Here we describe a new trait-based model for cellular resource allocation that we use to investigate the relative importance of different drivers for small cell size in phytoplankton. Using the model, we show that increased investment in nonscalable structural components with decreasing cell size leads to a trade-off between cell size, nutrient and light affinity, and growth rate. Within the most extreme nutrient-limited, stratified environments, resource competition theory then predicts a trend toward larger minimum cell size with increasing depth. We demonstrate that this explains observed trends using a marine ecosystem model that represents selection and adaptation of a diverse community defined by traits for cell size and subcellular resource allocation. This framework for linking cellular physiology to environmental selection can be used to investigate the adaptive response of the marine microbial community to environmental conditions and the adaptive value of variations in cellular physiology.A key challenge facing marine biogeochemical modelers is how best to represent the important role that diverse, rapidly evolving microbial populations play in marine biogeochemical cycles. Simple models, which may include just a single state variable for all phytoplankton (e.g., Fasham et al. 1990), often ignore the distinct functional role that different taxa perform (e.g., silicifying, or calcifying organisms), while more complicated models, which attempt to explicitly resolve multiple functional types (Le Quéré et al. 2005), can face severe practical problems in terms of the number of organism-level measurements and parameters required to describe the model (Anderson 2005;Flynn 2006). A still more fundamental problem lies in how best to represent adaptive or evolutionary processes, which are typically ignored in current state-of-the-art marine ecosystem models.Recent approaches have attempted to address some of these issues (Follows and Dutkiewicz 2011). For example, Follows et al. (2007) used a Monte-Carlo sampling method with a marine ecosystem model that included a diverse phytoplankton community, described by a set of empirically motivated traits with trade-offs. Emergent patterns in phytoplankton biogeography were in broad agreement with observations, illustrating the importance of environmental selection in determining spatial and temporal patterns in phytoplankton biogeography. The same model has also been used to examine drivers for latitudinal patterns in biodiversity (Barton et al. 2010) and the effect of chromatic adaptation on community structure in oligotrophic environments (Hickman et al. 2010). Meanwhile, Bruggeman and Kooijman (2007) described a biodiversity-based marine ecosystem model in which species were defined by a generic model with continuous trait values for investment in nutrient-and light-harvesting machinery. In their model, trade-offs between traits emerged naturally as a result of different allocation strategies. When configured for a representative oligotrophic open-ocean sit...