Environmental control of open-ocean phytoplankton groups: Now and in the future

Boyd, Philip W., Robert Strzepek, Feixue Fu, and David A. Hutchins

Limnol. Oceanogr., 55(3), 2010, 1353-1376 | DOI: 10.4319/lo.2010.55.3.1353

ABSTRACT: Climate change will alter concurrently many environmental factors that exert control over oceanic phytoplankton. Recent laboratory culture work, shipboard experiments, and field surveys reveal many remaining unknowns about the bottom-up controls for five globally important algal groups. Increasing uncertainties exist, respectively, for picocyanobacteria, diatoms, Phaeocystis spp., N2-fixing cyanobacteria, and coccolithophores. This missing information about current environmental controls will hinder progress in modeling how these phytoplankton will be influenced by climate change. A review of conceptual approaches used to elucidate the relationship between environmental controls and phytoplankton dominance, from Margalef's mandala to functional traits, uncovered limitations regarding their application to climate-change scenarios. For example, these previous approaches have insufficient scope or dimensions to take into account the confounding effects of synergistic and antagonistic interactions of multiple environmental change variables. A new approach is needed that considers all of the different environmental properties altered by climate change and their interactions while at the same time permitting a subset of the most significant controls for a specific phytoplankton group to be isolated and evaluated in factorial matrix perturbation experiments. We advocate three new interlinked approaches, including environmental clusters that incorporate all factors (temperature, CO2, light, nutrients, and trace metals), which both exert control over present-day floristics and will be altered by climate change. By carefully linking a holistic conceptual approach to a reductionist experimental design, the future responses of open-ocean phytoplankton groups to a complex, rapidly changing environment can be better predicted.

Article Links

Please Note