Perchlorate (ClO₄⁻) is a contaminant that occurs naturally throughout the world, but little is known about its distribution and interactions in terrestrial ecosystems. The objectives of this Amargosa Desert, Nevada study were to determine (i) the local-scale distribution of shallow-soil (0–30 cm) ClO₄⁻ with respect to shrub proximity (far and near) in three geomorphic settings (shoulder slope, footslope, and valley floor); (ii) the importance of soil, plant, and terrain variables on the hillslope-distribution of shallow-soil and creosote bush [Larrea tridentata (Sessé & Moc. ex DC.) Coville] ClO₄⁻; and (iii) atmospheric (wet plus dry, including dust) deposition of ClO₄⁻ in relation to soil and plant reservoirs and cycling. Soil ClO₄⁻ ranged from 0.3 to 5.0 μg kg⁻¹. Within settings, valley floor ClO₄⁻ was 17× less near shrubs due in part to enhanced leaching, whereas shoulder and footslope values were ∼2× greater near shrubs. Hillslope regression models (soil, R² = 0.42; leaf, R² = 0.74) identified topographic and soil effects on ClO₄⁻ deposition, transport, and cycling. Selective plant uptake, bioaccumulation, and soil enrichment were evidenced by leaf ClO₄⁻ concentrations and Cl⁻/ClO₄⁻ molar ratios that were ∼8000× greater and 40× less, respectively, than soil values. Atmospheric deposition ClO₄⁻ flux was 343 mg ha⁻¹ yr⁻¹, ∼10× that for published southwestern wet-deposition fluxes. Creosote bush canopy ClO₄⁻ (1310 mg ha−1) was identified as a previously unrecognized but important and active reservoir. Nitrate δ¹⁸O analyses of atmospheric deposition and soil supported the leaf-cycled–ClO₄⁻ input hypothesis. This study provides basic data on ClO₄⁻ distribution and cycling that are pertinent to the assessment of environmental impacts in desert ecosystems and broadly transferable to anthropogenically contaminated systems.