Abstract. Salt structures and their surroundings can play an important role in the energy transition related to a number of storage and energy applications. Thus, it is important to assess the current and future stability of salt bodies in their specific geological settings. We investigate the influence of glacial loading and unloading on subsurface salt structures using physical models, based on the example of the Scandinavian Ice Sheet in northern Germany. Apparent spatial correlations between subsurface salt structures in northern Germany and Weichselian ice marginal positions have been observed before and the topic is a matter of ongoing debate. Recently described geomorphological features – termed surface cracks – have been interpreted as a direct result of ice sheet induced salt movement resulting in surface expansion. The spatial clustering and orientation of these surface cracks was so far not well understood, owing to only a limited number of available studies dealing with the related salt tectonic processes. Thus, we use four increasingly complex physical models to test the basic loading and unloading principle, to analyze flow patterns within the source-layer salt and within salt structures, and to examine the influence of the shape and orientation of salt structures with respect to a lobate ice margin in a three-dimensional laboratory environment. Three salt structures of the northern German basin were selected as examples that were replicated in the laboratory. Salt structures were initially grown by differential loading, and buried before loading. The ice load was simulated by a weight that was temporarily placed on a portion of the surface of the models. The replicated salt structures were either completely covered by the load, partly covered by the load, or were situated outside the load extent. In all scenarios, a dynamic response of the system to the load could be observed: while the load was applied, the structures outside the load margin started to rise, with a decreasing tendency with distance from the load margin and, at the same time, the structures under load subsided. After the load was removed, a flow reversal set in and previously loaded structures started to rise, whereas the structures outside the former load margin began to subside. The vertical displacements during the unloading stage were not as strong as during the load stage, and thus the system did not return to its pre-glaciation status. Modeled salt domes that were located at distance from the load margin showed a comparably weak reaction. A more extreme response was shown by modeled salt pillows whose margins varied from sub- parallel to sub-perpendicular to the load margin, and were partly covered by the load. Under these conditions, the structures showed a strong reaction in terms of strain and vertical displacement. The observed strain patterns at the surface were influenced by the shape of the load margin and the shape of the salt structure at depth, resulting in complex deformation patterns. These physical modeling results provide more evidence for a possible interplay between ice sheets and subsurface salt structures, highlighting the significance of three-dimensional effects in dynamic geological settings. Our results lead to a better understanding of spatial patterns of the surface cracks that were mapped at the surface above salt structures.