How transport shapes copepod distributions in relation to whale feeding habitat: Demonstration of a new modelling framework.

Highlights • New model represents Calanus life history, vertical behavior & ocean circulation. • Model simulates high surface layer prey abundance in the sGSL in summer & fall. • Transport of active stages is important to late stage Calanus abundance in sGSL. • Model intercomparison & groundtruthing identify future coupled model improvements. Abstract Two copepod species, Calanus finmarchicus and Calanus hyperboreus , constitute the main prey of endangered North Atlantic right whales in eastern Canadian shelf waters. Estimating the spatial distribution of Calanus spp. and understanding how it changes in response to environmental variability are crucial information requirements for mitigating risks to the right whale population. Here we present a new coupled dynamic model that builds on observations from the Atlantic Zone Monitoring Program (AZMP) and statistical modelling of copepod spatial distributions using Generalized Additive Models (GAMs). The coupled model is initialized with the GAMs climatological abundance and explicitly represents transport of the two species C. finmarchicus and C. hyperboreus via particle tracking, considering Calanus vertical swimming behavior and along-track mortality and temperature-dependent development. Assessment of model performance and the role of advection in determining the modelled Calanus distributions is focused on the southern Gulf of St. Lawrence (sGSL), an area regularly visited by right whales, and where local Calanus abundance is likely dependent on the seasonal transport from deep upstream areas. The coupled model simulates high Calanus abundance in the sGSL in summer and fall, distinct from the spatial distribution estimated by GAMs, which do not consider the effect of transport in the region. The exercise of building and testing this model allows us to investigate the factors producing high concentrations of late stage Calanus , and in the cases of model-data mismatch, infer important processes that are not adequately represented in the model. The model constitutes an important step forward in the ongoing effort to improve estimates of Calanus availability in the region, guiding future modelling efforts (e.g., a population development model) and potentially informing North Atlantic right whale management. [ABSTRACT FROM AUTHOR]