|
Document: PET-3-87-14
Relationship between decadal anomalies in climate and model-estimated AET, NPP, total carbon, and net ecosystem carbon exchange, 1895-1993: Results from VEMAP Phase 2. THORNTON, P.E.* 1, S.W.RUNNING 1 and V.MEMBERS 2
University of Montana, Missoula, MT 59812 USA 1 Vegetation Ecosystem Modeling and Analysis Project, multiple affiliations 2
Abstract: Thanks to the combined efforts of the VEMAP Phase 2 group, we have available gridded datasets of surface weather conditions and the results of terrestrial ecosystem simulations from six models with very different representations of biogeochemical processes. We present a summary of the VEMAP results for the period of historical record, 1895-1993, aggregated to the annual and decadal level, for a region covering the conterminous United States. We explored the relationships between climate anomalies and simulated surface process anomalies at the decadal scale. Our objective was to describe the patterns in these relationships for each model, and to compare these patterns between models. The purpose of such comparisons was to identify ecosystem processes for which different simulation approaches give significantly different results, and to further identify the climate drivers that were of greatest importance for different approaches. We found that model estimates of actual evapotranspiration (AET) have very similar responses to climate anomalies, and that simple regressions with climate anomalies can explain most of the variability in AET at the decadal scale. Relationships between climate and estimated net primary production (NPP) were less similar between models, and simple regressions captured substantially less variability in the decadal NPP estimates. Climate effects on net ecosystem exchange of carbon (NEE) were quite variable between models, and were weak for most models in most vegetation types. Anomalies in precipitation and radiation anomalies were more highly correlated with NEE than were anomalies in temperature and humidity. Drier vegetation types showed generally stronger relationships between NEE and climate anomalies than did wetter types. The implication of these results is that predictability in net ecosystem fluxes of carbon does not necessarily follow from the predictability of other important ecosystem water and carbon flux processes. This result has importance for the assessment and monitoring of terrestrial sources and sinks of carbon under variable climatic conditions.
Keywords: VEMAP, climate, ecosystem model, net carbon exchange
|







This abstract is being presented at: 4:15 PM in session: Oral Session #66: Large Scale Climate Change. |