AUTHOR=Weber Sarah C. , Subramaniam Ajit , Montoya Joseph P. , Doan-Nhu Hai , Nguyen-Ngoc Lam , Dippner Joachim W. , Voss Maren TITLE=Habitat Delineation in Highly Variable Marine Environments JOURNAL=Frontiers in Marine Science VOLUME=6 YEAR=2019 URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2019.00112 DOI=10.3389/fmars.2019.00112 ISSN=2296-7745 ABSTRACT=

The structure of the phytoplankton community in surface waters is the consequence of complex interactions between the physical and chemical properties of the upper water column as well as the interaction within the general biological community. Understanding the structure of phytoplankton communities is especially challenging in highly variable and dynamic marine environments. A variety of strategies have been employed to delineate marine planktonic habitats, including both biogeochemical and water-mass-based approaches. These methods have led to fundamental improvements in our understanding of marine phytoplankton distributions, but they are often difficult to apply to systems with physical and chemical properties and forcings that vary greatly over relatively short spatial or temporal scales. In this study, we have developed a method of dynamic habitat delineation based on environmental variables that are biologically relevant, that integrate over varying time scales, and that are derived from standard oceanographic measurements. As a result, this approach is widely applicable, simple to implement, and effective in resolving the spatial distribution of phytoplankton communities. As a test of our approach, we have applied it to the Amazon River-influenced Western Tropical North Atlantic (WTNA) and to the South China Sea (SCS), which is influenced by both the Mekong River and seasonal coastal upwelling. These two systems differ substantially in their spatial and temporal scales, nutrient sources/sinks, and hydrographic complexity, providing an effective test of the applicability of our analysis. Despite their significant differences in scale and character, our approach generated statistically robust habitat classifications that were clearly relevant to surface phytoplankton communities. Additional analysis of the habitat-defining variables themselves can provide insight into the processes acting to shape phytoplankton communities in each habitat. Finally, by demonstrating the biological relevance of the generated habitats, we gain insights into the conditions promoting the growth of distinct communities and the factors that lead to mismatches between environmental conditions and phytoplankton community structure.