Calculating how many acres of Amazon forest are cleared each year is relatively easy thanks to satellite imaging. Determining how much carbon is stored in that forest is another matter.
Unlike forest cover, carbon cannot be seen from satellites in space. Scientists have to physically measure tree trunks in order to calculate how much of the greenhouse gas is stored in their woody limbs and green leaves.
Measuring every tree on the planet is clearly impractical, and this has posed a challenge for those seeking to establish financial incentives to reduce emissions from deforestation, such as the United Nations' Reduce Emissions from Deforestation and Degradation (REDD) program. The carbon released when trees are cut down is estimated to contribute 10 to 15 percent of global greenhouse gas emissions.
For years, scientists and policymakers have questioned whether it is possible to cost-effectively measure forest carbon with high enough precision to make REDD programs viable.
Ecologist Greg Asner and his partners have found a way to do just that.
Combining airborne laser technology with satellite mapping and ground surveys, Asner and a team of scientists from the Carnegie Institution, World Wildlife Fund and Peru's Ministry of the Environment precisely measured the carbon stored and emitted across 11 million acres of tropical forest for about 3 cents per acre.
The findings were published last week in the journal Proceedings of the National Academy of Sciences.
"[The study] shows you can do this on a pretty large scale and at a reasonable cost," said Lydia Olander, a research scientist and director of the Ecosystem Services Program at Duke University's Nicholas Institute for Environmental Policy Solutions. "It's saying this is a viable option."
To calculate the above ground carbon stocks of such a large area (about the size of Switzerland), Asner's lab has developed a process whose key ingredient is light imaging and ranging, or LiDAR. While flying over the forest canopy, an invisible light beam is shot toward the ground 100,000 times per second. Every time it hits something, such as a branch or leaf, a signal is sent back to a computer. The result is a 3-D map of the forest showing the size and shape of individual trees and plants.
"From that, we can calculate with very high precision and accuracy the biomass," said Asner, who is a principal investigator at Carnegie's Department of Global Ecology.
It is not feasible to scan every square inch of forest, so the team only sweeps a sample area of different vegetation types. Team members also verify that LiDAR measurements are accurate by manually measuring trees and plants in select sites. Based on the biomass, or how much plant material is in a site, they can model how much carbon is stored there. These values are then applied across larger areas with similar vegetation types to produce the final carbon map.
"The outcome is we are able to cover large areas with statistical certainty that's been not possible in the past," said Asner, who has researched the effects of human- and nature-driven land-use change in the tropics for 13 years.
The map revealed this lowland Amazon forest in southern Peru contains 395 million metric tons of carbon, and emissions from felled trees reached about 630,000 metric tons per year.
The method distinguished between emissions from deforestation -- complete clearing for uses such as cattle ranches, mining and agriculture -- versus degradation, which can be caused by selective logging or low-intensity fires. The study found carbon emissions from degradation contributed about a third of total emissions in the region.
"It's a huge contributor," Asner said. "If you don't deal with it you are missing a lot of carbon."
There was also an intriguing difference in the amount of carbon stored in older, versus geologically younger ground. Layers of earth created around the time dinosaurs were dying off appear to be less fertile and support less vegetation growth and carbon storage than soil just a few thousand years old.
The results showed older, more diverse forest stored more than three times as much carbon than secondary growth forest. But natural regrowth on abandoned sites did offset about 18 percent of carbon emissions from deforestation or degradation over 11 years.
"It's a wonderful demonstration of the ability to monitor carbon stocks, which is required to implement policies such as REDD," said Ruth DeFries, an environmental geographer at Columbia University.
Having a precise estimate is key to building confidence in REDD programs, agreed Brian Murray, director for Economic Analysis at the Nicholas Institute for Environmental Policy Solutions.
The study found the carbon stock for this region is about a third less than estimated by the Intergovernmental Panel on Climate Change (IPCC). The IPCC's tier-1 estimates are based on one average number assigned to different biomes and as such, the monetary value of the carbon would be discounted if based on the rough default, Murray said.
It was not surprising to many scientists that Asner's more precise calculation was different from the IPCC estimate because there is much greater variability within the region. Even though the amount of stored carbon may end up being lower than initially estimated, having a more accurate estimate means participants in REDD programs would receive full price for carbon.
"The barrier to REDD is more along the political dimension than the technical dimension," Murray said.
Whether or not REDD ever gets off the ground, Asner said the carbon mapping benefits Peru and other countries.
Peru will use this technical information to achieve its goal to protect 135 million acres of forest, as well as guide other land-use decisions, said Doris Rueda, director of land management for the Peruvian Ministry of the Environment.
"This technical information helps us better manage our land," Rueda said. "It also generates better information so we can channel public and private investment to different strategies for mitigation and adaptation to climate change."
The Carnegie Institution strives to work with governments and other organizations so its research has real-world applications and benefits. Peruvian scientists trained alongside the research team to learn how to use the mapping method and technology. Peru plans to survey other areas as part of an ongoing partnership with Carnegie.
Asner is heading to Colombia and Ecuador next. While the scientists initially work with the host country, the goal is to enable the countries to do the monitoring themselves. Carnegie is offering the carbon analysis method and software for free to noncommercial organizations.
"Nobody really has these technologies anywhere," Asner said. "Rather than hold them up north in the U.S., we're trying to develop them and train and give them away at the same time in the host countries."