Fifty Years of Exploration and Innovation: How Landsat Launched the Remote Sensing Era
An experimental satellite called Earth Resources Technology Satellite (ERTS) launched into orbit in 1972 carrying the hopes of geologists, cartographers, agronomists, and governments around the world.
This single satellite, the first explicitly designed to observe the Earth’s surface, also launched a global science and technology revolution of remote sensing that continues today.
Hundreds of miles above Earth, ERTS, later renamed as the first Landsat, peered down upon land on which few, if any, people had set foot, such as vast ice sheets and oceans, and temperate lands that had been occupied for thousands of years. The imagery revealed much about both kinds of places.
Now, the ninth Landsat has launched, and scientists and researchers use Landsat’s unparalleled archive of 10 million scenes to explore changes over time in ecosystems around the world, from forests to coasts to urban areas.
The Launch of Land Observation from Space
The Soviet Union launched the first satellite into space in 1957, Sputnik 1, which transmitted radio signals. The United States followed in 1958 with its first satellite, Explorer 1, which carried a cosmic ray detector. Many others followed with various intents, especially reconnaissance and meteorology.
U.S. Geological Survey Director William Pecora saw satellites as a potential solution to the challenge of viewing the Earth’s surface from above, in a manner more thorough and inexpensive than aerial photography. Upon Pecora’s advocacy, in 1966, Interior Secretary Stewart Udall announced Project EROS (Earth Resources Observation Satellites).
“The time is now right and urgent to apply space technology towards the solution of many pressing natural resources problems being compounded by population and industrial growth,” Udall said in the 1966 Department of Interior announcement.
The USGS and NASA worked together to develop a program that would launch Landsat 1, then capture and process the data sent back to Earth and distribute the imagery. On July 23, 1972, ERTS carried 570 miles into space the first Earth observation satellite sensor, the Multispectral Scanner (MSS), which was designed by “Mother of Landsat” Virginia Norwood. MSS images had 60-meter spatial resolution and four spectral bands: visible green, visible red, and two near-infrared.
ERTS’ initial data were received by ground stations in California and Alaska, which sent tapes to the NASA Goddard Space Flight Center in Maryland. Goddard converted the digital data into film negatives, which were flown to the USGS Earth Resources Observation and Science (EROS) Center in South Dakota to be processed into photographs and distributed upon request to users around the world. ERTS launched on a Sunday; negatives of the first images arrived at EROS late the following Tuesday.
After 50 years, the Landsat Program reveals more than just what the Earth’s surface looks like with each satellite overpass. Today, with decades of continuous and consistently calibrated data offered for free, anyone with an internet connection can discover changes over time in ecosystems from forests to coasts, urban areas to glaciers—and even predict how they might keep changing.
Early Experiment Results
Landsat started out as an experiment. Before ERTS soared into orbit and began sending back its views of Earth, scientists didn’t know quite what to expect from its resolution. The satellite’s first images, including the Dallas area, encouraged them—they could make out details like roads, an airport, and farm fields.
“Some of those experimenters are like kids with new toys. They’re so excited about the resolution, saying, ‘Look at this,’ and, ‘Hey, look at that,’” Jack Hays of the Goddard Space Flight Center said in a Lompoc (California) Record article several days after the launch of ERTS from nearby Vandenberg Air Force Base.
Soon, scientists and government agencies were using Landsat for measuring the global food supply, locating mineral deposits, estimating timber volume, and monitoring flooding, snowmelt runoff, and glacier footprints around the world. An island off the coast of Labrador bears Landsat’s name because satellite imagery helped researchers discover it. A diving expedition between NASA and Jacques Cousteau tested Landsat’s ability to measure ocean water depth and gave birth to satellite-derived bathymetry.
As Landsat’s archive grew, it revealed changes over time: urban growth, deforestation, shrinking ice caps and lakes, fluctuating rivers and reservoirs, burned area and volcanic flow recovery, and any other circumstances people could think of to study. Landsat began to deliver answers about the Earth’s resources, their influence on humanity and ecosystems, and humanity’s influence on them.
Path to Data Use
The path from launching a satellite to seeing its data resound throughout the world wasn’t always smooth. In 1972, for example, computers for processing and analyzing digital data and mapping with geographic information systems (GIS) software tools were rare. So were the people who knew how to use that technology. However, anyone could order photographic prints or transparencies of Landsat scenes from the large photographic laboratory at EROS, then use a magnifying glass to scrutinize all four corners and everything in between.
Tom Loveland was a longtime USGS scientist at EROS. During an interview before his recent death, Loveland recalled that his first job with the South Dakota State Planning Bureau in 1976 involved bridging the gap between photographic and digital data. Before he started, South Dakota had made one of the first State land cover maps in the world with Landsat photographs. But officials wanted more detail. So, Loveland and Jeff Eidenshink, a longtime EROS employee and past EROS deputy director, remade the map digitally using Landsat MSS data, at that time stored on computer-compatible tapes.
Training was clearly needed to help the government agencies and scientists responsible for managing resources benefit from applying Landsat data to their work. So, EROS, home of the archive and also of scientists developing applications for data, invited people from U.S. agencies and other countries to attend monthlong training sessions at the facility through the late 1970s and early 1980s and take their new skills home. “It really had sort of a foreign policy, humanitarian aid flavor to it that brought people in from all over,” Loveland said. EROS helped other countries develop Landsat applications, as well.
Later, universities with remote sensing programs took over the role of training. Computers, and working with digital data, became more common, and people found plenty of uses for satellite data. For Landsat users, however, there were still some challenges to making maximum use of the data. Landsat 4 and Landsat 5 both had Thematic Mapper sensors in addition to MSS. The Thematic Mapper had seven spectral bands, which added to the size of data and made the data less compatible with earlier image processing software.
In addition, the 1980s efforts to privatize Landsat, turning it over to commercial vendor Earth Observation Satellite Company, resulted in the scene price rising to $4,400, up from $650. Projects using more than one or a few scenes were unaffordable for most, and data from the French SPOT 1 satellite had become an alternative option during that decade as well. The Landsat program would have perished if not for Congressional and global objections. Landsat returned to government operations with a cheaper image price and a new satellite on the horizon with plans for Landsat 7, which launched in 1999.
Data use picked up again. “Suddenly, we were in a time when the amount of imagery around the world was growing more rapidly, and the access to data improved significantly as well,” Loveland said.
Revolutionary Decision
Then in 2008, the USGS made a move that proved revolutionary to Landsat data usage. The new Landsat Data Distribution Policy opened up the entire Landsat archive to the world for free distribution. University researchers, geographers in developing countries, anyone yearning to look at decades of change in tens or even hundreds of Landsat scenes could download the data without worrying about a price tag. The examination of regional and even global patterns of change became possible. A USGS study on the economic benefit of Landsat imagery determined that it yields $2.06 billion in annual benefits to United States users alone, and $3.45 billion worldwide.
“That started the era of people using the data they needed, not the data they could afford. Single scenes aren’t enough to capture the diversity on the landscape,” Loveland said. “To me, from 2008 forward is when Landsat really started achieving the expectations, the potential, that was envisioned by Pecora and Udall in the 1960s.”
Most recently, cloud computing has been revolutionizing remote sensing by allowing people to access and analyze massive amounts of data in the cloud rather than their local computer. “You no longer have to bring all the data into a collection of your own and deal with it. Now you just send your question to the cloud, and the answer is sent back from the collection in the cloud,” said principal systems engineer Jon Christopherson, a 25-year contractor at EROS.
Based on decades of historical patterns, data users can model ecosystem change decades into the future. The applications for civil and commercial satellite data are now almost endless. “A lot of studies … have really proven that a community-based cloud approach is just brilliant. I think it’s not an overstatement to say that when Google made all free Landsat data available to the world on a free computing platform, we accelerated the pace of global learning of environmental conditions,” Loveland said.
“Landsat is just one data stream with many more joining it,” Christopherson added.
Key Themes of Landsat Use
Loveland pointed to three themes where Landsat has especially helped achieve comprehensive monitoring. The first is agriculture, in which the world has benefited from continual monitoring of the global food supply and early warning alerts for famine in vulnerable areas. The second is forestry, which is monitored through early warning systems for losses and gains, disturbances, and deforestation as well as longer-term trends.
“Now, I think the hot area for applications has to deal with water,” Loveland said. Efforts have included a global inventory of surface water, the monitoring of water quality, and studies of energy balance in the hydrologic cycle to show how much water is consumed in irrigation (through evapotranspiration), which relies on Landsat’s thermal data, among the first available.
Gabriel Senay, a USGS research physical scientist who models evapotranspiration (ET), calls it “the biggest component of the water budget. Really, ET maps tell you how much water and where and when it’s being used.” The implications stretch from water management decisions to water rights determinations and will become even more important in a changing climate, especially in drier regions of the world.
Landsat’s Continual Reliability
The Landsat Program was intended to produce a continuous record of data, preferably with two optical polar-orbiting satellites placed opposite each other. Doing this for 50 years was no small feat, especially when you consider it has relied on most of the eight satellites to transmit data well beyond their life expectancies.
One, Landsat 5, delivered data for a quarter-century longer than its three-year design, earning it the Guinness World Record of “Longest Operating Earth Observation Satellite.” For more than five years, after Landsat 4 ended its transmissions in 1993 and before Landsat 7 started in 1999, Landsat 5 worked alone.
Having 50 years of Landsat data is an achievement in satellite function and data archiving. But trusting that data enough to reliably compare scenes of the same location 20 or 30 years apart is an achievement in calibration. The EROS Calibration and Validation Center of Excellence (ECCOE) characterizes and calibrates the radiometric and geometric performance of Landsat satellite instruments and makes adjustments to present and past data. Missions are cross-calibrated to maintain the same quality reference. Civil and commercial satellite operators, meanwhile, trust Landsat’s standards of excellence to cross-calibrate with their own data.
“We’re among the best for data quality, dependability, reliability,” Christopherson said.
Landsat, meanwhile, relies on ground stations that downlink real-time and recorded science data and transmit the data to the processing location. The primary ground stations for Landsat 8 and Landsat 9 are at EROS and in Alaska, Norway, Germany, and Australia. Historically, additional international cooperators have received Landsat data directly for themselves in locations such as Russia, Pakistan, Kenya, and Brazil. Some still do, including China, South Africa, and Argentina.
Looking Beyond 50 Years
Landsat Next is the proposed successor to Landsat 9, with a potential launch later this decade. USGS physical scientist Zhuoting Wu was involved in collecting stakeholders’ land imaging needs to help define Landsat Next’s science objectives and requirements.
“The user community has expressed great interest in maintaining Landsat continuity, supporting synergy with the Copernicus Sentinel-2 mission, and enabling new emerging applications that are critical to tackle the challenges in today’s global environment,” Wu said.
Improved spatial resolution, an increased temporal revisit, and an increase to 26 spectral bands would provide enhanced detail for applications. In agriculture, this would translate to a better understanding of the effects of cover crop and rotational crop planting. In freshwater, this would further define algal blooms and other water quality issues. Measurements would reveal more about snowpack and polar ice changes. New emissivity measurements would increase the accuracy of surface temperatures.
“Landsat is truly a system of systems. It has very unique capabilities addressing user needs. It also nicely complements some of the other available missions and systems out there,” said Paul Haugen, USGS chief engineer and acting project manager for Landsat Next.
“Landsat Next is continuing to evolve the Landsat continuity mission, and harmonization and interoperability are really key elements. But we see—at least in the commercial sector—that these missions really depend on Landsat as a reference calibrated measurement to adjust or align their measurements to Landsat,” added Chris Crawford, a USGS research physical scientist who serves as the USGS Landsat project scientist, and the project manager and principal scientist for the EROS-Imaging Spectroscopy Project that informs the USGS’s Sustainable Land Imaging program for development of future Landsat missions.
Climate change affects different parts of the Earth in different ways, from rising seas, extreme droughts, and sweltering cities to intensified storms, dwindling rivers, and worsening wildfires. Satellites shift our perspective and let us look down on one spot, look across a broad area, look back in time, and even look forward. They help us learn from the past and present and give us a chance to change the future.
Between the launches of Landsat 1 in 1972 and Landsat 2 in 1975, then-NASA administrator James C. Fletcher said, “If I had to pick one spacecraft, one Space Age development to save the world, I would pick ERTS and the satellites which I believe will be evolved from it later in this decade.”
Decades later, hundreds of satellites have launched from all over the globe, but the prophecy still stands. Earth observation satellites hold a vital key to understanding our planet and preserving it for the future—and ultimately preserving humankind.