September 10th, 2013
Planktonic food web dynamics shape biogeochemical cycles and global patterns of ocean productivity across trophic levels. Primary production alone, for example, is a poor predictor of cross-ecosystem differences in fisheries yields. Predictive capability improves only after consideration of factors such as the number and efficiency of trophic links separating phytoplankton and fish. Limited representation and validation of planktonic food web dynamics within the present generation of Earth System Models limits both their resolution of biogeochemical processes and their utility for assessing climate impacts on living marine resources.
The Carbon, Ocean Biogeochemistry and Lower Trophics (COBALT) planktonic ecosystem model was developed to address these limitations within GFDL’s Earth System Models. The authors describe this model, which captures observed biome-scale patterns in the flow of energy through the planktonic food web to an extent that exceeds the skill demonstrated by other global models. Diagnosis of the simulations provides numerous new insights into energy transfer through planktonic food webs to fish. Climate projections with this model will improve quantitative confidence in assessments of climate impacts on fisheries and other marine resources.
This paper assesses COBALT simulations against diverse observations of energy and carbon flows through planktonic marine ecosystems, and derives holistic, quantitative and self-consistent estimates of energy and carbon flows through the planktonic marine ecosystem. Carbon/energy flow estimates are derived globally and for each of three broadly defined ocean biomes: oligotrophic oceans, continuously stratified upwelling regions, and seasonally stratified regions. To our knowledge, these are the first comprehensive estimates of this kind derived from a global Earth System simulation.
The model recreates both sharp contrasts and surprising regularities across ocean biomes. Continuously stratified upwelling regions, for example, are estimated to support 46% of global mesozooplankton production despite accounting for only 21% of ocean area. Bacterial production, in contrast, is maintained in remarkably constant proportion to primary production across vastly different systems. Models must simultaneously reproduce such diverse patterns to elucidate the dynamics that underlie them in a consistent manner. Further diagnosis of COBALT reveals muted differences in the trophic level of mesozooplankton across ocean biomes and broadly distributed respiration of organic material throughout the planktonic food web.
COBALT is intended as a robust baseline model to support continued study of a broad set of interactions between ecosystems and climate. A number of innovative model improvements are being pursued within GFDL and in collaboration with colleagues from other research institutions, such as a size-structured fisheries food web/movement model for the study of global fish distribution. A model of zooplankton diurnal vertical migration has been developed to understand the global biogeochemical and ecological implications of this behavior. With each new development, comparison against the baseline defined with this paper will be essential for understanding the implications of these important but poorly understood marine ecosystem dynamics.