Agroforestry and grass buffer practices are believed to reduce non point source pollution from row cropped watersheds, yet little is known about the processes and mechanisms involved. The objective of this study was to compare dynamics of soil water under row crop, grass buffer, and agroforestry buffer areas throughout the growing season to understand soil water use and recharge differences among the treatments. The study was conducted on two watersheds under a corn (Zea mays L.)-soybean (Glycine max (L.) Merr.) rotation with grass and agroforestry buffers at the Greenley Research Center, Knox County, MO. Soil water sensors were installed in 2003 in crop, grass, and agroforestry areas with six replications at 5, 10, 20, and 40 cm depths to record volumetric soil water at 10 minute intervals for four consecutive years. Initial soil moisture in the spring season was lower under tree and grass areas than crop areas probably due to water use by the permanent vegetation before crops were established. The differences were larger for shallower depths as compared to the 40 cm depth. The trend continued throughout the growing season. During rain events water content increased for all depths and treatments and the differences in water content among treatments diminished. At the end of the growing season soil water content increased when water use was low and rain occurred. Permanent vegetative areas tended to retain more water than crop areas during recharge probably due to improved soil properties. The results of the study suggest that establishment of grass and agroforestry buffers help reduce non point source pollution from row crop agriculture by improving infiltration and reducing runoff that can carry sediments, nutrients, and pesticides.