Assessing climate-driven variability in phytoplankton functional groups and dimethyl sulfur compounds from space in the northern Benguela upwelling system

Abstract

Marine phytoplankton form the base of oceanic ecosystems influencing food webs, biogeochemical cycling and climate dynamics. Different phytoplankton functional groups (PFGs) play distinct roles, contributing to varying climate and ecosystem functioning. For example, planktonic dimethyl sulfur compounds (DMSc), such as dimethyl sulfide (DMS) and its precursor dimethyl sulfoniopropionate (DMSP), play a key role in sulfur cycling, which influences cloud formation and albedo with significant climate feedback. The northern Benguela upwelling system (nBUS) is considered one of the most productive Eastern Boundary Upwelling Systems, with intense and regular phytoplankton blooms that sustain rich fisheries. It also features a semi-permanent stratocumulus cloud deck that strongly impacts radiative forcing and global climate. However, large-scale seasonal and multidecadal trends in phytoplankton biomass, functional composition and their role in DMSc production in response to climate change remains poorly studied. This knowledge gap is primarily due to limited data coverage from ship-based environmental monitoring efforts. This study develops and validates remote sensing algorithms for the nBUS and applies them to two decades (2003-2022) of daily satellite observations to examine the spatial and temporal variability in PFGs and DMSc in the context of environmental adjustments in sea surface temperature (SST) and equatorward winds (WEQ) that are associated with climate change. A multispectral remote sensing approach to detect the PFGs commonly observed in the nBUS, namely diatoms, dinoflagellates, flagellates, and coccolithophores, is presented. This is achieved using a large, microscopic phytoplankton dataset and coincident satellite matchups from the Moderate Resolution Imaging Spectroradiometer of aqua (MODIS-Aqua) to generate relationships between the spectral characteristics of in water constituents and community dominance of a specific PFG. The algorithm, validated with an independent dataset, is applied to two decades of daily satellite observations to reveal regional and seasonal characteristics and trends in the Spatial Extent and Frequency of dominant PFGs, with evidence of niche conditions favoring particular PFGs. PFG trends show distinct regional and seasonal variability. Diatoms exhibit positive trends in their Frequency and Spatial Extent in autumn, indicating prolonged dominance over time and an expansion of their spatial occupation. However, their Spatial Extent declines in summer. In contrast, dinoflagellates display increasing trends in Spatial Extent and Frequency in summer and, together with flagellates, shrink in Spatial Extent in autumn. Diatoms, as key carbon exporters, show the strongest trends, emphasizing their sensitivity to climate change with implications for carbon drawdown and energy transfer to higher trophic levels. The nBUS is impacted by general warming trends, whereas WEQ shows coherent offshore increasing trends and contrasting trends in alongshore patterns. Positive alongshore trends are observed off the Kunene and Central cells while negative trends are observed off the Northern and Lüderitz upwelling cells. The link between alongshore WEQ and diatom trends relative to upwelling cells strongly suggests an altered nutrient supply-driven response. That is, increasing alongshore WEQ trends aligned with increasing diatom trends (Kunene and Central cells) and vice versa for regions of decreasing WEQ (Northern and Lüderitz cells). Warming trends, which characterize the nBUS, are also likely to increase phytoplankton's ability to exploit events of enhanced nutrient availability. Declines in coccolithophores was evident offshore and linked to offshore accelerating WEQ.

Subsequently, a satellite-based DMSc detection algorithm using chlorophyll-a (Chl) and SST as predictors was developed (using in situ observations) and validated (using independent satellite matchups). This regional algorithm was applied to the same 20-year daily satellite datasets and facilitated the first spatially and temporally comprehensive assessment of the regional, seasonal and multidecadal trends in DMSc. The DMSc seasonal cycle in the nBUS displays a classical "summer DMS paradox", in which the DMSc peak in summer coincides with minimum Chl concentrations. Our results suggest that the seasonal cycle of DMSc is driven by phytoplankton community succession, with coccolithophores associated with the summer DMSc peak while the winter minimum in DMSc is associated with a maximum in Chl that is instead dominated by diatoms and flagellates. Annual trends in DMSc show a coherent increase offshore which contrasts with the heterogeneous alongshore trends. The increasing trend in DMSc offshore tends to be associated with warming and an expansion in the Spatial Extent of dinoflagellates whereas alongshore regions of declining DMSc are more aligned with declining Chl and diatoms relative to the Northern and Lüderitz cells.

Overall, this study advances our understanding of seasonal and long term PFG dynamics and their role in dimethyl sulfur cycling in the nBUS through the development, application and interpretation of novel satellite remote sensing algorithms. The results and insights gained from this study have the potential to advance environmental monitoring capabilities (through novel region-specific satellite algorithms) and enhance our understanding of large-scale phytoplankton community and dimethyl sulfur dynamics in the nBUS, one of the world's most productive and climatically sensitive marine ecosystems. The observed multidecadal trends in PFG and DMSc suggest potential shifts in ecosystem structure under a warming climate with altered winds and nutrient supply, with key implications for regional productivity, trophic interactions and climate regulation.