The water balance in the very shallow subsurface can be influenced by capillary rise due to a high groundwater table. Since moisture content is an important factor for the thermal conductivity of soils, this can also have an influence on the very shallow geothermal potential (vSGP). For this reason, the effect of spatial and seasonal variations in groundwater tables on moisture content in essential depth layers was investigated at a large-scale geothermal collector system (LSC) in Bad Nauheim, Germany. Quasi-one-dimensional simulations using the FEFLOW^® finite-element simulation system were employed to determine site-dependent and seasonally varying moisture contents, from which thermal conductivities were derived. The model setup was previously validated based on recorded moisture contents. The simulations resulted in groundwater-related maximum seasonal and spatial differences in thermal conductivity of 0.14 W/(m∙K) in the LSC area. Larger differences of up to 0.21 W/(m∙K) resulted for different soil textures at the same depth due to different thermal properties. The results indicate that an efficient design of LSCs requires a sufficiently detailed subsurface exploration to account for small-scale variations in grain size distribution and groundwater level.