Space weather requires a fast and accurate modelling of magnetic and flow structures in the heliosphere to anticipate their impact on Earth spatial environment. In particular, it is well known that the position of the current sheet is a crucial information to determine the interaction with the Earth magnetosphere and anticipate ICMEs propagation. Because modelling the entire heliosphere is so challenging, the current approach is to combine coronal and heliospheric models, as is done in the EUHFORIA 2.0 project. This however leads to the open question of the transmission of uncertainties between the models, which is not clearly answered yet. In particular, at the beginning of the chain of modelling lies a crucial choice that is not always obvious: the choice of the input solar observations to provide the magnetic field boundary conditions at the solar surface. To this day, there is a great variety of sources with different treatments, especially at the poles to fill the currently missing observations. The impact of the synoptic map source has started to be discussed for PFSS models, but a clear overview of the consequences for MHD models and their description of the corona and the heliosphere is still missing.  

We present here the newly developed MHD coronal model for the EUHFORIA 2.0 project, based on the COOLFluiD framework. This model has the advantage of using an implicit solver for speed and an unstructured mesh for accuracy, especially around the polar region. After briefly presenting its benchmarking and validation procedure for its polytropic version, we will use it to explore how various synoptic maps can affect the simulation results. We select the date of 2nd of July 2019 because of the low solar activity and the associated solar eclipse seen on Earth. We use data from all currently available sources (GONG, GONG-ADAPT, WSO, MWO, SOLIS, HMI) and perform the same simulation with the same pre-processing and the same physical parameters, to assess only the impact of the choice of the input synoptic map. We focus on the implications for the magnetic field configuration by comparison with white-light eclipse pictures, for the coronal hole locations by comparison with SDO/AIA and for the position of the HCS at 0.1 AU by comparison with standard WSA models. We demonstrate that even at minimum of activity the input synoptic map has a great influence on the output of coronal models, and that the modelling of the poles is crucial for the shape of the HCS. We finally discuss the future developments of the model such as the inclusion of heating terms to model CIRs.