Aerial surveys are commonly used to evaluate in-water sea turtle abundances. A correction is applied to account for turtles diving below the observed surface. Historically, observations of summer/fall surfacing behavior were used for this correction, assuming constant behavior among seasons. Using radio/acoustic telemetry, seasonal differences in sea turtle surfacing behavior were determined among Kemp's ridleys and loggerheads. Mean time spent at surface in the spring ranged between 9.9%-30.0% with significant differences among individuals. Observed surfacing times were higher than historic summer/fall observations (Byles 1988; 5.3%), indicating that historic springtime abundances were overestimated by 50%-80%. Aerial surveys (2001-2004) indicated a 65%-75% decline in the Chesapeake Bay sea turtle population since the 1980's. Current sea turtle estimates range between 2,500-5,500 turtles compared to 6,500-9,000 turtles observed in the Lower Bay alone in the 1980's.