Modeling the coupling of ocean ecology and biogeochemistry

Abstract

[1] We examine the interplay between ecology and biogeochemical cycles in the context of a global three-dimensional ocean model where self-assembling phytoplankton communities emerge from a wide set of potentially viable cell types. We consider the complex model solutions in the light of resource competition theory. The emergent community structures and ecological regimes vary across different physical environments in the model ocean: Strongly seasonal, high-nutrient regions are dominated by fast growing bloom specialists, while stable, low-seasonality regions are dominated by organisms that can grow at low nutrient concentrations and are suited to oligotrophic conditions. In the latter regions, the framework of resource competition theory provides a useful qualitative and quantitative diagnostic tool with which to interpret the outcome of competition between model organisms, their regulation of the resource environment, and the sensitivity of the system to changes in key physiological characteristics of the cells.

Phytoplankton niche generation by interspecific stoichiometric variation

“…Yet in practice, there is no steady state and species must avoid exclusion only for the timescale of the system under consideration, which is usually several orders of magnitude longer than the lifetime of a phytoplankton cell. Species with very similar R*s may coexist for long enough to survive in the ocean [Dutkiewicz et al, 2009].…”

“…[9] To explain these earlier findings and to explore the potential of a simple and plausible model alteration in enhancing coexistence, we simulated phytoplankton communities in a simple chemostat [Petersen, 1975;Huisman and Weissing, 1999;Shoresh et al, 2008] and in a global ocean model with a self-assembling phytoplankton community [Follows et al, 2007;Dutkiewicz et al, 2009]. In both models, each resource and phytoplankton species are modeled individually.…”

“…The impact of these parameters on coexistence are evaluated by numerical simulations of randomly assembled plankton communities. Particular attention is paid to the effects of varying the species' stoichiometric coefficients C j,i , since in global plankton models those are commonly parameterized according to the Redfield Ratio [Redfield, 1934;Gregg et al, 2003;Dutkiewicz et al, 2009], so that all species have the same stoichiometry (C j,i = C j for every species i). We compared modeled diversity in runs with identical stoichiometry (molar N:C = 0.15, P:C = 9.4 Â 10 À3 , Si:C = 0.15, Fe:C = 1.175 Â 10 À5 [Redfield, 1934;Follows et al, 2007], Si only in chemostat model) to modeled diversity in simulations with stoichiometry drawn randomly from a AE 25% range around those values.…”

“…For n resources, this implies that at most n species can coexist. Yet in many models with multiple potentially limiting resources, the number of surviving species rarely reaches the number of limiting resources [Follows et al, 2007;Shoresh et al, 2008;Dutkiewicz et al, 2009]. …”

“…[7] The reason for the extinctions in recent phytoplankton models [Bruggeman and Kooijman, 2007;Follows et al, 2007;Shoresh et al, 2008;Dutkiewicz et al, 2009] can be deduced using the R* concept [Dutkiewicz et al, 2009], which is part of Tilman's resource competition theory [Tilman, 1980]: In steady state, a monoculture of any species reduces the concentration of its limiting resource to the lowest concentration allowing for its survival (R*), hence growth rate equals losses. In a multi-species assemblage, the species requiring the lowest resource concentration will set the equilibrium resource concentration to its resource requirement R*, which is too low for any other species to survive.…”