Satellite-derived phytoplankton biomass dynamics in an equatorial Atlantic archipelago

Tropical oceans are typically oligotrophic, although physical mechanisms promote the upwelling of deeper waters, boosting phytoplankton biomass. Orbital remote sensing (RS) provides an alternative to infer phytoplankton biomass by estimating surface chlorophyll-a concentration (Chl-a), with unprecedented spatial and temporal coverage, especially in remote areas. The present work evaluated the spatial and seasonal surface patterns in phytoplankton biomass around the St. Peter and St. Paul Archipelago (SPSPA), based on Chl-a data estimated by the OLCI/Sentinel-3. In situ water sampling was conducted in 2022 to evaluate the RS sensors' performance and chemotaxonomic analysis. Estimates of Chl-a from 2020 to 2022 were used to comprehend spatio-temporal patterns in phytoplankton biomass in the oceanic region adjacent to the SPSPA using general linear mixed models. Two annual peaks in phytoplankton biomass were observed, a more pronounced one from June to August and a smaller one from November to January, suggesting that phytoplankton biomass around the SPSPA is linked to the seasonality of the large-scale pattern of the South Equatorial Atlantic Ocean. In situ chemotaxonomic analysis showed strong vertical variation in phytoplankton composition, with cyanobacteria dominating the surface (up to 90%) and haptophytes prevailing at depth. The general linear mixed model points out the significant effect of distance (off the archipelago) factor, sea surface temperature, mixed layer deep and the zonal currents velocity in estimating phytoplankton biomass in the region, indicating a localized IME process, governed by multi-level current–topography interactions, in regions closer to the SPSPA. • Surface phytoplankton biomass suggests an Island Mass Effect (IME) near SPSPA. • Sentinel-3/OLCI reveals a bimodal Chl-a pattern at SPSPA over a three-year period. • Model results suggest IME is governed by multi-level current–topography interactions. • OLCI sensor performance assessment for SPSPA shows small systematic errors for Chl-a. • Phytoplankton shift from cyanobacteria at the surface to haptophytes at the DCM.