High-Resolution Stratigraphy of a Sandy, Ramp-Type … et al., 2004.pdf · Sequence stratigraphic models have been used to predict sand-body distribution and facies character (Posamentier

25HIGH-RESOLUTION STRATIGRAPHY OF A SANDY, RAMP-TYPE MARGIN—APALACHICOLA, FLORIDA, U.S.A.

HIGH-RESOLUTION STRATIGRAPHY OF A SANDY, RAMP-TYPE MARGIN—APALACHICOLA, FLORIDA, U.S.A.

HEATHER A. MCKEOWNDepartment of Geology and Geophysics, Louisiana State University, Baton Rouge, Louisiana 70803-4101, U.S.A.

Present address: ExxonMobil Exploration Company, 233 Benmar, Houston, Texas 77060-2304, U.S.A.PHILIP J. BART

Department of Geology and Geophysics, Louisiana State University, Baton Rouge, Louisiana 70803-4101, U.S.A.AND

JOHN B. ANDERSONDepartment of Earth Sciences, Rice University, Houston, Texas 77251-1892, U.S.A.

ABSTRACT: High-resolution single-channel seismic data were collected on the middle shelf to upper slope, offshore Apalachicola, Florida,to characterize fluvial–deltaic stratigraphy on a sandy, ramp-type margin. On the west Florida shelf, deltas exhibit wedge-shapedgeometries. Clinoform angles are relatively steep (~ 3°), and bottomset bed development is minimal, suggesting deposition in a relativelyhigh-energy environment.

Chronostratigraphic control for the study area is absent, so ages of deltaic units were inferred from the updip pinch-out elevations, offlapbreak elevations, and stratigraphic relationships. From the stratal relationships, we infer that at least three glacioeustatic cycles (OIS-9 toOIS-1) were imaged. Five units were identified, with unit 1 being the youngest.

Sediment volume for unit 4 (OIS-7 to OIS-6) was four times greater than unit 2 (OIS-5 to OIS-4). Unit 2 drainage was line-sourced anddeposited deltas with a shore-parallel steep seaward margin. In contrast, unit 4 drainage was point-sourced and deposited elongate deltason the shelf.

Highstands were dominated by deltas deposited on the middle to outer shelf. Subaerial exposure of the inner shelf and the developmentof an extensive braid plain characterized lowstand deposition. Lowstand incised fluvial valleys, canyons, and deep-water fans are absent.The transgressive systems tract is characterized by valley fill and sand ridges on the inner shelf and backstepped deltas and slope wedgeson the middle shelf to upper slope.

Late Quaternary Stratigraphic Evolution of the Northern Gulf of Mexico MarginSEPM Special Publication No. 79, Copyright © 2004SEPM (Society for Sedimentary Geology), ISBN 1-56576-088-3, p. 25–41.

INTRODUCTION

Sequence stratigraphic models have been used to predictsand-body distribution and facies character (Posamentier andVail, 1988; Posamentier et al., 1988; Van Wagoner et al., 1990;Christie-Blick, 1991). These conceptual models are constrainedby studies of outcrop, well logs, and low-resolution seismic data.One key assumption is that sequence stacking patterns have aglobal signature that is repetitive as long as other controllingfactors, such as climate, discharge, and subsidence, etc., remainconstant. Yet, if controlling factors were not constants from cycleto cycle, then the stratigraphic architecture between sequencesshould vary.

With respect to other drainage systems entering the Gulf ofMexico, the Apalachicola fluvial–deltaic system is considered anend-member setting (Anderson et al., this volume). A moist-humid climate, steep incised streams, ramp morphology, mini-mal to absent growth faulting and salt tectonics, and relativelylow subsidence rates characterize this system.

Previous studies of the Florida inner shelf offshore theApalachicola River suggest that at least two major late Pleis-tocene glacioeustatic cycles are recorded above an unconformityat the top of the upper Miocene limestone (Schnable, 1966; Schmidt,1984; Locker and Doyle, 1987, 1992; Doyle and Locker, 1988).Preliminary seismic interpretation strongly suggests that thedata grid acquired for this study images strata from a minimumof three glacioeustatic cycles. Donoghue (1992, 1993) mappedpaleo-channels associated with the ancestral Apalachicola Riveron the inner shelf. These previous studies provided an importantframework for this investigation.

The objectives of this study include the following: (1) charac-terize fluvial–deltaic stratigraphy on a sandy, ramp-type margin;(2) evaluate timing of deposition on the shelf using the oxygenisotope curve as a proxy for sea level; (3) evaluate depositionalstyles between different glacioeustatic cycles; and (4) character-ize sedimentation associated with different systems tracts.

METHODS

Approximately 500 km of high-resolution single-channel seis-mic data were collected during the summer of 1998 aboard the RVLone Star (Fig. 1). The grid was designed to capture strike and diporientations of delta lobes identified on the basis of seawardprotuberances (Fig. 1). Seismic sources included an 800 joulesparker and 15 cubic inch water gun. A Benthos single-channelstreamer was used during the entire survey.

In addition to high-resolution seismic data, attempts to re-trieve sediment gravity cores were made along seismic lineAPL03 (Fig. 1). These attempts were mostly unsuccessful owingto the hard, sandy bottom. In place of cores, sediment grabsamples were collected in three locations (gs03, gs04, and gs05).

The digital acquisition system failed during the cruise, so onlyanalog records were acquired. Seismic interpretation and map-ping were conducted by hand on records generated from an EPCplotter. Bounding surfaces and units were placed within a strati-graphic context and related to glacial-eustatic cycles and systemstracts (lowstand, transgressive, and highstand) using an oxygenisotope record as a proxy for sea level (Fig. 2). The oxygen isotopestages (OIS) for the last glacial cycle are numbered 1 to 5. Thesystems-tract designations indicated in Fig. 2 follow the conven-

Copyright © 2012, Society for Sedimentary Geology (SEPM)

HEATHER A. MCKEOWN, PHILIP J. BART, AND JOHN B. ANDERSON26

APL11

29 15'

29 30'

28 45'

29 45'

85 30' 85 15' 85 00'

��

��

��

��

��

��

isobath interval = 10 m

APL tracklines

25 kilometers

17

14

02

03

19

01

��

��

��

drop cores/grabsamples

��

Study Area

04

11

06

12

15

13

07

08

1610

29 00'

convex seawardbathymetric contours

09

Apalachicola River

4

FIG. 1.—Map showing data set including high-resolution seismic tracklines and grab sample locations. Bathymetry is from USGSand NOAA published maps for Apalachicola (1979), Florida Middle Ground (1977), and Destin Dome (1975).


tion used in this volume. Studies by Abdulah (1995) and Banfield(1998) have shown that the method of discerning relative timingof sedimentation from the positioning of bounding surfaces andunits below the seafloor, and use of the sea-level proxy curve todiscern depths and timing of sedimentation, is reliable in thenorthern Gulf of Mexico. This is based on seismic-stratigraphicstudies on the Texas shelf where chronostratigraphic control isavailable (e.g., Abdulah et al., this volume; Banfield and Ander-son, this volume; Fillon et al., this volume).

The seafloor is the maximum flooding surface associated withthe current highstand (OIS 1). The first prominent boundingsurface beneath the modern seafloor is the OIS 1 transgressiveravinement surface (TS-1) (Fig. 3). This is a planar erosionalsurface. Beneath this surface is the sequence boundary associatedwith OIS 2 (SB-2) (Fig. 3). The sequence boundary is characterizedby fluvial incision on the inner shelf. The maximum floodingsurface of OIS 5 (MFS-5) is characterized by downlap on themiddle to outer shelf (Fig. 3). This same naming convention isused for bounding surfaces of previous glacioeustatic cycles.

Prograding clinoforms with successively lower toplap eleva-tions in a basinward direction represent deposition during rela-tive sea-level fall. The updip pinch-out of the unit represents theonset of deposition, and the offlap break represents the culmina-tion of progradation (Fig. 4). Elevations were related to the sea-level proxy curve to determine timing of deposition. Preliminaryresults from this study suggest that the offlap break elevationdoes not always correspond with the maximum eustatic lowstandas predicted by sea-level proxy curves. To convert two-way traveltime to depth, a velocity of 1500 meters per second (m/s) wasused (Anderson et al., this volume).

The elevation of the updip pinch-out and offlap break may beinappropriately deep, because subsequent changes in sea levelmay remove, by ravinement or fluvial erosion, the upper part ofthe unit. Because the updip pinch-out is located in interfluveareas, most of the erosion is probably caused by transgressiveravinement. The depth of erosion by the transgressive ravinement

surface has not been measured on the west Florida shelf. How-ever, the shelf morphology is similar to south Texas in terms ofgradient and width (Banfield, 1998). There, the depth of trans-gressive ravinement is approximately 10 m (Rodriguez et al.,2001). On the basis of results from south Texas, the elevation of theupdip pinch-out could be approximately 10 m deeper, owing toerosion by the overlying transgressive ravinement surface.

Isochron maps were used to calculate volumes of deltas.Calculated sediment flux values are minimum estimates, becausesediment volumes include only depositional bodies expressed bydeltaic foreset strata. In reality, sediment included in fluvial sheetcomplexes, coastal-plain environments, and transgressive sheetsands would increase volume estimates. Also, transgressive ero-sion and associated longshore transport are capable of removingsubstantial sediment volumes from the area. Results from thisstudy suggest that early highstand and lowstand deposits arevirtually lacking within stratal packages of the shelf and upperslope and transgressive deposits are relatively thin. This suggeststhat the most of the sediment delivered to the shelf during aglacioeustatic cycle ultimately ends up in late-highstand deltas.

STUDY AREA

The study area is located on the middle and outer Florida shelfin water depths ranging from 30 to 150 meters (Fig. 1). The surveyarea is between 85° and 86° W longitude and 28° 45’ and 29° 45’N latitude. This area is offshore Apalachicola and Panama City,Florida.

The Apalachicola River drainage basin covers an area ofapproximately 60,000 km2 and exists in Alabama, north Florida,and Georgia (Fig. 5). On the basis of water discharge, theApalachicola River is the largest river in Florida and the fourthlargest entering the Gulf of Mexico (Donoghue, 1993). The drain-age-basin geology is divided into two provinces: the PiedmontPlateau and the Coastal Plain (Fig. 5). The Valley and Ridgeprovince is beyond the northern limit of the drainage basin. The

230 Th age (ky)

Compositebenthic/planktic δ18Ocurve, Pacific ocean(Shackleton, 1987)

Benthic δ18O curve,Norwegian sea(Labeyrie et al., 1987)

U/Th dates ofBarbados corals(Bard et al., 1990)

OIS 2 OIS 3 OIS 5e

ca

OIS 6

b d

20

0

-20

-40

-60

-80

-100

-120

-1400 20 40 60 80 100 120 140

early HST

HS

T

LST

OIS 1

TS

T

late HST

sea

leve

l (m

)

U/Th dates ofHuon, NewGuinea corals(Chappell et al.,1996)

New Guinealowstand deposits(Chappell et al.,1996)

OIS 4

FIG. 2.—Sea-level curve derived from U/Th dates of Barbados corals (Bard et al., 1990), U/Th dates of Huon, New Guinea, corals(Chappell et al., 1996), New Guinea lowstand deposits (Chappell et al., 1996), Benthic δ18O curve from the Norwegian Sea(Labeyrie et al., 1987), and the composite curve from benthic/planktic δ18O curve from the Pacific Ocean (Shackleton, 1987). Alsoindicated are the oxygen isotope stages and systems tracts for the last glacial cycle.


geologic formations that crop out within the lower reaches of thedrainage basin are mainly Oligocene and Miocene clays, sands,marls, and limestones, although recent deposits cover these nearthe coast.

The present climate is humid semitropical with a mean tem-perature of 19°C (67°F) and mean summer temperature of 27°C(81°F) (Kofoed and Gorsline, 1963). The region experiences pro-longed dry periods and a summer rainy season. Modern precipi-tation varies from 127 cm/yr (50 in/yr) in central Alabama andFlorida to 178 cm/yr (70 in/yr) in Georgia (Fig. 5).

The Apalachicola River drainage basin contains more ruggedand steep topography relative to Texas drainage basins of similarsize. The Tertiary coastal plain of panhandle Florida is approxi-mately 200 kilometers wide (Fig. 5), compared to approximately450 kilometers for east Texas. The average gradient along theentire reach of the Apalachicola River is 0.15 m/km (Donoghue,1989).

Updip of ancient deltaic lobes on the inner to middle shelf,Donoghue (1993) recognized paleo-channels that are consider-ably deeper and wider (600 m wide) than the modern riverchannel. This is potentially the result of increased paleo-dis-charge at the last glacial maximum (LGM) (i.e., OIS 2). Thelocations of these paleo-channels on the inner shelf suggest thatlobes of the ancient Apalachicola Delta were west of the modernsystem. This is consistent with steep bluffs on the east side ofmodern streams. On the basis of these results, we expected to findfluvial valleys and depocenters to the west of the modern river.

The shelf is characterized by long-term tectonic stability, lowto moderate wave energy, a lobate coastline, and mild climate

(Donoghue and Tanner, 1992). The present gradient of the shelfis 1.3 m/km (inner shelf) to 3.5 m/km (outer shelf), and the shelfis ramp-like in character (Fig. 1). Subsidence is low, estimated atapproximately 0.5 mm/yr on the shelf, on the basis of the depthof the OIS 3 maximum flooding surface directly west of this studyarea. Long-term subsidence rates for the study area are indicatedby westward thickening of the post-Miocene sedimentary unitsfilling the Apalachicola Embayment, a negative structuralembayment underlying the study area (Schmidt, 1984). Quater-nary depocenters associated with the Apalachicola system havenot prograded beyond the lower Cretaceous carbonate platform,which reflects the lower subsidence rates compared to Quater-nary depocenters offshore Louisiana and Texas.

The predominant wind direction is from the south-southeast.Southeasterly winds generate waves that produce a net west-ward-flowing longshore current. The microtidal setting experi-ences diurnal to mixed tides with an average range of only 0.4–0.5m (National Oceanic and Atmospheric Administration, 1999).

Sediment deposited on the shelf includes medium- to coarse-grained quartzose sand, silt, and clay, and a large component ofshell material (Kofoed and Gorsline, 1963). The terrigenoussand probably originates in the upper reaches of the drainagebasin. Grain size increases offshore. Deposition of these sandson the shelf occurred when stream competence and capacitywere higher than present (Kofoed and Gorsline, 1963; Ludwick,1964; Doyle and Sparks, 1980; McBride et al., 1999; McBride etal., this volume). Tributaries dissected Tertiary and Quaternarysediments of the coastal plain, which were ultimately derivedfrom the Piedmont Plateau. The fine-grained material is from

one kilometer

0

100

200

two-w

ay trav

el time (m

sec)

SWNE

v.e. ~ 20x

multiple

MFS-5

unit 2

unit 4SB-6

TS-1

SB-2

MFS-7

unit 3

0

100

200 two

-way

tra

vel

tim

e (m

s)

W

unit 1

upper Miocene carbonates

SB-2/MFS-5 TS-1SB-6/MFS-7unit 3

E

one kilometer

v.e. ~ 32x

High

Low

Eust

asy

Time (Ka)

0 50 100 150 200

-120

5

Dep

th (

m)

SB-6

MFS-7

SB-2

MFS-5

TS-1

Dip Profile - Middle to Outer Shelf

Strike Profile - Inner ShelfEustatic Curve

onlap

toplap

downlap

FIG. 3.—Line drawings of dip (APL 08) and strike (APL 04) seismic profiles to illustrate ages of seismic stratigraphic boundaries. Thedip-oriented profile best illustrates the generalized stratigraphic architecture and flooding (downlap) surfaces, whereas strike-oriented profiles best image sequence boundaries (fluvial incision). Surfaces are also indicated on the eustatic curve showing theirapproximate time of formation. Ages are inferred from the relative positioning of surfaces beneath the sea floor, with TS-1 beingthe upper transgressive surface, SB-2 being the upper sequence boundary, and MFS-5 being the upper maximum flooding surface.


the sandy soils and unconsolidated Pleistocene sediments thatcover most of the drainage basin (Kofoed and Gorsline, 1963).The distribution of shell material on the outer shelf is the resultof sufficient wave energy for winnowing, biologic distributionof organisms, and paucity of terrigenous sediment (Kofoed andGorsline, 1963).

In contrast to Texas and Louisiana, canyons and fans are notfound on the slope. This is partially a function of the differentshelf and fluvial morphologies of the Apalachicola system.

RESULTS

Stratigraphic Surfaces

Seven prominent bounding surfaces were mapped (seafloor,OIS 2-1 transgressive surface, OIS 2 sequence boundary, OIS 5maximum flooding surface, OIS 6 sequence boundary, OIS 7maximum flooding surface, and OIS 8 sequence boundary).

Modern Maximum Flooding Surface.—

The modern maximum flooding surface is the seafloor asso-ciated with the present highstand. There is no evidence forfaulting or salt-related structures intersecting the seafloor. Ingeneral, the seafloor surface slopes basinward at 1.3–3.5 m/km(0.1–0.2°) (Fig. 1).

In water depths between -10 and -40 m, the seafloor is charac-terized by asymmetric elongate ridges, believed to represent thetops of sand ridges (McBride, 1997; McBride et al., 1999) (Fig. 1).On the basis of modern bathymetric maps, individual ridges are< 10 m high with 10 km wavelengths (Fig. 1).

On the middle shelf, below ~ 40 m water depth, the averageseafloor slope is ~ 0.2° (Fig. 1). Mounded features with low, broadprofiles (Fig. 6) and subtle terraces are associated with deltatopsets. These terraces probably represent wave-cut platformsand are located mainly at a water depth of approximately -65 m.This depth matches the elevations of flat top carbonate mounds,to the west, on the Mississippi–Alabama shelf (Sager et al., 1999).

The seafloor gradient steepens (~ 3.4°) at the distal limits ofbathymetric protuberances on the outer shelf (Fig. 1). Seismicanalysis demonstrates that convex-seaward bathymetric con-tours and steeper gradients represent the preserved depositionalfronts of fluvial deltas. Figure 7 is a dip profile that shows theremarkable preservation of these deltas. The upper surface ofwhat appears to be barriers, recognized on the inner shelf byTanner (1966), are the only bathymetric features on an otherwisesmooth sea floor that slopes gently offshore (Fig. 8). Seismicanalysis shows no evidence for late Quaternary slope canyons orfans. Seaward of deltaic foresets, the seafloor is smooth with anaverage slope of ~ 0.1° (Fig. 9).

OIS 2 to 1 Transgressive Surface.—

The OIS 2 to 1 transgressive surface (TS-1) mimics the mod-ern seafloor with convex-seaward contours in water depthsbetween -100 and -150 m (Fig. 10A). TS-1 is an erosional surfacethat dips seaward at approximately 3.2 m/km (0.2°). Relativelythin transgressive deposits, unit 1, are formed between theseafloor and TS-1 (Figs. 7, 8). On the inner shelf, TS-1 truncatesSB-2 and MFS-5; only the deepest incised valleys of SB-2 arepreserved below TS-1.

seaward

-60

-75

-90

met

ers

Downlap

Toplap

B

�

Sea

Lev

el (

m)

25

0

-25

-50

-75

-100

-125

0 50 100Time (Ka)

1

2

3

4

5

deposition

SB-2

A

FIG. 4.—Methodology used to constrain the timing of deposition from the elevations of topset beds. “A” marks the approximate waterdepth at the onset of delta progradation (assumed -60 m) and “B” is the approximate water depth at the end of progradation(assumed -90 m) or offlap break. Because topset elevations decrease seaward, it is inferred that sea level was falling duringdeposition of the delta. The time interval from -60 m to -90 m is highlighted on the sea-level curve to the right. This is theapproximate time span of deposition. This methodology assumes that onlap strata are coastal-plain deposits and not fluvialdeposits.


On the inner shelf, TS-1 is the base of the unit whose uppersurface exhibits ridges at the seafloor. Where paleo-channels arepresent, TS-1 overlies the valley fill. A change in seismic faciesoccurs above and below the surface, from chaotic (valley fill) toparallel above.

On the middle–outer shelf, TS-1 truncates high-angle clino-forms and/or caps distributary channel fill (Fig. 7). Where TS-1 isnot directly underlain by clinoforms, parallel reflectors charac-terize the underlying unit. The dips of the overlying strata aregenerally conformable to the dip of TS-1.

The upper slope is characterized by strata dipping gentlyseaward with an average gradient of 2.9 m/km (0.2°) (Figs. 9,10A). TS-1 is parallel to the dipping strata.

OIS 2 Sequence Boundary.—

SB-2 is difficult to map, because of chaotic seismic facies aboveand below the surface (Fig. 7). Updip of the 60 meter isobath, theOIS 2 sequence boundary (SB-2) is characterized by incision on

the inner shelf (Fig. 10B). Dimensions of channel complexes rangefrom approximately 350 m wide and 15 m deep to approximately3.5 km wide and greater than 15 m deep. Northeast–southwestoriented channels deepen, widen, and converge in an offshoredirection (Fig. 10B). Channel dimensions are similar to thosemapped by Donoghue (1992, 1993) on the inner shelf in waterdepths less than 30 m. Seaward of the -60 m isobath, channelsshoal abruptly in a basinward direction (Fig. 10B). Basinward ofthis location, the character of SB-2 changes to a flatter erosionalsurface.

SB-2 is characterized by truncation of underlying clinoformson the middle to outer shelf, downdip of the -60 m isobath (Fig.7). The surface shows less relief, and the unit above is character-ized by chaotic seismic facies with shallow (< 15 m deep)complex channel geometries. Judging by their depth and loca-tion seaward of the main trunk stream and updip pinch-out ofthe delta, these channels are interpreted as distributary chan-nels. The channels locally incise the depositional front of thedelta (Fig. 7).

100 Kilometers

85 84

30

31

32

33

34

7060

55

50

60

55

ALABAMA

GEORGIAFLORIDA

GULF OF MEXICO

CHATAHOOCHEE R.

FLINT R.

APALACHICOLA R.

Valley and Ridge

Province

Piedmont

Plateau

Coastal Plain

CamelLake

87 86 oo o o

o

o

o

o

o

FIG. 5.—Drainage-basin map constructed from topographic maps. Light gray shading represents modern coastal plain, and dark grayshading represents modern highlands. Precipitation contours (dashed lines) are inches/year (from: Department of Agriculture).Geologic provinces are indicated (from: Locker and Doyle, 1992).


On the outer shelf, SB-2 is characterized by a continuous,parallel reflector. Above and below SB-2, strata are acousticallylayered. On the inner shelf (< 60 m water depth), SB-2 is amalga-mated with overlying (TS-1) and underlying (MFS-5) surfaces(Fig. 10B).

OIS 5e Maximum Flooding Surface.—

The OIS 5e maximum flooding surface (MFS-5) is truncatedby either SB-2 or TS-1 landward of the- 60 m isobath (Fig. 10C).Downdip of the -60 m isobath, the overlying stratigraphic sectionis characterized by clinoforms (Fig. 7).

MFS-5 is a regional downlap surface for overlying clinoforms(Fig. 7). Strata beneath the surface are conformable and acousti-cally layered (Fig. 7). Locally, MFS-5 is conformable with mounded,parallel reflectors associated with the underlying unit (Fig. 6).The gradient of MFS-5 is approximately 3.3 m/km (0.2°) to thesouthwest in interdeltaic regions (Fig. 10C). In the vicinity of thedepositional front of the underlying unit, the surface steepens toapproximately 30 m/km (1.7°) (Fig. 10C).


Updip of the 60 meter isobath, the OIS 6 sequence boundary(SB-6) is characterized by incision and truncation of layeredstrata (Fig. 10D). Overlying the erosional surface is chaoticchannel fill. Channel dimensions are generally broader (aver-age width of 2 km) and deeper (average depth of ~ 30 m) than

those associated with SB-2. Stage 6 channels are less constrainedby sea level, and valleys were potentially occupied longer.Unfortunately, SB-6 channels are commonly obscured by theseafloor multiple. Where imaged above the water-bottom mul-tiple, SB-6 channels are often eroded by the overlying SB-2surface. The result is amalgamated and crosscutting channelsystems. Channels shoal out basinward and are not observedbelow -60 meters water depth (Fig. 10D). At the seaward pinch-out of the underlying unit, SB-6 amalgamates with the underly-ing MFS-7 (Figs. 9, 10E).

Downdip of the -60 m isobath, SB-6 is characterized by trun-cation of underlying clinoforms (Figs. 6, 8) and onlap (Fig. 9) ordownlap (Fig. 6) by overlying layered or prograded strata. Nearthe axis of the underlying delta, SB-6 is characterized by chaoticseismic facies, believed to represent distributary channels. Clino-forms of the underlying unit are best imaged near the basinwardlimits of the delta. Onlap onto SB-6 generally occurs where thedepositional front of the underlying unit steepens (Fig. 9), anddownlap onto SB-6 occurs where the surface is relatively flatowing to amalgamation of the sequence boundary and transgres-sive surface (TS-2, Fig. 6).

The time–structure map of SB-6 shows topography thatwas created by the progradational outbuilding of two under-lying delta lobes (Fig. 10D). Between these depocenters, SB-6 ischaracterized by incision (Fig. 10D). However, the incisiondoes not appear to connect with depocenters on the outer shelfor slope. On the upper slope, SB-6 is a relatively flat surface(0.1°) (Fig. 10D) with layered strata above and below, except

100

200

260

two-w

ay t

ravel

tim

e (m

s)

one kilometer

NE N S 0

100

200

300

MFS-7

MFS-5

unit 3

lateral pinch-out

of unit 4

unit 3

SB-6unit 4

unit 4 unit 3

SB-6

MFS-7

MFS-5

multiple

v.e. ~ 22x

APL14APL15

oblique

parallel

oblique

tangential

downlap

lens-shape

geometry

abrupt pinch-out

and amalgamation

of intermediate

bounding surfaces

wedge-shape

geometryshingled

TS-2SB-6/TS-2

two-w

ay t

ravel

tim

e (m

s)

FIG. 6.—Seismic lines APL14 and APL15 together constitute a dip profile on the middle–outer shelf (see Figure 1 for locations). Thisprofile shows units 3 and 4 with bounding surfaces. The lateral pinch-out of unit 4 is shown. Transgressive unit 3 fills thebathymetric low between older, unit 4 highstand delta lobes. Unit 3 is characterized by 3 generations of backstepped delta lobes.


where chaotic wedges onlap the surface. This indicates that theincision is a headward-eroded valley that was not connected toa major fluvial source of sediment. The sediments eroded fromthis valley were apparently distributed uniformly across theslope.

OIS 7 Maximum Flooding Surface.—

The OIS 7 maximum flooding surface (MFS-7) is character-ized by downlap of overlying clinoforms (Figs. 6, 8). Theunderlying unit is generally obscured by the seafloor multiple.Where MFS-7 is located beneath the seafloor multiple, map-ping the surface required greater interpolation. Where strataare imaged, they are generally conformable with the relativelyflat MFS-7.

On the inner shelf, MFS-7 is locally truncated by SB-6 (Fig.10E). Locally, on the middle to outer shelf, the underlying unit isabsent and MFS-7 is eroded by SB-6 (Fig. 9). Where the twosurfaces are amalgamated, strata are laminated above and belowthe surface, except where overlying chaotic slope wedges arepresent (Fig. 9).

On the middle to outer shelf, MFS-7 is a downlap surface foroverlying oblique parallel, oblique tangential, and sigmoidclinoforms (Figs. 6 and 8). To the east, where clinoforms are not

well imaged, mapping is based on seismic facies character. AboveMFS-7, the unit is chaotic (Fig. 9). Below MFS-7, reflectors areparallel and continuous. The overlying unit is locally character-ized by aggraded bottomset beds. Beyond the depositional frontof the overlying unit, these bottomset beds have the same seismiccharacteristics as the layered strata (Fig. 8). This makes mappingMFS-7 in these areas difficult.

The time–structure map of MFS-7 shows the gradient andmorphology of the seafloor prior to significant deposition associ-ated with younger units (Fig. 10E). The gradient of MFS-7 isapproximately 2.3 m/km (0.1°) (Fig. 10E).


The OIS 8 sequence boundary (SB-8) is imaged only in thewestern part of the mapped area (Fig. 10F). In most locations, it isobscured by the seafloor multiple.

On the inner shelf, SB-8 is locally truncated by SB-6 (Fig. 10F).Incision and truncation of underlying strata characterize thesurface.

On the middle to outer shelf, SB-8 is characterized by trunca-tion of underlying clinoforms and onlap of overlying layeredstrata. Outside the area of deltaic deposition, seismic facies aboveand below the surface are layered.

FIG. 7.—Seismic line APL03 is a dip profile from the middle shelf to the outer shelf (see Figure 1 for location). This profile shows unit2 with associated bounding surfaces. The profile shows the remarkable preservation of deltas at the seafloor, external wedge-shape geometry, absence of thick aggraded bottomset beds, distributary incision to the depositional front, mostly obliquetangential and oblique parallel clinoforms, and local sigmoid geometries associated with the progradational center. Delta lobeswitching and progradational outbuilding events are also indicated.

0

100

200

one kilometer

100

200

multiple

progradational

center

MFS-5

TS-1

MFS-5

TS-1

SB-2

unit 2

delta

delta lobe

switching

v.e. ~ 29x

N Sprogradational

center

unit 2

sea floor

steep foresets

(~3°)

wedge-shapedsigmoid

clinoforms

DLL

100

200multipleMFS-5

TS-1

SB-2

laminated

strata

beneath

MFS-5

unit 1

unit 2

two

-way

tra

vel

tim

e (m

s)

two

-way

tra

vel

tim

e (m

s)tw

o-w

ay t

rav

el t

ime

(ms)


Unit Descriptions

Five seismic stratigraphic units (units 1 through 5) wereidentified in this study. Seismic facies, bounding surfaces, andinternal and external morphologies characterize the units. Onceunits were identified and mapped, their volumes were calculatedfrom isochron maps so that long-term average sediment fluxvalues could be determined.

Unit 1.—

Unit 1 is confined mostly to the inner shelf and is character-ized by chaotic seismic facies that fills channels and/or formssand ridges (Fig. 7). The top of the unit is the seafloor, and the baseis amalgamated SB-2-TS-1 surface (inner shelf) on the SB-2 sur-face (outer shelf). The channel fill is capped by TS-1, which is aplanar surface. The upper surface is molded into sand ridges onthe inner shelf. Unit 1 deposits on the middle–outer shelf arerelatively thin (< 10 ms thick), layered strata that are conformablewith bounding surfaces.

McBride et al. (this volume) described the MAFLA (Missis-sippi–Alabama–Florida) sheet sand between the Louisiana shelfand this study area. The top of this extensive sheet sand can bereworked into sand ridges. The MAFLA sand sheet is character-ized by > 90% siliciclastic sand and < 10% carbonates. Sands aremoderately sorted and medium-grained (2.9–1.1 phi).

Tanner (1966) first recognized sand ridges on the inner shelf.The dimensions of sand ridges observed in this data set are notuniform. On average, ridges are less than a few meters high and

250 m wide, and have an asymmetric profile that is steeper on thewestern side. The asymmetric profiles suggest westward migra-tion. There is no clear indication of landward migration. Theirlow heights render them barely discernible unless the profiles arevertically exaggerated.

Unit 2.—

On the inner shelf, the top of unit 2 is defined by fluvialincision (SB-2). Interfluves probably represent subaerial expo-sure surfaces or transgressive ravinement surfaces. On the innershelf, unit 2 deposits are thin to absent. On the middle to outershelf, unit 2 is characterized by prograding clinoform seismicfacies and wedge-shaped external geometries (Fig. 7). Topsets areeroded by the SB-2 and/or TS-1 surfaces (Fig. 7). Associated withSB-2 is a chaotic-fill seismic facies, assumed to be distributarychannels. Clinoforms downlap MFS-5 (Fig. 7).

Clinoform geometries of unit 2 suggest a deltaic depositionalenvironment. Clinoform geometries grade from shingled andoblique parallel to oblique tangential seaward. Overlying sur-faces SB-2 and TS-1 truncate topsets of the underlying unit (Fig.7). Locally, sigmoid clinoform geometries are preserved, associ-ated with delta lobe switching and progradational centers (Fig.7). Dip line APL03 illustrates the steep foresets (~ 3°) at thedepositional front of unit 2 and absence of thick, aggradedbottomset beds (Fig. 7). Where bottomset beds are imaged, theyare not laterally extensive. Distributary channels extend to thedepositional front of the delta, as evidenced by the truncation ofunderlying clinoforms (Fig. 7). The offlap break, representing the

0

100

200

300

100

200

one kilometer

v.e. ~ 21x

NW SE

SB-6

TS-1

unit 4

MFS-5

unit 2 barrier

MFS-7

multiple

aggradedbottomset beds

progradationalcenter

delta lobeswitching

sigmoid clinoforms

SB-6

unit 4multiple

progradationalcenter

sigmoid clinoforms

MFS-7

TS-1

two-w

ay t

ravel

tim

e (m

s)

two-w

a y t

ravel

tim

e (m

s)

FIG. 8.—Seismic line APL16 is a strike profile on the middle shelf (see Figure 1 for location). Units 2 and 4 and associated boundingsurfaces are shown. Unit 4 shows significant erosion by distributary channels, several progradational outbuilding and delta lobeswitching events, and aggraded bottomset beds. Sigmoid clinoforms are common, associated with the numerous sub-lobes of unit4. To the southeast, there is a unit 2 barrier bar at the depositional shoreline break of unit 4.


culmination of regressive deposition, occurs at approximately -90 m below present sea level. Note the variable clinoform anglesand downlap onto localized flooding surfaces, which indicatesdelta lobe shifting.

Unit 2 is well preserved on the outer shelf, as evidenced byconvex-seaward bathymetric contours (Fig. 1). Unit 2 delta lobesare not capped or onlapped by extensive layered strata.

Two separate depocenters of unit 2 were mapped (Fig. 11A).To the west, the unit is characterized by isolated highs withoblique-parallel clinoform geometries that are interpreted aspaleoshorelines or barriers (Figs. 8, 11A). They presumably werelocated at the depositional shoreline break of older unit 4 deltas(Fig. 8). To the east, a delta is oriented parallel to depositionalstrike and is characterized by shingled, oblique parallel, andoblique tangential clinoform geometries (Fig. 11A).

The area of the unit 2 delta is approximately 23 km wide, 10 kmlong, and up to 80 ms (60 m) thick (Fig. 11A). The sedimentvolume is approximately 10.7 km3. According to the elevation ofthe updip pinch-out and offlap break of the unit, sea level rangedfrom approximately -60 m to -90 m below present during depo-sition of the delta (Fig. 7). The updip pinch-out is a minimumestimate, because of subsequent erosion and/or transgressiveravinement. The offlap break of the unit is a maximum estimate,also because of subsequent erosion and/or transgressiveravinement.

Unit 3.—

Unit 3 is characterized by variable thickness of transgressivestrata on the shelf. Thicknesses are greatest on the inner shelf influvial valleys (> 60 ms (45 m)). Unit 3 is characterized by chaoticfill seismic facies over an erosional surface associated with SB-6on the inner shelf. On the middle to outer shelf, unit 3 is charac-terized by mounded-onlap-fill seismic facies that fills topographiclows within unit 4 (Fig. 6). On the upper slope, unit 3 is character-ized by onlapping layered or chaotic seismic facies (Fig. 9). Theunit is underlain by SB-6 and overlain by MFS-5.

Unit 3 strata are generally conformable with the overlying SB-2 and MFS-5 amalgamated surface and either downlap at a verylow angle (Fig. 6) or onlap the underlying SB-6 surface (Fig. 9).Locally, SB-6, and hence unit 3, are truncated by SB-2. Where noobvious geometric relationships exist between unit 3 and bound-ing surfaces, the unit is mapped on the basis of seismic faciescharacter, because strata of unit 3 are more layered than boundingunits 2 or 4.

Inner-shelf unit 3 deposits are characterized by chaotic-fillseismic facies filling an erosional surface formed by the amalgam-ated SB-6 and MFS-7 surface. Channel dimensions are generallybroader (~ 2 km) and deeper (~ 30 m) than those associated withthe amalgamated SB-2 and MFS-5 surface. Above the level ofincision, strata are layered.

On the middle to outer shelf, unit 3 is characterized bybackstepped deltas, recognized by mounded-onlap-fill seismicfacies in topographic lows associated with antecedent topogra-phy on unit 4 (Fig. 6). Foresets downlap SB-6 at a relatively lowangle (Fig. 6). Intermediate downlap surfaces separate individuallobes. At least three generations of backstepped lobes are imagedin the eastern part of the study area (Fig. 6). Backstepped deltalobes extend as far landward as the data set extends and arethickest between sub-lobes of the unit 4 eastern lobe. Backsteppeddeltas on the middle to outer shelf are up to 40 ms thick (30 m).Unit 4 delta lobes are onlapped by unit 3 layered strata at theirdistal limits, referred to as slope wedges (Fig. 9). In the study area,unit 3 slope wedges are concentrated on the western lobe.Backstepped shorelines are not as easily recognized, because theyare thin and not laterally extensive.

Unit 4.—

On the inner shelf, unit 4 was subjected to subaerial exposureand fluvial incision. The unit is very thin to absent on the innershelf. On the middle to outer shelf, unit 4 is characterized byprograding-clinoform-fill seismic facies and wedge-shaped ex-ternal geometries (Figs. 6, 8, 9). Topsets are eroded by SB-6 and

FIG. 9.—Seismic line APL19 is a dip profile on the outer shelf to upper slope (see Figure 1 for location). The profile shows the ramp-like morphology of the margin. Unit 4 and associated bounding surfaces are indicated. Distributary channels, associated withSB-6, incise to the depositional front of the delta.

0

100

200

300

two-waytraveltime(msec)

N S

multiple

MFS-7

SB-6

unit 4

delta

unit 3

slope wedge

1 kilometer

v.e. ~ 33x

MFS-5

onlap

distributary

channelschaotic

facies

TS-1

two-waytraveltime(ms)


Study Area

contour interval = 10 ms

bathymetric contours

water depth (m)

tracklines

Time-Structure Map of SB-6Legend

��

��

��

��

��

��

��

��

Study Area

FL

GAALA


bathymetric contourswater depth (m)

tracklines

Time-Structue Map of TS-1

kilometers0 20

Legend

8585 30'

29 30'

29

60

150

100convex seaward

��

��

��

��

��

��

��

Study Area

FL

GAALA



water depth (m)

tracklines

Time-Structure Map of SB-2 Legend 29

kilometers0 20

50

100

29 30'

��

��

��

��

��

��

��

��

FL

GAALA

29

kilometers0 20

50

150

190100

29 30'

��

��

��

��

��

��

��

��

FL

GAALA

29

kilometers0 20

100

150

190stage 7-6delta lobes

Study Area



tracklines

Time-Structure Map of MFS-5Legend

amalgamated MFS-5and SB-2

85 30'

85 30' 85 30'

85

85 85

A B

C D

29 30'

FIG. 10.—Time–structure maps of A) TS-1, B) SB-2, C) MFS-5, D) SB-6, E) MFS-7, and F) SB-8.

20

60

100

140

180

220

260

Study Area

FL

GAALA



tracklines

Time-Structure Map of MFS-7Legend

8585 30'

29 30'

29

kilometers0 20

amalgamated MFS-7and SB-6

100

150

190

2 0

60

100

140

180

220

260

Study Area

FL

GAALA



tracklines

Time-Structure Map of SB-8Legend

8585 30'

29 30'

29

kilometers0 20

2 0 02 3 0

1 0 0

SB-8 truncated by SB-6

E F


locally by TS-2 (Figs. 6, 8). Associated with SB-6 is chaotic-fillseismic facies, assumed to be distributary channels. Clinoformsdownlap MFS-7 (Figs. 6, 8). Two separate deltas exist on theshelf, separated by an interlobate area (Fig. 11B). Both deltas aremanifested on seafloor bathymetry maps by convex-seawardbathymetric contours (Fig. 1).

Strike lines on the middle shelf capture an apparent deepen-ing of depositional systems to the west, associated with the lateralpinch-out of unit 4. The eastern part of the study area is character-ized by widespread incision, whereas the western part is charac-terized by oblique parallel clinoform geometries.

Clinoform geometries of unit 4 suggest a deltaic depositionalenvironment. Clinoform geometries grade from shingled andoblique parallel to oblique tangential seaward with foreset anglesapproximately 3° (Fig. 6). The western delta is also characterizedby sigmoid clinoform geometries associated with numerous lat-erally stacked progradational centers (Fig. 8). Two distinct deltalobes occur on the shelf that share common bounding surfaces.Both deltas are characterized by bird-foot morphologies, sug-gesting point-sourced deposition (Fig. 11B). Two distinct lobesthat were fed by separate distributary systems (Fig. 11B) repre-sent each delta. Local flooding surfaces separate sub-lobes of thedeltas and represent periods of avulsion. Relative ages of indi-vidual sub-lobes, based on intermediate downlap surfaces, sug-gest that the system migrated to the west (Fig. 8). Unlike the unit2, these deltas are capped by laminated strata.

A dip profile of the western delta images distributary chan-nels that locally incised to the depositional shoreline break atapproximately -120 m below present sea level (Fig. 9). At approxi-mately 80 ms (60 m), the western delta is thicker than the easterndelta, which is approximately 40 ms (30 m) thick (Fig. 11B). Theeastern delta is characterized by a single progradational center.Distributary channels of the eastern delta appear to extend far-ther basinward.

The area of the western delta is approximately 37 km wide, 34km long, and up to 80 ms thick (60 m) (Fig. 11B). The eastern deltais approximately 38 km wide, 37 km long, and up to 40 ms thick(30 m) (Fig. 11B). The unit 4 delta extended farther seaward thanthe unit 2 delta, suggesting shallower waters and less accommo-dation. The sediment volumes associated with the western and

eastern deltas are approximately 32 km3 and 21.8 km3, respec-tively.

Judging by the elevations of the updip pinch-out and offlapbreak of the unit, sea level ranged from approximately -90 to -120m during deposition of the unit 4 deltas.

Unit 5.—

Unit 5 is obscured by the seafloor multiple over most of thestudy area; therefore, its areal extent and thickness were notresolved. The unit is imaged in relatively shallow water depths(< 80 m). It is characterized by shingled and oblique parallelclinoform geometries. Where clinoform geometries are not im-aged, the fill seismic facies is chaotic. The unit is overlain by SB-8, characterized by the truncation of foresets. The base of thedelta is generally at or beneath the seafloor multiple. A localflooding surface exists near the updip pinch-out of the delta,suggesting a potential avulsion event.

DISCUSSION

Fluvial–Deltaic Deposition

Wedge-shaped morphology and the gradation from shingledto high-angle oblique clinoforms suggest that deltaic unitsprograded seaward into deep water (Fig. 7). The relatively steep(3°) clinoform angles suggest higher-energy depositional pro-cesses (Fig. 7). The absence of topset beds at the seaward marginof deltas suggests significant erosion by the coeval distributarynetwork and/or transgressive ravinement surface (Fig. 7). Weassume that erosion occurred as distributary channels activelyextended the depositional front throughout the late highstandand lowstand. The absence of appreciably thick bottomset beds(prodelta muds) indicates relatively high bedload sediment sup-ply (Fig. 7).

The ancestral Apalachicola Delta system experienced avul-sion and abandonment, as evidenced by numerous progradationalcenters and local flooding surfaces (Fig. 8). The external morphol-ogy of deltas varied between different eustatic cycles, from shore-parallel, elongate lens (unit 2) to bird-foot (unit 4). Fluvial systems

20

60

100

140

180

220

260

Study Area


water depth (m)

tracklines

Isochron Map of Unit 2Legend


8585 30'

29 30'

29

10

80

50

pinch-out of unitDLL

depositional frontof unit 2

unit 2barriers

20

60

100

140

180

220

260

Study Area

FL

GAALA



tracklines

Isochron Map of Unit4Legend

8585 30'

29 30'

29

kilometers0 20

8050

40

0

0

pinch-out of unit

T

T

T

T

T

offlapbreak

A B

kilometers0 20

FIG. 11.—Isochron maps of A) unit 2 and B) unit 4.


characterized by high bedload, relatively unstable channels, andclosely spaced channels deposited uniform delta fronts similar tothat of unit 2 (Postma, 1995). More stable, distinct feeder channelsdeposited mouthbar-type deltas similar to unit 4.

Timing of Deposition

On the basis of the seaward decrease in toplap elevationsfrom the updip pinch-out of the delta to its offlap break at thedelta front, it is inferred that units 2, 4, and 5 represent deposi-tion that occurred during sea-level fall. Ages were inferred fromelevations of the updip pinch-out and offlap break of the units(Fig. 4).

On the basis of the elevations of the updip pinch-out andofflap break for unit 2, sea level ranged from approximately -60m to -90 m during deposition of the unit (Table 1). Because weinterpret the bounding surfaces of unit 2 to be SB-2 and MFS-5, itis inferred that deposition of the unit occurred during sea-levelfall of the last glacioeustatic cycle (OIS 5 to OIS 2). On the basis ofthe correlation between offlap break elevations and the proxysea-level curve, deposition of the delta occurred between 25 kaand 75 ka (OIS 5 to OIS 2) (Table 1).

On the basis of elevations of the updip pinch-out and offlapbreak for unit 4, sea level ranged from approximately -90 to -120m during deposition of the unit (Table 1). Assuming that thebounding surfaces (MFS-7 and SB-6) correspond to OIS 7 to OIS6, and on the basis of the correlation between the updip pinch-out and offlap break elevations and the proxy sea-level curve,deposition of the unit occurred between 156 ka and 141 ka (OIS7 to OIS 6) (Table 1).

Unit 5 is bounded by SB-8 above, and the underlying maxi-mum flooding surface is obscured by the seafloor multiple.Toplap and offlap break elevations are not well imaged in unit 5.On the basis of its stratigraphic positioning beneath youngerunits and relationship to bounding surfaces, unit 5 is interpretedas having been deposited between OIS 9 and OIS 8.

Interpreted depositional environments for units 1 and 3,including sand ridges, valley fill, sheet sands, backstepped del-tas, and slope wedges, are consistent with transgressive deposits.The environments indicate that deposition occurred during rela-tive sea-level rise, suggested by onlapping-fill seismic facies and/or progressively younger strata located updip of older strata ofthe same unit. Ages of units 1 and 3 can be inferred from theirbounding surfaces. Unit 1 (bounded by the seafloor and SB-2)represents deposition between OIS 2 and OIS 1. Unit 3 (boundedby MFS-5 and SB-6) represents deposition that occurred betweenOIS 6 and OIS 5.

Comparison of Deposition Associatedwith Different Eustatic Cycles

Because unit 5 is only locally imaged, comparisons will bemade between units of the last two glacioeustatic cycles only. Tobegin, comparisons will be made between unit 2 (OIS 5 to OIS 2)and unit 4 (OIS 7 to OIS 6). Average sediment-flux values for theunits, based on estimated volumes and ages, are presented in

Table 2. The average sediment-flux values in Table 2, column 4,are based on the assumption that the updip pinch-out and offlapbreak elevations can be used to accurately predict the timing ofthe onset and culmination of deposition. The estimated durationof deposition is shown in column 2. Average sediment flux valuesgiven in Table 2, column 5, assume that ages of the units wererepresented by the entire highstand (approximately 100 kyr).This was based on the assumption that the deltas were composedlargely of sediment reworked from early highstand deposits.

Sediment volume for the unit 4 delta was five times greaterthan the sediment volume of the unit 2 delta. Although unit 4 ismanifested as two separate deltas on the shelf and individualdeltas are areally more extensive, thicknesses are locally con-centrated within their progradational centers. The unit 2 deltapinches out more abruptly, and the delta has a more uniformthickness.

Another difference between deltas of units 2 and 4 is theorganization of their distributary-channel networks. Distribu-taries that fed unit 2 deltas served as a line source for deltaprogradation (Fig. 12). During the prior highstand, unit 4 deltaswere more point-sourced, with more organized distributary sys-tems feeding separate lobes (Fig. 12).

On the basis of the extensive distributary network mappedfor unit 2, sedimentation appears to have been line-sourced bya laterally extensive distributary complex (Fig. 12). The easterndepocenter is confined by antecedent topography of the olderunit 4 delta system (Fig. 12). This illustrates that accommoda-tion was controlled mainly by antecedent topography. Thewestern depocenter probably developed mostly as barriers (Fig.12). The development of these barriers suggests minimal sedi-ment supply and extensive wave reworking from the unit 4delta. Steep depositional fronts and foreset angles, wave-cutplatforms, the presence of barriers, and elongate lens-shapedgeometries (Figs. 7, 8, 12) reflect the importance of waves in theformation of unit 2 deltas. At the height of their development,these deltas prograded into 70 m of water. Wave-cut platformsprobably developed during sea-level stillstand and/or littlesea-level rise.

The smaller thickness of unit 4 (~ 30 m) relative to unit 2, andits more basinward extension, suggests limited accommodation.The bird-foot morphologies and branching distributary systemsare indicative of greater fluvial influence on unit 4 deltas relativeto unit 2 deltas.

Deltas of the same glacioeustatic cycle share common bound-ing surfaces, suggesting the same time span of deposition.During both glacioeustatic cycles, deltas were deposited in twodifferent locations on the shelf. In relative terms, we infer thatunit 2 deltas were wave-dominated and unit 4 deltas werefluvial-dominated. The differences between transgressive de-posits of the two glacioeustatic cycles are less dramatic. Unit 1deposition occurred mainly on the inner shelf and in the form ofsand ridges, sheet sands, and valley fill. Unit 3 depositionoccurred on the middle shelf to upper slope, but sand ridges andsheet sands are not preserved in the subsurface. Rather, unit 3deposits of the inner shelf are mainly valley fill deposits. Unit 3deposits on the middle to outer shelf consist of backsteppeddeltas.

unitToplap

elevations (m)offlap breakelevation (m)

timing (ka) OIS

2 60–90 90 75–25 5–2

4 90–120 120 156–141 7–6

TABLE 1.—Summary of topset bed and offlap break elevationsand associated ages for units 2 and 4. Toplap elevations refer to

the updip pinch-out and offlap break of the respective unit.

TABLE 2.—Summary of volumes, ages, and average sedimentflux calculated based on the two different scenarios

described above for units 2 and 4.

unitvolume(km3)

age(kyr)

Average sediment flux (1)(km3/1000 yr)

Average sediment flux (2)(km3/1000 yr)

2 10.7 50 0.2 0.1

4 53.8 15 3.6 0.5


Deposition Associated with Different Systems Tracts

Highstand Systems Tract.—

Results from this study suggest that highstand deposits onthe inner shelf have largely been eroded by subaerial exposureduring the lowstand. Preserved late highstand deposits arelocated on the middle to outer shelf and consist mainly of deltasand their associated distributaries. These deltas form convex-seaward bathymetric protrusions on the shelf in water depths ofapproximately 50 m to 150 m. Preservation of these deltas on themiddle to outer shelf is possibly the result of hardgroundformation during the lowstand. Grab samples taken from theseafloor, above the unit 2 highstand delta, include a relativelyhigh carbonate component, but no cemented material was re-covered.

Calculated sediment-flux values, based on the entire 100,000years of highstand deposition, were within the same order ofmagnitude for both units 2 and 4 (0.1 km3/1000 yr and 0.5 km3/1000 yr, respectively). The calculated sediment flux for the mod-ern system, based on the volume and age of the modern delta, isapproximately 0.5 km3/1000 yr (Kofoed and Gorsline, 1963;Schnable and Goodell, 1968). Thus, unlike the rivers of Texas,whose sediment flux values were much higher during the mostrecent OIS 5 to 2 falling stage of sea level (Abdulah et al., this

volume; Banfield and Anderson, this volume), the sediment fluxof the modern Apalachicola River is inferred to be slightly greaterthan it was during the OIS 5 to 2 falling stage of sea level.

Estimates for ancient sediment flux of the Apalachicola depo-sitional system appear to be reasonable when compared withfluvial systems of similar size in the northern Gulf of Mexico. Forreference, highstand sediment flux values for the Rio Grande andBrazos–Colorado systems were 1.4 to 2.9 km3/1000 yr and 0.3 to3.2 km3/1000 yr, respectively (Banfield, 1998; Snow, 1998).

Lowstand Systems Tract.—

The lowstand systems tract is characterized by subaerialexposure of the inner shelf and development of an extensivebraid-plain system, composed of numerous, shallow amalgam-ated channels. As sea level fell, the streams and rivers of theregion flowed across a relatively high-gradient shelf covered bysand. The high gradient and high bedload content contributed tobraided fluvial morphology (Bridge, 1993; Bristow and Best,1993; van den Berg, 1995; Leeder, 1999). The fluvial systems feddeltas on the middle to outer shelf during the late highstand andlowstand, judging by the elevations of the updip pinch-outs andofflap breaks. These channels are locally incised into upperMiocene carbonates on the inner shelf. No evidence exists forcross-shelf incised valleys on the inner shelf or canyons and fans

unit 2barriers

High

Low

Eus

tasy

Time (Kyr)

0 50 100 150 200

-120

5

Depth (m

)

5e

2

unit 2 delta

unit 3 backstepped deltas

unit 4 deltas

fluvial network

fluvial network

unit 3slope wedge

FIG. 12.—Block diagrams of units 2 and 4. The upper diagram shows the unit 2 delta filling antecedent topography between unit 4deltas. Unit 2 barrier bars are shown at the depositional front of the abandoned unit 4 system. Unit 3 backstepped deltas are presentin the former position of unit 4 deltas on the shelf. Unit 3 slope wedges are localized on the western unit 4 delta. The lower diagramshows the unit 4 fluvial–deltaic system, characterized by two distinct bird-foot deltas on the shelf.


on the upper slope, as predicted by many general sequencestratigraphic models (Posamentier and Vail, 1988; Posamentier etal., 1988; Van Wagoner et al., 1990; Christie-Blick, 1991).

At times during the glacioeustatic cycle, braided channelsconverged into larger “trunk” valleys. These larger valleys wereflanked by terraces, which provide evidence for a system that waslaterally migrating and incising. Steep bluffs on the eastern sideof paleo-channels provide evidence that the system was migrat-ing to the east. Alternatively, the terraces may represent succes-sive stages of fluvial incision during an overall fall in sea level, butchannel geometries are not supportive of this interpretation. Twoseparate deltas of the same age on the shelf suggest the presenceof two separate feeder channels. Where these systems convergedupdip is not known.

An important distinction of this ramp margin is the absence ofcross-shelf incised fluvial valleys, which is a function of relativelyplanar fluvial and shelf gradients. No distinct shelf–slope breakexists to initiate incision and headward erosion. The only excep-tion is where topographic lows between late highstand andlowstand deltas served as centers for headward-eroding valleysduring lowstands.

Seismic data suggest that the unit 4 offlap break occurs at adepth of approximately 120 m. In contrast, the offlap breakelevation of the unit 2 delta is estimated at –90 m. The actual OIS2 lowstand, according to sea-level proxy curves, was approxi-mately –120 m. This poses the problem as to where the lowstandsystems tract manifests itself on the outer shelf to upper slope. Itis possible that the data set collected for this study did not imagethe lowstand lobe of unit 2, although no evidence exists for sucha lobe in modern seafloor bathymetry. Another possibility is thatthe remaining 30 m is represented by sediments deposited on theslope as the result of bypass. However, there is no evidence oflowstand deposits on the slope. It is more likely that extensiveerosion by the sequence boundary (SB-2) and transgressiveravinement surface (TS-1) removed a significant part of the topsetbeds and upper portion of foreset beds, and sea-level estimatesfor the culmination of deposition are underestimated. Otherwise,the OIS 2 lowstand, as predicted by calibrated sea-level proxycurves, is overestimated by ~ 30 m.

Distributary channels, associated with the sequence bound-ary, locally incise the delta to the depositional shoreline break(Figs. 7, 9). As expected, this is best observed in dip profiles whereprogradation was most active. Lateral limits of deltas show betterpreservation of clinoform geometries (Fig. 8). Depths of seawardtopset beds for the unit 4 delta are consistent with the lowstandpart of the sea-level curve. This evidence for lowstand deltas isdifferent from the Van Wagoner et al. (1990) definition, whichsuggests that lowstand deltas downlap the sequence boundary,making them a distinctly different system from highstand deltas.The deltas described for units 2 and 4 began development duringthe early highstand in more proximal shelf positions andprograded to the outer shelf during late highstand and lowstandtime. These deltas downlap the maximum flooding surface andare truncated by the sequence boundary and transgressiveravinement surface, respectively (Figs. 7, 9).

Planar strata, indicative of hemipelagic sedimentation, domi-nate the lowstand deposits of the upper slope. Slope canyons andfans are absent. Several reasons are given for this lack of slopecanyons and fans. According to models by Posamentier and Vail(1988), canyons and fans do generally not characterize rampmargins. Ramp margins are not conducive to incision at the “shelfedge” because no drastic change in gradient occurs. Slope insta-bility, which triggers large-scale mass wasting, is less likely tooccur where there is no appreciable height difference between theshelf edge and the base of slope, as is the case here. This data set

shows minimal evidence for slope instability in the form of slides,slumps, or debris flows. At the depositional front of deltas,bottomset beds are asymptotic, merging into a single reflector(Fig. 7). This very thin unit, developed at the depositional front ofthe delta, is representative of the basinward expression of thelowstand systems tract. The thin unit is suggestive of minimalsedimentation and absence of slides, slumps, and debris flows, aswell as prodelta deposits (Fig. 7).

Van Wagoner et al. (1990) generically consider the lowstandsystems tract as the most viable place for reservoir-quality sandbodies, on the basis of the presence of lowstand valleys, canyons,and fans. Lowstand depositional sand bodies are absent in thisstudy area.

Transgressive Systems Tract.—

Middle- to outer-shelf transgressive deposits of unit 3 arelocally preserved on the shelf, but transgressive deposits of unit1 are more extensive on the inner shelf. Commonly, thesedeposits fill valleys and aggrade the interfluve areas, but sandridges also occur. Sand ridges and sand-rich valley fill wereprobably derived from the reworked topsets of underlyingdelta lobes.

The asymmetric profile of subsurface sand ridges suggeststhat the dominant current transport direction was to the west,similar to the present. One would also expect to see landwardmigration of ridges. At present, the shelf is sediment starved, andreworking of sediment probably occurs only during storms andthe passage of cold fronts (McBride et al., this volume).

On the middle and outer shelf, backstepped deltas fill low-lying areas, generally at the lateral pinch-out of the underlyingdelta (Fig. 6). In dip and plan views, they overlie unit 4 deltas(Figs. 6, 12). Eroded topset beds of highstand deltas are probablya significant source of transgressive sediments on the shelf.

Slope wedges, which represent deposition during the trans-gressive systems tract, onlap the sequence boundary at thedepositional fronts of the late highstand to lowstand unit (Fig.9). Slope wedges in unit 3 are associated with the OIS 8 to 7transgression. The wedge may contain sands that were derivedfrom the topsets of deltas as distributary channels eroded to thedepositional front of deltas. The approximate thickness of theunit 3 slope wedges is 15–30 m, thinning abruptly basinward.Thicknesses are a function of the height of the preserved deltafront. These slope wedges appear to be more pronounced in thewestern part of the study area (Fig. 12). West of the study area,offshore Pensacola, slope wedges are more extensive (Bart,1998). It is suggested that these deposits were sourced bybottom currents at the base of slope during the transgression(Bart and Anderson, this volume). In the study area, unit 1 is notonlapped by transgressive deposits (slope wedges) on the shelfor upper slope.

CONCLUSIONS

Fluvial-Deltaic Deposition

• Seismic data from the study area indicate that late-highstandto lowstand fluvial channels and depocenters were located tothe west of the modern Apalachicola River.

• Extensive highstand deposits exist on the middle to outershelf as sandy, fluvial, and wave-dominated deltas with rela-tively steep clinoforms (~ 3°) and minimal bottomset beds,implying deposition by bedload-dominated rivers in a high-energy environment.


• The highstand deltas are manifested on the outer shelf asconvex-seaward bathymetric protuberances. The preserva-tion of these features was perhaps enhanced by hardgroundformation.

• Topset beds were deeply eroded by advancing distributarychannel system that actively extended the depositional frontof deltas.

• Numerous progradational centers and localized flooding sur-faces suggest that the system avulsed several times.

• The absence of highstand deposits on the inner shelf is in-ferred to be due to low subsidence and extensive erosion.

Timing of Deposition

• Chronostratigraphic control for the study area is absent, soages of deltaic units were inferred from updip pinch-out andofflap break elevations and stratigraphic relationships. Atleast three glacioeustatic cycles (OIS-9 to OIS-1) can be im-aged.

Comparison of Deposition Associated withDifferent Eustatic Cycles

• During OIS 7 to 6 (unit 4), two deltaic depocenters existed inseparate locations on the outer shelf, suggesting the existenceof two separate fluvial feeders of the same age. During OIS 6,a major fluvial valley was present near the eastern unit 4 delta.This valley is believed to represent the paleo–ApalachicolaRiver. By OIS 2, the western delta of unit 4 is expressed onlyby deltaic remnants, including barriers, suggesting that thewestern system of unit 4 was inactive by unit 2 time.

• Highstand deltas were fluvial-dominated during OIS 7 to 6and wave-dominated during OIS 5e to 2. Estimates forhighstand sediment volumes and flux values were signifi-cantly greater for unit 4 (OIS 7 to 6) than unit 2 (OIS 5 to 2).Sediment flux for the modern Apalachicola River is greaterthan that calculated for the unit 2 Apalachicola Delta (OIS 5 to2). In contrast, larger Texas rivers (Rio Grande, Brazos, andColorado rivers) had sediment-flux values that were signifi-cantly greater during falling sea level than at present. Onemajor difference between Texas rivers and the ApalachicolaRiver is the greater storage capacity for the low-gradientTexas rivers. The result was a more continuous sedimentsupply by the Apalachicola River during the highstand.

• Branching, relatively deep valleys sourced the OIS 7 to 6fluvial-dominated delta. The OIS 5e to 2 wave-dominatedsystem was sourced by a braid-plain system, characterized bynarrower, shallower channels.

Deposition Associated with Different Systems Tracts

Lowstand.—

• The seismic facies analysis suggests that the ApalachicolaRiver was braided during the lowstand. This was likely theresult of steeper valley slopes, increased discharge, and a highbedload sediment supply.

• The lowstand systems tract is expressed as an extension ofhighstand deltas. Distributaries incised to the depositional

front of deltas with depths of topset beds for the OIS 7 to 6delta at approximately -120 m.

• Offlap break elevations for the unit 2 deltaic wedge areapproximately -90 m. This is 30 m shallower than the sea-levelestimate for the last glacial maximum.

• Toplap elevations decrease seaward for both units 2 and 4.Comparable sediment volumes for both units, no evidence forbackstepped depositional systems, and an intermediate trans-gressive unit (unit 3) suggest deposition during sea-level fallfor both units 2 and 4.

• Shelf-wide incised fluvial valleys, slope canyons, or slope fansdo not occur in the lowstand systems tract.

Transgression.—

• The asymmetry of seafloor features, associated with sandridges, indicates dominantly westward-directed currentsduring the last transgression.

• Slope wedges are thicker and laterally more extensive to thewest of the study area (see Bart and Anderson, this volume,and Fillon et al., this volume). In these areas, there is noevidence for contour currents in upper-slope settings (Sydowand Roberts, 1994; Bart and Anderson, this volume). West-ward-directed currents, active during the transgression, po-tentially transported sediments from the Apalachicola systemto be deposited as thicker, more extensive slope wedges to thewest.

• Rapid sea-level rise overwhelmed sediment flux, resulting inbackstepped deltas on the middle to outer shelf. These deltasare located in interlobate depressions of the OIS 7 to 6 delta.

Highstand.—

• Results from this study on the west Florida shelf suggestthat the most extensive, laterally continuous, and volu-metrically significant sand bodies probably were producedduring the late highstand. Deposits of the lowstand andtransgressive systems tracts are less extensive, less con-tinuous, volumetrically less significant, and have low pres-ervation potential.

REFERENCES

ABDULAH, K.C., 1995, The evolution of the Brazos and Colorado fluvial/deltaic systems during the late Quaternary: an integrated studyoffshore Texas: Ph.D. dissertation (unpublished), Rice University,Houston, Texas, 284 p.

BANFIELD, L.A., 1998, The late Quaternary evolution of the Rio GrandeSystem, offshore south Texas: Ph.D. Dissertation (unpublished), RiceUniversity, Houston, Texas, 232 p.

BARD, E., HAMELIN, B., AND FAIRBANKS, R.G., 1990, U–Th ages obtained bymass spectrometry in coral from Barbados: sea level during the past130,000 years: Nature, v. 346, p. 456–458.

BART, P.J., 1998, Seismic-stratigraphic analysis of shelf-margin delta/slope fan and basin floor fan on high-latitude and middle-latitudemargins (Ross Sea, Weddell Sea and Alabama/West Florida shelf):paleoclimatic and eustatic implications: Ph.D. Dissertation (unpub-lished), Rice University, Houston, Texas, 184 p.

BRIDGE, J.S., 1993, The interaction between channel geometry, water flow,sediment transport and deposition in braided rivers, in Best, J.L., and


Bristow, C.W., eds., Braided Rivers: Geological Society of London,Special Publication 75, p. 13–71.

BRISTOW, C.S., AND BEST, J.L., 1993, Braided rivers: perspectives and prob-lems, in Best, J.L., and Bristow, C.W., eds., Braided Rivers: GeologicalSociety of London, Special Publication 75, p. 1–11.

CHAPPELL, J., OMURA, A., ESAT, T., MCCULLOCH, M., PANDOLFI, J., OTA, T., AND

PILLANS, B., 1996, Reconciliation of late Quaternary sea levels derivedfrom coral terraces at Huon Peninsula with deep sea oxygen isotoperecords: Earth and Planetary Science Letters, v. 141, p. 227–236.

CHRISTIE-BLICK, N., 1991, Onlap, offlap, and the origin of unconformity-bounded depositional sequences: Marine Geology, v. 97, p. 35–56.

DONOGHUE, J.F., 1989, Modern and ancient valleys of the ApalachicolaRiver and estuary, northwest Florida, in Tanner, W.F., ed., CoastalSediment Mobility: 8th Symposium on Coastal Sedimentology, Talla-hassee, Florida, Proceedings, p. 231–248.

DONOGHUE, J.F., 1992, Late Quaternary coastal and inner shelf stratigra-phy, Apalachicola Delta region, Florida, in Donoghue, J.F., ed., Qua-ternary Coastal Evolution: Sedimentary Geology, v. 80, p. 1–12.

DONOGHUE, J.F., 1993, Late Wisconsin and Holocene depositional history,northeastern Gulf of Mexico: Marine Geology, v. 112, p. 185–205.

DONOGHUE, J.F., AND TANNER, W.F., 1992, Quaternary terraces and shore-lines of the panhandle Florida region, in Fletcher, C.H., and Wehmiller,J.F., eds., Quaternary Coasts of the United States: Marine and Lacus-trine Systems: SEPM, Special Publication 48, p. 233–241

DOYLE, L.J., AND SPARKS, T.N., 1980, Sediments on the Mississippi, Ala-bama, and Florida (MAFLA) continental shelf: Journal of Sedimen-tary Petrology, v. 50, p. 905–915.

DOYLE, L.J., AND LOCKER, S.D., 1988, Neogene–recent mixed carbonatesiliciclastic deposition in the eastern Gulf of Mexico with some earlieranalogs: American Association of Petroleum Geologists, Bulletin, v.72, p. 180.

KOFOED, J.W., AND GORSLINE, D.S., 1963, Sedimentary environments inApalachicola Bay and vicinity, Florida: Journal of Sedimentary Pe-trology, v. 33, p. 205–223.

LABEYRIE, L.D., DUPLESSY, J.C., AND BLANC, P.L., 1987, Variations in mode offormation and temperature of oceanic deep waters over the past125,000 years: Nature, v. 327, p. 477–482.

LEEDER, M.R., 1999, Sedimentology and Sedimentary Basins; FromTurbulence to Tectonics: Oxford, U.K., Blackwell Science Ltd, p.307–329.

LOCKER, S.D., AND DOYLE, L.J., 1987, Stratigraphy of the northwesternFlorida inner shelf from high resolution seismic reflection data (ab-stract): Geological Society of America, Abstracts with Programs, v. 19,no. 7, p. 748.

LOCKER, S.D., AND DOYLE, L.J., 1992, Neogene to recent stratigraphy anddepositional regimes of the northwest Florida inner continental shelf:Marine Geology, v. 104, p. 123–138.

LUDWICK, J.C., 1964, Sediments in northeastern Gulf of Mexico, Chapter 11,in Miller, R.L., ed., Papers in Marine Geology—Shepherd Commemo-rative Volume: New York, Macmillan Co., p. 204–238.

MCBRIDE, R.A., 1997, Seafloor morphology, geologic framework, andsedimentary processes of a sand-rich shelf offshore Alabama andnorthwest Florida: northeastern Gulf of Mexico: Ph.D. dissertation,Louisiana State University, Baton Rouge, 509 p.

MCBRIDE, R.A., ANDERSON, L.C., TUDORAN, A., AND ROBERTS, H.H., 1999,Holocene stratigraphic architecture of a sand-rich shelf and the originof linear shoals: northeastern Gulf of Mexico, in Bergman, K.M., andSnedden, J.W., eds., Isolated Shallow Marine Sandbodies: SequenceStratigraphic Analysis and Sedimentologic Interpretation: SEPM,Special Publication 64, p. 95–126.

NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION, 1999, Tide Tables:U.S. Department of Commerce, Rockville, Maryland, 189 p.

POSAMENTIER, H.W., AND VAIL, P.R., 1988, Eustatic controls on clasticdeposition II—sequence and systems tract models, in Wilgus, C.K.,Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., and

Van Wagoner, J.C., eds., Sea-Level Changes: An Integrated Ap-proach: SEPM, Special Publication 42, p. 125–154.

POSAMENTIER, H.W., JERVEY, M.T., AND VAIL, P.R., 1988, Eustatic controls onclastic deposition I—conceptual framework, in Wilgus, C.K., Hastings,B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wag-oner, J.C., eds., Sea-Level Changes: An Integrated Approach, SEPM,Special Publication 42, p. 109–124.

POSTMA, G., 1995, Causes of architectural variation in deltas, in Oti, M.N.,and Postma, G., eds., Geology of Deltas: Rotterdam, AA. Balkema, p.3–16.

RODRIGUEZ, A.B., FASSELL, M.L., AND ANDERSON, J.B., 2001, Variations inshoreface progradation and ravinement along the Texas coast, Gulf ofMexico: Sedimentology, v. 48, p. 837–853.

SAGER, W.W., SCHROEDER, W.W., DAVIS, K.S., AND REZAK, R., 1999, A tale oftwo deltas: seismic mapping of near surface sediments on the Missis-sippi–Alabama outer shelf and implications for recent sea levelfluctuations: Marine Geology, v. 160, p. 119–136.

SCHMIDT, W., 1984, Neogene stratigraphy and geologic history of theApalachicola embayment, Florida, Tallahassee: Florida Bureau ofGeology, Bulletin 58, 146 p.

SCHNABLE, J.E., 1966, The evolution and development of part of thenorthwest Florida coast: Ph.D. dissertation (unpublished), FloridaState University, Tallahassee, Florida, 231 p.

SCHNABLE, J.E., AND GOODELL, H.G., 1968, Pleistocene–Recent Stratigraphy,Evolution, and Development of the Apalachicola Coast, Florida:Geological Society of America, Special Paper 112, 72 p.

SHACKLETON, N.J., 1987, Oxygen isotopes, ice volume, and sea level:Quaternary Science Reviews, v. 5, p. 183–190

SNOW, J.N., 1998, Late Quaternary highstand and transgressive deltas ofthe ancestral Colorado River: eustatic and climatic controls on depo-sition: M.A. thesis, Rice University, Houston, Texas, 138 p.

SYDOW, L., AND ROBERTS, H.H., 1994, Stratigraphic framework of a latePleistocene shelf-edge delta, northeast Gulf of Mexico: AmericanAssociation of Petroleum Geologists, Bulletin, v. 78, p. 1276–1312.

VAIL, P.R., MITCHUM, R.M., JR., AND THOMPSON, S., III, 1977, Seismic stratig-raphy and global changes of sea level, part 3: relative changes of sealevel from coastal onlap, in Payton, C.E., ed., Seismic Stratigraphy—Applications to Hydrocarbon Exploration: American Association ofPetroleum Geologists, Memoir 26, p. 63–81.

VAN DEN BERG, J.H., 1995, Prediction of alluvial channel pattern of peren-nial rivers: Geomorphology, v. 12, p. 259–279.

VAN WAGONER, J.C., MITCHUM, R.M., CAMPION, K.M., AND RAHMANAN, V.D.,1990, Siliciclastic Sequence Stratigraphy in Well Logs, Cores, andOutcrops: Concepts for High-Resolution Correlation of Time andFacies: American Association of Petroleum Geologists, Methods inExploration Series, 7, 55 p.


Documents

High-Resolution Stratigraphy of a Sandy, Ramp-Type … et al., 2004.pdf · Sequence stratigraphic models have been used to predict sand-body distribution and facies character (Posamentier