Using ocean models to predict spatial and temporal variation in marine carbon isotopes

Natural-abundance stable isotope ratios provide a wealth of ecological information relating to food web structure, trophic level, and location. The correct interpretation of stable isotope data requires an understanding of spatial and temporal variation in the isotopic compositions at the base of the food web. In marine pelagic environments, accurate interpretation of stable isotope data is hampered by a lack of reliable, spatio-temporally distributed measurements of baseline isotopic compositions. In this study, we present a relatively simple, process-based carbon isotope model that predicts the spatio-temporal distributions of the carbon isotope composition of phytoplankton (here expressed as dCPLK) across the global ocean at one degree and monthly resolution. The model is driven by output from a coupled physics-biogeochemistry model, NEMO-MEDUSA, and operates offline; it could also be coupled to alternative underlying ocean model systems. Model validation is challenged by the same lack of spatio-temporally explicit data that motivates model development, but predictions from our model successfully reproduce major spatial patterns in carbon isotope values observed in zooplankton, and are consistent with simulations from alternative models. Model predictions represent an initial hypothesis of spatial and temporal variation in carbon isotopic baselines in ocean areas where a few data are currently available, and provide the best currently available tool to estimate spatial and temporal variation in baseline isotopic compositions at ocean basin to global scales.