This paper presents a numerical modeling approach for assessing the impacts of geologic heterogeneity on groundwater recharge estimates derived from environmental tracers. Common to many of the environmental tracer methods used to infer recharge in arid environments is an assumption of one-dimensional, vertical downward flow of water and solutes. However, in recent years there has been a growing recognition that fluid flux rates through geologic materials can be spatially variable owing to heterogeneities in porous media properties. Previous studies have suggested that local flow directions are dependent upon the stratigraphic and sedimentologic characteristics of the medium and may not be vertical even when application of water at the surface is spatially uniform and hydraulic gradients are vertical. Consequently, environmental tracer movement will also be spatially variable, and the one-dimensional assumption invoked to interpret unsaturated environmental tracer concentration profiles may be unrealistic. Two different numerical simulation experiments were performed to investigate the problems with using environmental tracers to infer recharge. The first involves simulation of flow and tracer transport through heterogeneous flat-lying media, and the second considers dipping layered anisotropic media. The results of these simulations show that recharge inferred using environmental tracer methods is also highly spatially variable and that recharge estimates obtained by tracer profiles tend to overestimate recharge and should be considered accurate only to within an order of magnitude, particularly in situations with significant media heterogeneity. Consequently, recharge estimates obtained from tracer profiles should be critically evaluated with regard to impacts of spatially variable flow.