More than half of turtle species worldwide are threatened because of habitat loss, invasive species, environmental pollution, disease, unsustainable use, and global climate change. However, some turtles are capable of existing in highly modified habitats, including structures designed to benefit human populations such as reservoirs and canal systems. Examining turtle distributions in large canal systems can inform conservation plans protecting turtle populations within a potential reservoir network and expand our understanding of underlying mechanisms regulating populations. We conducted spatial capture–recapture on turtles inhabiting sections of the Chesapeake and Ohio Canal. We developed a Bayesian spatial capture–recapture model to estimate densities, sex ratios, and associated capture probability parameters for Chrysemys picta (Painted Turtle), Chelydra serpentina (Common Snapping Turtle), Sternotherus odoratus (Eastern Musk Turtle), and Pseudemys rubriventris (Red-Bellied Turtle) captured at 12 sites along 28 km of the canal. We examined the impact of canal depth and forest cover on population densities and the variation in capture probability between sites and sampling days. We found population densities to vary between sites and the associated sex ratios to vary between species, as did the effect of depth and forest cover. Overall capture rates decreased each day, but there was trap-happy behavior from all species except S. odoratus. Our information can set a baseline for understanding turtle populations and inform management in the Chesapeake and Ohio Canal National Historical Park. It is also one of the first studies to establish methods for using new spatial capture–recapture to quantify densities and aquatic space use of turtles.