Variability of DMS in the East China Sea and its response to different ENSO categories

Abstract

Dimethylsulfide (DMS) produced by marine phytoplankton is the most important volatile sulfur compound in ocean surface seawater and has a distinct contribution to global climate change. To assess DMS-climate feedbacks, the interannual variations in seawater DMS concentrations in the East China Sea (ECS) were examined, and their periodic characteristic patterns related to different El Niño-Southern Oscillation (ENSO) categories were discussed. Based on satellite surface seawater temperature (SST) and chlorophyll-a (Chl-a) datasets, a 23-year (1998–2020) dataset of DMS concentrations in the ECS was hindcasted by using a generalized additive mixed model (GAMM). The ENSO events from 1998 to 2020 were selected and divided into persistent, recession and transformed types according to their development process. The variability of surface seawater DMS concentration in the Yangtze River Estuary and its adjacent seas, the coast of Zhejiang and Fujian, and the coast of northern Jiangsu are more strongly influenced by ENSO events than the remainder of the ECS. Furthermore, ENSO alters the seasonal variability in DMS concentrations in the Yangtze River Estuary. The DMS concentrations of Apr-Jul were reduced (enhanced) by approximately 3.05 (0.61) nmol/L during Persistent El Niño (PEN) (Persistent La Niña (PLN)) events, which were reduced by approximately 0.51 and 1.10 nmol/L during recessionary El Niño (REN) event and recessionary La Niña (RLN) event, respectively. However, the DMS concentrations were enhanced by approximately 0.53 (1.29) nmol/L during El Niño-La Niña transition (ELT) (La Niña-El Niño transition (LET)) events. ENSO events affected the DMS concentrations in the surface seawater of the study area by influencing Chl-a and SST, where the short-period interannual variability (2–3 years) of DMS was mainly affected by SST, while the interannual variability of the long period (6 years) was mainly affected by Chl-a. The findings can be applied to the development of DMS concentration prediction methods.