Use of small arms during training is an important activity associated with the development and proficiency of soldiers. These weapons traditionally have used copper-jacketed lead projectiles; the copper facilitates the oxidation of the metallic lead resulting in more mobile oxides and carbonates. Consequently, many ranges at installations have high soil concentrations of lead. Many of these ranges are no longer used and have become habitat for wildlife. To address the potential for adverse effects from lead exposure in songbirds, we compared the outputs of traditional deterministic exposure models with a spatial model and compared the results of both with blood-lead levels from songbird species at two small arms range complexes. An integrative data collection procedure was used and incorporated into the spatially explicit exposure model (SEEM) for two small arms range sites. Site-specific data were used to refine model input parameters. These data included lead soil concentrations, analysis of lead concentrations in nestling food items, acid-insoluble ash content of feces (to estimate soil ingestion), location and mapping of singing males, and nest site location and characteristics. Territorial males also were spot-mapped to determine likelihood of breeding activity. Modeled estimates of risk were compared with blood and feather lead levels of adults and nestlings. Overall, edge species had higher blood-lead concentrations; however, most had concentrations below subclinical effect levels. Conventional deterministic methods produced risk estimates exceeding 10-fold the highest SEEM estimates. The spatially explicit exposure model provided good agreement with field observations and therefore produced more accurate risk estimates. The present study provides support for the application of spatial methods over conventional deterministic methods.