We used a daily time step, multi-layer simulation model of soil water dynamics to integrate effects of soils, vegetation, and climate on the recruitment of Bouteloua eriopoda (black grama), the historically dominant grass in the Chihuahuan Desert. We simulated landscapes at the Jornada ARS-LTER site with heterogeneous soil properties to compare: (1) a grass-dominated landscape in 1858 with the current shrub-dominated landscape (i.e., a change in vegetation structure), and (2) the current shrub-dominated landscape with future landscapes over a range of climate scenarios associated with the North American monsoon (i.e., a change in climate). A historic shift from high productivity grasslands to low productivity shrublands decreased simulated recruitment for most of the site; the amount of reduction depended on location-specific soil properties and changes in production. In some cases, soil properties interacted with vegetation structure: soils high in clay content maintained high recruitment even with a decrease in production. Wetter summers would increase recruitment in all vegetation types. Drier summers below 25% of current rainfall would decrease recruitment to negligible values (< 0.03) throughout the landscape. We used our results to identify the conditions where recruitment of B. eriopoda is possible with and without significant modifications to soil and vegetation.