Sensitivity of satellite-derived zooplankton grazing rates and biomass to input parameters

Zooplankton link primary producers to higher trophic levels and recycle particulate carbon and nutrients into dissolved pools. Zooplankton roles in food web models and biogeochemical cycles have been studied using ecosystem and biogeochemical models. As an alternative to models, remote sensing measurements can be used alone indirectly or in tandem with numerical models to quantify the spatio-temporal distribution of zooplankton. Here, we combined satellite-derived products (e.g., size-specified phytoplankton biomass and net primary productivity) with an inverted lower trophic level model (the North Pacific Ecosystem Model for Understanding Regional Oceanography (NEMURO) (Kishi et al., 2007)) to estimate zooplankton grazing rates and biomass. Our approach estimates zooplankton grazing in three size classes as a function of parameters which can be derived from remote sensing. Nevertheless, satellite-derived ocean-color products are still prone to large uncertainties. In order to better understand the strength and limitation of satellite-derived zooplankton grazing rates and biomass, the sensitivity of the model to mixed layer depth, net primary productivity, and size-fractionation methods was assessed. Results were compared against simulated zooplankton fields provided by a validated three-dimensional bio-physical ocean model in order to determine uncertainty. In addition, uncertainties related to error in mixed layer depth input from two different sources, instantaneous and climatological, were evaluated to understand their impact on zooplankton dynamics along the mesoscale circulation features such as eddies and filaments. This study demonstrates how remotely-sensed synoptic observations of zooplankton distribution may provide a means of assessing the role of fronts and eddies in driving trophic transfer in the ocean.