Modelling the effects of copepod diel vertical migration and community structure on ocean carbon flux using an agent-based model.

• Creation of a new agent-based model allows investigation of zooplankton community structure and behavior effects on fecal pellet carbon flux in the ocean. • Diel vertical migration behavior in the model resulted in model fecal pellets fluxes that were closer to that of field data, compared to scenarios with no diel vertical migration. • Validation with field data shows that trait-based relationships for physiological traits can be used to give insight into particle flux variability in the deep ocean. Fecal pellets are a significant component of total particulate organic carbon (POC) flux in the ocean and a key component of the biological pump. The effect of copepod community structure, feeding behavior, and diel vertical migration (DVM) on fecal pellet carbon flux was investigated using an agent-based model, which was applied to two contrasting open ocean regions: an oligotrophic (Hawaii Ocean Time-series site ALOHA) and a mesotrophic (Japanese time-series site K2) system. Data from the VERtical Transport in the Global Ocean (VERTIGO) project was used to parameterize the model, and modelled fecal pellet fluxes were compared to field-derived neutrally buoyant sediment trap data to examine how community structure and DVM impact fecal pellet carbon flux in the water column. The model produces copepod fecal pellet fluxes that are consistent with sediment trap data from K2 and ALOHA and with other modeling studies. The vertical distribution of copepods matching nighttime field data resulted in a greater magnitude of fecal pellet flux at all model depths compared to daytime distributions. Communities dominated by omnivores resulted in higher fecal pellet carbon flux and lower POC attenuation than those dominated by carnivores or detritivores. Incorporating DVM into the simulations produced lower fecal pellet fluxes in some cases, while also marginally improving the agreement between the modeled and observed median fecal pellet fluxes. The relatively simplified model presented here shows that an agent-based model that uses trait-based relationships for physiological rates can be used to give insight into the variability of particle flux in the deep ocean. [ABSTRACT FROM AUTHOR]