--Darrel Williams (darrel@ltpmail.gsfc.nasa.gov),
Landsat Project Scientist, NASA Goddard Space Flight Center
--Samuel Goward and Jeffrey Masek, Landsat Science
Team Office, University of Maryland
The Landsat-7 Science Team held their semi-annual team meeting April 14-16 at the NASA Goddard Space Flight Center (NASA GSFC). Participants from the 14 research teams, as well as NASA, NOAA, and USGS attended the meeting, which was co-chaired by Landsat-7 Science Team Leader, Samuel Goward, Landsat Project Scientist, Darrel Williams.
The first morning was dedicated to briefings on the status of the Landsat platform and ETM+ instrument by the Deputy Project Manager, Ken Dolan (NASA GSFC), and on the Landsat Ground System by Deputy Project Scientist, Jim Irons (NASA GSFC). The instrument briefing concentrated on the status of the ETM+ power supplies, which shut down anomalously during thermal-vacuum testing, forcing an indefinite delay in the launch of Landsat-7. It was reported that the power supplies were being reworked at Raytheon's Santa Barbara Remote Sensing (SBRS) facilities, and will be reintegrated with the instrument and spacecraft during the summer; both the instrument and spacecraft remained in residence at the Lockheed Martin Missiles and Space (LMMS) facilities in Valley Forge, Pennsylvania. Work at SBRS and GSFC had confirmed that the power supplies were overheating, and this overheating was traced primarily to "new" diodes that were used within the power supplies. These diodes met or exceeded all specifications. They were considered to be an improvement over diodes used previously, and they were chosen because the older style diodes are no longer in production. There was no way to predict the problem that we experienced with the diodes.
It has been decided that the revised, formal launch date will not be established until after the power supplies have successfully gone through thermal and vacuum testing at the instrument level, which should occur by mid-August. We hope that launch could still occur in March/April time frame in 1999. Dolan also reported that preparations of the Landsat-7 platform at the LMMS Valley Forge facility are proceeding normally. [NOTE: As this report was being submitted in the mid-June time frame, the rework of the power supplies at SBRS had been completed successfully. The power supplies had been shipped back to LMMS, re-integrated with the instrument, and baseline testing under ambient conditions was completed successfully. To obtain up-to-date status on the Landsat-7 Project, please access the Landsat homepage at: http://landsat.gsfc.nasa.gov/project/projects.htm.]
The Landsat-7 ground system consists of ground receiving stations, the Landsat Processing System (LPS) for producing Level 0R products, the Landsat Product Generation System (LPGS) for producing Level 1G products, and the EOSDIS Core System (ECS), which performs data archive and distribution functions. Development of both the LPS and LPGS are proceeding normally, although continued delays of ECS are jeopardizing data distribution. To mitigate these risks, NASA and the EROS Data Center (EDC) are developing a DAAC Emergency System (DES), capable of temporarily distributing 100 Landsat-7 scenes per day in the absence of the ECS.
James Ellickson (NOAA) detailed progress on signing up international ground stations to receive Landsat-7 data. NOAA has introduced a more flexible licensing policy for international ground stations, with the licensing fee based on the number of scenes acquired by year, rather than a single flat rate. This policy makes Landsat-7 a more attractive investment, and to date some 10 to 12 ground stations have indicated that they will participate. Ellickson noted that the level of participation by stations in the Far East will probably be lower than for Landsat-5, largely due to the current Asian economic crisis.
Martha Maiden (NASA GSFC) presented the results of the NASA Earth Sciences Information Partners (ESIP) awards, focusing on serving data products of interest to the terrestrial science community. Of particular note are projects headed by David Skole (Michigan State University), John Townshend (University of Maryland), and Berrien Moore (University of New Hampshire).
NASA's current planning activities for the next generation of EOS measurements were reviewed by Anthony Janetos (NASA HQ). He pointed out that there are currently no concrete plans to continue Landsat-class measurements beyond the Landsat-7, which has a nominal ending date of ~2004. During the next several months, NASA will be soliciting measurement priorities from the science community, and crafting focused missions for the first decade of the next century using this input. Janetos indicated that contributions from the Landsat Science Team would be particularly important in assessing the proper role of Landsat-class measurements for NASA's Earth Science Enterprise (ESE).
The second day extended this theme, exploring various directions for future land observing missions. Stephen Ungar (NASA GSFC) reviewed progress on the upcoming New Millennium Program (NMP) Earth Observer 1 (EO-1) mission. EO-1 was designed to prototype advanced technologies, including hyperspectral imaging, for future land observation missions. Ungar noted, however, that development of both the Grating Imaging Spectrometer (GIS) and Wedge Imaging Spectrometer (WIS) was behind schedule, and testing indicated that performance of these components would not meet specifications. As a result, development of both the WIS and GIS had been terminated*, leaving only an advanced pushbroom multispectral imager, although other options for replacing the hyperspectral imaging capabilities were being explored. Initial testing of the multispectral imaging module were promising, suggesting much improved signal-to-noise ratios compared to the Landsat-7 ETM+ instrument. *[NOTE: Since the mid-April meeting, a decision was made to incorporate a TRW supplied hyperspectral device on EO-1 similar to what was to have been flown on the Lewis HSI mission. The projected launch date for this revised EO-1 mission is no earlier than December 1999.]
A number of companies have announced plans to launch commercial remote sensing missions during the next decade. Vic Leonard presented an overview of the planned Resource-21 mission, which is targeted toward agricultural users. The current specification calls for several platforms with high-resolution (10-20 m) multispectral sensors, with launch in the 2001 timeframe. Leonard noted that farmers and agribusiness require high-level information products (e.g. crop type, crop health, drought condition) rather than low-level data, and that rapid dissemination of these derived products was key to the commercial success of Resource-21.
Anne Kahle (NASA JPL) and Ray Taylor (NASA GSFC) gave an overview of a proposed hyperspectral thermal-IR mission. Kahle noted that hyperspectral measurements in the TIR had never been made from satellite, and that this avenue showed considerable promise for monitoring surface temperature, emissivity, geological composition, and atmospheric properties. Taylor presented an array of advanced technologies for on-board band selection, on-board data processing, and inter-satellite communications. This "database in the sky" concept would leverage advances in high bandwidth optical data transmission to make the satellites themselves active participants in data processing and distribution.
Jeffrey Masek (University of Maryland) concluded the morning presentations by reviewing the Landsat Science Team recommendations for the Landsat-8 mission. Team members had been asked to submit measurement requirements for their scientific studies. The principal recommendation was a desire for higher radiometric precision, from the current 8-bit system to a 12-bit system for Landsat-8, although most team members were satisfied with other aspects of the current Landsat-7 configuration. A number of participants also suggested the addition of specific spectral bands, primarily to mitigate the effects of atmospheric contamination.
Following lunch, Roger Mitchell (Earth Satellite Corporation) presented EarthSat's Orthorectification Project, which was selected as a Phase 1 project under the NASA $50M data buy. Using NIMA ground control points and digital topography, EarthSat proposed to georectify a global collection of Landsat TM images to a precision of 60 meters. These data would be distributed to users through the EROS Data Center upon validation. During Phase 1, slated to end later this year, EarthSat was to generate a georectified Landsat data base for eastern Africa. Mitchell also outlined plans for producing global land cover assessments from these data, although Landsat Science Team members expressed some skepticism regarding the likely accuracy and utility of the land cover products.
The final day of the meeting began with presentations on outreach strategies for the Landsat-7 mission. Carolyn Merry (Ohio State U.), currently on sabbatical with the Landsat Project Science Office at Goddard, discussed approaches for using remotely sensed data in educational curricula. Her presentation included a comprehensive review of sources of support for K-12 and college education programs. Lynn Chandler, representing the Public Affairs Office at NASA GSFC, then introduced the NASA Goddard Landsat-7 public outreach strategy to the Science Team, concentrating on effective ways to communicate science findings to the local and national media.
The meeting wrapped up with presentations by Chris Justice (U. Virginia), Jeff Privette (NASA GSFC), Eric Vermote (U. Maryland), and Kurt Thome (U. Arizona) on EOS/Landsat-7 synergy. Jeff Privette presented an overview the EOS Core Validation sites, which will be monitored to provide validation of EOS science products. Landsat-7 data will be an important component for many of the validation projects. The Science Team took an action to review the site list, and suggest any additional sites that might be useful for calibration/validation of Landsat-7 data. Eric Vermote and Kurt Thome reviewed approaches for atmospheric correction of Landsat-7 and archival Landsat data. For the Landsat-7 era Thome recommended an approach based on Vermote's 6S model, with appropriate atmospheric inputs from the MODIS sensor, while for archival data Thome recommended a simple dark-object approach. Thome took the action to prepare software for the Science Team to use for both these cases.
The meeting also included presentations from all team members on their research activities during the last six months. Readers interested in specific science team investigations should consult the article on the October 1997 Landsat Team meeting in the February/March 1998 issue of the Earth Observer. The next team meeting will be held September 29-October 1, 1998 at the Patuxent Wildlife Visitor's Center in Maryland. Interested parties should contact Jeffrey Masek (jmasek@geog.umd.edu) for further information.