Abnormal cartilage loading after injury is believed to be an important factor leading to post-traumatic ankle osteoarthritis. Due to the viscoelastic behavior of cartilage, it is possible to measure localized cartilage strains from changes in thickness following dynamic activities. However, there are limited data characterizing in vivo cartilage mechanics under physiological loading conditions in the healthy ankle. Therefore, the objective of this study was to directly measure in vivo cartilage strains in the healthy ankle joint in response to a dynamic hopping exercise. Ten healthy subjects with no history of ankle injury underwent magnetic resonance imaging before and after a single-leg hopping exercise. Bony and articular cartilage surfaces were created from these images using solid modeling software. Pre-exercise and post-exercise models were then registered to each other, and site-specific cartilage strains (defined as the normalized changes in cartilage thickness) were calculated at grid points spanning the articular surfaces. The effects of both location and exercise on strain were tested using a two-way repeated measures analysis of variance. We did not detect any significant interaction effect between location and exercise for either tibial or talar cartilage. However, hopping resulted in significant decreases in tibial (p<0.05) and talar (p<0.05) cartilage thicknesses, corresponding to strains of 3% and 2%, respectively. Additionally, pre-exercise cartilage thickness varied significantly by location in the talus (p<0.05), but not in the tibia. These strain data may provide important baseline information for future studies investigating altered biomechanics in those at high risk for the development of post-traumatic ankle osteoarthritis.