Climate change in the 21st century is expected to cause warming of the surface ocean,melting of the sea ice and stratification of the mid and high latitudes with unknown consequences for the nutrient cycles and the marine ecosystems.To understand how marine DMS fluxes will respond to climate change,we develop and evaluate a prognostic DMS model within the framework of a global Dynamic Green Ocean Model that includes three phytoplankton functional types and two zooplankton functional types.Our prognostic DMS model takes into account the different pathways from particulate Dimethylsulphonioproprionate (DMSPp) within the phytoplankton cells to the DMS emission at the sea-air interface.We test the effect of different parametrisations on the model results and the dependence of DMS emission on the phytoplankton functional type diversity and we study the coupling behaviour of DMS and chlorophyll-a,comparing the high latitudes with the tropics.Particular emphasis is given to the representation of the seasonality and spatial distribution of DMS and the light dependence of the processes involved in DMS production.The model reproduces the observed inverse relationship between mixed layer depth and DMS,which is primarily explained by the light dependence of the DMSP/C cell ratio of the individual plankton functional types and partly by the ecosystem composition.