The low estimates from the models and inventories suggest that ecosystems in the US, and North America, are storing carbon, but much less than is being emitted by fossil fuel use. The uptake is widely distributed around the US, in forests, grasslands, farmlands and other ecosystems. This result implies that other regions of the world - Europe, Siberia or the tropical forests - must also be important.

The uptake of carbon was modeled from 1895 to 1993. Over this period, the fluxes were highly variable as the weather varied from one year to the next. This shows that carbon storage is very sensitive to the climate and if there is climate change in the future, ecosystem carbon storage will respond. In general, warmer and drier conditions caused carbon release to the atmosphere. The high variability of carbon storage from one year to the next has another implication. Under the Kyoto protocol, which is the international agreement to limit fossil fuel emissions, "commitment periods" are identified during which national carbon budgets must be compiled, including estimates of carbon uptake in deliberately managed forests. However, the models show that the amount of ecosystem uptake is greatly variable and can even go from uptake to release from year to year! Therefore, depending on the weather, the ecosystem balance in a given country could be positive or negative during the commitment period, and this will complicate the use of reporting intervals.

Until recently, scientists and policymakers have assumed that CO2 fertilization was the main mechanism for carbon uptake by ecosystems. Carbon fertilization is the stimulation of plant growth by increasing atmospheric CO2, a well-known phenomenon. Recently, other mechanisms have also been identified as having global importance. In particular, the regrowth of forests on abandoned farmland (
'"/>

Contact: David Schimel
dschimel@bgc-jena.mpg.de
49-3641-643729
Max-Planck-Gesellschaft
9-Mar-2000