Multivariate regression methods are applied to measurements of accumulation covering much of the interior of the Greenland ice sheet to evaluate the important factors that describe the current distribution of accumulation. Predictor variables considered in the regressions are geographical coordinates and three independent factors describing the geometry of the ice sheet. The results indicate that most of the variance in the data is explained by the combined effect of large-scale atmospheric circulation and ice sheet topography. This finding implies that climate change scenarios in which changes in accumulation are mostly associated with changes in temperature or some other parameter may only be correct if the pattern of atmospheric circulation remains unaltered. Comparison with values predicted with a precipitation retrieval model is favorable, suggesting that the model captures the most important features of Greenland precipitation.