Natural and human-induced spatial gradients provide a useful vehicle with which to better understand diverse marine processes. On the Palos Verdes margin (S. California), historical and ongoing waste-water discharge has created an along-shelf gradient in organic C and total N, as well as various trace metals and other pollutants (e.g., DDT). To better understand the impact of such pollution on bioturbation and to develop a more general understanding of the controlling factors of sediment bioturbation intensity, a series of stations representing severely, moderately and negligibly impacted sediments was studied. Vertical profiles of the naturally occurring radionuclide, 234Th, as well as the abundance and species composition of macrofauna were measured from box cores collected at three sites during July 1992. During a March 1993 cruise, radionuclide profiles were collected at an additional eleven sites on the margin. Excess 234Th profiles are, in general, consistent with a steady-state model that balances vertical biodiffusive mixing with radioactive decay. Biodiffusivities determined from the 234Th profiles yield a spatial pattern in which sediments near the outfall are mixed at intensities of ≈10 cm2/yr, and bioturbation intensities are five times as rapid at sites 5–7 km from the outfall. Average mixing intensities are between these extremes (28 cm2/yr) at a nearby unimpacted site. Despite the overall consistency of this pattern the reasons behind it remain unclear. Structural aspects of the macrofauna either do not vary between the three intensively studied stations (e.g., depth distribution, size) or do so in a manner that would suggest an opposite effect on the biodiffusivity (e.g., abundance). There is also little variability in trophic groupings along the enrichment gradient. Behavioral modifications, such as: (1) sublethal pollution effects caused by elevated contaminant (e.g., organic carbon and DDT) concentrations, and (2) inhibition by a tube-building polychaete, Mediomastus sp., are postulated to suppress mixing intensities near the outfall. The results of this study suggest that, at least in shallow-water settings, the general controls of bioturbation intensity are still poorly understood.