Data and Model Structure Impacts on Carbon-Cycle Projections

Steven J Smith, Pacific NW National Laboratory (PNNL), ssmith@pnl.gov (Presenter)
Andrew Rothwell, Pacific NW National Laboratory (PNNL), andrew.rothwell@pnl.gov

Land use, the carbon cycle, and energy supply/demand comprise a complex system that interacts to determine atmospheric carbon dioxide concentrations. Integrated assessment models (IAMs) incorporate carbon-cycle dynamics, along with representations of many other physical and socio-economic processes, in order to examine the interactions between these systems. Given the substantial uncertainty in current knowledge, it is important that IAMs consider a full range of potential responses in all of these systems. With respect to the carbon cycle, IAMs have generally considered only a limited set of parameters such temperature and CO2 fertilization feedbacks, although even only these uncertainties have substantial economic consequences (Smith and Edmonds 2006).

To address these questions we developed a flexible carbon-cycle model, whose flexibility is made possible by use of gridded data sets developed by various research groups fusing remote sensing and other data sources. We use this model to examine the impact of a range of different model configurations, input data sets, and parameter values to determine the relative impact of different assumptions on net terrestrial carbon flows. As expected, climate feedback parameters have a substantial impact on modeled uptake, however we also find that use of a more mechanistically sound carbon model and the incorporation of forest age classes also impact results. We conclude with implications for carbon-cycle data gathering and synthesis.

Presentation Type:  Poster

Session:  Science in Support of Decision Making   (Wed 10:00 AM)

Associated Project(s): 

• Smith, Steven: Projections of Land-Use Change and the Carbon Cycle: Atmospheric and Policy Consequences ...details

Poster Location ID: 246