Sampling bias misrepresents the biogeographical significance of constitutive mixotrophs across global oceans

Abstract

Aim Most protist plankton are mixotrophic, with potential to engage in photoautotrophy and phagotrophy; however, the ecology of these organisms has been misdiagnosed for over a century. A large proportion of these organisms are constitutive mixotrophs (CMs), with an innate ability to photosynthesize. Here, for the first time, an analysis is presented of the biogeography of CMs across the oceans. Location Global marine ecosystems. Time period 1970-2018. Major taxa studied Marine planktonic protists. Methods Records for CM species, primarily from the Ocean Biogeographic Information System (OBIS), were grouped by taxonomy and size to evaluate sampling efforts across Longhurst's oceanic provinces. Biases were evaluated through nonparametric tests and multivariate analysis. Biogeographies of CMs from OBIS data were compared with data from studies that specifically targeted these organisms. Results Constitutive mixotrophs of different taxonomic groups, across all size ranges, are ubiquitous. However, strong database biases were detected with respect to organism size, taxonomic groups and region. A strong bias was seen towards dinophytes. Species < 20 mu m, especially non-dinophytes, were least represented, with their recorded distribution limited to coastal regions and to temperate and polar seas. Studies specifically targeting these organisms revealed their distribution to be much wider. Such biases are likely to have occurred owing to a failure to capture and correctly identify these organisms in routine sampling protocols. Main conclusions Constitutive mixotrophs are dominant members of organisms traditionally termed "phytoplankton". However, lack of routine protocols for measuring phagotrophy in "phytoplankton" protists has led to widespread misrepresentation of the fundamental nature of marine planktonic primary producers; most express both "animal-like" and "plant-like" nutrition. Our results have implications for studies of the global biogeography of plankton, of food web dynamics (including models) and of biogeochemical cycling in the oceans.