Simulating vertical phytoplankton dynamics in a stratified ocean using a two-layered ecosystem model

Abstract

Phytoplankton account for around half of planetary primary production and are instrumental in regulating ocean biogeochemical cycles. Around 70 % of the oceans is characterized by either seasonal or permanent stratification. In such regions, it has been postulated that two distinct planktonic ecosystems exist, one that occupies the nutrient-limited surface mixed layer and another that resides below the mixed layer in a low-light, nutrient-rich environment. Owing to challenges observing the planktonic ecosystem below the mixed layer, it remains largely unexplored. Consequently, it is rarely characterized explicitly in marine ecosystem models. Here, we develop a simple, two-layered box model comprised of an ecosystem (nutrient, phytoplankton, and zooplankton - NPZ) in the surface mixed layer and a separate one (NPZ) in a subsurface layer below it. The two ecosystems are linked only by dynamic advection of nutrients between layers and controls on light attenuation. The model is forced with surface light (modelled from the top of the atmosphere) and observations of mixed layer depth. We run our model at the Bermuda Atlantic Time-series Study (BATS) site and compare results with a time series of more than 30 years for phytoplankton and nutrient observations. When compared with observations, the model simulates contrasting seasonal and interannual variability in chlorophyll in the two layers, reproducing the observed trends post-2011. A shoaling mixed layer post-2011, driven by ocean warming, increases light availability in both layers, which alters surface phytoplankton physiology while increasing subsurface phytoplankton biomass. Results lend support to the hypothesis that the euphotic zone of stratified systems can be described using two vertically separated planktonic ecosystems. Nevertheless, simulating the ecosystem in the subsurface layer was more challenging than the ecosystem in the surface mixed layer as less is known about model parameters and processes due to a lack of measurements, suggesting that more work is needed to study controls on subsurface planktonic communities.