Forest Carbon Sequestration and Global Land Use

US forests have been a net carbon sink for the past 50 years, sequestering 500–800 million tons of CO2 per year for the past 20 years. This sequestration has benefited the global climate system, offsetting 10–15% of gross GHG emissions in the US. Recent projections from the US Forest Service, however, suggest that the ability of US forests to sequester carbon may be imperiled over the coming decades, in part due to adverse climate impacts on forests, as well as the expansion of cropland. While there is nascent research into the future capacity of forest carbon sinks, many analyses either do not incorporate market responses to biophysical shocks; are static, failing to account for dynamic adjustments in harvest schedules or investments; do not address climate change, which will likely cause increased forest disturbance; or are deterministic and ignore the impact of future uncertainty on optimal decision making in the presence of significant irreversibilities and risk aversion.

vegetation projections (a) Mci-modeled potential vegetation under observed 1961-1990 cliamte. (b) mci-modeled potential vegetation under projected 2071-2100 climate where any of nine general circulation model-emissions scenario combinations projects a change. biomes, in (a) and (B), from poles to equator: ice (ic), tundra and alpine (UA), boreal conifer forest (bc), temperate conifer forest (TC), temperate broadleaf forest (tb), temperate mixed forest (tm), temperate shrubland (ts), temperate grassland (tg), desert (de), tropical grassland (rg), Tropical woodland (rw), tropical deciduous broadleaf forest (rd), and tropical everygreen Broadleaf forest (re). (c) confidence of biome projections calculated from fraction of general circulation model-emissions scenario combinations that project the same time of biome change.

vegetation projections (a) Mci-modeled potential vegetation under observed 1961-1990 cliamte. (b) mci-modeled potential vegetation under projected 2071-2100 climate where any of nine general circulation model-emissions scenario combinations projects a change. biomes, in (a) and (B), from poles to equator: ice (ic), tundra and alpine (UA), boreal conifer forest (bc), temperate conifer forest (TC), temperate broadleaf forest (tb), temperate mixed forest (tm), temperate shrubland (ts), temperate grassland (tg), desert (de), tropical grassland (rg), Tropical woodland (rw), tropical deciduous broadleaf forest (rd), and tropical everygreen Broadleaf forest (re). (c) confidence of biome projections calculated from fraction of general circulation model-emissions scenario combinations that project the same time of biome change.

In a new project, we're extending the FABLE (Forest, Agriculture, and Biofuels in a Land use model with Environmental services)  model to more accurately capture forestry management decisions. Forests are managed differently in different regions from intensive management in plantations forests to continued old growth extraction in boreal zones to selective logging with significant collateral damage in the tropics. Over 60% of the world’s timber harvests occur in temperate and boreal zones, which may experience increased forest fire activity. Adapting management to rising forest fire probabilities, or extreme fire seasons, could affect global prices and consequently management in the fast-growing plantation regions of the subtropics. Disaggregating FABLE by ecological zones will account for factors such as forestry rotations, management intensity, and the development of fast-growing plantations.

Further, we plan to use a dynamic global vegetation models to estimate uncertainties in timber yields due to climate change. Impacts from climate change will differ substantially by region. Boreal zones are expected to experience the largest shocks, although the species in the tropics are potentially the least adaptive and are more susceptible to smaller shocks. We will model natural disturbances due to forest fires as shocks to forest stocks, using the MC1 dynamic global vegetation model for these estimations. The change in timber yields will be assumed to be proportional to changes in net primary productivity projected by the MC1 model.

We will also incorporate improved estimates of land use change and land availability. Changes in the potential area of land available for forests to grow will be assumed to be proportional to changes in the ecological zones produced by MC1. While MC1 may predict a large increase or decrease in the area of particular species, the actual area of the species will be a function of economic conditions and crop yields.

People:

Yongyang Cai | Alla Golub | Thomas Hertel | Kenneth Judd Brent Sohngen | Jevgenijs Steinbuks