Changes in atmospheric pressure continuously ventilate soils and snowpacks. This physical process, known as pressure pumping, is a major factor in the exchange fluxes of H2O, CO2 and other trace gases between the soil and atmosphere. Thus models of pressure pumping are relevant to many areas of critical importance. This study compares the three principal models used to describe pressure pumping. Beginning with the fundamental physical principles and whether the flow field is compressible or incompressible, these models are categorized as linear parabolic (one model – compressible) or nonlinear hyperbolic (two models – incompressible). Using observed soil surface pressure data, measured vertical profiles of soil permeability and standard linear analysis and numerical methods, this study shows that nonlinear models produce advective velocities that are one to two orders of magnitude greater than those associated with the linear model. Incorporating soil temperature and moisture dynamics made very little difference to the linear model, but a significant difference in the nonlinear models suggesting that advective velocities induced by pressure changes associated with soil heating and moisture dynamics may not always be small enough to ignore. All numerical results are sensitive to the frequency of the pressure forcing, which was band-pass filtered into low, mid and high frequencies with the greatest model differences at low frequencies. Partitioning the pressure forcing and model responses helped to establish that mid-frequency weather-related phenomena (empirically identified as inertia gravity waves and solitons) are important drivers of gas exchange between the soil and the atmosphere.