Topography strongly regulates soil formation at the hillslope scale through its effects on sediment redistribution and biological activities. Spatially explicit land surface parameters (LSPs) such as slope and curvature hold potential for modeling the resulting soil carbon (C) and nitrogen (N) distributions, but their representation of deep soil profiles remains largely unexplored. In this study we examine relationships between deep soil profile C and N stocks and LSPs derived from a fine-resolution digital elevation model (DEM) on prototypical rolling hillslope catenas. Consistent with other studies we found that soil thickness was the primary controller of soil organic C and N stocks and was best predicted by mean curvature. Specifically, subsoil thickness, instead of A horizon thickness, explained variability of soil C and N on hillslopes. In addition, our results suggest that, along ridge to toeslope catenas, the processes mediating soil C and N distribution varied from convex to concave positions. Convex ridge positions appeared to favor processes that enrich soil profiles with high C and N concentrations despite their drier position, while concave hollow and toeslope positions favored cumulic processes, despite their conceptually moister conditions in which enrichment processes would be favored. Our data also point to slope aspect as a weak but potentially geomorphically important covariate in modeling soil thickness and C and N stocks using LSPs. Overall, LSPs of curvature and aspect explained 51% of the variability in soil thickness, while curvature and aspect explained 50% of the variability in soil organic C stocks. Our results suggest that diffusive sediment transportation likely exerts a first-order control on soil thickness and soil organic C and N stocks in many semi-arid landscapes. Our data also highlight the importance of subsoil in mapping soil C and N stocks and other soil properties. Quantitative modeling of soil C and N as in our study supports examination of additional ecosystem properties at fine spatial scales.