Cold water from the depths of the ocean has long been known to come to the surface in the Pacific, along Peru's coast, for example. Now scientists are learning new lessons about upwelling in the Atlantic, off the coast of New Jersey.
by S. M. Glenn, M. F. Crowley, D. B. Haidvogel, and Y. T. Song, Institute of Marine and Coastal Sciences, Rutgers University, New Brunswick, N.J.
For three summers, researchers offshore of Tuckerton, New Jersey, have observed coastal upwelling in action. Each year, real-time satellite images of wind-driven upwelling events trigger shipboard surveys to study the structure of the upwelled water just below the ocean's surface and its relationship to bottom dissolved oxygen concentrations (see figure (1)). Recurrent upwelling centers develop on the downslope sides of topographically high areas of the seafloor, in the same locations as previously observed regions of low dissolved oxygen. Recent modeling suggests that because of variations in the depth of the water along shore, a cyclonic eddy in the upwelling center forms that may concentrate nutrients, leading to phytoplankton blooms and reductions in dissolved oxygen as the organic material decays below the thermocline.
(1) Satellite-derived sea surface temperature image for July 27, 1995, showing the midsummer 1995 upwelling event along the southern New Jersey coast. The temperature section in the upper left corner is derived from a concurrent temperature survey (black line) through the Mullica River upwelling center.
During the summer of 1976, the New Jersey continental shelf was devastated by a severe low dissolved oxygen event, resulting in over $550 million in estimated potential losses to the shell fishing and related industries. This prompted many research programs to begin monitoring dissolved oxygen concentrations off the Jersey shore. The researchers found that there is a correlation between the strong summer thermoclines that isolate the bottom water, upwelling events that provide nutrients, and the subsequent late-summer reductions in bottom dissolved oxygen. They also found that the low dissolved oxygen events were clustered offshore of selected inlets and estuaries.
Four regions of recurrent low dissolved oxygen were identified offshore. They are the Hudson-Raritan estuary (A), Barnegat Inlet (B), the Mullica River estuary (C), and Townsend-Hereford Inlets (D) (see figure (2)). While early results suggested that additional nutrients from inlets and estuaries control the locations of recurrent low dissolved oxygen, there are vast differences in these New Jersey watersheds. The Hudson and Raritan Rivers are polluted, while nearly the entire watershed of the pristine Mullica River is protected by the New Jersey Pinelands National Reserve. Barnegat Inlet is the entrance to a long bay fed by several rivers, while Townsend and Hereford inlets serve several small bays that no major rivers feed. Finally, there is no region of recurrent low dissolved oxygen offshore of the Great Egg Harbor River estuary ((E) in figure (2)), which drains a relatively unspoiled area similar to the Mullica.
(2) The New Jersey coast, with tributaries and ocean depths shown. The gray shaded areas identify the four historical regions of recurrent low dissolved oxygen. Letters designate specific inlets and estuaries discussed in the text. White lines indicate rivers.
Under the initial hypothesis that regions of recurrent low dissolved oxygen are related to the three-dimensional structure of upwelling features, several coastal upwelling events were documented off New Jersey each summer between 1993 and 1995. The upwelling response to winds from the southwest usually begins with the offshore transport of the warm surface water and its replacement by a narrow, fairly uniform cold band along a section of the New Jersey coast. This forms a strong temperature front along the offshore side of the upwelled water. Subsequent evolution of the upwelling front occurs as illustrated in the figure.
After 3 to 4 days of persistent southwesterly winds, an approximately 50-km-long wavelike pattern develops along the front, and the near-shore band of upwelled water forms a series of isolated cold surface patches. Each year, these upwelling centers are located off the same inlets and estuaries previously identified as regions of recurrent low dissolved oxygen. Shipboard temperature surveys across the Mullica River upwelling center reveal the same complicated thermal structure below the surface each year. Rather than a single upwelled bulge of cold water in the center of the feature, the coldest upwelled water is split into one bulge along the offshore front and a second along the shore. Wind forcing following the midsummer upwelling events differ from year to year. After the 1993 upwelling event and its phytoplankton bloom, the winds reversed, causing a strong thermal cap to form as the warm surface water was transported back toward the shore. Offshore of the Mullica estuary, dissolved oxygen levels below the thermocline plummeted and remained well below normal until Hurricane Emily thoroughly mixed the region and returned oxygen values to their normal levels in September. In August of 1994, however, the midsummer upwelling event was immediately followed by a northeaster. In 1995, the upwelling was followed by a series of hurricanes. In both cases, the storms thoroughly mixed the layers of water beneath the surface, the late-summer warm water cap never formed, and bottom dissolved oxygen levels remained elevated.
As the observations above attest, locations of recurrent low dissolved oxygen may be related to the development of recurrent upwelling centers in which nutrients are concentrated independent of their source. But what causes the upwelling centers to recur in approximately the same locations year after year? The bathymetry illustrated in The figure (2) may hold the answer. Three topographic highs are observed along the southern New Jersey shore spaced approximately 50 km apart in the alongshore direction. The three upwelling centers in the figure (1) are located on the northern sides of these topographic highs.
To examine the influence of this alongshore topographic variation on coastal upwelling, a model was used that incorporates typical wind and temperature values common to the Jersey shore in summer. Three days into the model run, a broad coastal jet is meandering northward with a cyclonic eddy forming on the downstream side of each topographic high. The complicated structure observed in the temperature survey is simply the subsurface thermal signature of this eddy. Warm water on the offshore side of the upwelling front is transported downwind, but the cold water within each upwelling center is temporarily trapped by the eddy. Preliminary model results suggest the new hypothesis: the topographically controlled upwelling centers concentrate nutrients and phytoplankton, but subsequent depletion of the bottom dissolved oxygen depends on the timing of the late summer mixing storms.
Source: Eos, June 18, 1996, p. 233.
I grew up in Avon, Connecticut; majored in geology and mechanical
engineering at the University of Rochester; earned a Ph.D. in the Massachusetts Institute of
Technology and the Woods Hole Oceanographic Institution Joint Program in Oceanography; and
pursued post-graduate work in ocean forecasting at Harvard University. My initial interest in
science was sparked in the 8th grade by the teachers at the Talcott Mountain Science Center who
guided me through a research project on the acquisition and interpretation of images from one of
the earliest weather satellites. I am currently an oceanography professor at Rutgers University
where my research interests focus on the development of new satellite remote sensing, shipboard
and sub-sea observation techniques and their combination with state-of-the-art mathematical
models for coastal ocean forecasting.
Back to Science for Everyone
