The Louisiana-Texas Shelf Physical Oceanography Program (LATEX) has as its principal objective the identification of key dynamic processes governing the circulation, transport, and cross-shelf mixing of the waters on the Texas-Louisiana shelf. Sponsored by the Minerals Management Service (MMS) of the Department of the Interior, LATEX is the largest shelf physical oceanography research project yet undertaken. MMS is responsible for managing federal mineral resources on the continental shelf. To meet this responsibility, the service needs to understand physical processes and circulation on the shelf and how they may affect the stability of structures and the transport of pollu-tants. In addition, shelf circulation data will be used in MMS's oil spill risk-analysis models.
The LATEX component at Texas A&M, LATEX A, has been carried out in three phases. From October 1991 through April 1992, we prepared and assembled equipment and personnel for the field work. From April 1992 through December 1994, we executed the field program, and until late 1997 we will interpret and synthesize our results.
The field phase was most challenging due to frequent maintenance needed by instruments moored offshore and the constant threat of theft or damage by seafaring vandals. Mother Nature added to the challenge-LATEX equipment registered the devastating passage of Hurricane Andrew in 1992, the record flooding of the MississippiAtchafalaya river system, and the explosive development of the "Storm of the Century" in 1993.
During the field phase, instruments deployed in an array of 33 moorings on the shelf recorded horizontal ocean currents, temperature, and conductivity, and the LATEX team conducted ten hydrographic surveys of the region. During each cruise, acoustic Doppler current profilers measured vertical profiles of horizontal currents as the ship moved over the shelf. These field measurements have since provided details about the vertical structure of shelf currents and revealed shelf-wide patterns of horizontal currents as a function of depth. On some cruises, surface drifters were released with parachute-like drogues to carry them with the water. These drifters reported their position via satellite and proved quite useful in tracking medium-sized ocean phenomena, such as ring-shaped ocean currents that detach from the Loop Current (a permanent feature in the eastern gulf) and transit the Gulf of Mexico.
Texas-Louisiana shelf circulation
Shelf researchers assume that three principal external mechanisms force general circulation over the Texas-Louisiana shelf. These are: wind direction and intensity, or wind stress; buoyancy effects that result when water discharged from rivers overlays higher-salinity seawater; and anticyclonic (clockwise) and cyclonic (counterclockwise) ring currents that interact with the continental slope.
In 1986, John Cochrane and Frank Kelly of Texas A&M University suggested that an annual cycle governs shelf-wide circulation. They attributed circulation over the inner shelf, where water depth is less than 50 meters, principally to the effects of wind stress. They proposed that when the wind is directed alongshore and downcoast (from the Mississippi River toward Brownsville, Texas), nearshore currents also flow downcoast. Likewise, currents flow upcoast in response to upcoast wind.
The alongshore component of surface wind (l0 meters above sea level) is generally downcoast except for a period during summer when it reverses. The early summer transition from downcoast to upcoast is generally completed by June along the middle to upper Texas coast and is characterized by numerous episodic reversals. Based on 30-year records that we examined from coastal weather stations, winds typically blow upcoast during July and August, then shift abruptly downcoast at the end of August. On average, they remain downcoast from September through May. There are numerous short periods during which the alongcoast direction of the wind stress is reversed relative to its average direction, and we have found that currents respond to such reversals quite rapidly (in less than 24 hours).
Considering this temporal pattern of alongshore wind stress, Cochrane and Kelly projected that the nearshore ocean flow should be downcoast from September through May, transition to upcoast in June, and remain upcoast during July and August. They projected these patterns by statistically analyzing differences in geopotential. Geo-potential is a measurement of gravity's ability to do work (in this case, move water) based on the distance between the surface and a specified reference plane.* Cochrane and Kelly calculated geopotential for various locations on the shelf using hydrographic data collected from 1963 to 1965 aboard the R/V Gus III and surface salinity data taken in 1964.
LATEX observations
Using current measurements from the LATEX A field study, we confirmed Cochrane and Kelly's projected wind effects and subsequent flow patterns over the inner Texas-Louisiana shelf. The LATEX database presently occupies more than two gigabytes of computer disk storage. To obtain meaning from such an agglomeration, specialized data products have been created from the database. These products include calculations and plots of mean currents and wind stress over the study area for each month, contoured plots of surface salinity, and contoured plots of differences in geopotential. Analysis of these data products is confirming the Cochrane-Kelly schema for shelf circulation off Texas and Louisiana and is illuminating the forcing mechanisms of wind, buoyancy, and off-shelf circulation.
Wind as a forcing mechanism on the inner Texas-Louisiana shelf
Plots of statistically analyzed LATEX current and wind-stress data illustrate the pattern of shelf circulation and the transition between the downcoast and upcoast regimes. Mean wind stress and current vectors for each month from April 1992 through November 1994 were produced by averaging two figures: measurements from 31 LATEX current meters suspended approximately 10 meters below the sea surface and Gulf of Mexico meteorological data. Statistical analyses then yielded fields of current vectors from those averages.*
We think the analyzed fields reasonably represent the mean observed behavior of currents and winds. In non-summer months, average wind stress over the inner shelf had a downcoast component everywhere north of about 27.5°N. Currents over the inner shelf likewise flowed down-coast. In summer months, average wind stress had an upcoast component over the entire inner shelf, and currents over the inner shelf likewise flowed upcoast.
Interannual variations occur in the summer pattern, depending on when the wind shifts from downcoast to upcoast. For example, a variation in the summer pattern was evident during August 1992, when the wind stress over the inner shelf shifted from upcoast to downcoast. The resulting average current field showed upcoast flow along the shore of Louisiana and downcoast flow along the shore of Texas.
In general, after the August transition in alongshore wind-stress direction, wind stress over the inner shelf is all downcoast north of about 27°N, as is coastal flow. Thus, the non-summer circulation regime is fully restored by September.
Buoyancy as a forcing mechanism on the inner shelf
We can determine the buoyancy effects of river discharge on the shelf circulation by comparing mean distributions of surface salinity and differences in geopotential for fall, spring, and summer. The large year-to-year variability of the Mississippi-Atchafalaya system discharge is illustrated by comparing daily discharge rates for the LATEX field years 1992, 1993, and 1994 with one another and with the 64-year average. Average daily discharges of 11 Texas rivers, based on data series ranging from 20 to 77 years in length, are quite small and are expected to have only minor effects on the general shelf circulation. Average Mississippi-Atchafalaya discharge peaks in April, but is still near maximum in May. It reaches a minimum in September. The period from July to August and the month of November appear to have similar daily discharge rates.
Using historical files of Gulf of Mexico research acquired and maintained by LATEX, we selected hydrographic surveys with good quality data that covered major portions of the Texas-Louisiana shelf, and then calculated mean distributions of temperature, salinity, and differences in geopotential.* From these we determined the relative effects of river dis-charge and wind stress on the distribution of salinity and, consequently, on differences in geopotential for November, May, and July-August.
In May the mean salinity is much lower than in November, both near the source of the fresh water and over the entire inner shelf. Alongshore winds are downcoast in both spring and fall, so we conclude that the salinity distributions differ principally because of the difference in river discharge volume. The buoyancy effect of the larger amount of fresh water over the inner shelf during May leads to greater differences in geopotential during that month, contributing to enhanced downcoast flow nearshore.
The surface salinity pattern for July and August looks completely different from that for May, although separated in time by only one month. Salinities greater than 36 extend halfway up the Texas coast in July and August. This indicates that upcoast alongshore wind influences upcoast flow. As noted earlier, river discharge rates are about the same in July and August as they are in November. The fresh water discharge in November, however, is distributed along the inner shelf by downcoast flow, while in the summer the discharge is held near the mouth of the river system by the upcoast flow.
Off-shelf circulation as a forcing mechanism on the shelf
During LATEX-A cruise HO5 (April 26 - May 10, 1993) nearshore flow was downcoast, as expected for April or May. At the same time there was cyclonic circulation east of about 95°W that may not have been closed to the southeast-again as expected from Cochrane and Kelly's averages and those developed by the LATEX team. Over the outer shelf and slope in the southwest there was a large anticyclonic ring between 94°W and 96°W and a cyclonic ring between 26°N and 27.5°N. Those rings caused considerable on-shelf flow near 95.5°W to 96°W and major off-shelf flows between 94°W and 95°W and south of 26.5°N. The entire circulation pattern of the outer, lower Texas shelf was dominated by the effects of these rings.
The resulting flow onto and off of the shelf is considerably more prominent than what is evident in Cochrane and Kelly's monthly circulation averages, but is typical of situations observed on several occasions during LATEX field work. The anticyclonic ring seen in May remained over the continental slope and continued to influence outer-shelf circulation through August. In September, its effects were not evident in the monthly average current field.
Concluding hypotheses
Our principal conclusions may be framed as hypotheses related to the low-frequency circulation patterns previously proposed for the Texas-Louisiana continental shelf. The Cochrane-Kelly pattern for the low-frequency circulation is essentially correct for the inner shelf, and forcing for the inner shelf is essentially by wind and buoyancy contrast. The transition at the beginning of summer to downcoast flow over the inner shelf (less than 50 meters depth) and the return to upcoast flow at the end of August are direct results of the wind regime.
There is more on- and off-shelf exchange than previously reported. The upcoast (eastward) flow at the shelf edge, envisioned by Cochrane and Kelly, may be the result of integrated effects of anticyclonic eddies impinging on the shelf edge.
References
Cochrane, J.D. and F.J. Kelly, 1986: Low-frequency circulation on the Texas-Louisiana continental shelf. Journal of Geophysical Research. 91, 10,645-10,659.
Oey, L.-Y., 1995: Eddy- and wind-forced shelf circulation. Journal of Geophysical Research. 100(C5), 8,621-8,637.
Map of the Gulf of Mexico showing the locations of LATEX instruments.
Comparison of wind and water current intensity and direction along the Texas-Louisiana
coast.
Yearly discharge of Gulf Coast rivers.
Seasonal distribution of salinity along the coast is due to Mississippi-Atchafalaya
River discharge.
