The efficiency of heat transfer in the outer shell of icy satellites is important to determine the evolution and thermal state of their interior with major implications for the cooling behavior of an internal ocean. In this study, we systematically investigate thermal convection in the ice shell of Europa using an Arrhenius viscosity and accounting for ice I material that is dependent on both grain size and strain rate. To this end, we employ the geodynamical code GAIA [1] with a mixed rheology approach similar to [2], and perform calculations in a 2D Cartesian box and spherical annulus geometry for two values of the ice shell thickness (i.e., 30 and 70 km). In our simulations, we test various constant grain size values. In a first serie of simulations, we tested the importance of the dislocation creep mechanism for modeling convection in Europa’s ice shell. Our results show that, in a mixed diffusion-dislocation creep rheology, diffusion creep is the dominant heat transfer mechanism, similar to the study of [3]. A pure dislocation creep rheology leads to a conductive ice shell. Dislocation creep may become dominant if its rheological prefactor increases by about 5 orders of magnitude, which even taking into account the uncertainty associated with rheological measurements is considered unrealistic. Additional simulations that use a mixed diffusion-basal slip rheology show that for ice shells, basal slip may be a relevant deformation mechanism in addition to diffusion creep. Another important aspect is that the efficiency of heat transfer is larger for a thick ice shell (70 km, compared to a thinner one (i.e., 30 km)). However, the dimensional surface heat flow obtained for a thin ice shell is larger than for a thicker one. This is caused by the rescaling of non-dimensional parameters to a dimensional heat flow. References: [1] Hüttig et al., PEPI 2013; [2] Schulz et al., GJI 2019; [3] Harel et al., Icarus 2020.