LAUSR

Organismal distributions are largely mediated by temperature, suggesting thermal trait variability plays a key role in defining species’ niches. We employed a trait-based approach to better understand how interand intraspecific thermal trait variability could explain diatom community dynamics using 24 strains from 5 species in the diatom genus Skeletonema, isolated from Narragansett Bay (NBay), where this genus can comprise up to 99% of the microplankton. Strain-specific thermal reaction norms were generated using growth rates obtained at temperatures ranging from −2 C to 36 C. Comparison of thermal reaction norms revealed interand intraspecific similarities in the thermal optima, but significant differences approaching the thermal limits. Cellular elemental composition was determined for two thermally differentiated species and again, the most variation occurred approaching the thermal limits. To determine the potential impact of interspecific variability on community composition, a species succession model was formulated utilizing each species’ empirically determined thermal reaction norm and historical temperature data from NBay. Seasonal succession in the modeled community resembled the timing of species occurrence in the field, but not species’ relative abundance. The model correctly predicted the timing of the dominant winter–spring species, Skeletonema marinoi, within 0–14 d of its observed peak occurrence in the field. Interspecific variability approaching the thermal limits provides an alternative mechanism for temporal diatom succession, leads to altered cellular elemental composition, and thus has the potential to influence carbon flux and nutrient cycling, suggesting that growth approaching the thermal limits be incorporated into both empirical and modeling efforts in the future. Temperature is a principal driver of global organismal distributions in both terrestrial (Angilletta 2009; Sunday et al. 2012) and marine (Poloczanska et al. 2013; García Molinos et al. 2015) environments and one of the most important environmental factors shaping microbial composition in the euphotic ocean (Sunagawa et al. 2015). Temperature differentially influences growth and cellular metabolism between organisms (Eppley 1972), which results in thermal niche differentiation among species (Hardin 1960). In microbes, the influence of temperature on growth has been used to characterize the thermal niche of a species, define thermal traits, and predict a species’ ability to respond to environmental variability (Litchman et al. 2012). For example, thermal traits have been utilized to interpret species’ thermal ranges on a global scale (Thomas et al. 2012; Boyd et al. 2013). However, the utility of thermal traits remains relatively uncharacterized in terms of their contribution to community dynamics, such as succession and seasonality. Thermal reaction norms, or performance curves, describe individual or species’ responses to a wide range of temperatures and are parameterized by the thermal traits. They peak at the thermal optima and extend to the thermal limits. Between species or individuals, thermal reaction norms can vary along the temperature axis, both in their position horizontally and in their magnitude vertically (Kingsolver 2009); two theoretical examples are depicted in Fig. 1 (adapted from Bolnick et al. 2003). In one example, species display unique growth optima along a thermal gradient that results in clear niche differentiation between species (Fig. 1a). In another example, species display similar thermal optima and niche widths resulting in less niche differentiation (Fig. 1b). While thermal reaction norms can be differentiated in a multitude of ways, greater differentiation, like that depicted in our first example, is readily observed on the global scale as species tend to assort by optima across latitudes (Thomas et al. 2012). However, on regional scales, there are insufficient data to characterize how species structure *Correspondence: [email protected] This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. Additional Supporting Information may be found in the online version of this article.