building desing and windows

Open Data: The Global Effort for Open Access to Satellite Data

Mariel Borowitz’s new book, Open Space: The Global Effort for

Open Access to Environmental Satellite Data traces the history of environmental satellite data sharing policies, offering a model of data-sharing policy development, case studies and practical recommendations for increasing global data sharing. Below, she writes about why some countries have adopted an open data policy, while others have not. 

Satellites view the Earth as a whole—collecting data without regard to political boundaries. Given their unique perspective, perhaps it isn’t surprising that many governments make this data freely available to anyone in the world. In fact, early satellite programs were some of the first to use the term “open data” to describe their data policies.

But not all governments make data from their unclassified government satellites freely available. In fact, as 2016, data from less than half of the 458 satellites launched between 1957 and the beginning of 2016 was made available on a free and open basis.

Open Space: The Global Effort for Open Access to Environmental Satellite Data examines the factors that caused some nations to embrace open data sharing while others continue to restrict access to their environmental satellite data.

When examining the history of the agencies that have embraced open data sharing—like the U.S. National Aeronautics and Space Administration (NASA), the United States Geological Survey (USGS), the European Space Agency (ESA), and others—a pattern emerges. Many began their satellite programs by sharing data freely. In many cases, nations’ first satellites were focused on weather monitoring, a field that had a long history of international collaboration. Many agencies wanted to get the data to as many researchers as possible, to help better understand how this new technology could be used.

Politicians saw strategic benefits as well—in 1961, one Congressman called data sharing an “excellent opportunity for the United States to win many friends in many lands” and thought it could be one of “the greatest peaceful weapons against communism.” [1]

It quickly became clear that satellite data was valuable—but it also became clear that building and operating satellites was expensive. Some began to suggest that the government should offset costs by charge a fee for access to data, or that it should promote commercial satellite activities, removing the cost of developing satellites from government responsibility altogether.

These revenue-generating possibilities were tempting to legislators and politicians trying to reduce pressure on the budget. Space and meteorological agencies around the world were obliged to give it a try. They tried different models: government data sales, licensing private entities to sell data, or fully privatizing whole satellite systems, and they experimented with these arrangements for different periods of time, but for the vast majority of satellite data, the experience was the same: satellite data sales did not result in significant revenues, but it did reduce data use by scientists, researchers, commercial firms, and even the government itself.

The U.S. Landsat program illustrates this point well. When the satellite program was privatized by an act of Congress in 1984, the company given the right to license and sell Landsat data set prices at thousands of dollars per scene, pricing many government and research users out of the system. When the Landsat program was returned to government control almost a decade later—also by an act of Congress—the government continued to charge fees in an effort to recover costs. Although data costs were reduced by a factor of ten, prices were still too high for many users—including the U.S. Department of Agriculture (USDA). In 2006, USDA stopped its purchases of Landsat imagery, noting that it was a “price-sensitive purchaser of satellite imagery” and concluding that Landsat was “not the best value for USDA.” USDA then became the largest commercial purchaser of India’s lower-cost Resourcesat-1 data in 2006 and 2007.[2]

Experiences like these led the U.S. Geological Survey to abandon its cost-recovery efforts and adopt an open data sharing policy in 2007. The results were dramatic. Before adopting an open data policy, Landsat’s greatest annual distribution was in 2001, with 25,000 scenes sold at about $600 each. By 2012, USGS was distributing 25,000 scenes a month.[3] (Some portion of this total was accessed by USDA, which returned to use of Landsat data after it was made freely available.)

This pattern explains why many agencies adopted open data policies, but why are some agencies still restricting access to data now?

In many cases today, it is newer, smaller space agencies that don’t share their data. Out of the 35 nations that have Earth observation programs, 25 of them have owned five or fewer satellites. In many cases, these nations are focused on training new engineers and establishing new indigenous capabilities. Collecting the most useful types of data and ensuring data is widely used simply are not the primary goals of the mission.

This has important implications for those who wish to increase data sharing. Demonstrating that there is demand for data – even from small technology demonstration satellites – will help to engage these agencies. Technical or logistical assistance in making data available can also play an important role. Nations that already share their own satellite data may consider arrangements to host foreign data as well, so that nations with new programs can make their data available without having to devote resources to developing a data portal of their own.

Economic motives for restricting access to data have not disappeared altogether. In fact, there has been a recent wave of innovative new remote sensing companies—like Planet, Spire, and GeoOptics—that plan to sell satellite data. Some government leaders—particularly those in Congress—are enthusiastic about the potential for government cost savings, and have encouraged agencies like NOAA and NASA to investigate purchasing data from these companies, rather than developing similar satellites on their own.

Past experience shows that these arrangements will be most successful if agencies are able to find ways to purchase data while preserving the ability to share that data to the greatest extent possible—by allowing data to be made openly available after a set period of time, for example. In the end, innovations in policy design may be as important as technological innovations in determining the long-term success of government engagement with the “new space” sector.

While Open Space focuses on data collected from space, the model that is developed and the lessons that are learned in the space sector are applicable to data collected or produced on the ground—environmental and otherwise. Significant progress has been made across sectors in the effort to make government data openly available to all users, resulting in benefits for science, public policy, and private enterprise—but there is still a significant amount of work yet to do.

 

[1] United States House of Representatives Committee on Science and Astronautics, “National Meteorological Satellite Program” (Washington, DC, 1961).

[2] Robert Tetrault, “Access and Availability of Resourcesat-1 AWIFS Data for Agriculture,” in USDA FAS Agriculture Applications Seminar 2006, ed. USDA Satellite Imagery Archive (2006).

[3] Michael A Wulder et al., “Opening the Archive: How Free Data Has Enabled the Science and Monitoring Promise of Landsat,” Remote Sensing of Environment 122 (2012).