Determining ecological provinces from optical cytometric data in the North Pacific Ocean

Phytoplankton community structure is an important determinant of marine ecosystem functionwith far-reaching implications for the cycling of energy and matter in the marine environment. Phytoplankton communitiesdevelop as a function of biological, chemical, and physical processes. As such, spatial and temporal variability in their properties can act as biological indicators providing insight into mechanisms controlling the organization of phytoplankton communities. In this study, we leverage an extensive optical (scatter, fluorescence) dataset collected in the North Pacific using the SeaFlow flow cytometer, in conjunction with machine learning tools, to identify statistically unique phytoplankton communities which in aggregate delineate coherent and distinct ecological provinces encountered during in situ sampling of the marine environment. Examination of resulting phytoplankton communities in the context of environmental data indicates a strong association with water mass characteristics, identifying known large-scale gradients (e.g. polar / transition / subpolar), alongside higher frequency shifts. Mapping of lower-resolution rate measurements, as well as comparison with model and satellite-derived seascapes / provinces, yields additional insights into the organization of marine primary producers and the impact of environmental variability on lower trophic functioning.