HR: 1340h
AN: B13B-1208    [Abstracts]
TI: Geospatial modeling of plant stable isotope ratios - the development of isoscapes
AU: * West, J B
EM: jwest@biology.utah.edu
AF: Department of Biology University of Utah, 257 South 1400 East, Salt Lake City, UT 84112, United States
AU: Ehleringer, J R
EM: ehleringer@biology.utah.edu
AF: Department of Biology University of Utah, 257 South 1400 East, Salt Lake City, UT 84112, United States
AU: Hurley, J M
EM: hurley@biology.utah.edu
AF: Department of Biology University of Utah, 257 South 1400 East, Salt Lake City, UT 84112, United States
AU: Cerling, T E
EM: cerling@earth.utah.edu
AF: Department of Geology and Geophysics University of Utah, 135 South 1460 East, Salt Lake City, UT 84112, United States
AB: Large-scale spatial variation in stable isotope ratios can yield critical insights into the spatio-temporal dynamics of biogeochemical cycles, animal movements, and shifts in climate, as well as anthropogenic activities such as commerce, resource utilization, and forensic investigation. Interpreting these signals requires that we understand and model the variation. We report progress in our development of plant stable isotope ratio landscapes (isoscapes). Our approach utilizes a GIS, gridded datasets, a range of modeling approaches, and spatially distributed observations. We synthesize findings from four studies to illustrate the general utility of the approach, its ability to represent observed spatio-temporal variability in plant stable isotope ratios, and also outline some specific areas of uncertainty. We also address two basic, but critical questions central to our ability to model plant stable isotope ratios using this approach: 1. Do the continuous precipitation isotope ratio grids represent reasonable proxies for plant source water?, and 2. Do continuous climate grids (as is or modified) represent a reasonable proxy for the climate experienced by plants? Plant components modeled include leaf water, grape water (extracted from wine), bulk leaf material ( Cannabis sativa; marijuana), and seed oil ( Ricinus communis; castor bean). Our approaches to modeling the isotope ratios of these components varied from highly sophisticated process models to simple one-step fractionation models to regression approaches. The leaf water isosocapes were produced using steady-state models of enrichment and continuous grids of annual average precipitation isotope ratios and climate. These were compared to other modeling efforts, as well as a relatively sparse, but geographically distributed dataset from the literature. The latitudinal distributions and global averages compared favorably to other modeling efforts and the observational data compared well to model predictions. These results yield confidence in the precipitation isoscapes used to represent plant source water, the modified climate grids used to represent leaf climate, and the efficacy of this approach to modeling. Further work confirmed these observations. The seed oil isoscape was produced using a simple model of lipid fractionation driven with the precipitation grid, and compared well to widely distributed observations of castor bean oil, again suggesting that the precipitation grids were reasonable proxies for plant source water. The marijuana leaf δ²H observations distributed across the continental United States were regressed against the precipitation δ²H grids and yielded a strong relationship between them, again suggesting that plant source water was reasonably well represented by the precipitation grid. Finally, the wine water δ¹⁸O isoscape was developed from regressions that related precipitation isotope ratios and climate to observations from a single vintage. Favorable comparisons between year-specific wine water isoscapes and inter-annual variations in previous vintages yielded confidence in the climate grids. Clearly significant residual variability remains to be explained in all of these cases and uncertainties vary depending on the component modeled, but we conclude from this synthesis that isoscapes are capable of representing real spatial and temporal variability in plant stable isotope ratios.
UR: http://isoscapes.org
DE: 0454 Isotopic composition and chemistry (1041, 4870)
DE: 0466 Modeling
DE: 1622 Earth system modeling (1225)
DE: 1819 Geographic Information Systems (GIS)
DE: 9820 Techniques applicable in three or more fields
SC: Biogeosciences [B]
MN: 2007 Fall Meeting