California’s cap-and-trade marketplace allows private forest landowners to enroll “improved forest management” projects into the carbon market. Conceptually, these forests are supposed to sequester more carbon under this “improved” management—less intense logging and more forest protection, for example—than under their ordinary management, with the increased carbon stored available to sell to polluters elsewhere in the state. This requirement of “additionality”—for projects to achieve carbon reductions in addition to what would happen under business-as-usual without the project—is at the core of the California carbon market.
Utilizing satellite imagery and machine learning, researchers compared the forests enrolled in carbon markets to nearby forests. Using data contained in the images, such as the color of a pixel, researchers are able to figure out things like species composition—a tanoak would present as X in the data while a redwood would look like Y, for example—and the stored biomass in the forest. Looking at all the data, researchers came to important conclusions.
The “additional” carbon credit received might not be from improved management, but larger changes in timber markets. The carbon market system assumes a flat baseline—that harvest and growth are generally equal—and that increased carbon accumulation is a result of enrollment of a forest in a carbon program. The fault in this? Across California’s forests, carbon stocks are generally increasing—the baseline is not flat. Since the days of liquidation logging, timber practices have changed and forests are generally increasing in biomass. By failing to account for this general trend of increasing biomass, the carbon market sets an artificially flat baseline and projects are receiving credit for carbon increases that are likely not attributable to enrollment in the carbon market.
Timber companies are also enrolling forests in carbon markets that are at lower risk of logging in the first place. Based on their data, researchers found that Green Diamond Resource Company was selectively including areas with high tanoak concentrations and low amounts of redwood in improved forest management projects. The problem? Tanoak is not a commercial timber species, so these areas would not likely be at risk for logging in the first place. There is no “additional” carbon if species composition inherently limits logging. Sierra Pacific Industries seems to preferentially select for younger forests that have experienced logging recently. This timber company would then get credit for naturally recovering forests as they won’t experience logging pressure while they are pre-commercial in size—there is no “additional” carbon.
This research interests EPIC for two reasons. First, it calls into question many forest carbon projects. If the parameters are flawed and timber companies are selling carbon credits that don’t reflect actual additional carbon sequestration, then our larger statewide cap-and-trade system is not actually reducing carbon emissions in the amount it assumes. Critics of forest carbon offsets have long worried about the lack of additionality and this research largely confirms suspicions. Second, this research demonstrates the potential for objective remote monitoring of forest conditions through satellite imagery. This has utility beyond ensuring that carbon projects are effective but can also help us understand broader trends in carbon sequestration on forestlands.