Landsat 7 Malfunction Hampers Global Remote Sensing Capability

A thirty-year legacy of continuous remote sensing data that scientific, commercial and worldwide government users have come to rely upon, and which is largely responsible for the civilian satellite remote sensing industry, has been choked to a trickle and is threatened to terminate. The Landsat Program that has been the lifeblood of the global remote sensing industry, providing a steady stream of moderate resolution, multispectral imagery since the launch of Landsat 1 in 1972, is in jeopardy as a result of a technical malfunction to Landsat 7, the latest iteration which was launched, and has operated splendidly, since 1999.
According to the US Geological Service (USGS) website (landsat7.usgs.gov), "On May 31, 2003 at approximately 21:45 GMT, unusual artifacts began to appear within image data collected by the Enhanced Thematic Mapper plus (ETM+) instrument on board the Landsat 7 spacecraft. The artifacts are consistent with a failure of the instrument's scan line corrector (SLC), an electromechanical device that compensates for the forward motion of the spacecraft," which further analysis has confirmed to be the problem with Landsat 7.
The USGS web site states that "The spacecraft itself appears to be in no danger. However, it is believed that all data collected by the ETM+ since the initial failure contain the defect. The affected data (acquired since May 31) have been archived and removed from the publicly accessible data ordering systems. Operations, other than those required to support the current investigation, have been limited to routine housekeeping procedures since the data anomaly was discovered. At this time, it is unknown why the SLC ceased functioning."
I contacted Ronald E. Beck, public affairs specialist for the USGS EROS Data Center, to find out just how dire the Landsat 7 situation is. He prefaced his statement to me with a reminder that "while Landsat 7 is currently experiencing difficulties, any assumption of its demise would be premature and inaccurate. USGS and NASA engineers, along with the instrument's manufacturer, Raytheon Santa Barbara Remote Sensing, are working on the problem and have yet to determine the prospects for a full recovery. It has been shown, however, that the instrument can operate without the component, but with the loss of some geographic coverage within a scene. Testing is being conducted to verify the accuracy and utility of the data carrying the artifacts.
The USGS and the National Aeronautics and Space Administration (NASA) jointly develop the Landsat 7 Project to collect medium resolution satellite remote sensing data of the earth. Operating for thirty years, the Landsat Project is the longest running civilian remote sensing program with millions of Earth images having been collected for a broad range of applications—from agriculture to global change to resource management—with thousands of users world wide. NASA developed and launched the satellites, and the USGS manages the program from the Earth Resources Observation Systems (EROS) Data Center in Sioux Falls, South Dakota. The EROS Data Center is responsible for operating the satellites and managing all the ground segment activities such as receiving, processing, archiving, product development and distribution of the data and information products. One of the greatest benefits of the Landsat Project is the acquisition of consistently calibrated Earth imagery on a frequent and repeated basis which is necessary to qualify and quantify changes to Earth's surface and atmosphere.
"The occurrence of an anomaly now is particularly unfortunate as we are in the northern hemisphere's agricultural growing season—a monitoring opportunity for which the Landsat mission observations are particularly suited," Beck said. "Secondly, coupled with the limited use of Landsat 5, eight-day coverage cycles were possible using both satellites. This has allowed for higher temporal resolution observations for time-critical applications such as agriculture. Also, monitoring of natural hazards such as potential grassland and forest fires or weather-related disasters benefitted from the more frequent imaging opportunities."
He added, "A total loss of Landsat 7 would eliminate our ability to collect Landsat-quality medium resolution data over all the earth's land masses. Landsat 5 continues to be used to collect data over the U.S. and China and, in fact, a number of International Cooperators are preparing to resume acquiring data over their areas of interest. However, as the Landsat 5 mission does not have an image data recorder onboard, coverage will be limited to areas in direct line of sight with an operational ground station."
On a positive note, Beck discussed the Landsat 7 Long Term Acquisition Plan (see article in this issue, pages #10-20), which was designed to ensure global land coverage, capture seasonal change, and avoid acquiring cloud-contaminated imagery, selecting the 'best' of up to 250 scenes per day for the U.S. archive. Consequently, a base of high-quality data has been collected; data that supports studies in seasonal change and provides significant coverage over most of the earth's land masses. Since the start of the Landsat 7 mission, more than 300,000 scenes for the U.S. archive have been acquired and made available to the user community.
In discussing the current state of operations, Beck said the Enhanced Thematic Mapper Plus (ETM+) instrument is turned on for about forty minutes a day to maintain a nominal thermal balance within the instrument's electronics and optics. While that is being done, test data are being collected. The test data are being analyzed by a group of scientists and engineers tasked with determining the quality and accuracy of the data. The users studying the data will submit a report to the Landsat Anomaly Team and that report will influence decisions on the future of the data collection. Until managers are confident that the accuracy and utility of the data are acceptable, and when appropriate processing systems are in place, data will not be available to the broad user community. It is an important goal of the Landsat managers to restore data access as soon as possible.
Beck suggests that the wider remote sensing user community should check the Landsat website for examples and appropriate descriptions of the data with the anomaly problems as well as additional updates and current news on the status of Landsat 7. At this time, the center of the images appears to be acceptable, though tests on the radiometric and spatial accuracy are being done to confirm this. The left and right thirds of the images illustrate the problem. Both sides of the image contain wedges of no data. This represents areas where, due to the malfunctioning scan line corrector (SLC), the ETM+ is unable to 'see' portions of the land.
"The Landsat system has an impressive 30-year record of continuous data collection. No other system offers an equivalent record of moderate resolution observations of the land masses of the earth," Beck said. "The problem with the scan line corrector endangers that thirty-year series of observations. There is no simple replacement for Landsat 7. Other commercial and international systems, while providing useful observations, do not offer the consistent, calibrated, global observations available from the Landsat Program."
Weather and military programs routinely build in redundant spacecraft which are available should the primary system fail. Landsat has never had such a redundancy. Further, the Landsat Data Continuity Mission, designed to provide Landsat type data to the global science community, is far from operational. At the earliest, the Landsat Data Continuity Mission is not expected to be providing data before early 2007.
"The need for continuous, calibrated medium-resolution observations of the globe is not a capricious academic notion," Beck said. "The global science community has come to rely on the Landsat Program for moderate resolution observations available quickly and economically. One of the major lessons to be learned from the current situation is that back-up systems need to be in place to continue providing the continuous observations that have made the Landsat Program a global science success."
The USGS, scientists and other users are putting a brave face on the potential negative impact of this anomaly to Landsat 7 which could potentially become a disaster for the global remote sensing community. Possible panaceas such as post-processing techniques to make the damaged data efficacious, or relying on other satellites such as the French SPOT system or India's satellites in the event of total failure of Landsat 7, or the continued trickle of data flow would leave the US totally dependent upon foreign sources for this type of remote sensing data, and I doubt that would be acceptable to the American remote sensing industry.

Until next time . . . Cheers!

Roland Mangold
Editor and Publisher

Back