Given current concern about the stability of ice sheets, and potential sea level rise, it is imperative that we are able to reconstruct and predict the response of ice sheets to climate change. The Intergovernmental Panel on Climate Change (IPCC), amongst others, have highlighted that our current ability to do so is limited. Numerical ice sheet models are a central component of the work to address this challenge. An unresolved key issue in this work concerns the volume and rate of ice mass loss needed to explain the large difference between late glacial and interglacial global sea levels. Some 20% of observed sea level rise since the Last Glacial Maximum (LGM) cannot be attributed to any known former ice mass, indicating that this inconsistency arises from the deficiencies in modelled reconstructions of ice sheet volumes and postglacial rebound. Ice sheet models are tested and refined by comparing model predictions of past ice geometries with field-based reconstructions from geological, geomorphological and ice core data. However, on the East Antarctic Ice sheet, Dronning Maud Land (DML) presents a critical gap in the empirical data required to reconstruct changes in ice sheet geometry. In addition, there is poor control on regional climate histories of ice sheet margins, because ice core locations, where detailed reconstructions of climate history exist, are located on high inland domes. This leaves numerical models of regional glaciation history largely unconstrained. MAGIC-DML is a Swedish-US-Norwegian-German-UK collaboration with a focus on filling the critical data gaps that exist in our knowledge of the timing and pattern of ice surface changes on the western Dronning Maud Land margin. Here we describe a series of high-resolution modelling experiments to help identify those areas across western Dronning Maud Land that are the most sensitive to uncertainties in the regional climate history and the choice of model parameters. For this we employ a wide range of climate and ocean histories combining published outputs of 18 general circulation models for the LGM and mid-Holocene with ice core records. The modelling results together with remote sensing mapping of glacial landforms is informing and guiding cosmogenic nuclide sampling campaigns in western Dronning Maud Land starting 2016/17. Successful integration of numerical modelling and field investigations in an iterative manner is key to achieving the anticipated outcome of the MAGIC-DML project, a reconstruction of the long-term pattern and timing of vertical changes in ice surface elevation since the mid-Pliocene warm period, which will provide the missing empirical data required to constrain numerical ice sheet models.