Machine learning techniques to characterize functional traits of plankton from image data.

Plankton imaging systems supported by automated classification and analysis have improved ecologists' ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite of lab-based and in situ tools, collecting imaging data at unprecedentedly fine spatial and temporal scales. But these data have potential well beyond examining the abundances of different taxa; the individual images themselves contain a wealth of information on functional traits. Here, we outline traits that could be measured from image data, suggest machine learning and computer vision approaches to extract functional trait information from the images, and discuss promising avenues for novel studies. The approaches we discuss are data agnostic and are broadly applicable to imagery of other aquatic or terrestrial organisms.

Modelling dimethylsulfide diffusion in the algal external boundary layer: implications for mutualistic and signalling roles.

Dimethylsulfide (DMS), a dominant organic sulfur species in the surface ocean, may act as a signalling molecule and contribute to mutualistic interactions between bacteria and marine algae. These proposed functions depend on the DMS concentration in the vicinity of microorganisms. Here, we modelled the DMS enrichment at the surface of DMS-releasing marine algal cells as a function of DMS production rate, algal cell radius and turbulence. Our results show that the DMS concentration at the surface of unstressed phytoplankton with low DMS production rates can be enriched by <1 nM, whereas for mechanically stressed algae with high activities of the enzyme DMSP-lyase (a coccolithophore and a dinoflagellate) DMS cell surface enrichments can reach ~10 nM, and could potentially reach µM levels in large cells. These DMS enrichments are much higher than the median DMS concentration in the surface ocean (1.9 nM), and thus may attract and support the growth of bacteria living in the phycosphere. The bacteria in turn may provide photoactive iron chelators (siderophores) that enhance algal iron uptake and provide algal growth factors such as auxins and vitamins. The present study highlights new insights on the extent and impact of microscale DMS enrichments at algal surfaces, thereby contributing to our understanding of the potential chemoattractant and mutualistic roles of DMS in marine microorganisms.

Carbonate Disequilibrium in the External Boundary Layer of Freshwater Chrysophytes: Implications for Contaminant Uptake.

The interplay between biological and chemical reactions in the freshwater phytoplankton phycosphere and the resulting modulations of contaminant speciation and uptake is poorly characterized. Here we modeled the effect of algal C and N uptake on carbonate cycling and speciation of selected contaminants in the phycosphere (external boundary layer) of chrysophytes, a key phytoplankton group in oligotrophic systems. We calculated an enrichment in H+ concentration relative to that in the bulk solution (pH 7.0) of approximately 40% or a depletion of approximately 30% for NH4+ or NO3--grown cells, respectively, at the algal membrane surface of a 5-µm radius cell. Such changes are mainly due to direct H+ uptake or release at the plasmalemma if NO3- or NH4+ is the N source, respectively. Due to these pH changes in the external boundary layer, competition between H+ and metals for uptake is enhanced, for NH4+-grown cells which contributes to a decrease in potential metal uptake. Our model suggests that the uptake of protonated weakly acidic organic acids (HA) is greater in NH4+-grown cells compared to that in NO3--grown cells. The account of chemical reactions in the algal external boundary layer could improve ecological risk assessments for a wide range of contaminants.

First principles of copepod development help explain global marine diversity patterns.

A major goal of modern ecology is to understand macroecological patterns based on their mechanistic underpinnings. The metabolic theory of ecology predicts a monotonic increase of biodiversity with temperature based on the principles of metabolism. For marine copepods, observations have shown that while biodiversity does increase with temperature, the theory's prediction overestimates the slope of this relationship by a factor of two. By relaxing the theory's assumption that size is invariant with respect to temperature, and by incorporating a mechanistic description of copepod development into the theory, we provide an adjusted prediction that agrees with the observed relationship. The addition of development into the theory adds the potential to refine the prediction for a wider range of taxa, to account for discrepancies between prediction and observations, and to describe a wider variety of temperature-richness relationships.