South Atlantic Bight (SAB) winds are typically well-correlated over large alongshore scales and closely coupled with coastal sea level (CSL) fluctuations. However, this correlation is greatly reduced when meteorological dominance is changing between the Ohio Valley High of winter and Bermuda-Azores High of summer. During the sping 1981 transition, strong persistent wind in the northern portion of the SAB (e.g., at Charleston) was highly coherent with local CSL (coefficient of determination, R2=0.70). In the southern SAB, however, a weak poorly organized wind regime was decoupled with local CSL. Daytona Beach CSL was uncorrelated with local alongshore wind, contrary to frictional shelf models, but was coherent with alongshore wind stress and CSL at Charleston (R2=0.74), lagging events in the north by ~15 hours. The results suggest that a CSL signal forced by wind in the northern SAB propagates southward with a phase speed of ~700 km d-1. Barotropic shelf wave theory for an exponential shelf profile representing SAB topography predicts a phase speed of ~500 km d-1. Thus the potential importance of shelf wave activity in the SAB during mixed wind conditions should be reassessed. Analysis of SAB wind-CSL relationships provides, for the first time, direct evidence of remote forcing on the broad shallow shelf between North Carolina and Florida. Previous observations in the SAB from other seasons have not afforded evidence of nonlocal propagations, suggesting that conditions appropriate for their observation have a strong seasonal dependence.