UMD Researchers Create First System to Track Near-Real Time Changes to Global Land Cover

Scientists worldwide can now receive updates about the how the earth’s surface is changing more quickly than ever before, thanks to a new satellite-based monitoring system created and operated by University of Maryland researchers.

The system, known as the OPERA Land Surface Disturbance Alert (DIST-ALERT), was described earlier this month in Nature Communications. It is the first to track lands that are being changed by various causes, human activity, weather events and fires in a truly global manner. 

Previously, only particular areas were closely monitored for changes to their land cover—such near the equator, where rainforests are concentrated—and only for specific types of changes, like forest loss or fire.

In addition to broader coverage, DIST-ALERT also produces this information more quickly than other monitoring systems. Most utilize imagery from one satellite system, but DIST-ALERT leverages data from both Landsat 8/9 and Sentinel-2A/B/C, totaling five satellites. Since it takes several days for a single satellite to revisit an area once it has passed over, DIST-ALERT’s use of five satellites can shorten these intervals to as little as one day, giving scientists access to updated information far more quickly.

“These satellites are affected by cloud cover, and if we don’t see the ground, we don’t know whether disturbances have happened,” said Amy Pickens, an assistant research professor in the University of Maryland’s Department of Geographical Sciences (GEOG) who leads DIST-ALERT. “So with more observations we get more opportunities to get a cloud-free observation, and each observation can matter so much, particularly in cloudy areas,” 

Pickens is also the first author of the new Nature Communications paper, which also used DIST-ALERT data to compare how much land changed around the world over four years.

Comparing land in 2023 to the same area during the same month in the previous three years, the team found, for instance, that the largest share of land change resulted from natural climatic variations, including both unusually hot or dry conditions in some areas and extra rainfall elsewhere.

The second-largest driver of land surface changes was shifts in human-led management—mostly because of agricultural decisions like planting a different crop or harvesting at different times.

Accounting for all the ways that humans might convert land surfaces—plant new farms, cut down forests, add new buildings, etc.—the researchers found that the sum of human-led land conversion in 2023 totaled around 28.6 million hectares, or greater than the area of Ecuador or Colorado. Half of that land was long-lived forests, shrublands, grasslands or wetlands.

“The conversion of natural lands to land use is a significant contributor to climate warming,” said GEOG Professor Matthew Hansen, one of the paper’s coauthors. “The result is increasing anomalous weather at the global scale, often seen in our data as large-scale vegetation loss due to fire, drought, floods, and windthrow events.”

DIST-ALERT data can be accessed by the public via the Land Processes Distributed Active Archive Center. Some organizations are adding DIST-ALERT data to existing systems that send out electronic notifications about land changes that are of interest to their audiences. Global Forest Watch, for instance, used to only provide users with alerts about changes to tropical forests, but now provides updates for users concerning changes to all kinds of forests.

“There are a lot of aspirations to protect land and a lot of need to manage land, to make sure it is being used in the most valuable ways. But you can’t do that unless you have data about what’s going on,” said Pickens. “DIST-ALERT monitoring in such a low latency mode enables that data to be more actionable, not just for policymaking, but for actually responding to individual events.”

Other paper coauthors include Department of Geographical Sciences Assistant Research Professors Anna Komarova, Zhen Song and André Lima; Associate Research Professor Alexandra Tyukavina; Faculty Specialists Andrew Poulson, Antoine Baggett, Theodore Kerr and Carolina Ortiz Dominguez; and Research Graduate Assistant Aleksandra Mikus.