AbstractOne of the challenges in evaluating and applying regional climate models (RCMs) is the nonlinear behavior of atmospheric processes, which is still poorly understood. The nonlinearities induce chaos which leads to an internal variability in the model. Therefore, an ensemble of RCM simulations has been run and a budget study for potential temperature has been applied to investigate the internally generated variability. Hence, the physical processes associated with diabatic and dynamical terms inducing the intermember variability have been analyzed. The study is applied over the Arctic on an ensemble of 20 members, differing in their initial conditions, simulated with the RCM HIRHAM5 during summer 2012. This time period is of particular importance because of the melting sea ice and its influence on atmospheric circulation and the resulting effect on the intermember variability. The amplitude of the intermember variability of the simulations fluctuates strongly both temporally and spatially. During the beginning of August 2012 the intermember variability is strongest and coincides with the great Arctic cyclone event. The most important contributions for the intermember variability tendency are the horizontal and vertical “baroclinic” terms. Both terms have largest absolute values along the coastlines of the Arctic Ocean which are associated with the Arctic frontal zone leading to the cyclone maximum over the Arctic Ocean during summer.