Sea-surface dimethylsulfide (DMS) concentration from satellite data at global and regional scales

• Published in: Biogeosciences
• Main Authors: Galí, Martí, Levasseur, Maurice, Devred, Emmanuel, Simó, Rafel, Babin, Marcel
• Other Authors: Natural Sciences and Engineering Research Council of Canada, Ministerio de Economía y Competitividad (España), Generalitat de Catalunya
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: European Geosciences Union 2018
• Subjects: Southern Ocean; North Atlantic
• Online Access: http://hdl.handle.net/10261/172890; https://doi.org/10.5194/bg-15-3497-2018; https://doi.org/10.13039/501100002809; https://doi.org/10.13039/501100000038; https://doi.org/10.13039/501100003329

23 pages, 9 figures, 4 tables, 1 appendix, supplementary material https://doi.org/10.5194/bg-15-3497-2018-supplement The marine biogenic gas dimethylsulfide (DMS) modulates climate by enhancing aerosol light scattering and seeding cloud formation. However, the lack of time-and space-resolved estimates of DMS concentration and emission hampers the assessment of its climatic effects. Here we present DMSSAT, a new remote sensing algorithm that relies on macroecological relationships between DMS, its phytoplanktonic precursor dimethylsulfoniopropionate (DMSPt) and plankton light exposure. In the first step, planktonic DMSPt is estimated from satellite-retrieved chlorophyll a and the light penetration regime as described in a previous study (Galí et al., 2015). In the second step, DMS is estimated as a function of DMSPt and photosynthetically available radiation (PAR) at the sea surface with an equation of the form: log10DMS Combining double low line α + βlog10DMSPt + 3PAR. The two-step DMSSAT algorithm is computationally light and can be optimized for global and regional scales. Validation at the global scale indicates that DMSSAT has better skill than previous algorithms and reproduces the main climatological features of DMS seasonality across contrasting biomes. The main shortcomings of the global-scale optimized algorithm are related to (i) regional biases in remotely sensed chlorophyll (which cause underestimation of DMS in the Southern Ocean) and (ii) the inability to reproduce high DMS • DMSPt ratios in late summer and fall in specific regions (which suggests the need to account for additional DMS drivers). Our work also highlights the shortcomings of interpolated DMS climatologies, caused by sparse and biased in situ sampling. Time series derived from MODIS-Aqua in the subpolar North Atlantic between 2003 and 2016 show wide interannual variability in the magnitude and timing of the annual DMS peak(s), demonstrating the need to move beyond the classical climatological view. By providing synoptic time series ...