USGS visual identity and link to main Web site

This report is preliminary and has not been reviewed for conformity with U.S. Geological Survey editorial standards or with the North American Stratigraphic Code. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

U.S. Geological Survey Open-File Report 00-306: Chapter 3

Marion-Dufresne Coring in Chesapeake Bay: Geophysical Environment at Sites MD99-2204 and 2207

by Peter R. Vogt1, Michael Czarnecki1, and Jeffrey P. Halka2

Abstract

Chirp sonar profiles (2-15 kHz) across or near Marion-Dufresne cores MD99-2204 (770 cm recovered) and 2207 (2070 cm recovered) help interpret and extrapolate the stratigraphy at these core sites. Near 2204, the Holocene section comprises an upper, acoustically transparent unit (A), overlying a middle, acoustically laminated unit with strong reflectors (B), overlying another laminated unit (C, with weak reflectors), overlying stiff middle (?) Miocene clayey silt (D). The C-D boundary is a very prominent reflector, the Cape Charles erosion surface. Although the core barrel was bent upon entering this Miocene "basement", all four units were sampled at site 2204. On the basis of Holocene sea level curves (e.g., Fairbanks, 1989) we propose that the three 14C ages (Colman and others, this volume) from lower unit B (10.3 ka), the BC transition (12.3 ka) and mid-C (12.7 ka) may be in error, with probable actual sediment ages ~7-9 ka for these units. Units B and C are interpreted as dipping "foreset" beds of a rapidly (~10 m/1000 yr) infilling early, narrow paleochannel, while Unit A represents slower hemipelagic sedimentation deposited under modern, open estuarine conditions based on their stratigraphy and micropaleontology. The overall stratigraphy at site MD99-2204 is probably similar at site 2207, although most of the structure is obscured by methane gas bubbles, which leave only the uppermost 200 cm (0-~500 yr BP) gas-free. Core ages combined with extrapolation of the base of the transparent unit into the gassy region suggests that modern hemipelagic sedimentation began roughly 6500-7500 yr BP. The age and significance of intermittent deep reflectors at ~13 m bsf (38 m bsl) in the area around MD99-2207 remain uncertain.

Introduction and Background

In this chapter we discuss the "geophysical" (seismoacoustic) data in the area of Marion-Dufresne Calypso piston cores MD99-2204 and 2207 in relation to preliminary stratigraphic and radiocarbon dating results from the cores. Cores MD99-2204 and 2207 were collected June 20-21, 1999, in the Maryland portion of the Chesapeake mainstem, east of Point Lookout and slightly north of the confluence of the modern Potomac estuary with the Chesapeake Bay (see Vogt and Chapter 2, Figs. 2.1, 2.2 and 3.1). The core sites were chosen on the basis of chirp profiles collected in the region two months prior to the cruise (see below). The main objective at 2204 was to recover a complete stratigraphic record of the Holocene in the modern Chesapeake channel off the mouth of the Potomac estuary, to verify the Miocene age below the Holocene section, and to "ground-truth" the chirp profiles. The 2207 site, located near the earlier core PTMC-3 (Fig. 3.1), was placed over the putative paleochannel at a location which, although mostly obscured by biogenic gas, revealed intermittent seismic reflectors below the top of the gas. This site was hoped to return a detailed, high-resolution Holocene paleoenvironmental record in a distal part of the Chesapeake estuary, off the mouth of the Potomac River/estuarine system.

Geophysical Surveys

Two seismoacoustic survey data sets are available in the area of MD99-2204 and 2207, informally named POT (Potomac) (Fig. 3-1): The USGS and MGS conducted reconnaissance mapping in the area, as elsewhere in the Maryland part of the Chesapeake Bay, in 1987, simultaneously running Ocean Research Equipment (ORE) Geopulse and 3.5 kHz profilers (Colman and Halka, 1989). The tracks were laid out along LORAN-C lines to facilitate navigation, with WNW lines spaced about 3 km apart and intersecting NE tracks ~ 6 km apart.

In late April, 1999, NRL conducted a "site survey" in preparation for the Marion-Dufresne coring. The NRL data, not yet fully analyzed, include EdgeTech 2-15 kHz chirp profiles and sidescan at both 100 and 500 kHz. Of these data, only the chirp profiles have been examined, being most relevant to interpretation of coring and placement of core sites. In the shallow waters of the Chesapeake Bay a short (5 ms) pulse was selected for maximum resolution of the immediate subbottom. The transducer "fish" was towed at various depths, generally between 2 and 5 mbsl, depending on water depth. The EdgeTech system software also computes the acoustic reflection coefficient, assuming spherical spreading of the chirp signal. Meaningful values for this coefficient were recovered for about 50 to 75 % of the profiler data from the POT and PAX (Patuxent) areas.

The POT surveying was conducted on 21, 22, and 27 April, with similar surveying off the mouth of the Patuxent River estuary (box in Fig. 2-2) conducted on 20 and 26 April. Although some candidate MD piston core sites were also chosen in the Patuxent (PAX) area, these were "bumped" by higher priority sites in the other areas of interest (Fig. 2.2). Surveying in both areas was carried out aboard the R/V Discovery, recently renamed R/V ß, and utilized differential GPS positioning throughout. The April 1999 NRL surveys returned about 305 km chirp profiles (105 km in the PAX area and 200 km in the POT area shown in Fig. 3-1), and mapped 40 km2 Chesapeake Bay floor (20 km2 in PAX and 40 km2 in POT). The survey tracks were mainly either north-south or east-west, to initiate generation of a homogeneous, interpretable data grid. Unlike the region off Parkers Creek, sidescan sonar coverage in the POT and PAX areas was less than 100%, even in the area of close-spaced east-west tracks (Fig. 3.1). In both POT and PAX areas, NRL tracks were chosen to cross previous core sites at least on one heading. In the POT area (Fig. 3.1) core PTMC-1 was crossed on an east-west heading, PTMC-2 on a north-south heading, and PTMC-2 on both headings. The chirp and sidescan data were continuously recorded digitally on DAT tapes, but only the chirp data were printed on Raytheon "paper."

The Maryland Geological Survey continuously ran their 200 kHz sediment classification system on the NRL survey operations, but these data have not yet been analyzed. Because only the bay floor and immediate subbottom sediments are classified by this system, the results are not of great relevance to MD core interpretation. While the NRL sidescan sonar data have not yet been analyzed, the chirp profiles (e.g., Figs. 3.2 and 3.3) were carefully examined by P. Vogt and J. Halka to chose the optimum sites for MD piston coring. Data quality was judged excellent for most of the POT and PAX chirp data, with high resolution down to 10-30 m subbottom, except where obscured by shallow methane gas or, in shallow water, a strong return from the Cape Charles erosion surface (CCES) or younger interfaces. Excellent returns were locally registered not only from the Holocene section, but also from Eastville and Exmore paleochannel/paleotributary infill sequences and from the almost horizontal Miocene strata.

MD Cores -2204 and -2207: Relation to Acoustic Stratigraphy

Site MD99-2204

Marion-Dufresne core MD99-2204 was taken in 18 m water depth (ship's log) at 38° 03.16' N, 76° 13.25' W (ship's GPS). Several unsuccessful attempts were made to position the vessel at intended location POT-2 on the north-south chirp profile C4-1 (Fig. 3.2), immediately south of the methane bubble zone, so as to sample the thickest possible gas-free section. The vessel made several passes but was unable to reposition over POT-2, probably due to the steep SSE-dipping slope of the Cape Charles erosion surface, i.e., the eroded channel slope of the Wisconsinan-aged paleochannel of the Susquehanna River. Despite the relatively poor quality of the ship's 3.5 kHz profiles, the methane gas reflector and the adjacent CCES reflector were clearly seen. In order to avoid using still more time to core precisely at POT-2, the shipboard scientists decided to core at a nearby site where the CCES reflector was clearly defined, but in somewhat shallower water and with a thinner Holocene section.

Using the ship's GPS-determined location for the actual core MD99-2204, we found small but significant discrepancies between the water depth reported by the Marion-Dufresne, the depth indicated on the NOAA charts for this location, and the subbottom chirp data. Our choice for the core location (Fig. 3.1) is an attempt to compromise among these discrepancies. MD99-2204 is the only core for which significant positioning discrepancies were found. We attribute the problems to the relatively steep slopes of both the CCES and the modern bay floor.

A core location somewhat west of chirp profile C4-1 (Figs. 3.1 and 3.2) is consistent with the relatively thinner Holocene section recovered, compared even to the south end of the chirp profile. The 1710 cm long core barrel penetrated 1000 cm, recovering 770 cm sediment, of which the lowest 110 cm was stiff clayey sand of Miocene age. Thus, the core confirmed the suspected Miocene age of sediment unit D below the strong subbottom reflector (Fig. 3.2) and proved the latter to be the Cape Charles erosion surface (CCES). If the recovered section is assumed to represent the interval from the bay floor to a subbottom (bsf) depth of 770 cm, the remaining 230 cm, penetrated but not recovered, must represent more stiff Miocene sediment. This inferred additional 230 cm penetration presents a problem, as the stiff Miocene stopped the core barrel after short penetration. The Miocene clayey sand explains why the retrieved core barrel was significantly bent, although it still proved possible to extrude the recovered sediment from the bent core barrel.

The acoustic stratigraphy displayed in profile C4-1 (Fig. 3.2) can be described in terms of three units, or sediment packages: The youngest unit (A), ca. 250 cm thick, comprises acoustically transparent sediments. This unit grades downward sharply into unit B, which is strongly acoustically laminated. The deepest Holocene unit (C) is also acoustically laminated, but the reflectors are generally weaker than those in unit A. The three-unit Holocene pile was judged to lie unconformably on middle Miocene sediments (unit D) similar to those described by Vogt and Eshelman (1987). A strong acoustic reflector, the Cape Charles erosion surface of Colman and Halka (1989), separates units C and D (Fig. 3.2). Up to 40 individual reflectors are resolved in the combined laminated part (units B and C) of the Holocene section, or an average of about 3 reflectors per vertical meter. Although the apparent dip of these reflectors is modest in a north-south section (Fig. 3.2), the true dip is much steeper (e.g., Fig. 3.3), of the order 30:1000 to 120:1000. By contrast the present more gentle western channel slope and the steeper eastern slopes are respectively 10:1000 and 30:1000. The steeper dip of the laminated reflectors may indicate stiffer, more terrigenous sediment. Differential sediment compaction in the central parts of the channel would also have steepened dips at depth, however.

We interpret the laminated reflectors as "foreset" beds, which represent the rapid early stages of infill of the Cape Charles paleochannel. Whether any of these reflectors can be traced over an extended area, and possibly reflect short-period climate or oceanographic variability, cannot be said without further analysis and more detailed chirp profiling.

Preliminary 14C ages (Colman and others, this volume) are available from shells from 407 cm bsf (10,000 +/- 300 yr BP calendar age) and 464 cm bsf (10,310 +/- 800 yr BP calendar age), and wood samples from 544 cm bsf (12,340 +/- 460 yr BP calendar age) and 626 cm bsf (12,750 +/- 160 yr BP calendar age) in core 2204. If these ages are linearly interpolated to the right edge of profile C4-1 (Fig. 3.2), the 464 cm shell sample corresponds to the middle of the band of about five close-spaced strong reflectors in the lower part of unit B, and the 407 cm bsf sample correlates to the top of this band. The 544 and 626 cm bsf wood samples correlate to the middle and lower part of Unit C. These ages, if correct, imply that units B and C are late Glacial in age. This result seems consistent with the 5.7-6.5 ka age for the AB transition inferred from core 2207 results (see below). However, if the basal C unit represents tidewater onlap of the Cape Charles erosion surface, the ages are significantly too old to correspond to the depths of the erosion surface below modern sea levels. If the sealevel rise curve of Fairbanks (1989) is assumed to be at least approximately correct for the Chesapeake Bay, it implies a sealevel of about 62 m bsl for the ~10,600 yr BP radiocarbon age (~12,800 yr BP calendar age) of the sediment directly overlying the CD unconformity (Fig. 3.2) at core site MD99-2204. This is 37 m deeper than the unconformity (25 m bsl) at the core site.

Sediment ages in the 7-9 ka age for units B and C would be more consistent with published sea level curves (e.g., Fairbanks, 1989). Pollen sampling could readily establish whether or not units B and C are late Glacial or, as we suspect, early Holocene. If the Fairbanks (1989) curve is assumed to apply to the oldest Holocene sediment (assumed to be estuarine) onlapping the CD unconformity in Figure 3.2, the age of this oldest sediment is predicted to increase from ~9.0 to 9.4 ka (radiocarbon age) from the right edge of Figure 3.2 (35 m bsf) to the last appearance of the unconformity as a seismic reflector (42 m bsf), where it disappears under the gas return. The A-B transition lies ~ 27 m below modern sea level, and probably corresponds to the time when rising sea level overtopped the narrow Cape Charles paleochannel. The estimated depth of the corresponding transition from laminated up to transparent sediments at 2207 (see below) occurred at ca. 5.7-6.5 ka (based on extrapolation of acoustic stratigraphy to 2207, and the sediment ages) or at 7.5-8.5 ka (based on the shape of the Cape Charles paleochannel and the sea level rise curve). On the basis of this we estimate that the A-B transition near site 2204 is also of about this age. It is likely that the underlying laminated units B and C were rapidly deposited, representing mainly terrigeneous materials eroded from the nearby channel edges. If estuarine sedimentation began immediately at an arbitrary point on the Cape Charles erosion surface when sea level had risen to that point, then the rapid rise of sea level during the early Holocene (~10 m per 1000 years) means that the oldest laminated sediment at the intended site POT-2 (Fig. 3.2) cannot be more than about 1000 years older than the Holocene sediment directly above the Miocene at the actual core site 2204. If it is assumed that the two laminated units were deposited at a rate of the order 10 m per 1000 years, the average spacing between the acoustic "laminae" (Fig. 3.2) is of the order 30 years.

Preliminary environmental interpretation of MD99-2204, based on foraminifera and ostracodes, shows that Holocene unit A contains estuarine assemblages similar to those in the mid-late Holocene Chesapeake Bay, unit B contains few to no calcareous microfossils but some gastropods, unit C contains no calcareous micro- or macrofaunas and unit D contains well-preserved Miocene marine microfaunas. These data are consistent with the interpretation that sediments in units B and C represent a fluvial/deltaic environment deposited during the initial sea level rise that flooded Chesapeake Bay, and the transition into unit A signifies the onset of true open estuarine sedimentation.

Site MD99-2207

Marion-Dufresne core MD99-2207 was placed in 25 meters of water (ship's log) at 38° 01.83' N, 76° 12.88'W (ship's GPS), slightly northeast of an earlier core, PTMC 3-2 (Fig. 3.1) and approximately in the axis of the Susquehanna River paleochannel as mapped by Colman and Halka (1989). The water depth at this location is 27 m according to NOAA bathymetry. The actual core site is close to the targeted location (POT-1: 38° 01.85'N, 76° 12.89' W, at about m 1075 on profile C7-1, is somewhat to the west of the actual core (Fig. 3.3)). POT-1 was chosen on the basis of two criteria. First, we wanted to sample the maximum Holocene section, presumed to overlie the paleochannel axis; and second, we looked for a site with some subbottom reflectors visible, albeit faint and intermittent, below the main gas reflector, which is close to the bay floor (Fig. 3.3). Although not as well developed as at POT-1, an intermittent faint reflector is present near MD99-2207 at about 13 m bsf (38 m bsl), below the strong gas return about 2 m subbottom (27 m bsl) (Fig. 3.3). MD99-2207 recovered 2070 cm mud, most of it gassy, and penetrated a further 530 cm. As the total penetration of 2600 cm is close to the core barrel length (2660 cm), we conclude that the core did not reach the stiff or gravelly sediments of the paleochannel thalweg. However, at a maximum core catcher depth of 52 m bsl, these thalweg sediments could not have been much farther below: As shown in Figure 2.2, the paleochannel bottom should be at about 55 m bsl.

The chirp profile across MD99-2207 (Fig. 3.3) shows that methane gas obscures most of the acoustic stratigraphy that otherwise would be evident. The top of the methane gas return is a sharp, rather than diffuse reflector, suggesting the methane gas bubbles may be trapped at a lithological, relatively impermeable boundary, which would appear as an "enhanced reflector" in seismic profiles. The age of sediments, based on the table of Colman and others (this volume), is about 500 yr BP at the level of the top of the gas.

Both east and west of the gassy region, the acoustic stratigraphy consists of an upper, mostly acoustically transparent unit which ranges from 2-3 meters thick at the east end of profile C7-1 (Fig. 3.3) and on profile C4-1 (unit A in Fig. 3.2) to as much as 10 meters thick just west of the channel (ca. m 600-800 in Fig. 3.3). We interpret this unit as hemipelagic sediments deposited relatively slowly under modern open-estuary type conditions. The thickness of this unit can be extrapolated westward to core site 2207 from its last gas-free expression at about m 1500 (Fig. 3.3), yielding a thickness of about 8 m and, based on Colman and others (this volume) an age of ~6.5-7.0 ka. A major pollen change is observed with a preliminary age of about 6500 yr BP or younger (Willard and others, this volume), which is near or at this inferred stratigraphic boundary. Based on the core ages and the chirp stratigraphy we suggest that open estuarine conditions may have begun to characterize this site about 6.5 ka. This inference is roughly consistent with sea level rise: Figure 3.3 shows that the Cape Charles paleochannel could not have been wider than ca. 1500 m when local sea level was 30 m bsl, and no wider than ca. 2000 m when sea level had risen to 25m bsl. The Chesapeake at the time was no wider than the modern Patuxent estuary, with nearby eroding shorelines providing plentiful sediment. Holocene sea level rise curves (e.g., Fairbanks, 1989) imply that the 30 and 25 m bsl sea levels were passed about 7.5-8.5 ka, not much older than the inferred sediment transition to a wide Chesapeake dominated by water-rich hemipelagic sediment deposition. The shape of the Cape Charles erosion surface, particularly east of the channel in Figure 3.3, suggests that the bay widened rapidly as sea levels rose through and above 25 m bsl.

At greater depths, ca. 13-14 m bsf (38-39 m bsl), several close-spaced intermittent reflectors are present below the gas return in Figure 3.3. The calibrated age of the sediments at the level of these reflectors is about 10,500-10,700 yr BP (Colman and others, this volume), suggesting a possible association with latest Pleistocene climate events such as the Younger Dryas. However, pollen analysis (Willard and others, this volume) suggests the oldest sediments in the core are post-glacial, i.e., younger than 10,000 years. The deep reflectors near 2207 may correspond stratigraphically to the strong reflectors at ~12 mbsf (37 m bsl) between m 1750 and m 1900 in Figure 3.3. Those reflectors may be acoustically enhanced by biogenic methane generated from the oldest infill of the Cape Charles (Susquehanna) paleochannel.

Summary and Conclusion

Preliminary results from MD cores 2204 and 2207 have been correlated with the acoustic stratigraphy near the core sites, as based on an April 1999 chirp (2-25 kHz) sonar site survey, previous reconnaissance seismic work (Colman and Halka, 1989), and cores with less penetration. To first order, we are confident that the upper 2-10 meters of relatively acoustically transparent sediments in this area represent water and clay-enriched hemipelagic sedimentation under modern, open estuarine conditions. On the basis of extrapolated ages from core MD99-2207, and acoustic stratigraphy from gas-free parts of the area, we estimate that the transition to open estuarine conditions occurred roughly 6.5-5.5 ka. We consider the underlying, acoustically laminated sediments to have been rapidly deposited and to be rich in terrigeneous components, as the narrow paleochannel filled by sediment from erosion of the nearby shores. Reliable dating of this earlier Holocene period is not yet available.

Acknowledgments

The Office of Naval Research supported the two NRL (Naval Research Laboratory) authors. The NRL Platform Support fund helped pay for the use of the vessel RV Discovery (now RV Kerhin). We thank Rick Younger, captain of Discovery, for his skills in carrying out the geophysical site survey. Irene Jewett and Jan Morrow helped prepare the illustrations, and C. Larsen was the technical reviewer.

References Cited

Colman, S.M., and Halka, J.P., 1989, Maps showing Quaternary geology of the southern Maryland part of the Chesapeake Bay: U.S. Geological Survey Miscellaneous Field Studies Map MF-1948-C, scale 1:125,000.

Colman, S.M., Halka, J.P., Hobbs, C.H., Mixon, R.B., and Foster, D.S., 1990, Ancient channels of the Susquehanna River beneath Chesapeake Bay and the Delmarva Peninsula: Bulletin of the Geological Society of America, v. 102, p. 1268-1279.

Colman, S.M., Halka, J.P., and Hobbs, C.H., III, 1992, Patterns and rates of sediment accumulation in the Chesapeake Bay during the Holocene rise in sea level, in Fletcher, C., and Wehmiller, J.F., eds., Quaternary coasts of the United States - marine and lacustrine systems: Society of Economic Paleontologists and Mineralogists Special Publication No. 48, p. 101-111.

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.

Hagen, R.A., and Vogt, P.R., 1999: Seasonal variability of shallow biogenic gas in Chesapeake Bay: Marine Geology, v. 158, p. 75-88.

U.S. Department of Interior, U.S. Geological Survey
URL of this page: https://pubs.usgs.gov/openfile/of00-306/chapter3/
Maintained by: Eastern Publications Group Web Team
Last modified: 03.28.01 (krw)