Oxygen deficiency negatively affects crop yield and has a major impact on soil biological activity. This work was performed to ascertain the factors affecting the O-2 concentration in a soil profile, and to analyze the hypothesis used in O-2 transport models. Oxygen concentration was measured using platinum microelectrodes. Soil porosity, water content, temperature, groundwater depth, and microbial respiration were analyzed as the driving variables affecting O-2 concentration. Soil respiration was well described by a Michaelis kinetics; the maximal rate of respiration for the 0- to 0.1-m layer (2.5 X 10(-5) mol m(-3) s(-1)) was five times higher than for the 0.3- to 0.5-m layer. This implies that the simplification of vertical homogeneity and zero-order kinetics for O-2 consumption are invalid to estimate O-2 transport. In the topsoil, the amplitude of the changes in O-2 concentration after rainfall decreased in the order autumn > winter > summer. At 0.2 m for rainfall of approximate to 40 mm, the O-2 concentration decreased 0.09 m(3) m(-3) within 3 d after rainfall in autumn, and 0.05 m(3) m(-3) within 6 d in winter. The fluctuations observed in O-2 concentration after rainfall indicated that the steady-state approximation used in models is too restrictive to describe O-2 transport in soil. In the subsoil, O-2 concentration varied slightly due to low O-2 consumption (approximate to 10(-7) mol m(-3) s(-1)). Oxygen concentration in the topsoil was negatively associated with the water-filled pore space, except for measurements taken immediately after rainfall >30 mm d(-1), which indicated the presence of entrapped air. This suggests that the hypothesis of instantaneous equilibrium between water content and soil O-2 concentration is a reasonable simplification except for the 12- to 24-h following high-intensity rainfall.