Mapping the country's changing biomass
- By Patrick Marshall
- Feb 23, 2009
There are a lot of reasons to be concerned about how much biomass the country has. For starters, land cover — whether it’s grass or old-growth forest — consumes carbon dioxide, one of the biggest contributors to the greenhouse effect.
Unfortunately, we haven’t had a good idea of how much biomass the country has, and we have even less insight into how much it is changing.
Researchers at the Woods Hole Research Center — an environmental science, education and policy institute with a Cape Cod, Mass., location — are rectifying that situation. Through the National Biomass and Carbon Dataset project, they are using data gathered on space shuttle missions and satellites to produce a map of the country’s biomass.
The project’s baseline is the National Elevation Dataset compiled by the U.S. Geological Survey, which records ground elevations for the continental United States, Alaska and Hawaii. The team at Woods Hole overlaid that information with data gathered during a space shuttle flight in 2000, which used radar to record the topography of the continental United States.
“It was the first radar mission and used two antennae, one inside the shuttle bay and one deployed 60 meters outside the shuttle bay on a boom,” said Josef Kellndorfer, the team’s leader and an associate scientist at Woods Hole. “From that mission, we got an interferometric signal that would allow us to retrieve topographical information.”
Kellndorfer said creating the National Biomass and Carbon Dataset was simply a matter of lining up the other two datasets and subtracting one from the other. “If you take ground elevation, which is available from the [U.S.] Geological Survey in the National Elevation Dataset, and subtract it from the elevation reading of the vegetation from the space shuttle, you get a reading of vegetation height,” he said.
However, to determine biomass from measurements of vegetation height, you need to know the density and type of vegetation in each map zone, Kellndorfer said. So the team turned to LandFire, a project funded by the Forest Service and the Interior Department, and the Forest Service’s Forest Inventory and Analysis project.
The team then used an open-source tool from the R Project for Statistical Computing to merge and model the data from the various sources and used PCI Geomatics to manage and analyze the immense amount of imagery. Finally, researchers relied on a number of applications from ESRI to perform the final integration of the data and deliver it to users via the ArcGIS Server 9.2 map service.
The result is that the team can derive estimates of biomass from satellite readings with about a 10 percent margin of error, Kellndorfer said.
“What’s nice is that at the end of the day, it’s going to be a fairly high-resolution dataset for the entire nation, except Alaska and Hawaii,” Kellndorfer said. “Whereas the Forest Service dataset is a statistical dataset, we’re actually mapping it out. It will be the first spatially explicit assessment of biomass for the country.”
However, the map’s full potential won’t be realized until the team goes through the process again. “That will allow a comparison over time,” Kellndorfer said. “We can hopefully detect regrowth, where carbon is accumulating. We can compare and get an idea of the carbon flux in the northeastern forests.”
Kellndorfer said the Agriculture Department and NASA have provided funding so the team can take a follow-up set of readings for the northeastern part of the country.
Patrick Marshall is a freelance technology writer for GCN.