The variability of currents and sea levels in the upper Delaware estuary are examined based on measurements from bottom mounted acoustic Doppler current profilers (ADCP) deployed at two sites (New Castle and Tinicum) from 18 March to 10 June 2003. New Castle is located 104 km from the mouth, and Tinicum is located another 32 km up-estuary. Supplemental data, including sea level at the mouth of the estuary, river discharge, and wind speed and direction, were also obtained from various federal agencies. The instantaneous current represents a superposition of variability driven by the tide, wind, and river discharge. Over the short (<36 hr) time scale, the tide is the dominant forcing mechanism, with M2 being the principal tidal constituent. The amplitude of the M2 tide increases from the mouth to the upper estuary and gives rise to a vigorous M2 current of the order 80 cm s–1. On time scales of 36 to 120 hr, the effect of wind drives a weak subtidal current with a standard deviation of 2 cm s–1 in the upper estuary. At time scales longer than 120 hr, the subtidal current variability, with a standard deviation of 6 cm s–1, is dominated by the barotropic response of the upper estuary to variations in the river discharge. The upper estuary exhibits a strong down-estuary mean current of the order—15 cm s–1. At Tinicum, river discharge accounts for more than half of the mean current, which is characterized by down-estuary flow throughout the water column. The magnitude of the river discharge-induced mean current is reduced at New Castle, in direct response to the down-estuary increase in the cross-sectional area. Tidally rectified current accounts for the remainder of the overall mean flow at Tinicum, and the effect of tidal rectification may be more important than river discharge in producing the mean flow at New Castle. There is no evidence of a baroclinic gravitational circulation, as the salt intrusion generally does not extend into the upper estuary.