were made, and the impact on the forecasts was slight, or even negative, over data-rich regions. Eventually, by the mid-1990s, researchers found a much better way of using the observations—assimilating the radiances directly in the models—and for the first time the infrared sounding observations showed a consistent positive impact in the Northern Hemisphere (Kalnay et al., 1998). Operational Status Infrared sounders have become a valuable part of the global observing system, and the observations from the NOAA polar orbiters are now being used by all major weather-forecasting centers in the world. Using the radiances directly in models through three-dimensional and four-dimensional variational analyses has been so successful that the accuracy of forecasts in the Southern Hemisphere has approached that of forecasts in the Northern Hemisphere (Figure B.2). Plans are under way to incorporate data from infrared sounding sensors on geostationary satellites, thus increasing the temporal frequency of data over the region scanned. The assimilation of satellite soundings over Northern Hemisphere land areas, not done now so as to avoid a potential conflict with radiosonde information, should also improve forecasts as a result of enhanced spatial and temporal resolution of the analyses used to initialize the prediction process. Advanced infrared sounders to be placed on future geostationary satellites (e.g., Geosynchronous Imaging Fourier Transform Spectrometer [GIFTS]) will be able to vertically resolve the wind field, as well as provide temperature and moisture profiles with high spatial and temporal resolution, thereby providing improved observations of atmospheric dynamics as needed in order to improve the forecast operation. Lessons Learned The extended period between the first flight of an infrared sounder and the full operational use of soundings in numerical models can be attributed to a number of factors: Inadequate scientific research to determine an appropriate way of using the observations. The discovery of how to use the radiance values rather than the retrieved profiles in numerical modeling took about 25 years. Resistance to change. A major potential user insisted that the data look like radiosonde data. Lack of a technology transition plan or process. The operational centers were not prepared to use the data when they first became available.

OCR for page 101
VERY HIGH RESOLUTION RADIOMETER/ ADVANCED VERY HIGH RESOLUTION RADIOMETER Imaging radiometers used by NASA and NOAA are commonly designed to produce calibrated multichannel images of Earth and the atmosphere in the visible and infrared portions of the spectrum. These radiometers measure reflected solar radiation in the visible and near-infrared wavelengths and emitted radiation in the shortwave and thermal wavelengths from Earth’s surface and the atmosphere. Radiometer data have become a mainstay for remotely sensing sea surface temperature and atmospheric temperature and moisture profiles. This case study captures the successful transition of research data from the Very High Resolution Radiometer (VHRR) into operational service. The early involvement of both research and operational managers at NOAA, as well as the development of a plan to “market” the benefits of the data to users, fostered the widespread use of the data. Research Origin/Heritage As part of the Operational Satellite Improvement Program, the Improved TIROS Operational Satellite (ITOS-1), the first three-axis stabilized polar-orbiting weather satellite, was launched on January 23, 1970. This satellite, developed and launched under NASA funding, carried both the older-generation vidicon instruments for real-time Automatic Picture Transmission and a new, two-channel scanning radiometer for both day and night imaging. An upgraded version of the ITOS, ITOS-D (later renamed NOAA-2), was launched in October 1972. For this satellite, NASA replaced the vidicon cameras with the redundant SRs and added a new instrument, the two-channel VHRR, which was to provide high-resolution day and night imaging. There was close cooperation between NASA and NOAA. For example, because of budget difficulties at NASA, NOAA agreed to support some of the costs of developing these new instruments. Also, NASA developed the ground-based processing software for the VHRR data, and NOAA supplied the computers. Transition Process Although satellite data were becoming more widely accepted by users in the early 1970s, the ad hoc way during the early years of moving a new remote sensing capability into operations was recognized as inadequate. NOAA’s National Environmental Satellite Service (NESS) created a project management arrangement within NESS, with the project manager selected from research and the project coordinator from the operations area. The project manager was to assure that the goals of ability

OCR for page 101
to calibrate, accuracy, and so on, were being met, while the project coordinator was to assure that VHRR data could be processed and delivered in a timely manner. Since the existing operational activities were saturated with meeting the needs of the operational weather community, NESS established an Environmental Products Branch to respond to the needs of users outside the operational weather community—primarily oceanographers, hydrologists, and others—for products not being developed by the operational community. This research community of users was small at the time, and focused on using the data from the recently launched Earth Resources Technology Satellite (later renamed Landsat 1). Since VHRR data were of significantly lower spatial and spectral quality than Landsat data, they did not compete well for the attention of these users. However, VHRR data had one attribute not offered by Landsat—near-real-time availability to users. To remedy the fact that there were few nonoperational users of VHRR data, the Environmental Products Branch set about to develop products that would take advantage of the near-real-time availability of VHRR data and then went out to the user communities to “sell” this capability. Some products—such as Polar Ice Charts, which depict the coverage and estimate the age and thickness of the polar ice sheet—were quickly accepted by the Navy Ice Center (which was located in the same building as NESS). Other products—such as Gulf Stream Analysis, which offered information on the location of the Gulf Stream and its eddies; West Coast Upwelling Charts, which provided details on the position and extent of the upwelling cold water along the coast; and Snow Cover Charts, which were important for managing the water resources of the western United States—had to be brought to potential users and marketed as supporting their needs. Soon these nonoperational users became accustomed to the products, and an operation was born. As part of the meteorological upgrade process, the two-channel VHRR became a four-channel advanced VHRR (AVHRR/1), with the addition of a near-infrared (IR) channel to separate clouds from snow and ice and another IR channel to aid in determining sea surface temperature. The splitting of the original IR channel into two nonoverlapping intervals for further improvement in determining sea surface temperatures was designated as AVHRR/2, a five-channel instrument. Operational Status Currently, the six-channel AVHRR/3 is the operational instrument on the NOAA-15 and the subsequent POES (Polar-orbiting Operational Environmental Satellite) series; it is providing real-time data for sea surface temperature, vegetation index, snow and ice cover, fire detection, aerosols, and volcanic ash detection and tracking.

OCR for page 101
Lessons Learned The lessons learned on transitioning research to operations from this case study on VHRR/AVHRR are these: The operational agency (in this case NOAA) cared enough about the potential utility of the instruments to partially fund their development. It is important to involve the research and operations community early in a mission. The researchers communicated with potential users, educated them about the observations and their potential use, and “marketed” the mission. Near-real-time availability of research data to users is a valuable part of the transitioning process. DEFENSE METEOROLOGICAL SATELLITE PROGRAM The Defense Meteorological Satellite Program (DMSP)2 has served the meteorological needs of the armed services for some 40 years, enabling the military to receive timely weather data in support of planning for both routine and mission-critical operations such as the Cuban missile crisis and the Vietnam War. While not a typical research program but an operational development program, DMSP developed and flew research instruments that were intended to provide operational capabilities, and it did so in an efficient and cost-effective way. This case study illustrates the key role that people play in the successful management and development of a satellite program. The impressive results of DMSP and the satellites that it produced proved beneficial for a series of research satellites and further DMSP satellite developments that were continued as “heritage” throughout the history of the program. Research Origin/Heritage In the mid-1950s, the Rand Corporation warned the U.S. Air Force (USAF) that successful operation of overhead photoreconnaissance satellites depended upon accurate and timely meteorological forecasts of the Sino-Soviet landmass. An interdepartmental study in April 1961 revealed that NASA held the U.S. franchise to establish requirements and develop meteorological satellites for both the Department of Commerce and the Department of Defense. A single National Operational Meteorological Satellite System (NOMSS) for both departments was intended to 2   This section is based mostly on verbatim transcripts and summaries of the history of DMSP written by R. Cargill Hall. Those interested in more detail on the history of DMSP are referred to the original document, A History of the Military Polar Orbiting Meteorological Satellite Program (Hall, 1985).

OCR for page 101
avoid duplication, produce a less costly national capability, and meet all of the forecasting needs of the civil and military sectors. The television camera of NASA’s first experimental “wheel-mode” TIROS weather satellite, spin stabilized to inertial space, was launched on April 1, 1960. It viewed an oblique swath of Earth’s surface occasionally in each orbit, instead of once each time it revolved. In 1962, NASA officials did not believe that a spin-stabilized satellite which would keep its spin axis perpendicular to its orbit plane could be developed in time to meet the need for strategic meteorological forecasts for reconnaissance operations over the Sino-Soviet block. The undersecretary of the Air Force therefore created an “interim” meteorological weather satellite program for the National Reconnaissance Office (NRO). On June 21, 1961, he asked the director of the Office of the Secretary of the Air Force Special Projects in El Segundo, California, for a minimum proposal for four Earth-referenced weather satellites to be launched on NASA Scout boosters. A proposal was provided for a small fixed budget and first launch in 10 months. The proposal was approved, and a program director for the new DMSP was appointed. The new program director, who was a meteorologist and an electrical engineer, accepted the assignment on condition that he would not have to use the resident Systems Engineering and Technical Direction (SE&TD) contractor, that he could select a small number of his own staff, and that he could use fixed-price, fixed-delivery contracts throughout the program. He believed that a SE&TD contractor could only justify its existence by introducing changes. Since changes involved time and money, SE&TD support was incompatible with fixed-price, fixed-delivery contracting. Transition Process DMSP incorporated two management approaches that proved decisive in the program’s success and that may provide a model approach for the transition from research to operations in other endeavors of similar size and complexity: (1) a slim management team and (2) the use of a firm fixed-price contract. The fixed-price, fixed-delivery contract proved valuable in December 1961 when a major structural component of the weather satellite, the base plate, failed during tests, and Radio Corporation of America (RCA) officials requested a 3-month delay for redesign. RCA was advised that it had 10 days to produce a fix or the contract would be terminated, at no cost to the government. The RCA program manager appeared 3 days later with revised internal schedules that met the original launch date. The Air Force “blue suit” program team met its 10-month schedule, although, as the high-risk aspects of the effort suggested, without immediate success. Two Scout vehicles, one with the first NRO weather satellite onboard, failed in rapid succession

OCR for page 101
in April and May 1962. However, on August 23, 1962, the first successful DMSP satellite was launched from Vandenberg Air Force Base in California. Although the ground control team had tracking problems early in the operational period of the satellite, each day at high noon the vehicle took pictures as it passed over the Soviet Union. Weather pictures of the Caribbean returned by this vehicle 2 months later, in October 1962, proved crucial during the Cuban missile crisis, permitting effective aerial reconnaissance missions over the region. The nimble management structure and the fixed-price contract enabled flexibility and responsiveness from both the internal program team and the contractor. Although the team encountered significant challenges with the launch vehicle, DMSP’s success in achieving a rapid and clean program with few technical problems cemented the nation’s approach toward a two-pronged military and civil meteorological satellite system. The first DMSP satellite ceased operation on March 23, 1963, but by then the program had received a new life. NASA had delayed the planned Nimbus series of the NOMSS, and the NRO authorized the DMSP director to extend his interim program by at least a year. Flight number four was launched on February 19, 1963, with many more to follow. In addition to the management and contract mechanisms used for the DMSP, the program benefited from the transition of research sensors into the DMSP program. The manager of the TIROS weather satellite program in NOAA’s Weather Bureau, who was one of the few persons in NOAA cleared to know about DMSP, referred various experiments to the NRO-Air Force program, including a novel one conceived at the University of Wisconsin. The Wisconsin instrument weighed about 6 ounces and produced useful data on the radiated heat of cloud cover, from which the heat balance of Earth could be determined. Many other research sensors were incorporated on DMSP satellites, leading to improved meteorological capability and forecasting in the defense community. When the first DMSP director stepped down in April 1965, DMSP had eclipsed all other overhead meteorological endeavors. Initial NASA skepticism notwithstanding, DMSP had pioneered the space technology so well, so quickly, and so inexpensively that the space agency (prodded firmly by the Department of Commerce) now embraced a carbon copy of the DMSP wheel-mode Block 1 satellite, called the TIROS Operational System (TOS), as an interim, polar-orbiting weather satellite. The first TOS, renamed ESSA-1, was launched in February 1966—4 years after DMSP proved the concept. Nine TOS civil satellites were launched between 1966 and 1969. A Nimbus first launch scheduled for June 1962 slipped to 1964; the Nimbus satellites were eventually directed to research purposes, never to become part of the NOMSS. DSMP sped through four blocks of satellite platforms in the 1960s. By the late 1960s, Block 5 was on the drawing boards. The USAF Air Weather Service (AWS)

OCR for page 101
was largely responsible for a payload design that made Block 5 especially user-friendly—for example, by formatting the imagery to standard AWS weather chart scales. The decision to develop a user-friendly design for the Block 5 series of satellites was a strategy that contributed significantly to the transition from research to operations. The Block 5 spacecraft departed entirely from the TIROS technology and took the form of an integrated system in which the space and ground segments were designed together. The AWS representative and his engineering partner visited meteorologists at work and then examined what the industry could produce. Instead of starting with a sensor and determining what it might tell the user about the weather, these two individuals based the Block 5 design on the users’ wish to receive a product in a form that approached as closely as possible the weather charts and maps that they, the meteorologists, employed—an example of “pull” transitioning. The DSMP case study also teaches some important lessons about the value of involving university researchers and students in satellite missions.3 During the mid-1960s, the university–USAF collaboration with DSMP was greatly aided by a team from the Electrical Engineering Department at the University of Wisconsin (UW), Madison, working with Professor Vern Suomi. Several of Suomi’s meteorology graduate students and some USAF officers spent considerable time on the project at the Air Force Global Weather Central (Offutt Air Force Base, Omaha, Nebraska) and probably as much time in the electrical engineering and space astronomy laboratories as they did in the meteorology department. Their experiments onboard the early DMSP satellites were (1) those of USAF (primary) and (2) joint USAF/UW and sole UW (secondary). The graduate students helped with all of the experiments (design, calibration, data processing, algorithms, and so on). One experiment on the Earth radiation budget flew on several of the satellites from 1964 to 1967. Using black and white flat-plate radiometers, the experiment resulted in the first global Earth radiation budget measurements. The experiment revealed a warmer and darker planet Earth with a lower albedo in tropical regions than had been estimated in pre-satellite papers. This discovery had major ramifications in future global energetics and circulation studies. Operational Status Over time, DMSP was declassified, and program management was moved into the mainstream like other Air Force programs. This change led to an increase in the 3   Personal communications of Thomas Vonder Haar, Colorado State University, with committee member George L. Frederick, May-August 2002.

OCR for page 101
munications blackouts, and power-grid failures. Observations of the solar disc in the x-ray portion of the spectrum provide critical information about solar activity, but they can only be made from space. In contrast to ground-based observations, x-ray images allow easy identification of solar regions called coronal holes, which are closely correlated with the sporadic production of high-energy particles. Full-disc x-ray observations thus provide an important tool for predicting the impact of solar activity on the near-Earth space environment. The research value of regular x-ray solar observations was first demonstrated on Skylab in 1973. NOAA Space Environment Center (SEC) forecasters quickly recognized that an operational follow-on based on imaging the entire solar disc was a high priority. Over the next decade, additional x-ray instruments flew on research missions, but low data rates and the research focus on higher spatial resolution meant that full-disc images were rare. In 1991, the Japanese Yokhoh mission flew a full-disc x-ray imager, the Solar X-ray Telescope (SXT), which reconfirmed the value of daily full-disc images. Transition Process By the early 1980s, the NOAA/SEC interest in flying a full-disc operational x-ray monitor was becoming widely known. One impediment to transition was purely technical: a solar x-ray imager is best suited to a nonspinning platform, and the GOES spacecraft were spin-stabilized prior to the launch of GOES-8 in 1994. The largest impediment to transition, however, turned out to be an inadequate budget. Space weather instruments, such as particle detectors and magnetometers, had historically been small and relatively inexpensive. NOAA/NESDIS was willing to fly a solar x-ray imager on the GOES-NEXT satellites being designed at the time, but had no budget to do so. Because they were aware of similar solar-monitoring needs within the DOD, NOAA/SEC decided to seek the support of the Air Force. Personnel at NOAA/SEC and the Air Force’s ground-based solar observatory in New Mexico started working toward a solar x-ray instrument on GOES. This advocacy process proceeded slowly because of limited human resources throughout the late 1980s, and GOES-NEXT design decisions were made without consideration of the needs for a solar x-ray imager. With a demonstrated need for full-disc solar x-ray imagery but having no operational sensor, NOAA/SEC turned to the Japanese in the early 1990s for access to SXT data. After considerable negotiation, the Japanese agreed to provide access to SXT data with two provisions: (1) images were not to be shared (particularly with military users), and (2) new ideas or discoveries made using the data could not be made public for 1 year. In 1990, the Air Force identified $18 million to apply to the project, but schedule

OCR for page 101
constraints limited inclusion of the x-ray sensor to only the last of the five spacecraft within the original GOES-NEXT procurement. Because spacecraft design decisions had been frozen, it was discovered that about half of the Air Force budget of $18 million would be required for spacecraft modifications alone. To work within the available budget, it was decided that the instrument would be built by NASA at the Marshall Space Flight Center, where some of the costs could be absorbed by nonproject funds. Although the design was to be made available to industry to build follow-on instruments, budget constraints resulted in performance limitations that reduced the value of this instrument for future designs. The instrument was finally launched on GOES-12 in 2001. Two follow-on instruments are being built by industry through a contract initiated in 1996. Operational Status The first operational NOAA Solar X-ray Imager, SXI, was launched on GOES-12 in 2001, nearly 30 years after the research justification for the operational need was identified on Skylab. Following successful checkout, the spacecraft was placed in on-orbit storage, ready to replace one of the currently operating GOES. Lessons Learned The main lesson learned in the SXI experience is that inadequate financial and human resources can prolong the transition from research to operations for many years after a technology has been demonstrated and a need established: Limited personnel left NOAA scientists overburdened during the development of SXI, requiring them to both do science and support the SXI development. These same human resource limitations resulted in use of NOAA personnel with primary expertise in data analysis as the technical advisers responsible for understanding and establishing the SXI instrument design and operational requirements. Difficulties with establishing funding for SXI prolonged the transition process to nearly 30 years (NRC, 2003). VOLCANIC ASH MAPPER The Volcanic Ash Mapper (VOLCAM)9 instrument was designed to conduct research on volcanic clouds and eruption precursors, providing measurements of 9   This case study is condensed from materials provided through personal communications from Arlin J. Krueger, University of Maryland, Baltimore County, to committee liaison William B. Gail, April 4, 2002.

OCR for page 101
volcanic ash and sulfur dioxide (SO2) clouds, SO2, total ozone, smoke, and dust. The data to be collected by VOLCAM would assist in monitoring volcanic ash clouds and would provide valuable information for aviation safety. VOLCAM is an example of a measurement that has demonstrated strong operational potential but, despite substantial effort and interest in both the research and operational communities, has not successfully been transitioned to operational status. Research Origin/Heritage In 1982, data from the NASA Total Ozone Mapping Spectrometer (TOMS) showed a surprising anomaly in the vicinity of the El Chichon volcano in southern Mexico. Investigation of the anomaly led to the first demonstration that SO2 in volcanic eruption clouds can be detected with satellite sensors operating in the ultraviolet (UV) portion of the spectrum, and that the larger-than-expected quantity of SO2, rather than ash, clearly is the driver of volcano-climate effects. At about the same time, the first incidents were being reported of commercial aircraft becoming disabled after encountering volcanic ash clouds. A British Airways 747 lost all power after flying through the ash cloud from the Gallunggung volcano in Indonesia; its engines restarting only after heroic measures, the airliner was forced to land with a windshield that had been rendered nearly opaque by the ash. A KLM 747 also had all four engines flame out when flying through the ash cloud of the Mt. Redoubt eruption in the Aleutian Islands. Again, after heroic measures the crew managed to restart the engines, after losing 10,000 feet of altitude. The aircraft landed safely at Anchorage, Alaska, with damage to the engines, flight surfaces, and windscreen. According to the U.S. Geological Survey (USGS, 1997), at least 15 aircraft were damaged from 1980 to 1997 while flying through volcanic ash clouds along North Pacific air routes. In addition, at least 80 ash cloud encounters occurred worldwide in the same time period, causing hundreds of millions of dollars in damage and lost revenue. These incidents led to great interest in TOMS data by the U.S. Geological Survey (USGS), the Federal Aviation Administration (FAA), and NOAA. A fast turn-around system was developed by the principal investigator of TOMS to respond to requests about reported volcanic eruptions. It soon became clear that SO2 was a unique discriminator between eruptions that produce large, dangerous clouds and smaller eruptions that represent little threat to aircraft. NASA continued to support this quasi-operational capability for a number of years on a best-efforts basis. In the 1990s, theoretical developments indicated that the sensitivity of the TOMS UV technique could be greatly enhanced by a better selection of wavelengths. This meant that the scientific output could be extended to the monitoring of pre-eruptive gas emissions, which were predictive for eruptions. However, the value of this

OCR for page 101
technique from polar orbiting satellites was limited, because the probability of seeing these low-altitude emissions depends on catching cloudfree moments. Thus, it became clear that geostationary satellites provide ideal platforms for volcano monitoring as well as for observing the drift of volcanic clouds. Transition Process Recognizing the operational value of the TOMS data, the FAA, NOAA, and NASA established a Memorandum of Understanding (MOU) to provide for the transfer of NASA technology to NOAA for processing TOMS data and for the provision of raw data to NOAA. The production software was incorporated into the NOAA/NESDIS near-real-time operational system. Special real-time processing codes were developed and delivered to the National Weather Service in Anchorage for immediate readout of the NASA Earth Probe TOMS instrument during satellite passes. The system was also planned for the QuikTOMS mission, but that satellite was destroyed in a launch failure in 2001. Meanwhile, the Office of the Federal Coordinator for Meteorology (OFCM) set up a working group to coordinate the activities of the operational agencies for dealing with volcanic ash clouds in aviation safety. Participants included the FAA, NOAA, USGS, DOD, the Smithsonian Institution, NASA, and the airline industry, represented by the Air Line Pilots Association and Air Transport Association. The primary satellite tools for the detection of volcanic clouds were the NASA/TOMS and the NOAA POES/AVHRR and GOES/sounder instruments. In addition, visible satellite imagery was used to detect plumes by their shapes. One of the immediate concerns of the OFCM Volcanic Ash Working Group was a NOAA plan to change the GOES sounder wavelengths, with the consequence that an important technique for retrievals of ash would no longer be possible. A second concern was the growing risk to aircraft from the increased number of flights, especially in the North Pacific, with its high density of active volcanoes. With this history, NASA made several attempts to develop new capabilities to detect volcanic eruptions. Two pertinent missions were proposed in the first Earth System Science Pathfinder Announcement of Opportunity (AO) in 1996, one by NASA Langley Research Center and one by the Jet Propulsion Laboratory. The proposal from Langley Research Center was selected in Step 1, with the plan to fly as a payload-of-opportunity on a commercial communications satellite, but it had to be withdrawn when the commitment with the satellite provider could not be completed. Goddard Space Flight Center, with the principal investigator (PI) of TOMS as VOLCAM PI, decided to propose the VOLCAM mission to the second ESSP AO in 1998 when OFCM indicated that substantial support would be available if the UV

OCR for page 101
sensor were augmented with an infrared (IR) sensor for nighttime detection. It was estimated at the time that this mission could be accomplished for a NASA cost of $45 million by flying on an existing spacecraft such as GOES. Part of the proposal strategy was to include other agencies as partners, contributing according to their means and capabilities. NASA would do mission development, flight hardware, software development, and scientific research; NOAA would do data ingest, processing, and analysis; FAA would do aviation control planning and education; and USGS would do eruption prediction and diagnosis. NESDIS agreed to contribute the processing costs, and the NWS endorsed the proposed concept, although reservations were expressed about the limited resolution of the IR camera. The FAA also endorsed the proposal, although attempts to elevate the level of commitment within the FAA failed. The FAA uses the NWS as its source of environmental data for air traffic control. The perceived lack of direct involvement in production of the data was a factor that limited the FAA’s contribution to the in-kind training costs following successful launch of the mission. The USGS administrator strongly supported the proposal, but warned that direct funding and substantial commitments could not be provided owing to the Office of Management and Budget’s concerns over “mission creep.” However, the USGS offered in-kind support during the operational test phase of the mission. NASA selected the VOLCAM proposal on the basis of its scientific merits, and asked for a full Step 2 proposal. During the Step 2 study, agreements were reached with the Tracking and Data Relay Satellite (TDRS) project for flight service on either TDRS I or J and with NOAA/NESDIS to carry VOLCAM on either GOES N or O. Two commercial satellite owners also indicated interest in working with NASA. At the end of the Step 2 studies, VOLCAM was one of two candidate missions selected to conduct extended assessment studies, with one of the two to be selected for flight based on the extended study. The strong science, low risk, and flight heritage were cited as the major strengths of VOLCAM. Weaknesses were cited in the data flow plan in the partner agencies, uncertainty in spacecraft integration costs, and lack of maturity in the IR camera design. The weaknesses were addressed in the extended study report. During the oral phase of the report, the question of the commitment of other agencies was raised, and it became apparent that in-kind contributions did not meet the expectations of the associate administrator of NASA’s Earth Science Enterprise (ESE). However, none of the attendees from other agencies was able to commit resources to the instrument and mission development. The prime issue was a commitment for continued operational funding support beyond the scientific demonstration mission rather than support for the proposed ESSP mission. VOLCAM was not selected for flight, but the ESE associate administrator expressed an intention to explore a cooperative program with NOAA for flight of the

OCR for page 101
UV sensor on a future GOES satellite. The cooperative program was later defined as a joint mission with a proposed NOAA instrument, the Special Events Imager (SEI), which had partial funding in the FY 2000 NOAA budget. Owing to the limited funding, both teams were asked to determine whether the functions of SEI and VOLCAM could be combined in a single instrument. Technical issues ended up eliminating this possibility, but a compromise plan was submitted to merge the electronics portions of the instruments. The joint project ultimately failed when Congress eliminated the SEI item from the NOAA budget. Without a NOAA contribution to the hardware costs, the joint mission was abandoned. Operational Status The VOLCAM instrument has not been successfully transitioned to operational status, although some of the VOLCAM eruption-monitoring algorithms developed for use with the TOMS sensor have been transitioned to NWS. Lessons Learned Lessons learned from the VOLCAM case study are largely related to a lack of sufficient agency commitment. In spite of relatively strong interest by a number of agencies, no one agency was sufficiently supportive to lead the transition. VOLCAM was developed out of a long-standing interagency collaboration at the working level, a result of the NASA-led geophysical science and natural hazards program’s having produced information of value to the operational agencies. An MOU was established between NASA and the operational agencies to cooperate in the area of volcanic hazard data. Nevertheless, neither the MOU nor the collaboration between agencies at the working level was sufficient to establish agency commitment for the VOLCAM transition. NOAA has a very limited capacity or budget to evaluate new sensing concepts internally, so advancements in observational measurements are difficult to make unless they involve extending the capabilities of NOAA’s few core instruments. Other than the GIFTS mission, NASA has not funded any geostationary sensor proposals in recent years. Opportunities to evaluate new research sensors or measurements for geostationary operational use have thus been extremely limited. The lack of an organizational transition mechanism between NASA and NOAA makes direct transfer of technologies between the agencies difficult.

OCR for page 101
TROPICAL RAINFALL MEASURING MISSION The Tropical Rainfall Measuring Mission (TRMM) is a low-Earth-orbiting satellite with an orbit that oscillates about the equator between roughly 35°N and 35°S. A joint mission between the United States and Japan, TRMM’s primary stated goals were (1) to improve the understanding of crucial links in climate variability that are due to the hydrological cycle, (2) to improve the large-scale numerical models of weather and climate through assimilation of TRMM data, and (3) to advance our understanding of cloud ensembles and their impacts on larger-scale circulations. Shortly after launch, scientists recognized that TRMM would also provide valuable new information on hurricanes in all stages of development. Research Origin/Heritage In 1987, the U.S. Department of Defense (DOD) became the first to fly a passive multifrequency microwave imager on a meteorological satellite. The Special Sensor Microwave/Imager (SSM/I) channels penetrate to Earth’s surface unless the signal is attenuated by precipitation or large aerosols. One of the SSM/I channels senses at a frequency of 85 GHz, with a spatial resolution of 13 to 15 km. This channel is able to penetrate nonraining clouds. However, larger, frozen hydrometeors (e.g., hail, graupel) and raindrops associated with vigorous convection dramatically scatter radiation at this frequency. Thus, the sensor can detect intense rain associated with hurricane rainbands and the eyewall, owing to the lowered brightness temperatures created by the intense scattering. A time series of 85-GHz data can reveal a storm’s internal convective structure and evolution by mapping the organization and vigor of the convection around the storm center. Building on the success of the DOD program, NASA launched a special satellite for measuring meteorological quantities over the tropics using passive and active microwave sensors. The TRMM satellite completed 4 years of successful data collection in November 2001. The primary TRMM sensors include a precipitation radar, TRMM Microwave Imager (TMI), and Visible/Infrared Scanner. TRMM’s precipitation radar is an active sensor and the first successfully deployed civilian rainfall-rate-monitoring radar to operate from space. The precipitation radar can provide three-dimensional profiles of precipitation through storm cloud patterns. (Kummerow et al. [2000] provide more detail on the TRMM instruments, algorithms, and a wide range of early results.) The TMI is a multichannel, dual-polarized, conically scanning passive microwave instrument similar to the SSM/I. The purpose of the visible/infrared instrument is to enable TRMM to be a “flying rain gauge.” The TRMM satellite radar and radiometer combination is intended to obtain high-quality vertical profiles of precipitation as well as surface rainfall estimates. TRMM’s rainfall-

OCR for page 101
rate observations from the combined radar and passive microwave instruments allow the calibration of empirical rain estimates from the IR sensors. As a result, uncertainty in tropical rainfall has been greatly reduced from earlier space-based estimates. Currently, these techniques are being applied to estimate hurricane rainfall. Transition Process TRMM observations are also being used to provide better initial conditions for numerical models. Krishnamurti et al. (2000) have developed a complex modeling approach using TRMM data that is showing promise for improving 3-day hurricane forecasts of both track and intensity. The approach uses multiple analyses and multiple models to create a “super-ensemble” forecast. TRMM data are being provided in near real time to hurricane forecasters. The high-resolution imagery is being used to help locate the centers of hurricanes and to assess convective organization trends. The value of TRMM to the forecast and research communities is evidenced by its wide usage in operational tracking and forecasting of tropical systems, along with its contribution to the increased understanding of the global water cycle. Figure B.3 shows TRMM-derived rainfall rates and surface winds obtained from QuikSCAT for Hurricane Floyd in 1999. Operational Status TRMM is not considered an operational satellite; however, TRMM observations are being used by operational forecast centers as noted above. The Global Precipitation Measurement mission, a joint Japan-U.S. mission scheduled for launch in 2007, is a follow-on to TRMM.10 Lessons Learned The lessons learned from this case study are these: The involvement of the operational community in preparing for TRMM observations before launch allowed for the rapid testing of TRMM data in operational models. Satellites designed primarily for research or for proof of concepts can provide data that are useful for operations if the operational centers are prepared for the data and if the data are provided in real time. 10   Additional information is available online at <http://gpm.gsfc.nasa.gov>. Accessed January 22, 2003.

OCR for page 101
FIGURE B.3 Example of QuikSCAT surface wind data overlain on TRMM rainfall signatures during Hurricane Floyd in 1999. SOURCES: Liu et al., 2000 (Plate 12 in Simpson et al., 2003). Reprinted with permission. REFERENCES Atlas, R., R.N. Hoffman, S.M. Leidner, J. Sienkewicz, T.-W. Yu, S.C. Bloom, E. Brin, J. Ardizzone, J. Terry, D. Bungato, and J.C. Jusem. 2001. The effects of marine winds from scatterometer data on weather analysis and forecasting. Bull. Amer. Meteorol. Soc. 82:1965-1990. Barrick, D.E., and C.T. Swift. 1980. The Seasat microwave instruments in historical perspective. IEEE J. Ocean. Eng. OE-5:74-79. Black, P.G., R.C. Gentry, V.J. Cardone, and J. Hawkins. 1985. Seasat microwave wind and rain observations in severe tropical and midlatitude marine systems. Adv. Geophys. 27:197-277. Burger, J.J. 1991. ERS-1 ready for launch. ESA Bull. No. 65:13-15. Chelton, D.B. 2001. Report of the High-Resolution Ocean Topography Working Group. Ref. 2001-4. College of Oceanic and Atmospheric Sciences, Oregon State University, Corvallis. Chelton, D.B., J.C.Rèis, B.J. Haynes, L.-L. Fu, and P.S. Callahan. 2001. Satellite altimetry. Satellite Altimetry and Earth Sciences, L.-L. Fu and A. Cazenave, eds. Academic Press, San Diego, Calif.

OCR for page 101
Christian, H.J., K.T. Driscoll, S.J. Goodman, R.J. Blakeslee, D.A. Mach, and D.E. Buechler. 1996. The Optical Transient Detector (OTD). Pp. 368-371 in Proceedings of the 10th International Conference on Atmospheric Electricity, Osaka, Japan, June 10-14, 1996. Christian, H.J., R.J. Blakeslee, S.J. Goodman, D.A. Mach, M.F. Stewart, D.E. Buechler, W.J. Koshak, J.M. Hall, W.L. Boeck, K.T. Driscoll, and D.J. Bocippio. 1999. The Lightning Imaging Sensor. Pp. 746-749 in Proceedings of the 11th International Conference on Atmospheric Electricity, Guntersville, Ala., June 7-11, 1999. ESA (European Space Agency). 1995. New Views of the Earth—Scientific Achievements of ERS-1. ESA SP-1176/1. ESA, Paris, France. Fea, M. 1991. The ERS ground segment. ESA Bull. No. 65:49-61. Freilich, M.H. 1985. Science Opportunities Using the NASA Scatterometer on N-ROSS. JPL Publication 84-57. Jet Propulsion Laboratory, Pasadena, Calif.. Freilich, M.H., D.G. Long, and M.W. Spencer. 1994. SeaWinds: A scanning scatterometer for ADEOS-II— Science overview. Pp. 960-963 in Proc. Int. Geosci. Rem. Sens. Symp., Pasadena, Calif., August 8-12, 1994. Fu, L.-L., W.T. Liu, and M.R. Abbott. 1990. Satellite remote sensing of the ocean. The Sea: Ocean Engineering Science, Vol. 9, Pt. B, B. Le Méhauté and D.M. Hanes, eds. John Wiley & Sons, New York. Gaiser, P.W. 1999. Windsat—Satellite-based polarimetric microwave radiometer. 1999 IEEE MTTS-Dig. 1:403-406. Graf, J., C. Sasaki, C. Winn, T. Liu, W. Tsai, M. Freilich, and D. Long. 1998. NASA scatterometer experiment. Acta Astronautica 43:397-407. Grantham, W.L., E.M. Bracalante, W.L. Jones, and J.W. Johnson. 1977. The SeaSat—A satellite scatterometer. IEEE J. Ocean. Eng. OE-2:200-206. Hall, R. Cargill. 1985. A History of the Military Polar Orbiting Meteorological Satellite Program. National Reconnaissance Office History Program. Originally classified 1985 (declassified 2000). National Reconnaissance Office, Washington, D.C.. Hawkins, J.D., and P.G. Black. 1982. Seasat scatterometer detection of gale force winds near tropical cyclones. J. Geophys. Res. 88:1674-1682. Kalnay, E., M. Kanamitsu, R. Kistler, W. Collins, D. Deaven, L. Gandin, M. Iredell, S. Saha, G. White, J. Woollen, Y. Zhu, M. Chelliah, W. Ebisuzaki, W. Higgins, J. Janowiak, K.C. Mo, C. Ropelewski, J. Wang, A. Leetmaa, R. Reynolds, R. Jenne, and D. Joseph. 1996. The NCEP/NCAR 40-year reanalysis project. Bull. Am. Meteorol. Soc. 77(3):437-431. Kalnay, E., S.J. Lord, and R.D. McPherson. 1998. Maturity of operational numerical weather prediction: Medium range. Bull. Am. Meteorol. Soc. 79(12):2753-2769. Krishnamurti, T.N., C.M. Kishtawal, Z. Zhang, T. LaRow, D. Bachiochi, E. Williford, S. Gadgil, and S.Surendran. 2000. Multimodel ensemble forecasts for weather and seasonal climate. Journal of Climate 13(23):4196-4216. Kummerow, C., et al. 2000. The status of the tropical rainfall measuring mission (TRMM) after two years in orbit. J. Appl. Meteorol. 39(12):1965-1982. Liu, W.T., H. Hu, and S. Yueh. 2000. Interplay between wind and rain observed in Hurricane Floyd. EOS Trans. 81:253-257. Moore, R.K., and A.D. Fung. 1979. Radar determination of winds at sea. Proc. IEEE 67:1504-1521. Naderi, F.M., M.H. Freilich, and D.G. Long. 1991. Spaceborne radar measurement of wind velocity over the ocean—An overview of the NSCAT scatterometer system. Proc. IEEE 79:850-866. NASA (National Aeronautics and Space Administration). 1985. Scatterometer Research in Oceanography and Meteorology Announcement of Opportunity. A.O. OSSA-1-85, January 31, 1985. NASA, Washington, D.C. NRC (National Research Council). 1995. Earth Observations from Space—History, Promise, and Reality. National Academy Press, Washington, D.C. NRC. 2003. The Sun to the Earth—and Beyond: A Decadal Research Strategy in Solar and Space Physics. National Academies Press, Washington, D.C., in press.

OCR for page 101
O’Brien, J.J., and members of NASA Satellite Surface Stress Working Group. 1982. Scientific Opportunities Using Satellite Surface Stress Measurements Over the Ocean—Report of the Satellite Surface Stress Working Group. Nova University/N.Y.I.T. Press, Fort Lauderdale, Fla. Offiler, D. 1994. The calibration of ERS-1 satellite scatterometer winds. J. Atmos. Ocean. Technol. 11:1002-1017. Orville, R.E. 1987. Meteorological applications of lightning data. Rev. Geophys. 25:411-414. Pierson, W.J. 1983. Highlights of the Seasat-SASS Program: A review. Pp. 69-86 in Satellite Microwave Remote Sensing (T.D. Allen, ed.). Halstead Press, New York. Simpson, R.H., R.A. Anthes, M. Garstang, and J.M. Simpson, eds. 2003. Hurricane! Coping with Disaster: Progress and Challenges Since Galveston, 1900. American Geophysical Union, Washington, D.C.. Spencer, M.W., C. Wu, and D.G. Long. 1997. Tradeoffs in the design of a spaceborne scanning pencil beam scatterometer: Application to SeaWinds. IEEE Trans. Geosci. Rem. Sens. 35:115-126. Stoffelen, A. 1998. Toward the true near-surface wind speed: Error modeling and calibration using triple collocation. J. Geophys. Res. 103:7755-7766. Stoffelen, A., and D.L.T. Anderson. 1995. The ECMWF Contribution to the Characterization, Interpretation, Calibration and Validation of ERS-1 Backscatter Measurements and Winds, and Their Use in Numerical Weather Prediction Models. ESA Contractor Report. European Centre for Medium-Range Weather Forecasts, Reading, U.K. Stoffelen, A., and D. Anderson. 1997. Ambiguity removal and assimilation of scatterometer data. Q. J. R. Meteorol. Soc. 123:491-518. Stoffelen, A., and G.J. Cats. 1991. The impact of Seasat-A scatterometer data on high-resolution analysis and forecasts: The development of the QE-II storm. Mon. Wea. Rev. 119:2794-2802. Taverna, M.A. 2002. U.S. seeks to regain edge in climate issue. Aviation Week 157(2):62-63. Turman, B.N. 1978. Analysis of lightning data from the DMSP satellite. J. Geophys. Res. 83:5019. USGS (U.S. Geological Survey). 1997. Volcanic Ash—Danger to Aircraft in the North Pacific. USGS Fact Sheet 030-97. USGS, Reston, Va. Yueh, S.H., W.J. Wilson, and S. Dinardo. 2002. Polarimetric radar remote sensing of ocean surface wind. IEEE Trans. Geosci. Rem. Sens. 40:793-800.