One of the largest factors of uncertainty in the surface mass balance of the polar ice sheets is the amount of snow accumulation. Despite a large number of field campaigns, the entire spatial extent of the ice sheets cannot be covered by in-situ measurements at a sufficient spatial and temporal resolution.Because of independence of solar illumination and cloud cover, microwave remote sensing is a well-established method of gathering data from the Polar Regions and thus has the potential to fill the spatial gaps between snow accumulation field measurements. Due to the complexity of snow-microwave interaction, large uncertainties still exist in the methods developed to retrieve snow accumulation from microwave remote sensing.This study will present a coupled snow microstructure/radiative transfer model to examine the interaction between electromagnetic radiation in the microwave frequency region and the snow volume in the dry snow zones of the polar ice sheets. Data from active and passive microwave sensors at different frequencies and resolutions are combined with snow microstructure data to verify the model and evaluate its sensitivity to different snowpack parameters.Theoretical assumptions made in earlier snow accumulation retrieval algorithms are reviewed using snow microstructure data measured at different firn cores. The description of snow densification and grain growth processes is refined on the basis of available field data. The radiative transfer model is improved to consider dense-medium effects in order to describe the microwave interaction with the snow pack in a more realistic fashion.The aim of this work is to establish a more robust method to estimate snow accumulation in the dry snow regions of the polar ice sheets.