Previous observations with a bottom-mounted, radially scanning sonar (BAMS) at 40 kHz suggested that macrofaunal activities influence low-angle, acoustic backscatter from seafloor sediments. In order to test that possibility experimentally, we measured and modeled time series of backscatter strength at both 40 and 300 kHz prior to manipulation and then introduced several macrofaunal species at known abundances to randomly selected locations within the ensonified area. We worked in West Sound, Orcas Island, Washington, at a water depth of 20.4 m and for the more frequently recorded 40-kHz series extracted effects by the time-series method known as "intervention analysis," wherein the intervention was the experimental alteration. We observed increased backscatter from patches of the small protobranch bivalve Acila castrensis, and of the cockle Clinocardium nuttali, from bait used as chum for fishes and crabs, and from tethered crabs (Cancer magister); other treatments showed no significant change. All of the effective treatments involved increased backscatter at 300 kHz from animals that have obvious hard parts or air bladders. Power calculations for intervention analysis and geoacoustic modeling suggest that failure of other treatments to show significant effects on backscatter strength stems from the small size of the organisms and structures used relative to the 40-kHz wavelength (3.7 cm) and to low sound-speed contrasts between surficial sediments at this site and overlying water (at both frequencies), producing low backscatter levels from both volume heterogeneity and surface microtopography. This experiment demonstrates, however, that low-angle acoustic backscatter can be used to observe at least some populations of benthic animals over a large area (ca. 8000 m2) and that intervention analysis can be a useful tool where logistics permit repeated observation but few or no spatial replicates—frequently the case in ecological manipulations.