ABSTRACT. The effects of food type and concentration on fecal pellet characteristics of the calanoid copepod Acartia tonsa were examined in the laboratory. Copepods were fed several autotrophic and heterotrophic diets, including the diatoms Thalassiosira weissflogii and Chaetoceros neogracile, the photosynthetic flagellates Rhodomonas lens and Tetraselmis sp., the heterotrophic dinoflagellate Oxyrrhis marina, the heterotrophic flagellates Cafeteria sp. and Oikomonas sp., and the scuticociliate Uronema sp. Copepods were fed both a low (-100-300 pg C l-l) and a high (-500-1500 pg C l") concentration of these diets. Length. width and density of the resulting fecal pellets were measured. Sinking rates were calculated from a semi-empincal model based on these parameters. In general, diets that resulted in large pellets also resulted in the least dense pellets. Ciliate and diatom diets produced the largest pellets and resulted in the fastest estimated sinking rates. Heterotrophic flagellate and heterotrophic dinoflagellate diets resulted in the most dense and slowest sinking pellets Within a diet, significant differences in pellet characteristics were often found between food concentrations, but there was no consistent pattern of increasing or decreasing pellet size or density with an increase in food concentration. The coefficient of variation of pellet sinking rates across all diets in this study was nearly 40%. This indicates the uncertainty in estimated sinking rates if dlet is not considered. Combining the sinking rates from this study with published diet-specific fecal pellet degradation rates, we define an L-ratio, the fraction of pellet degradation per unit length of sinklng. The L-ratio may be useful in predicting the degree of recycling of pellets within the mixed layer. Diatoms show the lowest L-rat~os and photosynthetic fldgellates the highest L-ratios.

No abstract

Zooplankton fecal pellet flux is a highly variable component of the biological carbon pump. While fecal pellets can comprise 0 to nearly 100% of particulate organic carbon collected in sediment traps, mechanisms for this variability remain poorly understood. Fecal pellet carbon flux is a complex function of several variables. We present a model that incorporates individual‐scale metabolic processes to determine fecal pellet production rate, the relationship between body size and fecal pellet size, the relationship between fecal pellet size and sinking rate, and a function representing the breakdown of particles in the water column. When applied to copepod communities sampled by the continuous plankton recorder in the Gulf of Maine over 25 years, a seasonal pattern of fecal pellet carbon flux emerges. The interannual flux time series produced by the model reflects known oceanographic perturbations and shows how organism‐scale processes can be scaled up to explain ecosystem level variability. We conclude that fecal pellet carbon flux in the Gulf of Maine is driven by copepod community size structure and copepod abundance, and that the fraction of fecal pellet carbon that reaches depth is a function of copepod size, rather than abundance. Changes in the physical environment which alter the size composition of the copepod community lead to variability in fecal pellet carbon flux. Our results indicate that incorporating size composition into biogeochemical models can more accurately constrain zooplankton‐mediated carbon flux.

No abstract