Recent observations have indicated rapidly increasing mass loss from the Greenland Ice Sheet. To explore the interactions and feedbacks of the ice sheets in the climate system, it is important to develop coupled climate-ice sheet models. The integration of an ice sheet model in a global model is challenging, and, currently, relatively few climate models include a two-way coupling to a dynamical ice sheet model. In this work package, we have continued developing the coupled ice sheet-climate model system comprising the global climate model EC-Earth and the Parallel Ice Sheet Model (PISM) for Greenland. The new model system, EC-Earth3-GrIS, is upgraded to include the recent model versions, EC-Earth3 and PISM version 1.2. In addition, a new module has been developed to handle the exchange of information between the ice sheet model and EC-Earth using the OASIS3- MCT software interface. The new module reads output from the ice sheet model and exchanges the fields with the relevant EC-Earth components. The ice sheet mask and topography are provided to the atmosphere and land surface components. The heat and freshwater fluxes from basal melt and ice discharge are provided to the ocean module via the runoff-mapper that routes surface runoff into the ocean. The new module also prepares the forcing fields for the ice sheet model, i.e., subsurface temperature and surface mass balance. These fields are calculated in EC- Earth3 using a land ice surface parameterization, developed explicitly for the Greenland ice sheet. The parameterization contains a responsive snow and ice albedo scheme and includes land ice characteristics in the calculation of heat and energy transfer at the surface. Experiments with and without the land ice surface parameterization have been carried out for preindustrial and present-day conditions to assess the influence of the surface parameterization on the calculated surface mass balance. The results show that the ice sheet responds stronger and more realistically to forcing changes when the new surface parameterization is used. Besides the model development, the results from experiments with the first model version, EC- Earth-PISM, have been analyzed. These results stress that a decent surface scheme with a responsive snow albedo scheme is necessary for investigating mass balance changes of the Greenland Ice Sheet. Overall, our results indicate that the feedbacks induced by the interactive ice sheet have a significant influence on Arctic climate change under warming conditions. In warm scenarios where the CO2 level is raised to four times the preindustrial level, the coupled model has a colder Arctic surface, a fresher ocean, and more sea-ice in winter.