Statement of Interest

Christopher Potter
Research Scientist
Ecosystem Science and Technology Branch
NASA Ames Research Center

Modeling Land Use Change and Ecosystem Processes in the Amazon Basin

I am a terrestrial ecologist by training. My main research interests are global change, biogeochemistry, and land use. I've been at NASA's Ames Research Center for the past six years developing simulation models that couple these three areas of interest. The main tools of this trade are super computers, remote sensing, GIS, field research, and (hopefully) some creative thinking.

My main interest with respect to this Land Use Modeling workshop is on coupling our ecosystem modeling knowledge to land use modeling at the regional scale. Our current geographic focus area is the Amazon Basin. I am considering a series of model-based questions that can best be addressed through joint research among several disciplines: What are the predominant land uses in different part of the Brazilian Amazon?

How is land use and management changing in the Brazilian Amazon?

How does land management influence the abundance and distribution of pastures and secondary forest growth, and the balance between clearing and re-growth?

How do changes in natural ecosystem processes influence these land-use changes and management practices?

Our regional geographic information system (GIS) for the Brazilian Amazon serves as the data source of land cover, climate drivers, satellite greenness images, and soil properties for input to the NASA-CASA (Carnegie-Ames-Stanford Approach) model at a 8-km grid resolution. Simulation results already reveal regional effects of forest conversion on plant production potential and soil carbon content, especially in seasonally dry areas. These results are being used to formulate a series of research hypotheses for testing in the next phase of regional modeling, which ideally will include linkage to land use simulations.

The background for our work is global change. Humans are causing environmental alterations of planetary significance. One of the clearest signals of human impact is the rapidly rising concentration of greenhouse gases like carbon dioxide (CO2) in the atmosphere as a result of combustion of fossil fuels and changes in land use. Other trace gases such as methane (CH4) and nitrous oxide (N2O) can significantly influence the energy balance of the Earth. Moreover, these compounds are linked to atmosphere-biosphere feedbacks (CO2 fertilization effects on vegetation), tropospheric chemistry (CH4 reactions with OH, O3 and NOx dynamics) and stratospheric chemistry (N2O-mediated destruction of ozone).

By the use of recently assembled satellite images of the global land surface, we have developed the simulation model called NASA-CASA to study of the role of terrestrial plants and soils in the cycling of carbon to and from the atmosphere. This research has produced a dynamic and detailed picture of the contemporary balance between photosynthetic fixation and microbial respiration of CO2. In the NASA-CASA Biosphere model, a greenness index from satellite sensors is combined with modeled climate stress to estimate plant production. Carbon and nitrogen fluxes from decomposition of plant residues are simulated in the soil and at the surfaces of forest and grassland soils.

The results of our latest multi-year simulation of global ecosystems directly infer the presence of a net sink for carbon dioxide in the terrestrial biosphere over the period 1985-1988. This is, to our knowledge, an original research result that has not been demonstrated before at the global scale using actual remote sensing observations and land surface climate data. By combining medium-term (decade long) climate data with ecosystem modeling, new answers emerge with respect to biosphere-atmosphere exchange of trace gases resulting from climate fluctuations. However, the results from this ecosystem modeling study, along with others that have used satellite observations to infer plant production, must be qualified to point out that conclusions cannot be extrapolated directly to infer long-term (several decades) response of the biosphere to global warming or land use change. The processes of disturbance and land management must be included in additional modeling analyses that include potential geographic shifts occurring over several decades in plant functional types, physiological responses, and soil carbon turnover rates.