I I I 2 ACKNOWLEDGEMENTS In response to House Bill 2 (1985) and Senate bill 683 (1987), as enacted by the I Texas Legislature, the Texas Parks and Wildlife Department and the Texas Water Development Board must maintain a continuous data collection and analytical study I program on the effects of and needs for freshwater inflow to the State's bays and I estuaries. As part of the mandated study program, this research project was funded through the Board's Water Research and Planning Fund, authorized under Texas Water I Code Sections 15.402 and 16.058(e), and administered by the Department under I interagency cooperation contract No. TWDB 9-483-706, IAC[88-89]1434. I I I would like to acknowledge the following people who helped with the field work: Pam Plotkin, Hayden Abel, Noe Cantu, Rick Kalke, Todd Olson, John Turany, and Terry Whitledge. Thanks is also due to the Abraxis Oil Company for the use of their platform and Homer Roberson for recovering a lost meter. I am most grateful to I The University of Texas at Austin, Marine Science Institute, for providing additional I funds to purchase and replace some of the equipment deployed during this project. I I t I I I I I I Acknowledgements I Introduction 3 CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 I Hydrography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 I Currents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 I 1. Short-Term . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2. l..ong-Term . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 I (a) Currents, Salinity and Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . 17 I (b) Currents and Tides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 I Tides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Weather . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Analysis and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 I Currents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 I 1. Short-Term . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 I 2. l..ong-Term . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 I Tides and Weather . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 The Monthly Regime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 The Yearly Regime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 t Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 I References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 I I I I I I I I I I I I I I I t I I I Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Figure 9. Figure 10. Figure 11. Figure 12. Figure 13. Figure 14. Figure 15. Figure 16. Figure 17. Figure 18. Figure 19. Figure 20. Figure 21. Figure 22. Figure 23. Figure 24. TABLES Chronology of events during the NIPS hydrography program . . . . . . . . 5 Example of conductivity, temperature, depth (CTD) listings . . . . . . . . . 7 Mooring history and data recovery: long-term current meters . . . . . . . 14 Format of current meter data submitted on 5-1/4" diskettes . . . . . . . . 16 Channel orientations at each long-term current meter mooring site . . . 20 Direction of flow as shown in Part 1 (Fig. 5) and Part 2 (Fig. 7) . . . . 20 Format of tide data submitted on 5-1/4" diskettes . . . . . . . . . . . . . . . 22 Daily printout of Pier Lab weather station (Aransas Pass Channel) . . . 26 Format of weather data submitted on 5-1/4" diskettes . . . . . . . . . . . . 27 FIGURES Example of conductivity, temperature, depth ( CTD) vertical profiles . . . 8 Example of current rose, short-term current meter data . . . . . . . . . . 11 Example of stick diagram, short-term current meter data . . . . . . . . . . 12 Example of current vectors, current speed, salinity, and temperature for long-term current meter data . . . . . . . . . . . . . . . . . . . . . . . . . 18 Example of currents, along-channel and cross-channel, actual and predicted tides for long-term current meter data . . . . . . . . . . . . . . 19 Example of actual and predicted tides at Port Aransas jetties . . . . . . . 23 Example of weather data from the Aransas Pass Ship Channel . . . . . 28 Periodogram, tides at the Aransas Pass Ship Channel (PIERLAB) . . . 36 Periodogram, E-W component, currents at PIERLAB/fOP . . . . . . . . 37 Periodogram, E-W component, currents at NUECES/BOTIOM . . . . . 38 Periodogram, E-W component, currents at SNOOPYS/BOTIOM . . . . 39 Periodogram, E-W component, winds at PIERLAB . . . . . . . . . . . . . . 41 Periodogram, E-W component, winds at NUECES . . . . . . . . . . . . . . 42 Periodogram, barometric pressure at PIERLAB . . . . . . . . . . . . . . . . 44 Periodogram, barometric pressure at NUECES . . . . . . . . . . . . . . . . . 45 Periodogram, sea temperature at PIERLAB!fOP . . . . . . . . . . . . . . . 46 Periodogram, sea temperature at NUECES/BOTIOM . . . . . . . . . . . 47 Periodogram, air temperature at PIERLAB . . . . . . . . . . . . . . . . . . . 48 Periodogram, air temperature at NUECES . . . . . . . . . . . . . . . . . . . . 49 Monthly scatter diagrams: currents at the Aransas Pass (PIERLAB!fOP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Monthly scatter diagrams: currents at the Aransas Pass (PIERLAB/BOTIOM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Monthly scatter diagrams: currents at Nueces Bay (NUECES/BOTIOM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Monthly scatter diagrams: currents at the northern Laguna Madre (SNOOPYS/BOTIOM) . . . . . . . . . . . . . . . . . . . . . 55 Net monthly flow in the Corpus Christi/Nueces Bay Estuary (a) Flow between Corpus Christi Bay and the Gulf of Mexico . . . . . 56 (b) Flow between Nueces Bay and Corpus Christi Bay . . . . . . . . . . 56 (c) Flow between Laguna Madre and Corpus Christi Bay . . . . . . . . 57 (i) I I I I I I I I I I I I I t I I I Figure 25. Figure 26. Figure 27. Figure 28. Figure 29. Figure 30. Sea level, Aransas Pass Ship Channel (a) August-December 1985 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 (b) January-December 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 ( c) January-December 1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 ( d) January-December 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 ( e) January-July 1989 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Quarterly sea level variations, Aransas Pass Ship Channel, 1985-89 (a) Predicted tides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 (b) Measured tides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Non-tidal flow in the Corpus Christi/Nueces Bay Estuary (a) Aransas Pass Ship Channel (PIERLAB!fOP) . . . . . . . . . . . . . . 68 (b) Nueces Bay Channel (NUECES/BOTIOM) . . . . . . . . . . . . . . . 69 (c) Northern Laguna Madre (SNOOPYS/BOTIOM) . . . . . . . . . . . 70 Non-tidal fluctuations in current speed in the Corpus Christi/ Nueces Bay Estuary (a) Aransas Pass Ship Channel (PIERLAB!fOP) . . . . . . . . . . . . . . 71 (b) Nueces Bay Channel (NUECES/BOTIOM) . . . . . . . . . . . . . . . 72 (c) Northern Laguna Madre (SNOOPYS/BOTIOM) . . . . . . . . . . . 73 Long-term wind patterns over Corpus Christi/Nueces Bay Estuary (a) Aransas Pass Ship Channel (PIERLAB) . . . . . . . . . . . . . . . . . . 75 (b) Nueces Bay (NUECES) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Long-term wind speed fluctuations over Corpus Christi/ Nueces Bay Estuary (a) Aransas Pass Ship Channel (PIERLAB) . . . . . . . . . . . . . . . . . . 77 (b) Nueces Bay (NUECES) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 (ii) I -.~-::-/ ··· -· I I I 28° I I I I I I I I 27° I I I / 30·-1.........................................1111111~..................llfl.............................,......... 30° 30° I } 98° 97 ° W -WEATHER STATION c CURRENT METER T TIDE GAUGE I = I I I 4 INTRODUCTION I This report covers the hydrography and circulation within Corpus Christi and I Nueces Bay and, to a lesser extent, the San Antonio Bay system. A major part of this component of NIPS-II involved an attempt to measure year-long time-series of currents, sea temperature and salinity, tides and weather conditions at key locations in the I I Corpus Christi/Nueces Estuary. The interpretation of these data is not complete because the instruments were not removed until August 1989 and the complex analysis of such long time-series is still in progress. A supplement to this report will be I submitted in 1990. The bulk of this present report is devoted to the long-term data series, which so far has proven to be a most valuable and revealing information source. .I The report is in two parts. Part 1 is a description of the field work, methods, I analysis, and discussion of the data, and Part 2 contains the data in the form of listings and plots. Part 1 is divided into three sections: Hydrography, Currents, Tides and I Weather. Table 1 gives a chronology of events during this project. I METHODS I HYDROGRAPHY I Survey trips to Corpus Christi and Nueces Bays were made monthly from September 1987 to August 1988. On most of these surveys a CTD was used with t varying degrees of success (see Table 1). The ODEC CTD proved to be a difficult and I often useless instrument and its employment was eventually abandoned. Where we knew no CTD would be available, salinity samples were collected and later analyzed I using a precision laboratory salinometer. I 5 I TABLE 1. CHRONOLOGY OF EVENTS: HYDROGRAPHY DATE BAY MEASUREMENT(S) REMARKS I I 6 APR 87 SAN ANTONIO SET CMs at A,B,C 10 APR 87 SAN ANTONIO RECOVER CMs 1 JUN 87 SAN ANTONIO SET CMs at A,B,C 4 JUN 87 SAN ANTONIO RECOVER CMs NOT YET ANALYZED 13 JUL 87 SAN ANTONIO SET CMs AT A,B,C 17 JUL 87 SAN ANTONIO RECOVER CMs at B&C LOST METER AT 'C' I 23 SEP 87 NUECES & CC ODEC CID CID DATA DOUBTFUL I 20 OCT 87 CORPUS CHRISTI ODEC CID CID DATA GOOD 21OCT87 NUECES ODEC CID CID DATA QUESTIONABLE 19 NOV 87 NUECES ODEC CID CID DATA QUESTIONABLE 20 NOV 87 CORPUS CHRISTI ODEC CID CID DATA GOOD I 8 DEC 87 CORPUS CHRISTI ODEC CID: CM C&D CID FAILED 9 DEC 87 NUECES ODEC CID: CM A&B CID FAILED 21 JAN 88 NUECES & CC HYDRO 16 FEB 88 NUECES SEACAT CTD (ON LOAN) CTD DATA GOOD CMA&B 17 FEB 88 CORPUS CHRISTI SEACAT CTD (ON LOAN) CID DATA GOODI CMC&D I 23 MAR 88 CORPUS CHRISTI HYDRO 24 MAR 88 NUECES HYDRO 12 APR 88 CORPUS CHRISTI HYDRO: CM C&D 13 APR 88 NUECES HYDRO: CMA COULD NOT GET TO 'A' I 10 MAY 88 NUECES SEACAT CID: CM A&B NEW CTD 11MAY88 CORPUS CHRISTI SEACAT CID: CM C&D 15 JUN 88 NUECES SEACAT CID 16 JUN 88 CORPUS CHRISTI SEACAT CID 17 JUN 88 ARANSAS PASS WEATHER STATION START RECORD AT UTMSI 7 JUL 88 SAN ANTONIO SEACAT CTD: CM C CM DATA NOT ANALYZEDI 8 JUL 88 SAN ANTONIO SEACAT CID: CM C CM DATA NOT ANALYZED I 12 JUL 88 NUECES SEACAT CID: CM A&B CM DATA NOT ANALYZED 14 JUL 88 CORPUS CHRISTI SEACAT CID:CM D ROTOR BROKE AT 'C' 15 JUL 88 CORPUS CHRISTI LONG-TERM CM START RECORD AT UTMSI PIERLAB I 22 JUL 88 CORPUS CHRISTI LONG-TERM CM START RECORD AT NUECES CAUSEWAY 22 JUL 88 CORPUS CHRISTI LONG-TERM CM START RECORD AT JFK CAUSEWAY 09 AUG 88 NUECES HYDRO 10 AUG 88 CORPUS CHRISTI HYDRO 30 SEP 88 NUECES WEATHER STATION START RECORD AT NUECES BAY PLATFORM ·~ I 31MAR89 CORPUS CHRISTI LONG-TERM CM RECORD ENDS AT NUECES BAY CAUSEWAY 14 JUN 89 NUECES WEATHER STATION END RECORD AT NUECES I BAY 8 AUG 89 CORPUS CHRISTI LONG-TERM CM END RECORD AT UTMSI PIERLAB (TOP METER ONLY) I I I 6 In May 1988 the successful use of the Sea-Bird Electronics model SBE-19 CTD I ("SEACAT") with an attached Seatek 25-cm transmissometer was started. One survey I to San Antonio Bay was also made in July 1988 using this instrument package; the SEACAT was not available for the August 1988 survey. By comparing CTD, salinity I sample, Hydrolab, reversing thermometer, and refractometer data, some large I discrepancies in measurement accuracy in salinity have been revealed (see Whitledge, 1989). The most accurate comparison is between the SEACAT CTD and salinity I I samples run on a standard laboratory salinometer. The deployment of the long-term in situ instrumentation was not accomplished until July 1988. And, because logistics did not permit continuing the monthly hydrographic surveys in Corpus Christi and Nueces I I Bays beyond August 1989, we were not able to observe the horizontal salinity field, for example, during the time period that the long-term instrumentation was in place. I The CTD data set is extensive, as one data point is collected several times a I second. Data were downloaded to an IBM-AT computer from the internal memory of the CTD after each daily cruise. CTD data are presented in two forms, listings and vertical profiles (Part 2, Table 1, pp. 1-30 and Figs. 1 through 4, pp. 31-165) and I I examples are given here in Part 1 as Table 2 and Figure 1. Listings are averaged at 25-cm levels for the shallower, and 0.5-m levels for the deep stations. Erroneous values have been replaced by dashes. Reproduced in Part 2 are CTD data from May, June, t July 1988 in Corpus Christi Bay and July 1988 in San Antonio Bay. Still to come are I "difficult" CTD data from October, November 1987 and February 1988. I I I 7 TABLE 2. EXAMPLE OF CTD LISTINGSI (See Part 2, Table 1, pp. 1-30) Listings are averaged at 25-cm levels for shallow, and at 0.5-m levels for deep stations. Erroneous values replaced by dashes I NUECES BAY MAY 1988 I CTD DATA (0.25 m averaged) I STATION 01 meters temp salinity sigma-t % light 0.25 26.4037 34.8762 22.8153 41.41 0.50 26.3874 34.7391 22.7173 11.68 STATION 02I meters temp salinity sigma-t % light 0.50 26.6134 35.6583 23.3389 0.98 I STATION 03 meters temp salinity sigma-t % light 0.50 26.5649 34.5603 22.5266 2.10 I STATION 04 meters temp salinity sigma-t % light I 0.25 26.8336 33.4746 21.6243 o.oo 0.50 26.8231 33.4070 21.~765 o.oo 0.75 26.8156 33.3965 21.5710 0.00 I STATION 04A meters temp salinity sigma-t % light 0.25 27.7486 23.9732 14.2026 0.02 I 0.50 27.6607 24.0230 14.2723 0.02 0.75 27.6452 24.0459 14.2943 o.oo STATION 05 meters temp salinity sigma-t % light 0.25 25.9020 33.3164 21.7938 91.18 0.50 25.9734 33.6239 22.0056 35.38 0.75 25. 974·3 33.8053 ' 22.1421 8.10 1.00 25.9705 33.8346 22.1654 0.01 1.25 25.9636 34.0194 22.3069 o.oo 1.50 25.9511 34.1080 22.3773 o.oo STATION 06 meters temp salinity sigma-t % light 0.25 26.6790 33.5417 21.7241 7.72 0.50 26.7079 33.6448 · 21. 7919 0.29 0.75 26.7137 33.8084 21.9132 0.00 I I 0.0Q0 .::: .......,. ~ a.. ~ 5.000 ~. I' TEMPERATURE SALINITV LIGHT 25.00 35.00 10.00 50.00 0.900 10Q.9 p = 0.03 t =25.1817 s =9.6231 sent =62 .dat co STATION 5 -MAV 1988 NUECES BAY Figure 1. Example of conductivity, temperature, depth (CTD) vertical profiles (see Part 2, Figs. 1-4, pp. 31-165). I 9 I CURRENTS Both short-and long-term current measurements were made in connection with I the ~IPS-11 program. 1. Short-Term I In Nueces and Corpus Christi Bays a single AANDERAA model RCM4 current I meter was emplaced at each of the four experimental sites for approximately 24 hours. I I The AANDERAA meter records temperature and conductivity as well as current speed and direction. The recording interval was 10 minutes in each instance. In San Antonio Bay, ENDECO model 105 current meters were deployed, recording current speed and direction only. Some problems were encountered with the aging ENDECO meters I The light sources which produce the data on film are I deployed in San Antonio Bay. fading and some records are difficult to decipher. The lights are irreplaceable (they are I a radioactive source and not allowed to be sold under present DOT regulations). I These data will be recoverable, but the data are not yet fully analyzed and corrected. Unfortunately, the July 1988 deployment of the AANDERAA meter in all three bays I failed to produce results when the meter's tape came out blank. Otherwise these I meters produced excellent results. The tapes from the AANDERAA meters are processed by the company and results supplied to us on diskettes; recording rate was set at 10 minutes. ENDECO t meters record on 16 mm film; these records were processed by us and the data I extracted by digitizing the various light-bars using a horizontally mounted microfiche reader to project the film onto a flat screen flush with a bench surface. A Grafbar I sonic digitizer was used to convert the positions of the light bars into X-Y coordinates I I 10 I proportional to current speed and direction and the data stored on diskettes in the IBM-AT system. These data were recorded at 1/2-hour to 1-hour intervals. I Data are presented in two forms: current roses and stick diagrams (Part 2, Fig. 5, pp. 167-200) and examples are presented here in Part 1 as Figures 2 and 3. The I current roses are useful in envisioning the ellipticity of the tidal flow, while the stick I diagrams show the relationship between current vectors and variations in temperature and salinity. I 2. Long-Term I General Oceanics type 6011 current meters were deployed in July 1988 at three key locations in the Corpus Christi Bay system. These meters measure currents by the I I degree of tilt and the angle of rotation they assume when suspended from a mooring. They record internally on a non-volatile RAM chip and can be left in situ for periods of up to a year. In this environment, however, biofouling is a serious problem and meters I I must be recovered and serviced at monthly to six-weekly periods. The meters also record temperature and conductivity. They are set to record at ten-minute intervals. Meters were deployed at: (1) The pier laboratory of UTMSI on the south jetty I I of the Aransas Pass (PIERLAB). There were two meters here: one at the surface and one at the bottom, approximately 7 meters deep. The top meter did not have a conductivity sensor. (2) Under the Nueces Bay Causeway of U.S. Highway 181, at the t bottom of the main channel connecting Nueces to Corpus Christi Bays (NUECES). (3) I Under the JFK Causeway bridge of South Padre Island Drive at the bottom of the Gulf Intracoastal Waterway, the main connection between Laguna Madre and Corpus Christi I Bay (SNOOPYS, after the name of a well-known nearby local restaurant). Recording I I 10 I proportional to current speed and direction and the data stored on diskettes in the IBM-AT system. These data were recorded at 1/2-hour to 1-hour intervals. I Data are presented in two forms: current roses and stick diagrams (Part 2, Fig. 5, pp. 167-200) and examples are presented here in Part 1 as Figures 2 and 3. The I current roses are useful in envisioning the ellipticity of the tidal flow, while the stick I diagrams show the relationship between current vectors and variations in temperature and salinity. I 2. Long-Term I General Oceanics type 6011 current meters were deployed in July 1988 at three I key locations in the Corpus Christi Bay system. These meters measure currents by the I degree of tilt and the angle of rotation they assume when suspended from a mooring. They record internally on a non-volatile RAM chip and can be left in situ for periods of I up to a year. In this environment, however, biofouling is a serious problem and meters I must be recovered and serviced at monthly to six-weekly periods. The meters also record temperature and conductivity. They are set to record at ten-minute intervals. I Meters were deployed at: (1) The pier laboratory of UTMSI on the south jetty I of the Aransas Pass (PIERLAB). There were two meters here: one at the surface and one at the bottom, approximately 7 meters deep. The top meter did not have a conductivity sensor. (2) Under the Nueces Bay Causeway of U.S. Highway 181, at the I I. bottom of the main channel connecting Nueces to Corpus Christi Bays (NUECES). (3) Under the JFK Causeway bridge of South Padre Island Drive at the bottom of the Gulf Intracoastal Waterway, the main connection between Laguna Madre and Corpus Christi I Bay (SNOOPYS, after the name of a well-known nearby local restaurant). Recording I ".~ N 20 15 DEC 1987 w E -20 -15 15 20 I-' I-' VELOCITY IN CM/SEC -15 -20 s SITE 'A' Figure 2. Example of current rose, short-term current meter data (see Part 2, Fig. 5, pp. 167-200). The current rose is useful in envisioning the elipticity of tidal flow. I I I 13 started on 15 July 1988 at PIERLAB bottom, and 22 July 1988 at NUECES and SNOOPYS. The second meter at PIERLAB surface was deployed on 3 August 1988. I Equipment failure caused a loss of data at PIERLAB bottom from 20 July to 13 September. Fouling by marine organisms, especially acorn barnacles (despite use of the I most expensive and carefully applied anti-foulant paint), caused a gradual decrease in I the conductivity reading. Anticipating this, a CTD cast was taken before and after each deployment and will be used to correct the calculated salinity. Fouling was not too bad I at SNOOPYS, worse at NUECES, and worst at PIERLAB. Because of the nonI linearity of barnacle growth, however, the problem of correcting the salinity data proved I I to be very difficult. In this report, the salinity is left uncorrected. An attempt will be made to rectify this and the corrected salinities will appear in the Supplement. An unexpected problem at the Nueces Bay site was the major construction work I on the causeway: it is being widened to four lanes and extensive heavy construction I was ongoing in the channel where the meter was moored. This work started after the meter was deployed. On some occasions huge barges and jackups were working there. I The construction company was informed about the meter and promised to let us know I when they were scheduled to pull the pilings to which the meter was chained. They did not. On 31 March 1989, just before cessation of work for the day, they spudded one of I the jackup rig legs down on the meter, destroying the flotation sphere and severing the t meter from its mooring. A hired professional diver recovered the meter, however, in May. The data was intact, but the meter required extensive repair. It was not redeployed. I I I 14 I TABLE 3. DATA RECOVERY FROM CURRENT METER MOORINGS I (a) ARANSAS PASS CHANNEL CURRENT METER (PIERIABtrOP) MONTH DAY BOUR EVENT HOURS HOURS EXPLANATION I RECORDED LOST AUG 2 2000 DEPLOY 0 0 SEP 12 1500 RECOVER 979 0 I SEP 12 1800 DEPLOY 0 3 SERVICING OCT 16 2000 RECOVER 818 0 OCT 16 2200 DEPLOY 0 2 SERVICING I NOV 22 1600 RECOVER 882 0 NOV 22 1800 DEPLOY 0 2 SERVICING JAN 11 1500 RECOVER 1197 0 JAN 11 1700 DEPLOY 0 2 SERVICING FEB 23 1600 RECOVER 1031 0 FEB 23 1900 DEPLOY 0 3 SERVICING I MAY 5 1700 ------1726 0 MAY 18 1200 RECOVER 0 314 RAN OUT OF MEMORY MAY 18 1400 DEPLOY 0 2 SERVICINGI JUL 8 2100 -------1231 0 JUL 14 1800 RECOVER 0 141 RAN OUT OF BATTERY JUL 14 1900 DEPLOY 0 1 SERVICINGI AUG 8 1400 RECOVER 595 0 END DEPLOYMENT 8459 470 (94.7% DATA RECOVERY) I (b) ARANSAS PASS CHANNEL CURRENT METER (PIERLAB/BO'ITOM) MONTH DAY HOUR EVENT HOURS HOURS EXPLANATIONI RECORDED LOST JUL 15 1100 DEPLOY 0 0 TEST DEPLOYMENT JUL 20 1500 RECOVER 124 0I JUL 20 1700 DEPLOY 0 2 SEP 12 1600 RECOVER 0 1295 SET SWITCH WRONG SEP 12 1800 DEPLOY 0 2 SERVICINGI SEP 13 0100 -------0 7 COMPUTER FOULUP OCT 17 1400 RECOVER 829 0 OCT 17 1600 DEPLOY 0 2 SERVICINGI NOV 22 1600 RECOVER 864 0 NOV 22 1700 DEPLOY 0 1 SERVICING JAN 11 1500 RECOVER 1197 0 JAN 11 1700 DEPLOY 0 2 SERVICING I~, FEB 23 1600 RECOVER 1031 0 FEB 23 1900 DEPLOY 0 3 SERVICING MAY 4 1700 ------1702 0 I MAY 18 1200 RECOVER 0 331 RAN OUT OF MEMORY MAY 18 1400 DEPLOY 0 2 SERVICING MAY 23 0600 -------112 0 JUL 14 1800 RECOVER 0 1260 RAN OUT OF BATTERY JUL 14 1900 DEPLOY 0 1 SERVICING AUG 8 1400 RECOVER 595 0 END DEPLOYMENT I ---- 6454 2909 ( 68. 9% RECOVERY) I 15 I (c) NUECES BAY CURRENT METER (NUECES/BOTTOM) I MONTH DAY HOUR EVENT HOURS HOURS EXPLANATION RECORDED LOST I JUL 22 1400 DEPLOY 0 0 INITIAL DEPLOYMENT SEP 14 0900 RECOVER 1291 0 SEP 14 1000 DEPLOY 0 1 SERVICING I NOV 21 1200 RECOVER 1634 0 NOV 21 1300 DEPLOY 0 1 SERVICING JAN 12 1400 RECOVER 1249 0 JAN 12 1600 DEPLOY 0 2 SERVICING FEB 24 1600 RECOVER 1032 0 FEB 24 1800 DEPLOY 0 2 SERVICING I MAR 31 1600 ------838 0 I MAY 5 1700 RECOVER 0 840 DESTROYED BY BARGE; TEMP & SALIN ONLY; END DEPLOYMENT I 6890 846 (87.7% DATA RECOVERY) (99.9% TEMP & SALIN) I ( d) LAGUNA MADRE CURRENT METER (SNOOPYS/BOTTOM) MONTH DAY HOuR EVENT HOURS HOURS EXPLANATION RECORDED LOSTI JUL 22 1700 DEPLOY 0 0 INITIAL DEPLOYMENT SEP 14 1100 RECOVER 1290 0 SEP 14 1300 DEPLOY 0 2 SERVICINGI NOV 21 1400 RECOVER 1609 0 NOV 21 1500 DEPLOY 0 1 SERVICING JAN 12 1000 RECOVER 1243 0 I JAN 12 1100 DEPLOY 0 1 SERVICING I FEB 24 1100 RECOVER 1032 0 FEB 24 1300 DEPLOY 0 1 SERVICING APR 21 1000 RECOVER 1341 0 I APR 21 1300 DEPLOY 0 3 SERVICING JUN 13 1300 RECOVER 1272 0 JUN 13 1400 DEPLOY 0 1 SERVICING JUN 23 1800 ------244 0 RAN OUT OF BATTERIES JUL 10 0900 RECOVER 0 399 MOORING PULLED JUL 13 1200 DEPLOY 0 0 MOORING BACK IN I t JUL 21 0100 ------181 0 SEP 11 1100 RECOVER 0 1259 MOORING DRAGGED; FLOTATION DESTROYED; TEMP & SALIN ONLY I 8212 1667 (83.1% DATA RECOVERY) (95.9% TEMP & SALIN) I I I I 16 Other problems plagued the efforts to obtain a year-long series. These included batteries running down before their rated replacement dates, internal memories I becoming filled before servicing could be accomplished, fouling by huge clumps of Sargassum weed and seagrasses, and moorings being dragged by shrimp trawls. Table 3 I lists the periods when data was successfully collected. I Although data are recorded at 10-minute intervals (the meters internally average a one-minute period within each interval), current data presented here are vectorially I averaged over one-hour intervals. Simple means of conductivity and temperature are I used to obtain the hourly data. Current meter data are recorded on 5-1/4" diskettes in I ASCII files, one file per meter/per month. The naming convention is CMNNP .MMM, I where CM is Current Meter, NN is the mooring place name (Pl=PIERLAB, NU=NUECES, SN=SNOOPYS), Pis the position in the water column (T=TOP, or surface, B=BOTIOM) and MMM=Month represented alphabetically by the first three I letters of the month's name. A sample is given in Table 4. TABLE 4. FORMAT OF CURRENT METER DATAI SUBMITTED ON 5-1/4" DISKETIES I YHR JUL TIME MN DY HR SEA SALIN SPEED DIR E-W N- s 8040 336 0 12 1 5 19.3 37.79 34.5 304.2 -28.6 19.4 8041 336 100 12 1 6 19.3 37.84 31.3 303.7 -26.0 17.4I 8042 336 200 12 1 7 19.3 37.85 19.4 297.5 -17.2 9.0 8043 336 300 12 1 8 19.2 37.70 8.8 300.7 -7.6 4.5 8044 336 400 12 1 9 18.9 37.89 5.1 286.9 -4.9 1.5 8045 336 500 12 1 10 18.9 37.68 8.7 94.2 8.7 -0.6t 8046 336 600 12 1 11 18.9 37.70 22.7 101.6 22.2 -4.6 8047 336 700 12 1 12 18.8 37.81 34.9 105.6 33.6 -9.4 I Where YHR= "Year-hour" (starting at 0000 hours CST, January 1), JUL=Julian day, TIME=local time (CST), MN=month, Dy=day,HR=hour (Hr is in GMT), SEA=sea temperature (C), SALIN=salinity (ppt), SPEED=current speed (cm/sec), DIR=current direction (degrees magnetic), E-W=east-west comEonent, N-S=north-south comEonent. I I I I I I I I I I I I I I I I I~, I I I 17 Current meter data are illustrated in two sets of diagrams in Part 2 of this report (Part 2, Fig. 6, pp. 201-302; Fig. 7, pp. 303-406) for each installation. Examples are included here in Part 1 as Figures 4 and 5. The moorings are identified as PIERLAB{fOP and PIERLAB/BOTTOM for the two meters in the Aransas Pass by UTMSI's Pier Lab (there is a meter at the surface and one at 7 m depth). At the entrance to Nueces Bay, the mooring is identified as NUECES/BOTTOM, and at the northern Laguna Madre site it is called SNOOPYS/BOTTOM. Both of these meters are moored about one-meter above the channel bottoms at 8-and 9-meters, respectively. (a) Currents, Salinity and Temperature (Part 2, Fig. 6 & Part 1 Example, Fig. 4) Each page represents approximately two weeks of data and pages open to show a full month. The top panel shows current vectors orientated to true north which is positive and pointing up. The next panel is current speed. Both are in cm/sec. Next is salinity in parts per thousand (ppt) calculated from conductivity. The PIERLAB/fOP meter does not have a conductivity cell. Because of the biofouling problem mentioned above, most of the salinity records show substantial jumps each time the meter is serviced and the conductivity heads cleaned. In this report the data are reproduced uncorrected for these excursions. A spline function must be fitted to each segment of data and the final salinities will be corrected according to this "barnacle function". At each servicing interval, a CTD profile and salinity sample was collected and adjustments will be made accordingly. At this time work on the spline function is progressing. The bottom panel is temperature in degrees C. Scales in the diagrams for each mooring have been kept the same for ease of comparison. Because a wide range of conditions ~ '~~ CURRENTS (BOTTOM). NUECES MAR 1989 100 100 50 50 0 0 -100 -50 I ~-· .~. -~ _ .~-·~'--· --·~-· ~-·~'-.~.~~:':_. ~-~~ =~~o 100 10075 7550 5025 25 0 0 45 45 1--' 40 40 co35 30 0 30 - 25 25 I 1sl 171 ~191 2:0 I I :r:r:r: I I I I~ 35 SALINITY (ppt) 35 3530 30 25 2520 2015 1510 10 5 50 0 161 171 181 191 201 211 221 231 241 251 261 271 281 291 301 31 SEA TEMPERATURE (C) Figure 4. Example of current vectors, current speed, salinity, and temperature for long-term current meter data (see Part 2, Fig. 6, pp. 201-302). TIDES & CURRENTS (BOTTOM) : NUECES MAR 1989 100 100 0 -50 -100 50 ~1 I2 I3 I4 I 7 I 101 ul 121 131 141 J=~;o 5I6I 8I9I DIFFERENCE (cm); ACTUAL minus PREDICTED 100 100 50 50 0 0 -50 . ~ v v :v v ·· -· : ,: ·· ·· ·· ~ I -5o -100 1 • I I I I I ::: -100 t--' \.0 100 50 0 -50 -100 100 100 50 50 0 0 -50 -50 -100 -100 I~l-2·1: ~~·i~~~7-1·~·~ :f10~~:27:)1x~ CURRENTS CROSS-CHANNEL (cm/sec)-UP•NE Figure 5. Example of currents, along-channel and cross-channel, actual and predicted tides for long-term current meter data (see Part 2, Fig. 7, pp. 303-406). I I 20 was encountered, some detail is compressed. (b) Currents and Tides (Part 2, Fig. 7 & Part 1 Example, Fig. 5) I In this set of diagrams, the currents are shown as "north" and "east" components rotated to the orientation of the various channels in which the meters are moored. I Conseque!ltly, they represent along-channel and cross-channel flow. Table 5 gives the I channel orientations at each mooring. I TABLE 5. CHANNEL ORIENTATIONS WHERE THE LONG-TERM CURRENT METERS WERE MOORED I PIERI.AB (TOP and BO'ITO M) . . . . . . . . . . 305/125 degrees true I NUECES . . . . . . . . . . . . . . . . . . . . . . . . . . 320/140 degrees true SNOOPYS ...... ~ . • . . . . . . . . . . . . . . . . 020/200 degrees true I I Starting from the bottom, these figures show the cross-channel flow and then the along- channel flow. Table 6 explains the direction of flow relative to each bay and the Gulf I of Mexico. Positive is above the zero line in each of the panels. I TABLE 6. DIRECTION OF FLOW AS SHOWN IN PART 2, FIG. 7 & PART 1 EXAMPLE, FIG. 5 I MOORING ALONG-CHANNEL CROSS-CHANNEL POSITIVE NEGATIVE POSITIVE NEGATIVE PIERLAB into CC Bay into Gulf to the NE to the SW ·~, NUECES into Nueces Bay into CC Bay to the NE to the SW SNOOPYS into Laguna into CC Bay to the NW to the SE I I For comparison with the tides at the Aransas Pass, the top two panels show the height of the tide in centimeters (solid curve) as measured at the U.S. Army Corps of I I I I I 21 Engineers tide tower on the south jetty. I have been operating a Stevens type tide gage on this tower since August 1985. The tide tower refere!1ce level was tied to the I benchmark system in September 1987 by Texas Water Development personnel. In these diagrams and all other tide diagrams in this report, the zero line is Mean Sea I Level (MSL). Also shown here is the height of the predicted tide (dotted curve) which I is referenced to the standard Gulf of Mexico Datum, which is Mean Lower Low Water I (MLLW). Predictions were made using NOAA's standard harmonic analysis which gives a datum for this station as 0.97 ft (29.6 cm) above MLLW and are the same as in the I published tables. The uppermost panel is the difference between real and predicted tides (real minus predicted) in centimeters. I I TIDES I Each four-day chart record is digitized using a Grafbar sonic digitizer at as many I intervals as is necessary to describe the real tidal fluctuations. This is usually several points per hour. Simple hourly averages are then computed and the height on the chart corrected to real sea level as described below. This file is then merged with the I hourly predicted tide data for every hour of the year. Gaps are filled (see below) so I that there are no missing data. Data are recorded on yearly ASCII files on 5-1/4" diskettes using the following naming convention: CTIDYYYY, where CTID is t Corrected Tide and YYYY is the year. I I I I I TABLE 7. I I I 22 FORMAT OF TIDE DATA SUBMIITED ON 5-1/4" DISKEITES YHR ACTUAL PRED M 0 -25.1 10.1 0 1 -30.7 4.6 0 2 -34.8 0.1 0 3 -34.5 -2.8 0 4 -32.0 -3.3 0 5 -30.3 -1.4 0 I Where YHR="Year-hour", ACTUAL=actual tide height (cm), PRED=predicted tide height, and Mis a code to indicate when gaps occurred and the actual tide is e.5timated (O=normal, l=a gap). I The tide data are presented in diagrammatic form (Part 2, Fig. 8, pp. 407-452), I with an example here (Part 1, Fig. 6). Included in Part 2 is all of the tide data I collected at the Aransas Pass gage for 1986, 1987, 1988, and up to February 1989. Data from 1985 includes a few segments that need correcting and is not presented here. I Each month of the year is presented on a separate page with two panels for each half-I monthly segment of the record. The bottom panel for each half month gives the actual and the predicted tide, while the upper panels show the difference, actual minus I predicted. I There are gaps in the record, occasionally as long as four days. The chart on the Stevens recorder is changed every four days. While this has been a burdensome task and has caused many of the gaps, I believe this record is as accurate a measure of sea level in the pass attainable by this type of gage. Every chart is calibrated at the start and end using a direct method of measuring the distance to the water surface. This method, developed by the author, uses a color-coded beaded cable and a compass-rose I scale to measure the distance in the stilling well to an accuracy of 0.3 cm. The gaps I ",~ TIDES PORT ARANSAS ~ETTIES: ~UL 1989 100 10050 50 0 -50 -100 9 I 101 111 12 minus PREDICTED 100 0 ~::-=:-:::::: '. :::::=::=';'~ =~go 100 50 50 0 0 -50 I 'u' \/ \J · 'a' \ -l , ' "' ... v: 'V' V \:A -50 -100 -+-~..---~--~--~-----~--~--~-----~--~-----.~--~-Po -1006 I 7 I 0 I 9 I 101 111 121 131 141 15(cm) -DOTTED-PREDICTED: SOLID-ACTUAL N w 100 10050 50 0 0 -50 _.... ,.,_ _J"'\. ~ A •A.. --50-100 ~ ~ I I I :;:~Tr; :;I I I I I I I /1 -100 161 171 181 191 201 211 221 231 241 251 261 271 281 291 301 31 DIFFERENCE (cm): ACTUAL minus PREDICTED 100 10050 50 0 -50 I ..Vi \;°I \,..': \< .,: : • • W' a .., • ._. •._. • ._. -, -50 -100 I I I I I I I I i -100221 231 241 251 261 271 281 291 301 31-DOTTED•PREDICTED: SOLID•ACTUAL Figure 6. Example of actual and predicted tides at Port Aransas jetties (see Part 2, Fig. 8, pp. 407-452) I I I 24 which were discovered in the digitization of these are filled in this record using the predicted heights with the trend removed for each missing data point adjusted to the I slope of the missing segment. In the diagrams these segments are dotted and the I differences appear as straight line segments in each upper panel. On 21 April 1989, an ENDECO model 1029 water level recorder was deployed at I the pier lab facility and, starting then, tidal heights in the diagrams come from this instrument. This recorder was purchased with funds provided by UTMSI. The Stevens I gage was run for several more months for a comparison, but unfortunately, some errors I records. While the data are still intact, the problem was not resolved in time to include in this report. There is a gap in the measured tides from 16 February 1989 to 21 April I 1989. The gap has been filled in with predicted data, hence the differences (top panels, I Part 2, Fig. 8) appear as straight lines (Part 2, pp. 397-401). I I WEATHER Two weather stations were emplaced in the Corpus Christi/Nueces Bay Estuary in 1988. One was at the UTMSI pier lab facility (PIERLAB) and one on a platform in I I Nueces Bay (NUECES). In June 1988 a weather station was put into operation on the UTMSI pier laboratory that extends over the Aransas Pass. The system sensors include a Texas Electronics vane and anemometer, Weathertronics air temperature, relative t, humidity, barometric pressure, sea temperature probe, and tipping-bucket rain gauge. I Data acquisition is accomplished by Weathertronics signal conditioning units for most sensors feeding into a Hewlett-Packard model 3214A multiplexer and a Hewlett-Packard I 71C computer with Corvallis Microtechnology non-volatile RAM memory. Data are I I I I 25 collected in real time at 10-minute intervals. Daily printouts update at hourly intervals (Table 8) and data must be dumped to a computer disk at two-weekly intervals. This I system is a bit of a hodgepodge, but was put together from existing equipment, some funding from TWDB and contributions from UTMSI. It took time to get all the I problems sorted out, but operated routinely with some periods of data loss. It survived I Hurricane Gilbert, but a power failure, which the system was supposed to have been able to ''weather", put it out of commission for 24 hours just after the storm. Other I problems included computer and printer malfunctions, a degradation of the relative I humidity sensor, and a period when the barometer stuck at 1016.2 Mb. The barometer I was repaired but the humidity sensor could not be replaced due to cost. A Coastal Climate Weatherpak self-contained system was put into operation on an Abraxas Oil I Company platform in Nueces Bay on 30 September 1988. Half the cost of the I Weatherpak was provided by UTMSI. This system stores data internally and comprises I a vane and anemometer, air and sea temperatures and relative humidity, barometric pressure and tipping-bucket rain gauge. At six-week intervals data was dumped into a I Hewlett-Packard 110 portable computer taken to the platform. These data are then I transferred to an IBM-AT computer for permanent storage on disk. The major problem with this system was the sea temperature probe which was non-operational for I several months. t Both weather stations record data at ten-minute intervals. At PIERLAB (Part 2, Fig. 9, pp. 453-482), data are downloaded weekly to an H-P Portable and then to an IBM-AT where they are merged into monthly files, averaged (vectorially for winds) to I give one-hour intervals. NUECES (Part 2, Fig. 10, pp. 483-503) data are similarly I I I 26 I I TABLE 8. DAILY PRINTOUT OF PIERLAB (ARANSAS PASS SHIP CHANNEL) WEATHER STATION INFORMATION DURING PASSAGE OF A STRONG "NORTHER" ON 31 OCTOBER 1988. Units are in "English" notation and made available to public. I UTMSI Pier Lab Weather Station 88/10/31 TEMPERATURE RELATIVE BAROMETRIC WIND RAINFALL TIME DATE AIR SEA HUMIDITY PRESSURE SPEED DIRECTION INCHES I 0000 88/10/31 75.7 78.6 95.2 30.03 20.6 ENE 0.00 I I I I 0100 88/10/31 75.8 78.6 95.3 30.01 19.3 ENE 0.00 0200 88/10/31 75.0 78.6 95.8 30.00 23.4 E .03 0300 88/10/31 73.8 78.5 98.2 30.00 16.3 E .06 0400 88/10/31 74.0 78.S 98.2 29.98 17.4 ENE .02 0500 88/10/31 73. 1 78.S 98.2 29.97 11.S ENE . 12 0600 88/10/31 73.3 78.S 98.2 29.98 11.3 E 0.00 0700 88/10/31 73.B 78.6 95.0 29.98 5. 1 ENE 0.00 0800 88/10/31 74.0 78.5 94.0 29.99 1.3 NNE 0.00 0900 88/10/31 72.S 78.4 98.2 30.02 7.9 NNE 0.00 1000 88/10/31 72.4 78.4 98.2 30.03 8.5 NNE 0.00 1100 88/10/31 65.9 78.3 98.3 30.03 43.4 N .07 1200 BB/10/31 65.S 77.9 98.3 30.04 26.4 NNE 0.00 1300 88/10/31 69.3 77.6 92.S 30.03 20.S ENE 0.00I 1400 88/10/31 70.7 77.S 89.8 30.02 12.3 NE 0.00 1500 88/10/31 69.8 76.9 90.2 30.03 14.3 N 0.00 1600 88/10/31 70. 1 76.6 85.8 30.02 22.S N 0.00 1700 88/10/31 71. 1 76.3 ·10.1 30.02 20.0 N 0.00I 1800 88/10/31 70.0 76.4 83.6 30.04 15.S NNE 0.00 1900 88/10/31 68.0 76.3 85. 1 30.06 18.7 NNE 0.00 2000 88/10/31 67.3 76.2 89.0 30.08 19.S N 0.00 2100 88/10/31 66.3 75.9 87.6 30.09 20.4 N 0.00 2200 88/10/31 66.2 _75.8 86.3 30. 11 20.3 N 0.00 I 2300 88/10/31 65.9 76.1 91.5 30. 12 17. 1 NNW 0.00 I t I I I I I 27 averaged but are in a different format. Final data files are converted to a standard format and are recorded in ASCII on 5-1/4" diskettes using the following naming I convention; WTHNN.MMM, where WTH stands for Weather, NN is place name (PI=PIERLAB, NU=NUECES), and MMM is month as described above for current I data files. I Results are presented in Part 2, Figure 9 (PIERLAB), pp. 453-482 and Part 2, Figure 10 (NUECES), pp.483-503, with an example here in Part 1 as Figure 7. In these I I TABLE 9. DATA FORMAT FOR WEATHER DATA I SUBMITTED ON 5-1/4" DISKETTES YHR JUL HOUR DY MN HR AIR SEA RH BARO RAIN SPEED DIR E-W N-S 8040 336 0 12 1 0 14.9 19.44 55.7 1028.4 0.0 5.2 16.7 1.5 4.9 8041 336 100 12 1 1 14.0 19.40 57.5 1028.3 0.0 5.4 16.5 1.5 5.2 Where YHR="Year-hour", JUL=Julian day, HOUR=Hour (CST, "Nautical" form), MN=Month, I I DY=Day,HR=Hour (CS1), AIR=Air temperature (C), SEA=Sea temperature (C), RH=Relative humidity(%), BARO=Barometer (Mb), RAIN=Rainfall (mm), SPEED=Wind speed (m/sec), DIR=Wind direction (degrees true), E-W=East-West wind component (m/sec), N-S=North-south wind component (m/sec). I diagrams, from top-to-bottom the panels are as follows: wind vectors in m/sec. (north I is positive), barometric pressure in millibars, relative humidity (%), sea and air temperature (degrees C). The area between sea and air temperatures has been shaded: I t light (stippled) shading is when the air temperature is colder than the sea temperature and dark shading is the reverse. The PIERLAB relative humidity sensor became I I unreliable by 20 December 1989 (Part 2, p. 465) and the barometer failed from 24 April 1989 (Part 2, p. 473) to 20 June 1989 (Part 2, p. 477) with occasional periods of "lucidity" between. The water temperature probe at NUECES failed at the end of October 1989 (Part 2, p. 487). It took until 23 April (Part 2, p. 499) to replace. I I WEATHER. PIERLAB MAR 1989 15 15 0 0 -15 -15 16 17 18 19 23 24 25 26 27 28 29 30 31 WINO VECTORS (m/sec) 1040 1040 1020 1020 1000 1000 100 100 75 7550 5025 25 N 0 ~ 0 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 RELATIVE HUMIDITY (%) 30 30 20 20 10 10 0 0 161 171 181 191 201 211 221 231 241 251 261 271 281 291 301 31 AIR ANO SEA TEMPERATURE (C) Figure 7. Example of weather data from the Aransas Pass Ship Channel, PIERLAB site (see Part 2, Fig. 9, pp. 453-482). From top to bottom, these diagrams are: wind vectors in m/sec (north is positive); barometric pressure (mb); relative humidity (%), sea and air temperature (°C). Light stippling in area between air and sea temperatures shows when air is colder than sea, dark shading is the reverse. I I I • I I I I I I I I I I I t I I I 29 ANALYSIS AND DISCUSSION The bulk of the analysis in this report is devoted to the long-term measurements. It must be regarded as preliminary, however, because data collection continued until August 1989, giving scant time to fully analyze this large data set. One of the major goals of this project was to see, by direct measurement of water movement in the main connecting channels to the Corpus Christi/Nueces Estuary, what were the flow patterns over a full year. In the past, water exchange between Corpus Christi Bay, Nueces Bay, the Laguna Madre and the Gulf of Mexico, has largely been determined by monitoring water levels at key locations and deducing flow indirectly. The studies of Smith (1977, 1978, 1985) are particularly noteworthy and have contributed much to our knowledge on this system. His numerical bay-shelf exchange model (Smith, 1985), using long-term water level data, shows that tidal flushing alone can replace 50% of Corpus Christi Bay water in three months. Older hydrodynamic models (Masch, 1972) show a net outflow through the Aransas Pass in all months except September and October, and strong flow from Laguna Madre into Corpus Christi Bay in every month. This would seem to be confirmed for the spring and summer months by the salinity distribution (see Whitledge, 1989). One of the critical questions in determining the role of freshwater inflow in this bay system is the contribution of high salinity water from the hypersaline Laguna Madre. Flow out of Nueces Bay would also be indicated as the Nueces River is the major freshwater source to the Estuary. Yet during the hydrographic study period (September 1987 to August 1988) we found little freshwater outflow from Nueces Bay as shown by parameter distribution (Whitledge, 1989). As important as the direct measurement of flow was considered, we wished to I I 30 determine the relative importance of the forcing mechanisms on the flow. To that end we also monitored the tides at the Aransas Pass Ship Channel (the Gulf tides) and the I atmospheric conditions at the same location and in the upper part of Nueces Bay. An analysis of the results of the currents, tides and weather will be followed by an initial I look at the interactions of all three sets of measurements. I CURRENTS \ I I 1. Short-Term I start with a brief analysis of the results of short-term current measurements in I Corpus Christi Bay.I Nueces Bay: (Part 2, Fig. S, pp. 169-183) • Currents are much more active at the Nueces Bay sites than at those in Corpus Christi Bay. • Currents seem to be I tidally driven rather than wind-driven although the 24-hour periods of the measurements I are insufficient to allow low pass filtering to examine the wind-field effects. • Currents seldom exceed 20 cm/sec. • At site A (pp. 169-174), near the head of Nueces Bay the tidal ellipses are more circular, while at B (pp. 175-183) they are elliptical indicating a I I certain channelization of the flow in an east-west direction. Corpus Christi Bay: (Part 2, Fig. 5, pp. 184-199) • In Corpus Christi Bay I currents are seldom above 5 cm/sec and for many hours are often below threshold t (approx 1.5 cm/sec). • Currents are not tidal. • Site C (pp. 184-191), in the northwestern part of the bay near Nueces Bay, is practically stagnant. • Site D (pp. 192-199), in the southeastern part nearer the Laguna Madre, when it has currents at all, I has a southerly flow towards the Laguna Madre. I I I 31 It must be emphasized that these are short-term measurements; indeed, taken only in December, February, April and May, and at times of those months chosen I without regard to the state of the tide. 2. Long-Term I The series in examined first, location by location, using the one-hour averaged I data presented in Part 2, Figures 6 and 7. Pier Lab: At the PIERLAB (Part 2, Figs. 6 and 7), surface and bottom currents I are strongly correlated, the deeper currents having slightly less amplitude than those at I the surface. • Current speeds often exceed 100 cm/sec (averaged hourly), and reached a maximum value of 185 cm/sec (3. 7 knots) following Hurricane Gilbert in September I I 1988 (Part 2, Fig. 6, pp. 209, 237 and Fig. 7, pp. 310, 311 and 338, 339). Currents are predominantly tidal, floods and ebbs represented by the peaks in the speed record and I slack tides when the speeds approach zero (there is some cross-channel flow during I tidal reversals). Other storm-related flow can clearly be seen superimposed on the tidal flow associated with the passage of cold fronts, particularly in early February 1989. I On 19 September, about the time of maximum hurricane flow, current amplitude I is high because the moon is at its southernmost declination. Conversely, the new moon on 10 October produces little tidal flow because at that time the moon had just crossed the equator. The tidal nature of flow in the pass and the effect of Hurricane Gilbert I t, can be even more dramatically illustrated by looking at the current speed and direction (Part 2, Fig. 5, p. 209). Diurnal and semidiurnal excursions in sea temperature and salinity (bottom panels, Part 2, Fig. 6, pp. 234-255) are a combination of the daily I warming cycle and the ebb and flow of bay and Gulf waters through the pass. The I I I I I I I I I I I I I I I I~, I I I 32 steady drop in salinity is an artifact fouling by barnacles. The meter at the bottom mirrors the top meter with a slight clockwise rotation and reduction in current velocities. The high degree of coherence between top and bottom meters gives confidence in the accuracy of these meters in measuring the current field. When the axes are rotated to show flow relative to the channel (Part 2, Fig. 7, pp. 308-357), it can clearly be seen that the along-channel motion dominates flow in the Aransas Pass. It can also be seen that outflow into the Gulf slightly exceeds inflow (dark shading is outflow). Peak velocities are in phase with both ebb and flood tides, but generally have much sharper peaks than the tidal heights would indicate. This is particularly true at flood tide when high water "stands" for several more hours than it does at the ebb tide. Nueces: Compare the above with currents at the Nueces Bay entrance (Part 2, Fig. 6, pp. 259-275). Once again currents are tidally dominated. Speeds here also exceed 100 cm/sec with some short-lived events in August, September, December and March. The same pre-hurricane flow into the bay and post-hurricane flow out of the bay can clearly be seen. A similar pattern in temperature and salinity (only minor barnacles here) is present (Part 2, Fig. 6, bottom panels), although excursions may be more of a diurnal nature rather than the result of tidal flushing (no semi diurnal variations in early October). When axes are rotated to that of the channel, along-channel flow dominates but there is more significant cross-channel flow during certain wind-driven events. Flow out of Nueces bay exceed that into the bay (dark shading in bottom panels of Part 2, Fig. 7, pp. 361-377). I I 33 Snoopys: Current speeds are much less at this location at the entrance to the I Laguna Madre, seldom exceeding 25 cm/sec (Part 2, Fig. 6, pp. 279-301). Only I following hurricane Gilbert do speeds approach 75 cm/sec. Tidal flow can barely be discerned here. Most of the flow appears to be storm-or wind-driven and occurred in I the middle of September. • This flow is primarily in a north-south direction, into the I Laguna prior to Gilbert and out of the Laguna for several days following the storm. Strong diurnal fluctuations (which were suppressed by Hurricane Gilbert) can be seen I I here in sea temperature (Part 2, Fig. 6, p. 283, bottom panel). A remarkable pattern of salinity (second panel from bottom, same page) ranging from near-zero during the storm to 50 ppt at the end of September (no barnacles grew in the conductivity sensor I here). After rotation of axes to align with the channel (Part 2, Fig. 7, pp. 382-405), the I flow is not nearly as aligned with the channel as in the other two locations. There is some ambiguity here about the actual channel orientation. There appears to be little I correlation with tides here. I I TIDES & WEATHER I Tides measured at the Aransas Pass tide gage differ from those predicted (see Part 2, Fig. 8, pp. 410-452) in two major respects (disregarding the difference in datum levels used; actual tides are tied to MSL, predicted to MLLW). First the measured I tidal amplitudes are less than those predicte~ and secondly flood tide peaks are far ~-;., I flatter in reality than those predicted. Times of highs and lows are generally in agreement, but the time of maximum flood is often hard to discern because of the I "stand" of the floods. These differences show as generally diurnal "bumps" on the I I I I 34 I difference plots (top and second from bottom panels in each of the monthly diagrams, Part 2, Fig. 8). The predictions are more successful during two-weekly periods of I semidiurnal tides of low amplitude. Wind tides are not possible to predict and do show in the record, e.g. during hurricane Gilbert in September 1988. I Part 2, Figure 9 (PIERLAB), pp. 456-481 and Figure 10 (NUECES), pp. 486-503 I show that the wind regimes at Nueces Bay and on the coast differed during the period I I October through June during which the Nueces Bay system was in operation. Winds over Nueces bay are more variable, but often exceed those at the coast in magnitude, although the situation is reversed at times. Diurnal variations are more pronounced over the bay and bay waters respond more rapidly than do those in the pass to air I I temperature changes. Relative humidity at the coast is higher than over the bay although the pier lab sensor failed in December 1989. Most dramatic was the cold I front event of early February, 1989 when water in the pass dropped to 4.5 degrees (Fig. I 9, p. 468 and Fig. 10, p. 494). This was not as cold as in the "big freeze" of 1983, but a second front late in the month (not shown here) cooled sea temperatures again, making I a large contrast between air temperatures at the coast and at Nueces Bay. Spectral analyses and coherence spectra were computed for the various time I I series using the Cooly-Tukey Fast Fourier Transform (FFT), modified by the author for use with an IBM-AT computer. Gaps in the series were filled with zeros. The one-I year record (August 1988 through July 1989) of two-hour averages were transformed ~:.r I using 800 lags. Figure 8 (herein) is a periodogram for the tides measured at the Aransas Pass. Both diurnal tides, 0 1 and K1 form prominent peaks at 25.82 hours and I 24.07 hours, respectively. The principal semidiurnal tide (M2, with a period of 12.42 I I I 35 hours) is also prominent, and the K2 tide at 11.97 hours is also significant. Longer I period astronomical tides cannot be resolved due to the length of the series. There is I some suggestion of a peak around 170 hours which may be atmospherically driven. How does this compare with the current spectra? I transformed the East I I component of the current measured by the PIERLAB{fOP meter (Figure 9). No attempt was made to rotate the axes here to align with the channel orientation. Both I diurnal and both semidiurnal tidal currents have considerable energy and it is clear that the currents are strongly tidal in the Pass. Peaks at other diurnal periods can also be I seen (01 and perhaps, 001). Coherence squared values (with the tides) are as follows: 0 1, 0.835; Ou 0.952; K1, 0.966; 001, 0.383 (not significant); M2, 0.959; K2, 0.515. A peak I at 12.0 hours has a value of 0.834 and may be due to the sea breeze. Other significant I peaks at 8.40 hours, 8.18 hours and 6.21 hours are not readily explicable at this stage of I the analysis. They do have some coherence with the tides (cf. Figure 8). Some peaks I at longer periods may be significant, but note how almost all the energy is in the tidal components and not in long-term variations. At NUECES/BOTIOM, Figure 10, the two principal diurnal and the M2 tidal I currents still have considerable energy. The other diurnal constituents have been attenuated by the time the tidal wave reaches the Nueces Bay entrance, as do the I mysterious short period currents. There is much higher energy in the long-period t currents, indicating that in Nueces Bay, atmospheric forcing is more significant than in the Aransas Pass. The situation is entirely different at the northern end of the Laguna I Madre (Figure 11, SNOOPYS/BOTIOM). Here, no peaks of significance can be I discerned. A rise in the energy around 24 hours is the only indication of any tidal I ~ '~ 10 9 FREQUENCY (cph) -4 -3 -2 10 10 10 ~ I +-+-H+H-t-+·-H+t+H I I ~ ± -1 10 I ++++H-1 I I I I I 10 I I I' 0 >rH CJ) z w 0 >c..Q a: w z w c..Q 0 _J 10°l± 10 71 6l 10 t J_ 10 sJ_ !4 i 10 t w ~ 10 3LI I 111111 I I I 111111 1 I I 1'111'1'1"1 I I I Figure 8. Periodogram of tides 10000 1000 100 10 1 measured at the Aransas Pass Ship Channel (PIERLAB). PERIOD (hr) ~ \~ FREQUENCY (cph) 10 9 10 -4 -3 10 -2 10 -1 10 10 ° $ _j_ 10 sJ_ I >rH CJ) z w 0 >CQ rr w z w lQ 0 _J 10 10 10 t 7 I $ f j_ 6 l-1:::t: +± ! T 5 I -=!::! +T w -.....! 4 10 111111 I 10000 I t++++-1 I I 1000 I -~-11111 I 100 I 1111 10 1 Figure 9. Periodogram, E-W component, currents measured at Aransas Pass (PIERLAB(fOP meter). PERIOD (hr) ~ '~ FREQUENCY (cph) -4 -3 -2 -1 0 10 10 10 10 10 9 10 ~--+ I I 111111 I I 1-t-ttHi 8 tI ~ 10 > fH (f) 10 7 -l z w w ± co 0 >6 c.!J 10 CI w z w (.!] 5 10 0 _J Figure 10. Periodogram, E-W 4 component, currents measured at 10 1 11111 I t-+ 111111+-t-t--flll I I I I I 111111 I 1-t- NUECES/BOTTOM. 10000 1000 100 10 1 PERIOD (hr) ~ \~ FREQUENCY (cph) -4 10 -310 -2 10 -1 10 10 0 10 7 1- I I I 1--1111 I ++ I I I 1111 I I I I I 1111 I I I I I 111 I ""'' I T >1H Cf) z w 0 -10 6 I-+! w ~ >(!) IT w z 10 I T 5 it w (_:) 0 _J iI _J_ I I I 10 J I II I I I I I I I111 I I I I I -I 11111 I I I Figure 11. Periodogram, E-W 10000 1000 100 10 1 component, currents measured at SNOOPYS/BOTTOM. PERIOD (hr) .--- I I 40 current at this location. The peak near 950 hours ( 40 days) is just significant and could be a salinity-driven phenomenon. I At this stage, I have not made mention of phase-relationships between currents and tides, or currents at different locations. First I have transformed the East-West I components of the currents. Because the currents are generally aligned (the exception I is at SNOOPYS) with the channels (see Fig. 22 herein), it will be necessary to transform currents rotated to align with channel orientation. Phase angles and lag or I lead times will be affected by the choice of vectors, rotated vectors or current speed I alone. Work on this is progressing and will be presented in the supplement. I I Two other driving mechanisms, the winds and the atmospheric pressure are known to have an effect on the water level and currents. Figures 12 and 13 are periodograms of the winds at PIERLAB and NUECES, respectively. The most I significant peak in both is one at 24 hours, which is the daily sea-breeze effect. Note I that this is more significant at the coast (PIERLAB) than it is over the inland bay (NUECES), where the sea breeze is not so prominent. Wind energy is higher over I Nueces Bay than at the coast. Longer term trends are not significant in either wind I regime, although the transform is not long enough to resolve the semi-annual wind-shift from southeasterly to a more northerly flow during the winter. I The reverse-barometric pressure effect is generally regarded to affect water levels, t near the coast and especially in the bays (Nowlin and Parker, 1974; Smith, 1977). When arctic air masses move over the Texas coast, pressures can approach 1050 Mb and correspondingly lower water levels at a time when the sea-level is already lowered I by the SSa constituent of the astronomical tide. Periodograms of barometric pressure I ,. ~~ FREQUENCY (cph) 4 -3 -2 10-10 10 -1 0 10 10 10 6 !---+-t-++++++i I I 111111----f I 111111 I I I I I U-U.L ± ~~ 5 ! 10 I 10 4 II rn I H nm 11 I 10 3 10 2 1 111111 I 11 I --f 11111111 I 11111111 I 1111~~1 I I I Figure 12. Periodogram, E-W10000 1000 100 component, winds measured at 10 1 PIERLAB. PERIOD (hr) > ~ H en z w 0 > (_Q IT wzw (_Q 0 _J . ~~ (c p h) FREQUENCY -2 10 0 -3 10-1 -4 10 ! 1010 I 11111111 I I~ I 11111111 I 11111111 6 J: 10 ~f\A ,11/i~ 10 5 .i:= N 10 4 1 10 3! 1111 Figure 13. Periodogram, E-W2~+t-H 111111 I t--+--t+ll 11 I I I 1 component, winds measured at 10 10 100 NUECES. 100010000 (hr) PERIOD > LI H CJ) z w 0 > l!J IT wz w l!J 0 _J I I I I I I I I I I I I I I I ..,•:,. I I I I 43 at the coast (Fig. 14) and over Nueces Bay (Fig. 15), show a marked contrast. At the Pier Lab energy drops almost linearly from lower to higher frequencies, and no spectral peaks can be discerned. In contrast, two peaks, one broad, centered on 24-hours and one narrow at 12-hours, are significant. These are daily pressure excursions, probably reflecting the diurnal atmospheric warming cycle, more prominent over the shallow inland bay, surrounded by urban, agricultural and industrial-use lands, rather than the strictly maritime coastal environment. To see what effect the atmospheric regime has on the channel waters, sea temperatures were also transformed and periodograms plotted (Figs. 16 and 17). At the Aransas Pass Ship Channel (Fig. 16, PIERLAB!fOP), no significant diurnal warming can be seen in the spectrum. Little energy is transferred from the atmosphere to the water in this tidally active channel, even at the surface. At NUECES (Fig. 17), however, the diurnal effect is highly significant, even though the depth is about 8 m at the channel bottom. Nueces Bay waters are shallow and strongly influenced by the daily air temperature cycle. Together with the strong wind field over the bay, waters are alternately warmed and mixed and then transported through the channel to Corpus Christi Bay. Periodograms of air temperature over both locations (Figs. 18 and 19) clearly show the diurnal warming with peaks at 24 hours. At the coast (Fig. 18), this effect is modified by the flow of air from the Gulf of Mexico, while over the inland bay (Fig. 19), the daily input of energy from the atmosphere is much more energetic . . Here, the diurnal temperature range is frequently 3 to 5 degrees above and below that at the coast. Two other peaks show in the periodogram for Nueces Bay; one at 12 hours, and the other at 500 hours (21 days). The significance of these has not been determined at this writing. , (~~ FREQUENCY (cph) -4 -3 -2 -1 10 10 10 10 10 11 lllll+ I 11111111 I 1111111 1o t4-~H+tt---i 10tt 10 I 109-t >81 I H U) z 10 lJJ 0 1071 > (_Q er =F w I z 10 SJ_ w =t:: 4= -+ (_Q :::t: 0 I 5 _J 10 10 4 I111 I -1-~--t+++-H+ I I l+t++H-+-t--+I I I I I I I I I I 10000 1000 100 10 1 PERIOD (hr) 0 .t:: .t:: Figure 14. Periodogram, barometric pressure measured at PIERLAB. " \~ FREQUENCY (cph) -3 -2 10-1 10 0 -4 10 10 I 1111111I 11111111 I 11111111 10 10 t-+++t+t+t 11 f 1010 -:::t:; 10 9 += U1 Figure 15. Periodogram, 1 barometric pressure measured 100 10 at NUECES. 1000 10000 PERIOD (hr) > t-HCJ) z w 0 > c..Q CI w z w lQ 0_J ,. \~ FREQUENCY (cph) -4 -3 -2 -1 0 10 10 10 10 10 8 10 ! I 111+-l-l+-i t-+-t ± -t 10 7 I I J_ f H >10 6-t UJ z +:" w O'l + 0 51 > (.!] ([ 10 w z w (.!] 4 10 0 _J 10 3 _ill+++++ I H1111 I I I ·-f+tt+1 I I I' I Figure 16. Periodogram, sea 10000 1000 100 10 1 temperature measured at PIERLABtrOP. PERIOD (hr) ,. ~~ FREQUENCY (cph) -4 -3 -2 -1 0 10 10 10 10 10 8 ! I 10 I 1111111 I I 1111111 I I 1111111 I I 11111 ~, 10 7 t > 6 H ~ 10 U) += z ± . I I \/\ . I I -......J w 0 5-i-! >~ 10 IT w z w (_Q 4 10 0 _J 3 1o -4+t-H-++--+---t-t+t-H-++--1--i+tt-H-l--~---t· Figure 17. Periodogram, sea 10000 1000 100 10 1 temperature measured at NUECES/BOTIOM. PERIOD (hr) ( FREQUENCY (cph) -4 -3 -2 -1 0 10 10 10 10 10 10°}-+-+++11111 111111+tt-111111111 '''""' + 7 10 > r 6 H 10 en z ~ 1 +:" 11 rvV\A. I I co w J 0 > (_!) 10 -± IT w z ! w 3 (_!) 0 10 l _J 3 10 I II I I I 111 II I I I I Figure 18. Periodogram, air temperature measured at 10000 1000 100 10 1 PIERLAB. PERIOD (hr) FREQUENCY (cph) -4 -2 -1 0 10 10 10 10 8 10 7 10 > t--- 6 H 10 c.n z .i:= l.D w ± IV ~ ~ I ~ I 0 >-5 (!) 10 IT w z w (!) 4 10 0 _J 3 10 111 Figure 19. Periodogram, air 10000 1000 100 10 temperature measured at 1 NUECES. PERIOD (hr) I I 50 THE MONTHLY REGIME I In Figures 20 through 23, monthly scatter diagrams of the currents are shown, I rotated so that flow along-channel is in the vertical axis and cross-channel is in the horizontal. Each dot represents the end-point of a one-hourly averaged vector. Numbers refer only to the along-channel flow. To the left of the ordinate are the I number of hours when flow is in flood (top) and ebb (bottom). To the right of the ordinate, the mean flow in cm/sec is given for floods and ebbs. Above each monthly I scattergram is the excess time (expressed as a percentage) that the tidal current is in I flood (positive) or ebb (negative). For example, at the PIERIAB(fOP (surface), flood I I tides last 7% longer than ebbs. The second number is the mean monthly along-channel flow in cm/sec (positive = floods; negative = ebbs). Examining the currents in the Aransas Pass (Fig. 20, PIERIAB{fOP; Fig. 21, I PIERIAB/BOTTOM), we see several interesting features. First (Fig. 20), there is I anisotropy in current direction and magnitude. Ebb tidal currents are almost exactly aligned with the channel axis, except when velocities exceed about 100 cm/sec. Floods I deflect to the right by about 10 degrees in all months. Net flow is seaward for all I months, with the greatest outflow occurring in September, January and February. In fall and early winter, ebb currents last longer than flood currents, while the rest of the I year the opposite is true, even though more water flows out of the bay than flows in. I The excessive outflow appear to be controlled by wind and storm events rather than astronomical tides. A similar pattern occurs at depth (approximately 7 m) in the Aransas Pass (Fig. I 21), but here there is some vertical shear in direction and magnitude. Currents are not I 7 % -6.5 385 39.1 334 1 -45.7 AUG .8 % -14~·2 339 If 42.2 333 i"-56.4 -~: "J.,( • ·t ...·· FE.:B -.9 % ..:14 l· 344 ·43.6 350 .,l -57.6 :.·:~\ )'; SE.P .2 % -1 373 11 45.8 37 1 "!I -46 . 8 .· :·· MAR 1.7 % -4.8 378 47 365 .·1 -51.9 ....... :; OCT 2. 2 % -3 ...7 368 II 44 .1 352 -47.8 APR -3.5 % -5.3 347 45.1 372 J-50.5 .....· NOV 3.3 % -1.1 ... :.:;· . 214 1l/45~1 200 : MAY -.6 % -3.9 370 46.8 374 I -50.7 .. ~ .. DEC 1.1 % -3 364 1.t 44.6 356 0UN -2. 1 % -11. 7 ~· 364 46.4 379 ·~ -58. 2 ~ ."i.' 0A·N U1 _. 13.1 % -4.8 3'4.1 11· 46 262 I .. -50. 9 ~~ .. ... 0UL Figure 20. Monthly scatter diagrams. Currents at the surface in Aransas Pass (PIERLAB!fOP). Flow along-channel is in the vertical axis; cross-channel is in the horizontal axis. Each dot represents the end-point of a one-hourlyaveraged vector; numbers refer only to the along-channel flow; to left of the ordinate are number of hours when flow is in flood (top)and in ebb (bottom); to right of ordinate, mean flow (cm/sec) for floods and ebbs; above each scattergram is excess time (expressed as percentage) tidal current is in flood (positive) or ebb (negative); and a second number (cm/sec) which is the mean monthly along-channel flow (positive = floods; negative = ebbs). ---------·--------- % ..:14.5 -.3 % -3.2 -3.9 % -3.1 1.6 % -4.9 3 % -13.3 225 32 371 : 36.6 346 36.9 378 34.4 383 }. 35.1 -46.6 373 -39.9 374 . -40 .1 366 -39.3 360 -48.5 ~ ·+ ·} AUG SEP OCT NOV DEC 0AN U"1 N 2.9 % -11.1 '; 0 % -1. 8 -3.4 % -.7 -3.3 % -10.2 3.ie % -2. 1 345 : tJ· 32. 8 372 It 36.4 348 It 31.9 90 I/ 29...,. 32.2 ' 325 tt· -44 ,'1 372 IJ. -38 . 3 '{. 4 I,•.: 372 ·~· -32. 7 ,.., 96 1~ -39. 3 '".,, 'r 199 -34.3 : FE .B MAR APR MAY 0UN 0UL Figure 21. Monthly scatter diagrams. Currents at the bottom in Aransas Pass (PIERLAB/BOTTOM). Flow along-channel is in the vertical axis; cross-channel is in the horizontal axis. Each dot represents the end-point of a one-hourlyaveraged vector; numbers refer only to the along-channel flow; to left of the ordinate are number of hours when flow is in flood (top) and in ebb (bottom); to right of ordinate, mean flow (cm/sec) for floods and ebbs; above each scattergram is excess time (expressed aspercentage) tidal current is in flood (positive) or ebb (negative); and a second number (cm/sec) which is the mean monthlyalong-channel flow (positive = floods; negative = ebbs). I 53 I aligned with the channel either during flood or ebb tides. While flood-tidal currents are I deflected at the same angle to channel orientation as at the surface, the ebbs are I deflected to the left (west) by about 10 degrees. Once again, the net flow is Gulfward in all months. At the Nueces Bay entrance channel (Fig. 22), where the meter was set at about I 7 m depth, the situation is reversed. A distinct anisotropy is still apparent, but the flood tide currents are parallel with the channel axis while the ebbs are deflected to the I right by 10 to 20 degrees. The greater scatter here is due to non-tidal effects, although I the tides still dominate the flow in this channel. A year's worth of data was not I obtained here because the mooring was destroyed by a construction barge and it took I over a month to locate and recover the meter. Mean flow is out of Nueces Bay in all months, but is more uniform in magnitude than at the Aransas Pass. Here the duration I of the ebb tidal currents and the magnitude of the mean flow are much more uniform. I The situation at the Kennedy Causeway (SNOOPYS) is entirely different. This meter, at a depth of 9 m, is moored in the Gulf Intracoastal Waterway, near the I northern terminus of the Laguna Madre. The channel orientation is not easily I determined here and the scatter is much greater (Fig. 23). The current magnitude is considerably smaller than at the other two channels. Low velocity currents occur with I no preferred direction, while higher velocities (20-30 cm/sec), which seem to be tidal, I are rotated 20-40° clockwise from the channel axis. Flow is directed into the Laguna in all months except November. Non-tidal motion dominates the flow pattern here. I have calculated the net flow at each of the mooring locations in the alongI channel directions only. Approximate cross-sectional areas of each channel have been I -.3 % -6.1 -7.4 %"-6.8 -19.8 % -11.6 -19.6 % -7 -9.7 % -7.6 -9.7 % -7.4 .~~ 371 :~~:··..~27. 9 334 ~; 20. 9 289 .16.2 290 .. 25.1 336 ~i 24. 7 336 ·Jf 25.4 .'I : • "' .. : 373 ~:1 ·-34 387 431 . ! ·-27 .9 431 408 -32.4 408 ~·1 -32.8 ·~{ : ··.· .. >: I I AUG SEP OCT NOV DEC 0AN U"l .i:: -8.2 % -1 -5.7 % -4.5 -2.8 % 2.2 308 351 350 I 5.6 363 393 370 I -3.5 FEB MAR APR Figure 22. Monthly scatter diagrams. Currents at the bottom (7 m) in Nueces Bay (NUECES/BOTTOM). Flow along-channel is in the vertical axis; cross-channel is in the horizontal axis. Each dot represents the end-point of a one-hourly averaged vector; numbers refer only to the along-channel flow; to left of the ordinate are number of hours when flow is in flood (top) and in ebb (bottom); to right of ordinate, mean flow (cm/sec) for floods and ebbs; above each scattergram is excess time (expressed as percentage) tidal current is in flood (positive) or ebb (negative); and a second number (cm/sec) which is the mean monthly along-channel flow (positive = floods; negative = ebbs). 51 % 2. 1 21.1 % .5 8 % 1. 8 -20.6 % -1.5 10.4 % 2 6.4 % 1.2 562 9.5 12.2 402 11 286 I 10. 1 411 10.3 396 11.5 ·i,:':··' .I " ·~:,;;,.-• I .s~.t:· ·~~· J,r·. -·' -·~>.;; :~ .·.. 182 I -7.5 283 342 -9.3 434 r.'-'11 .1 333 -8.3 ·349 I -10. 4 AUG SEP OCT NOV DEC 0AN U1 U1 4.3 % 1 4. 1 % 2.2 35.6 % 2 58.3 % 1 350 I 14 387 11.8 13.5 589 9.3 487, .,.J I . 1r;1'-. .. 321 I -13.1 356 '231 .. : -11.6 155 r-8.3 FEB MAR APR MAY Figure 23. Monthly scatter diagrams. Currents at the northern Laguna Madre (SNOOPYS/BOTIOM). Flow along-channel is in the vertical axis; cross-channel is in the horizontal axis. Each dot represents the end-point of a one-hourly averaged vector; numbers refer only to the along-channel flow; to left of the ordinate are number of hours when flow is in flood (top) and in ebb (bottom); to right of ordinate, mean flow (cm/sec) for floods and ebbs; above each scattergram is excess time (expressed as percentage) tidal current is in flood (positive) or ebb (negative); and a second number (cm/sec) which is the mean monthly along-channel flow (positive = floods; negative = ebbs). I 56 NET FLOW BETWEEN CORPUS CHRISTI BAY AND GULF OF MEXICO NEGATIVE-= FLOW INTO GULF I Figure 24 (a) I CUBIC METERS /SECOND 100.--~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~---, I 0 I -100 I -200 I -300 I -400 I AUG SEP OCT NOV DEC JAN FEB APR MAY JUN JUL 1988 1989 · MONTH I NET FLOW BETWEEN NUECES BAY AND CORPUS CHRISTI BAY NEGATIVE-FLOW INTO CORPUS CHRISTI BAY I Figure 24 (b) I CUBIC METERS /SECOND 10--~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I 0 MOORING DESTROYED I -10 BY BARGE METER LOST FOR 1 MONTH -DAMAGED I -20 -JO I -40 I I MONTH 1988 1989 57 NET FLOW BETWEEN LAGUNA MADRE AND CORPUS CHRISTI BAY NEGATIVE""'FLOW INTO CORPUS CHRISTI BAY I Figure 24 (c) I CUBIC METERS /SECOND 10...--~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I 0 I -10 I -20 I -JO -40 I AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL I MONTH 1988 1989 I I I I I I I I I I I I 58 calculated to arrive at a monthly volume flow in cubic meters per second. Figure 24(a,b,c) illustrates the results graphically. Aransas Pass dominates with a mean flow I into the Gulf of nearly 500 cubic m/sec in February 1989. Flow is gulfward in all months. Flow into Corpus Christi Bay from Nueces Bay dominates all months, but is I considerably less than at the Aransas Pass (note the scale difference in Fig. 24 a,b ). I The maximum outflow from Nueces Bay is about 20 cubic m/sec. Note that flow is at a minimum in February when it was at a maximum at the Aransas Pass. The flow is even I less at the Lagu~a Madre mooring ( <10 cubic m/sec) and for eleven of the twelve I months of recording here, flow is towards the Laguna Madre. Clearly there is a net flow out of the system to be accounted for. Stream flow of the Nueces River system I I has not been examined for the period during which these measurements were made. This will be necessary to fully analyze these data. It is also clear that a meter must be I placed at the entrance to Redfish/ Aransas Bay and Corpus Christi Bay junction, possiblyI in the Lydia Ann Channel. We intend to do this at a later date, and also to place a meter in the Corpus Christi Ship Channel near Ingleside. Both of these locations I present a technical problem as loss of any moorings here is a strong possibility. The I "Laguna Madre Problem" (flow into the Laguna rather than out as the salinity would indicate) needs to be further investigated. I THE YEARLY REGIME I Filtering techniques have been employed to examine the long-term fluctuations in I the tidal series (which may be both astronomically and atmospherically driven), the nonI tidal fluctuations in the current flow and variations of longer period than diurnal changes in the wind field. I I 59 Tides: The entire tide gauge record and the predicted tidal series has been "filtered", I initially using a 49-hour simple-moving averaging technique, to remove the diurnal and semidiurnal tidal signals. More sophisticated filtering techniques will be applied later to I this and the current series. In Figure 25 the difference in height between the averaged I real and predicted tides has been filled. Black shading shows where the measured sea level is less than the predicted, while dotted shading is the reverse. I When the diurnal and semidiurnal components are removed from the tide record I (Fig. 25 a-e ), an interesting picture emerges. Firstly, the regular semi-monthly I I components prominent in the predictions are barely discernible in the actual tide record. Instead, more frequent, irregular fluctuations, almost certainly wind-controlled show up year-round, less frequently in summer months when the wind field is more stable. The semi-annual components, readily visible in the predictions is only obvious in I I the real record in 1985 and 1986. Most interesting is the apparent steady decline in sea level shown in the real record. Note the decrease in the real tide shown as the area of I white above the zero line (MSL) as the years progress. To summarize, Figure 26 ( a,b) I has been constructed by averaging real and predicted tides quarterly from 3rd quarter 1985 through the end of 1988. Linear regression lines have been fitted to these data. I The predicted tide shows but a slight decrease, probably as a result of the 18.5 lunar I node component which reached a peak in 1986-1987. The actual tide shows a dramatic decrease of about 20 cm. Fitting a linear regression may exaggerate this decrease and more detailed analysis will be done later. What we may be observing here is a lowering I of water level due to the recent drought conditions in South Texas. A more I 50 50 25 25 0 0 -25 -25 0 -5o~0AN IFEB,MAR,APR,MAY l0uNl0uL IAUG,SEPlocTINovloEc r-50 ~ SEA LEVEL, 1985 (cm) , 49-HOUR MDV I NG AVERAGE ACTUAL v/s PREDICTED: dotted=actual>predicted black=predicted>actual Figure 25 (a). Sea Level, Aransas Pass Ship Channel, August-December 1985. 50 50 25 25 0 0 -25 -25 -50 -50 ~ 0AN IFEBIMAR IAPRIMAY l0UNl0UL IAUGISEPIOCT INOVIDEC SEA LEVEL, 1986 (cm), 49-HOUR MOVING AVERAGE ACTUAL v/s PREDICTED: dotted=actual>predicted black=predicted>actual Figure 25 (b). Sea Level, Aransas Pass Ship Channel, 1986. 50 50 25 25 0 0 -25 -25 -50 -50 ~ 0ANIFEBIMAR IAPRIMAY l0UNl0UL IAUGISEPIOCT INOVIDEC SEA LEVEL, 1987 (cm), 49-HOUR MOVING AVERAGE ACTUAL v/s PREDICTED: dotted=actual>predicted black=predicted>actual Figure 25 (c). Sea Level, Aransas Pass Ship Channel, 1987. 50 50 25 25 0 -25 -25 -50 -50 w 0AN IFEBIMAR IAPRIMAY l0UNl0ULIAUGISEPIOCTINOVIDEC CJ') SEA LEVEL, 1988 (cm), 49-HOUR MOVING AVERAGE ACTUAL v/s PREDICTED: dotted=actual>predicted black=predicted>actual Figure 25 (d). Sea Level, Aransas Pass Ship Channel, 1988. 50 50 25 25 0 0 -25 -25 -50 -50 ~ 0AN IFEBIMAR lAPRjMAY l~UNl0UL IAUG ISEPIOCT INOVIDEC SEA LEVEL, 1989 (cm), 49-HOUR MOVING AVERAGE ACTUAL v/s PREDICTED: dotted=actual>predicted black=predicted>actual Figure 25 (e). Sea Level, Aransas Pass Ship Channel, January-July 1989. MEAN SEA LEVEL (QUARTERLY) ARANSAS PASS 1985-1988 PREDICTED TIDAL HEIGHT Figure 26 (a) HEIGHT (CM) 40.--~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~- 20 o.___._~~_._~~_._~~_._~~_._~~_._~~_._~~--~~_._~~_._~~_._~~_._~~_._~~__.___, 3rdQ 4thQ 1 stQ 2ndQ 3rdQ 4thQ 1 stQ 2ndQ 3rdQ 4thQ 1 stQ 2ndQ 3rdQ 4thQ 1985 1986 1987 1988 QUARTER O'> U1 MEAN SEA LEVEL (QUARTERLY) ARANSAS PASS 1985-1988 ACTUAL TIDAL HEIGHT Figure 26 (b) HEIGHT (CM) 20.--~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 10 01 1-I ': I 7:'Jzc: ,« ': I -10 -2ol_...L_~--1~~J_~-L~~1--~_i_~~1__~-L:---~~~~~~:::--:-:1::::---~~---:~~ 3rdQ 4thQ 1 stQ 2ndQ 3rdQ 4thQ 1 stQ 2nsQ 3rdQ 4thQ 1 stQ 2ndQ 3rdQ 4thQ 1985 1986 1987 1988 QUARTER ~ ~ I 67 appropriate fit to the data may show that the decline started around the beginning of I 1987, as we well know from some very low tides encountered on several of the field I trips for this project. I Currents: I Figure 27 (a-c) are 49-hour moving averages of the currents (not rotated to the I I channel axes). At the Aransas Pass (Fig. 27 a), three or four events dominate, with several lesser events at irregular intervals. Up is flow to the north, or inflow and down is flow to the south or outflow. Note that the vectors are approximately aligned with I the channel axis. Hurricane Gilbert in September 1988 produced strong flow in both I directions with the most energetic being outflow. Two strong fronts in February and March 1989 created net seaward flow for several days. Dominance of long-period I outflow during fall and winter months is clear from the figure. The shorter record at I Nueces Bay (Fig. 27 b) shows almost exclusive outflow (into Corpus Christi Bay). Hurricane Gilbert can be seen in this record, but is less pronounced than at the I Aransas Pass. The situation is much more confused at the upper Laguna Madre I (SNOOPYS, Fig. 27 c). The scale here is expanded to twice that in Figures 27 a and b to show detail of the flow. Here Hurricane Gilbert pushed water into the Laguna. The I dominance of outflow is not clear from this diagram. I If we examine the 11on-tidal current speeds, without regard to direction (Figs. 28 a-c), we can see the mean flow with the events showing as spikes. At the Aransas Pass (Fig. 28 a), a mean speed of 10.7 cm/sec translates into a yearly net velocity of 5.9 I cm/sec towards 048 degrees. The events are restricted to the Gilbert and cold fronts I PIERLAB 1988 1989 40 ~ 40 J 20 20 I ~ I. I I I I I I I I~ I I ~ CX) 0 0 -20 -20 -40 -40 '""1UL AUG SEP CT NOV DEC wAN FE R APR MAY wUN wUL AUG I CURRENT VELOCITY (CM/SEC) 1988-89, 49-HOUR SIMPA Figure 27 (a). Non-tidal flow in the Corpus Christi/Nueces Bay Estuary. Aransas Pass Ship Channel (PIERLAB!fOP) NUECES 1988 1989 40 40 20 20 O"l l.O 0 0 -20 -20 -40 -40 ~ULIAUG ISEPI OCT INOVI DEC I~ANIFEB MARI APRI MAY I~UNI ~ULIAUG CURRENT VELOCITY (CM/SEC) 1988-89, 49-HOUR SIMPA I Figure 27 (b). Non-tidal flow in the Corpus Christi/Nueces Bay Estuary. Nueces Bay Channel (NUECES/BOTTOM) SNOOPYS 1988 1989 20 20 15 15 10 10 -.....J 5 5 0 0 0 -5 -5 -10 -10 -15 -15 -+-~-+-~-+-~-+-~-+-~+-~+-~+-~~~~~~~~~~~~---+--20 -20 ~ULIAUG I SEPI OCT INOVI DEC I~ANIFEB MAR IAPRI MAYI~UNI ~ULIAUG CURRENT VELOCITY (CM/SEC) • 1988-89, 49-HOUR SIMPA Figure 27 (c). Non-tidal flow in the Corpus Christi/Nueces Bay Estuary. Northern Laguna Madre (SNOOPYS/BOTIOM) PIERLAB t988 1989. 50 50 40 40 30 30 20 20 10 -....J 10 - 0 0 wUL 1AUG1 SEPI ocTI NOVI DEC IwAN IFE~MAR IAPRI MAYI wUNI wUL 1 AUG CURRENT SPEED (CM/SEC) • 1988-89, 49-HOUR SIMPA MEAN SPEED= 10.74 NET VELOCITY= 5.93 (CM/SEC) TOWARD 48 DEG Figure 28 (a). Non-tidal fluctuations in current speed in the Corpus Christi/Nueces Bay Estuary. Aransas Pass Ship Channel (PIERLAB{fOP) NUECES 1988 1989 50 50 40 40 30 30 20 20 ......i 10 10 "'-' 0 0 ~ULI AUG ISEPI OCT I NOVI DEC I~ANIFE CURRENT SPEED (CM/SEC) • 1988-89, 49-HOUR SIMPA MEAN SPEED= 8.59 NET VELOCITY= 6.19 (CM/SEC) TOWARD 140 DEG Figure 28 (b). Non-tidal fluctuations in current speed in the Corpus Christi/Nueces Bay Estuary. Nueces Bay Channel (NUECES/BOTTOM) SNOOPYS 1988 1989 50 50 40 --40 30 --30 20 -... 20 10 -. ~ft aA. ~~ ~A IL hL l .~ l J, JA' di11~. • .A.lta~ a -10 ---J w •1 '.1.-1 f ,. -"f''-M ,-,, • -r .. ,. 1-•ur· !~ r· I • -I ·I ,..I~ 0 I I I 0 wULI AUGI SEPI OCTI NOVI DEC IwANI FE CURRENT SPEED (CM/SEC) , 1988-89, 49-HOUR SIMPA MEAN SPEED= 6.95 NET VELOCITY= 2.43 (CM/SEC) TOWARD 199 DEG Figure 28 (c). Non-tidal fluctuations in current speed in the Corpus Christi/Nueces Bay Estuary. Northern Laguna Madre (SNOOPYS/BOTTOM) I I I 74 from November through April. The currents at Nueces Bay (Fig. 28 b) have a yearly mean speed of 8.6 cm/sec and a net velocity of 6.2 cm/sec towards 140 degrees. I Periods of stronger flow (September through November) and mid December to midJanuary are separated by relatively quiet intervals. Less activity is seen at the northern I Laguna Madre (Fig. 28 c) with a yearly mean speed of 7.0 cm/sec and net velocity of I 2.3 cm/sec towards 199 degrees. Events occur in September (Gilbert), November and February through April. The net velocities at each location are not aligned with I channel orientations. This is because of the anisotropy of the up-and down-channel I flows. Were these equal and opposite, the net flow would be zero. I Winds: I To see what effect the overlying winds have in determining this non-tidal motion, I the winds at PIERLAB and NUECES were similarly treated (Figs. 29 a and b ). In I Figure 29 (a) (the Aransas Pass Ship Channel), Hurricane Gilbert can be seen as a rotation of the winds from northeast to southwest (direction to which the wind is I blowing). Cold fronts show as wind flow to the south and the generally prevailing winds I out of the southeast dominate from April through July. Over Nueces Bay, the dominant wind is from the south, and slightly west of I south. These winds are much more prevalent at this location. Cold fronts interpose I the flow from November through March. Finally, the wind speed regime is illustrated at these locations in Figure 30 a and b. At the Aransas Pass (Fig. 30 a), the annual mean is 4.2 m/sec with a net velocity of I 2.0 m/sec from 147 degrees. November is the quietest month and the speed increases I PIERLAB 1988 1989 10 10 5 5 -....J U1 0 0 -5 -5 -10 -10 wUL I AUG! SEPI OCTI NOVI DEC lwANI FE WINO VELOCITY (M/SEC) , 1988-89, 49-HOUR SIMPA Figure 29 (a). Long-term wind patterns over the Corpus Christi/Nueces Bay Estuary. Aransas Pass Ship Channel (PIERLAB) NUECES 1988 :o j [:o I I I -....J O'l 0 0 -5 -5 -10 -10 wUL AUG SEP OCT NOV DEC wAN FE MAR APR MAY wUN wUL AUG WIND VELOCITY (M/SEC) , 1988-89, 49-HOUR SIMPA Figure 29 (b). Long-term wind patterns over the Corpus Christi/Nueces Bay Estuary. Nueces Bay (NUECES) PIERLAB 1988 1989 10 10 8 8 6 6 \ 4 4 2 2 -....,J ' -...J 0 0 '--'UL I AUG ISEPI OCT I NOVI DEC I '--'AN IFEB MAR IAPRI MAY I '--'UNI '--'UL I AUG WIND SPEED (M/SEC) , 1988-89, 49-HOUR SIMPA MEAN SPEED= 4.15 NET VELOCITY= 2.03 (M/SEC) FROM 147 DEG Figure 30 (a). Long-term wind speed fluctuations over the Corpus Christi/Nueces Bay Estuary. Aransas Pass Ship Channel (PIERLAB) NUECES 1988 1989 10 10 8 --8 - .... 6 ~ l -1~ ~ ~ 6 ~ 4 -~~ ' ~ \' \ -4 .... - I ~ ........ 2 ~~ 2 ~ co 0 0 wUL AUG SEP OCT NOV DEC wAN FEB MAR APR MAY wUN wUL AUG WIND SPEED (M/SEC) 1988-89, 49-HOUR SIMPA I MEAN SPEED= 4.38 NET VELOCITY= 2.54 (M/SEC) FROM 104 DEG Figure 30 (b). Long-term wind speed fluctuations over the Corpus Christi/Nueces Bay Estuary. Nueces Bay (NUECES) I I I I I I I I I I I I I I I I I I I I 79 to the windiest period (April, May and June). Again, Gilbert and the cold fronts in February, March and April are prominent. A generally more energetic wind-field occurs over Nueces Bay (Fig. 30b) with an annual mean of 4.4 m/sec and a net velocity of 2.5 m/sec from 104 degrees. May is the most energetic month here. The shear between winds at the coast and over the bay are due to the more maritime climate experienced at the PIERI.AB site. CONCLUSIONS What do these results tell about the water movements in and out of the Corpus Christi/Nueces Estuary? Far more water flows out of the system into the Gulf of Mexico through the Aransas Pass than comes in through the Nueces Bay channel and the Gulf Intracoastal Waterway channel at the northern end of the Laguna Madre. The measurements showed that 5.58 x 109 m3 of water flowed Gulfward through the pass from August 1988 through July 1989. During the period August 1988 through March 1989, when all three current meters were operating, the net flow (subtracting inflow from outflow) was 4.18 x 109 m3• This is more than three times the volume of Corpus Christi Bay (Armstrong, 1987). Using Smith's (1985) calculated "half-life" of water in Corpus Christi Bay of 105 days, it would take about two years to replace this volume by tidal flushing. How does this volume compare with the Nueces River streamflow? In 1984-1985 the daily streamflow was 9.4 x 105, or for a year 0.3 x 109 m3• This is an order of magnitude less than that measured through the Aransas Pass, but about the same as the outflow measured in the Nueces Bay channel in this present study. It seems unlikely I I I 80 that the excess flow Gulfward through the Aransas Pass could be accounted for by input from the only other entrance to Corpus Christi Bay, the Lydia Ann Channel to the I north. More likely, there is a tidal circulation within the pass, as indicated by the I difference in angle between the mean inflow and the mean outflow at the current meter site. A practical constraint to all these measurements is the problem of heavy maritime I and fishing traffic in the channels, making it impossible to locate sensors within the axes of the flow. Furthermore, the calculation of flow volume and direction of flow from the I current meters is sensitive to the orientation angle chosen for each of the channels. I There is undoubtedly additional flow from Nueces Bay and the Laguna Madre via the shallow connections and minor channels away from the main channels. I I In summary, there is a net Gulfward flow of water through the Aransas Pass in all months of the year. Flow is also Gulfward (into Corpus Christi Bay) in all of the I I seven months measured in the Nueces Bay channel, and in all but one month of the year at the Laguna Madre channel. Volume is an order of magnitude greater through the Aransas Pass than through the other two channels, but little tidal flow can be seen I at the Laguna Madre entrance. Atmospherically driven flow was important at all three I sites, especially during events such as Hurricane Gilbert and strong cold fronts in February and March 1989. Further analysis of these data is in progress. I I I I I I I 81 REFERENCES 1987. The ecology of open-bay bottoms of Texas: a community profile. I Armstrong, N.I U.S. Fish and Wildlife Service, Biological Report 85(7.12), 104 pp. Masch, F. D. 1972. Tidal hydrodynamic and salinity models for Corpus Christi and Aransas Bays, Texas. Report to Texas Water Development Board, I I #333.9/M37IT, 98 pp. Nowlin, W. D. Jr. and C. A. Parker. 1974. Effects of a cold-air outbreak on shelf waters of the Gulf of Mexico. Journal ofPhysical Oceanography 4:467-486. I I Smith, N. P. 1977. Meteorological and tidal exchanges between Corpus Christi Bay, Texas, and the northwestern Gulf of Mexico. Estuarine and Coastal Marine Science 5:511-520. I Smith, N. P. 1978. lntracoastal tides of Upper Laguna Madre, Texas. Texas Journal of I Science 30(1):85-95. 1985. Numerical simulation of bay-shelf exchanges with a one-dimensional Smith, N. P. I model. Contributions in Marine Science 28:1-13. I I Whitledge, T. E. 1989. Data Synthesis and Analysis, Nitrogen Processes Study (NIPS): Nutrient Distributions and Dynamics in Lavaca, San Antonio and Nueces/Corpus Christi Bays in Relation to Freshwater Inflow. Part II: Hydrography, Nutrient and Chlorophyll Data Tables. Final Report to Texas Water Development Board.I The University of Texas at Austin, Marine Science Institute, Technical Report I No. TR/89-007. I I I