Global ocean emission of dimethylsulfide predicted from biogeophysical data

Abstract

[1] Among the biosphere-atmosphere interactions that influence climate, the emission of dimethylsulfide (DMS) from the ocean plays a prominent role for its high potential in cloud albedo regulation. In order to advance in our understanding and quantification of this coupled ocean-atmosphere system, both synoptic and predictive capabilities must be largely improved. Hitherto, large-scale oceanic DMS has eluded being captured from remote sensing, correlated with synoptic variables, or simulated by mechanistic modeling. We have found a simple empirical relationship that permits global-ocean monthly distributions of DMS concentration to be computed from a combination of remotely sensed biospheric data (chlorophyll a) and climatological geophysical data (mixed layer depth). This relationship allows for the desired synopticity and predictability in the ocean-toatmosphere sulfur flux, which we have globally quantified as 23-35 Tg S yr À1 . Also, our algorithm stands in support of a biogenic-DMS/solar-radiation negative feedback and opens the door toward quantifying its strength and its response to global warming.

Long-Term Changes of Greenhouse Gases in the Ocean and Their Feedback Effects on the Climate

“…We have compared our DMS algorithm with a previous study's algorithm, which yielded results being the most similar to the observational DMS (Simó and Dachs 2002). The algorithm of Simó and Dachs (2002) was based on the ratio of Chl and the surface mixed layer depth with a difference of 0.125s q from the surface density (MLD, m).…”

“…The algorithm of Simó and Dachs (2002) was based on the ratio of Chl and the surface mixed layer depth with a difference of 0.125s q from the surface density (MLD, m). Thus, we here used the same data set of Chl as used to construct our DMS algorithm (see Subsection 3.2.2).…”

“…However, Belviso et al (2004a) demonstrated that there were large uncertainties and spatial differences of sea surface DMS between these studies. Especially, in the North Pacific, even the algorithm of Simó and Dachs (2002) yeilding the results of being the most similar to the observational sea surface DMS, gave values under half of the observational data.…”

“…The error bars indicate the standard errors of averaged values (SE). (b) Same as panel (a), but calculated from the algorithm of Simó and Dachs (2002) (R 2 = 0.19, RMSE = 0.96 (or ±2.6 nM), n = 504). The algorithm was based on the ratio of Chl and the surface mixed layer depth with a difference of 0.125 s q from the surface density (MLD, m).…”

“…Thus it is important to understand the spatiotemporal distribution of DMS flux for predicting future global climate change (e.g., Andreae 1990). To understand the global oceanic distribution of DMS, several attempts have been made to parameterize the global spatiotemporal distribution of sea surface DMS using oceanic climatological parameters (chlorophyll-a (Chl), light intensity (I), nutrients (N), surface mixed layer depth (MLD)) or using ocean models (Anderson et al 2001;Simó and Dachs 2002;Aumont et al 2002;Chu et al 2003;Belviso et al 2004b). However, Belviso et al (2004a) demonstrated that there were large uncertainties and spatial differences of sea surface DMS between these studies.…”