Desert Tortoise surveys and management in the Sonoran Desert typically exclude intermountain valleys, where tortoises are generally thought to be absent. Furthermore, few regional-level surveys have been conducted because of difficulties in sampling in the complex landscape of Sonoran Desert upland habitat. We used distance-sampling to document macrohabitat use and regional density of Desert Tortoises across the 76,800-ha Ironwood Forest National Monument in Arizona. We observed 42 tortoises on transects, and distance sampling produced an estimate of 17,997 tortoises ≥ 150 mm carapace length (coefficient of variation = 41.5%) on the monument. Stratification by habitat type (steep topography with boulders, incised washes, or neither of these components) improved precision slightly (37.2%). Detection probability contributed least to density variance, compared to encounter rate and tortoise detectability on the transect line, indicating that assumptions of the technique were met during sampling. We found tortoises or their sign on 92% of transects in boulder habitat, on 71% that included incised washes, and on 25% in habitat with neither of these features (up to 1.7 km away from the nearest slope). Our results indicate that Desert Tortoises in the Sonoran Desert occur at low density, but are not absent, from intermountain valleys, and the maintenance of these valleys for tortoise movement between local populations may be important for long-term population viability. With this in mind, concentrating survey effort in areas with steep topography and boulders will increase tortoise encounter rate, result in better precision of regional density and trend estimates, and may also reduce survey effort.