NOAA PROFESSIONAL PAPER 11 



bottom at station LT4, there was little temperature re- 

 sponse below the thermocline to suggest mixing of D.O. 

 across the pycnocline. 



We conclude that despite strong winds and surface 

 waves hurricane Belle was too small and passed too rapidly 

 to have had a lasting effect on either the advection or 

 vertical mixing processes in the Bight. 



SUMMARY 



Our principal objective was to explore the differences 

 between currents observed in 1975 and in 1976 as a pos- 

 sible cause of, or contribution to, the important differ- 

 ences in D.O. concentrations observed during these 2 

 years. Insufficient flushing of the coastal waters could be 

 an important factor in the depletion of the D.O. reservoir. 

 A distinct difference was observed in the flow beneath the 

 pycnocline in the area of anoxia off New Jersey. The weak- 

 ened flow and reversal of direction in 1976 altered the 

 usual pattern of material transports in the Bight. Much 

 is still unknown about mechanisms of shelf circulation and 

 its variability, but one major influence certainly was the 

 period of southerly and southwesterly winds from about 

 J.D. 125-215 (early May through early August). Although 

 these winds were not particularly unusual on any given 

 day, their direction was persistently opposite the normal 

 southwestward flow of shelf waters, and their cumulative 

 effect over more than 2 months must have been consid- 

 erable . Similar reversals of surface currents off New Jersey 

 in summer were reported during the 1960s (Bumpus 1969). 

 Bumpus primarily attributed these reversals to the low 

 river discharge at the time. Anoxia or mass mortalities 

 were not associated with these events. 



A likely reason that the southwesterly winds evoked so 

 little response in water below the pycnocline off Long 

 Island is that the water is about half again as deep off 

 Long Island as off New Jersey. The pressure gradient 

 response to wind stress in shallow water is expected to be 

 nearly inversely proportional to depth, hence the shoaler 

 waters off New Jersey are expected to be more responsive 

 to local winds. Another possible cause of the weaker per- 

 turbation of the current off Long Island is the spatial pat- 

 tern observed in the wind stress (ch. 3). Wind stress was 

 generally parallel to the shelf contours off New Jersey, 

 but crossed the coast at a considerable angle off Long 

 Island. Csanady (1976) explained that the alongshore com- 

 ponent of wind stress is more important than the cross- 

 shelf component in its influence on currents. 



During the period of observations, upwelling occurred 

 primarily in association with the current perturbation off 

 New Jersey, but not off Long Island. Before this time, 

 however, upwelling occurred consistently off Long Island 

 and may have contributed indirectly to the low D.O.'s 



observed by advection of nutrients into the region. Not- 

 able here (clearer in ch. 8) is that during late spring and 

 early summer 1976, circulation below the thermocline was 

 favorable off New Jersey but not off Long Island for con- 

 centration of Ceratium tripos by the mechanism described 

 in chapter 9, part 1. The average onshelf flow (nearly 1.5 

 cm/s) observed off New Jersey during the 2 months before 

 initial discovery of the benthic mortalities corresponds to 

 a virtual displacement of about one-third the total shelf 

 width per month. This is enough to have swept a large 

 amount of these organisms from the outer shelf onto the 

 inner shelf off New Jersey. 



We infer from the relatively minor response of water 

 below the pycnocline, first, that even though hurricane 

 Belle's winds were intense, its rapid rate of passage was 

 such that its impact on vertical mixing below the ther- 

 mocline in the Bight was not especially significant. To the 

 extent that wind-induced mixing may result in asymmetric, 

 upward, turbulent entrainment of water across the ther- 

 mocline, rather than symmetrical vertical mixing, then 

 very little downward flux of oxygen into the bottom water 

 would occur under even more extreme conditions than 

 those of hurricane Belle. Second, we infer that reduced 

 vertical mixing as a result of anomalously low summer 

 wind speeds is not a likely cause of the low D.O. concen- 

 tration observed during some years. In the New York 

 Bight every summer the density stratification is sufficiently 

 strong that normal or even greater than normal wind-in- 

 duced mixing cannot effectively transfer D.O. across the 

 pycnocline. This implies that, once the summer stratifi- 

 cation has become established, the only physical mecha- 

 nism for oxygen renewal is by advection or by mixing 

 along isopycnal surfaces that typically slope upward in the 

 offshelf direction, but at very small angles. There is, of 

 course, a period of time in spring and again in autumn 

 when establishment or destruction of the pycnocline is 

 critically dependent upon the occurrence and timing of 

 atmospheric events. 



ACKNOWLEDGMENTS 



The officers and crew of the NOAA ships Researcher 

 and George B. Kelez, are largely responsible for the cur- 

 rent meter data used herein. D. Ortman, N. Larsen, and 

 K. Gallery also assisted in this work. 



REFERENCES 



Beardsley, R L., Boicourt, W. C and Hansen. D. V.. 1976. Physical 

 oceanography of the Middle Atlantic Bight, Amer. Soc. Limnot. 

 Oceanogr. Spec. Symp. 2:2()-34. 



Bumpus, D. F. , 1969. Reversals in the surface drift in the Middle Atlantic 

 Bight area, Deep-Sea Res. 16 (suppl): 17-23. 



162 



