origin and evolution of sabine lake, texas-louisiana

21 downloads 0 Views 1MB Size Report
ORIGIN AND EVOLUTION OF SABINE LAKE, TEXAS-LOUISIANA. John B. ~ n d e r s o n l , Femando P. siringanl, Marco avia ani^, and James ~ a w r e n c e ~.
1,

12

T R A N S A C T I O N S 4 U L F COAST ASSOCIATION OF GEOLOGICAL SOCIETIES

VOLUME XLI, 199 1

ORIGIN AND EVOLUTION OF SABINE LAKE, TEXAS-LOUISIANA John B. ~ n d e r s o n lFemando , P. siringanl, Marco

avia ani^, and James ~ a w r e n c e ~

ABSTRACT High resolution (uniboom)seisrnic reflection profiles and vibracores acquired from Sabine Lake. Texas, during 1987 aid in the reconstmction of the bay's history. Seisrnic records show that the bay is located over an incised vailey. initially carved during an early .lowstand, tentatively assigned to a180 Stage 5d. Thus. the vailey fill sequence represents more than one stage of incision and fill. Sediment cores collected along the flanks of the vailey penetrated bay rnuds resting directly on Pleistocene clays. Radiocarbon dates suggest that the broad, shallow bay floor was flooded approxirnately 4,000 years ago. Prior to this tirne the bay was confined to the narrow, deep incised valley. None of the cores that penetrated beyond the Holocene fill contain transitionai (deepening upwards) deposits, implying either that flooding was rapid or that erosion had rernoved these transitional deposits. Cores frorn the upper ponion of the bay were characterized by coarsening-upward sequences which appear to reflect increasing fluviai input of sand with tirne. Cores frorn the rnarsh area at the southern end of the lake penetrated marsh deposits resting on flood-tidal delta and bay deposits. A strong dorninance of the bivalve Rangia cuneara (a species generaily found in fresh to brackish waters) occurs throughout the Holocene section of sediment cores. Carbon isotopic values of these rnolluscan shells are generally low and consistent with low salinities. The only exception to this is seen in the very base of these cores, where the rnicrofossil assernblage and isotopic results indicate a brief period during which saiinities were higher than at present. Othenvise, the saiinity stmcture of the bay and its fauna have changed little during the past 4,000 years.

INTRODUCTION Thomas and Anderson ( 1988) and Anderson and Thornas (1 99 1) have argued that the Holocene rise in sea level (the 1 s t 10,000 years) was punctuated by rapid, episodic events during which the rate of rise was on the order of 3 to 5 c d y r . Thomas (1990) has examined the offshore portions of the Trinity and Sabine river incised valleys and concluded that these rapid flooding events have resulted in dramatic landward shifts in the estuaries that occupied these valleys. This study looks at how these rapid flooding events influenced the evolution of Sabine Lake. The investigation includes analysis of changes in the salinity structure of the lake, based on fossi1 assemblages and stable isotope anaiyses.

BACKGROUND INFORMATION Fisk (1944) fust recognized that incised valleys occumd beneath Texas and Louisiana bays and estuaries and probably extend offshore. He proposed that these vaileys were formed during late Pleistocene lowstands in sea level. Bernard et al. (1962) suggested that these incised vaileys were backfilled with sedirnent during the ensuing sea-leve1 rise.

' Depamnent of Geology and Geophysics, Rice University, Houston, Texas 7725 1 Department of Geosciences, University of Houston, Houston, Texas 77204 Instituto per la Geologia Marina, Bologna, Italy 40127

Initiai incision of the Sabine River Valley was during aI80 Stage 5d (-110 ka), based on seisrnic stratigraphic analysis (Thornas, 1990). Later, during the a180 Stage 2 lowstand, the vailey was reincised; this was the deepest incision (ranging from-35 to-40 m offshore). The later incision left rernnants of older Deweyville fluvial terraces aiong the flanks of the vailey (Thomas, 1990). During the late Wisconsinan-Holocene rise in sea level, the offshore portions of the Sabine Valley were completely back-filled with fluviai, estuarine, and marine sedirnents. As sea level rose during the Holocene, sand spits rapidly developed and restricted the rnouth of the estuary. An imponant influence on the development of the present configuration of Sabine Lake was the development of the chenier plain. Byrne et al. (1959) have shown that the southwestern Louisiana chenier plain (including the Sabine Pass area) is underlain by a seaward thickening wedge of Holocene sedirnents (7-8 m thick) resting on top of the Holocene-Pleistocene unconforrnity. The lower ponion of this sediment wedge consists of coastal marsh deposits grading upward and seaward into nearshore and offshore marine sediments. The younger section grades upward and seaward into nearshore marine to mudflat and coastal marsh deposits. Byrne et al. (1959) and Gould and McFarlan (1959) infer that the lower transgressive unit corresponds to the latest postglaciai rise that ended when sea level reached its present stillstand. The upper offlapping sequence reflects the progradation of the chenier plain during the past 3,000 t0 2,800 years of sea-leve1stillstand (Gould and McFarlan, 1959). Gould and McFarlan's (1 959)

ANDERSON, SIRINGAN, TAVIANI, LAWRENCE C-14 dates of the chenier plain ridges imply that the estuary became resuicted about 2,000 years B.P.

METHODS The incised vailey was first mapped using high resolution seismic reflection (uniboom) data (Fig. 1) augmented by Kane's (1959) incised valley map, the latter was based on sediment cores and engineering borings. Areas where cores would likely penetrate the Pleistocene-Holocenecontact were then identified. Twelve vibracores were acquired from the lake for this study (Fig. 1). The Rice University research vessel RN Maragorda was used to acquire these seismic data and cores.

13

-

Thus, carbon isotopic analyses of these species offer a means of measuring paleosalinity.

RESULTS Figure 1 displays the seismic data and sediment core iocations. Seismic line 4 (Fig. 2) illustrates the seismic facies within the vailey. Older, presumably Deweyville, fluvial terraces occur along the flanks of the valley, as is the case in the down-dip portions of the valley (Thomas, 1990). The base of the a180 Stage 2 incision is poorly imaged. A change in the seismic character of the estuarine deposits occurs at an average depth of about 9.5 m, from discontinuous widely spaced thick reflectors to more laterally continuous reflectors that exhibit m'o're uniform, even spacing. This transition is marked by a fairl$'irregular surface having relief of a few meters. This change in seismic character may reflect a shift from upper baylbay head delta facies to middle bay facies brought about by a flooding event. A flooding surface recognized offshore, based on the occurrence of preserved tidal inlet and related lithofacies beneath Sabine Bank at about -12.5 m and Unmediately offshore of Bolivar Peninsula at about -10 m (Anderson er al., submitted), correlates to this depth and probably represents the same event. Correlation with Fairbanks' (1989) sea-leve1curve gives an approximate age of 6500 years B.P. for this event. No lower bay (flood tidai deltaltidal inlet) seismic facies are identified in the seismic profiles except hthe southern part of Line 2, in the area of the present tidal inlet/flood- tidai delta. The flood-tidai delta facies exhibit ebb-oriented clinofoms. Core SL-5 penetrates densely packed shells which exhibit imbrication and layering. It consists mostly of oyster shells and has a muddy sand matrix.

Figure 1. Map for the study area showing locations of seismic lines (solid lines) and sediment cores (triangles). Location of Sabine incised valley (contour -30 m), indicated by stipple pattern, is based on earlier work by Kane (1959) and results from this study. Location of line drawing of a segment of Line 4 (Fig. 2) is indicated by a bold line.

-

Sediment cores were cut and described before sampling. Sedimentologicai descnptions involved characterization of sediient color, texture, and identification of sedimentary shuctures. Paieontologicai analysis included identification of forarninifera and macroinvertebrates. Stable isotope anaiyses have been conducted to examine the development of the bay's salinity strutture. Carbon isotope values of fresh water bicarbonate are low compared to marine bicarbonate (Schwarz, 1969). Carbon isotope analyses of recent and living samples of bivaive shells of the species Mulinia lateralis and Rangia cuneara exhibit a positive correlation with salinity along the Texas coast (Lawrence er al., 1990).

Figure 3 provides lithologicai descriptions of vibracores in conjunction with paleontologicai data. Cores SL-9,10,11, and 12 have been acquired frorn the backside of the chenier plain. Core SL-12 is representative of the lithologic changes seen in these cores. It contains middle bay muds and muddy, loosely packed oyster sheii interbeds directly on top of the Pleistocene surface. Fiood tidal delta deposits consisting of interbedded to interlaminated sand and mud units overlie the middle bay muds, which are, in m,overlain by marsh plain rnuds. Sediment cores SL-3,4, and 7 were the only cores from the lake proper to penemte the Holocene bay fill into older Pleistocene deposits. Radiocarbon dates of Rangia shells resting just above the Pleistocene surface at --5 m in core SL-7 yielded ages of 3,890 +l20 and 3,440 f140, and provided an estimate of the age for the flooding of the shallow flanks of the valley. None of the cores from the bay encountered a transitionai, deepening upwards (onlap) sequence. Rather, there is little change in the character of these sediments from top to bottom of the bay fill. The same holds for the fossils that occur in these

14

TRANSACTIONS-GULF COAST ASSOCIATION OF GEOLOGICAL SOCIETES

o

i

i

-l

Bey Llne Fluvial

$

0.5

40

Kllorneters

Pnsent .%a Leve1

a

-

-

-I - . / 1

VOLUME XLI, 199 1

/--

.

1.O

-.- .

Sabine Lake Line 4

1 I

50

Figure 2. Line drawing of a segment of Line 4 across the Sabine incised valley showing valley fill sequence. FS denotes flooding surface. Location of line is denoted on Figure 1.

Figure 3. Sedimentological and paleontological data for sediment cores from Sabine Lake. Carbon isotope values for SL-1 and SL-3 are indicated by nurnbers on the left side of the two core logs.

cores. The fossi1 assemblage is dominated by Rangia. Minor amounts of Mulinia are present. Elphidium gunteri dominates the forarniniferai assemblage. Ammonia beccari is present in srnaller auantities. Ovsters are notably absent, except in core

SL-12; this core was acquired from the southem ponion of the bay, the only part of the bay where oysters,live today (White et al., 1987).

ANDERSON, SRINGAN, TAVIANI, LAWRENCE

-

-

Carbon isotopic analyses of shells of the bivalve Rangia cuneara have been measured in cores SL-1 and SL-3 (Fig. 3). With the exception of two intemals, the carbon isotopic values are lower than -3.5 per mi1 (relative to PDB), the lowest value being -8.7 per mil. Thesr isotope values can be compared with those of modem sarnples of bivalves along the Texa Coast (Lawrence et al.. 1990) for which a salinity-isotope relationship has been developed. Comparison indicates that the s a h i ties in Sabine Lake never exceeded 15 gmkgm when these shells formed. The only exceptions occur near the base of these cores, where higher isotope values suggest that salinities may have been as high as 25 gmkgm. This is supponed by work of Kane (1959). who observed that the microfaunai assemblage in the deeper portions of cores from the lake are characterized by an assemblage more tolerant to higher salinities-his "Rotali~"-Elphidiumassemblage.

CONCLUSIONS Sabine Lake occupies an incised valley that was initially formed during an older Pleistocene glacial eustatic lowstand and reincised during the late Wisconsinan ìowstand. Thomas (1990) and h d e r s o n and Thomas (1991) have identified older flooding events recorded in the offshore ponions of the Trinitylsabine River incised valley. These earlier flooding events are confined in the deeper ponions of the vailey beneath Sabine Lake. Extrapolation of flooding surfaces updip indicates that initial base level encroachment of the valley occurred between 10 to 8 ka ( h d e r s o n etal., submirted; Siringan et al., 1991). Initially, the bay was a narrow estuary that occupied only the deep incised ponions of the old nver valley. The early phase of bay evolution is poorly known due to a lack of sediment cores which penetrate the older vailey fill sequence. An estimate of -4,000 years B.P. for the flooding event which subrnerged the broad shallow portions of the bay comes from C-14 dates on core SL-7. The same timing and altitude of a flooding surface in Galveston Bay and other Texas bays have been found by Smyth (1991) and Siringan er al. (1991). A lack of change in sediment character with depth in the cores from the shallow flanks (--6 m) of the valley, including an absence of a deepening-upwards sequence, implies rapid flooding. This is collaborated by the results of the stable isotope analyses which show a very brief period of higher salinity foiiowed by establishment of a salinity strutture sirnilar to that of the present. There is no apparent change in the macrofossil assemblage of the bay following the 4,000 year B.P. flooding event. These results indicate that the mouth of the estuary became restricted shortly after 4,000 years B.P.

ACKNOWLEDGEMENTS Our research on the Texas continental shelf is made possible tìirough grants from the National Science Foundation (Grant

.

,

--

15

OCE-8908320). the Petroleurn Research Fund ( P R F 23194-AC8), and by the Industriai Associates Group consisting of the following oil cornpanies; Amoco, Arco, BP, Shell, Unocal, and Union Pacific.

REFERENCES CITED Anderson, J.B., M.A. Thomas, F.P. Siringan, and W.C. Smyth, Quatemary evolution of the east Texas coast and continental shelf, in J. Wehmiller and C. Fìetcher, eds.: SEPMDGCP Speciai Publication (subrniued). Anderson, J.B. and M.A. Thomas. 1991, Marine ice sheet decoupling as a mechanism for rapid, episodic sea level change: the record of such events and their influence on sedimentation: Sedirnentary Geology, v. 7, (in press). Bemard. H.A., R.J. LeBlanc, and C.F. Major, 1962, Recent and Pleistocene geology of southeast Texas, in Geology of the Gulf Coast and CentralTexas and Guidebook of Excursions: Houston Geologica1 Society, Houston, p. 175-224. Byrne, J.V., D.O. LeRoy, and C.M. Riley, 1959, The chenier plain and its suatigraphy, southwestern Louisiana: GCAGS Trans., v. 9, p. 237-260. Fairbanks, R.G., 1989, A 17,000-year glacio-eustatic sea level record: influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation: Nature, v. 342, p. 637-642. Fisk, H.N., 1944, Geological Investigation of the Alluvial Valley of the Lower Mississippi River: U.S. Army Corps of Engineers, Mississippi River Commission, Vicksburg, 145p. Gould, H.R. and E. McFarlan, 1959, Geologic history of the chenier plain, southeastern Louisiana: GCAGS Trans., v. 9, p. 261-270. Kane, H.E., 1959, Late Quatemary geology of Sabine Lake and vicinity, Texas and Louisiana: GCAGS Trans., v. 9, p. 225335. Lawrence, J.R., A.M. Perez-Guzman, and A. Cate, 1990, Paleosalimties from the carbon isotopic composition of fossils?: GSA Annua1 Meeting, Abstracts with Program, October 29-November 1, Dallas, p. A87. Schwarcz, H.P.,1969, The stable isotopes of carbon: in Handbook of Geochemistry, volume 1 111: Springer-Verlag, New York, p. 6-B-1 to 6-B-16. Siringan, F.P.. J.B. Anderson, and W.C. Smyth, 1991, Rapid flooding events due to Holocene episodic sea level rises as recorded in Texas bay deposits (abs): SEPM/lGCP Coastal Evolution Conference, May 8-1 1,199 1, Tallahassee, Fìorida (in press). Smyth, W.C., 1991, Seisrnic facies analysis and depositional history of an incised valley system, Galveston Bay area, Texas: Unpublished M.A. thesis, Rice University, Houston, 170p.

16

TRANSACTIONS-GULF COAST ASSOCIATION OF GEOLOGICAL SOCIETIES

Thomas, M.A., 1990, The impact of long-term and shon-term sea level changes on the evolution of the Wisconsinan-Holocene TrinityISabine incised valley system, Texas continental shelf: Unpublished Ph.D. dissenation, Rice University, Houston, 263p. Thomas, M.A. and J.B. Anderson, 1988, The effect and mechanism of episodic sea level events: the record preserved in

-

VOLUME XLI, 199 1

Late Wisconsinan-Holocene incised valley-fill sequences: GCAGS Trans., v. 38, p. 399406. White, W.A., T.R. Calnan, R.A. Morton, R.S. Kimble, T.G. Littleton, J.H. McGowen, and H.S. Nance, 1987, Submerged lands of Texas, Beaumont-Pon Arthur area: sediments, geochemistry, benthic macroinvenebrates, and associated wetlands: Bureau of Economic Geology, Austin, 110p.