Stochastic and Deterministic Processes Regulate Phytoplankton Assemblages in a Temperate Coastal Ecosystem.

Assessing the relative contributions of the interacting deterministic and stochastic ecological processes for phytoplankton community assembly is crucial in understanding and predicting community organization and succession at different temporal and spatial scales. In this study, we hypothesized that deterministic and stochastic ecological processes regulating phytoplankton, present seasonal and repeating patterns. This hypothesis was explored during a 5-year survey (287 samples) conducted at a small spatial scale (~15km) in a temperate coastal ecosystem (eastern English Channel). Microscopy and flow cytometry quantified phytoplankton abundance and biomass, while metabarcoding data allowed an extended evaluation of diversity and the exploration of the ecological processes regulating phytoplankton using null model analysis. Alpha diversity of phytoplankton was governed by the effect of environmental conditions (environmental filtering). Temporal community turnover (beta diversity) evidenced a consistent interannual pattern that determined the phytoplankton seasonal structure. In winter and early spring (from January to March), determinism (homogeneous selection) was the major process in the phytoplankton community assembly. The overall mean in the year was 38%. Stochastic processes (ecological drift) prevailed during the rest of the year from April to December, where the overall mean for the year was 55%. The maximum values were recorded in late spring and summer, which often presented recurrent and transient monospecific phytoplankton peaks. Overall, the prevalence of stochastic processes rendered less predictable seasonal dynamics of phytoplankton communities to future environmental change. IMPORTANCE While ecological deterministic processes are conducive to modeling, stochastic ones are far less predictable. Understanding the overall assembly processes of phytoplankton is critical in tracking and predicting future changes. The novelty of this study was that it addressed a long-posed question, on a pluriannual scale. Was seasonal phytoplankton succession influenced by deterministic processes (e.g., abiotic environment) or by stochastic ones (e.g., dispersal, or ecological drift)? Our results provided strong support for a seasonal and repeating pattern with stochastic processes (drift) prevailing during most of the year and periods with monospecific phytoplankton peaks.