The Seafloor Revealed

Introduction

Geologic maps depicting topography, surficial materials, geomorphology*, and bedrock play an important role in understanding the origin of the earth's surface and the ongoing processes that shape it. Such maps are also instrumental in aiding the sound economic development of natural resources in terrestrial environments. They further provide guidance to natural hazards that exist within the landscape. As people increasingly work in, on, and beneath the sea, the need to understand the regional geology of the seabed has also grown.

During the past ten years we have conducted many exploratory surveys of the seafloor of the western Gulf of Maine. Recently we compiled that information, along with previously published data, using a geographic information system (GIS) to produce a series of maps of the seafloor of the inner continental shelf of the western Gulf of Maine (Figure 1). The data compiled for this map series were originally collected for a variety of research projects, contracts, and graduate student theses. For this reason there are varying degrees of geophysical data and bottom-sample coverage from place-to-place along the coast. More detailed information regarding specific locations and original field descriptions exists in Maine Geological Survey open-file reports (Kelley and others, 1987a, 1987b, 1990, 1995a, 1995b; Kelley and Belknap, 1988, 1989; Barnhardt and Kelley, 1991; Dickson and others, 1994) and University of Maine theses (Barber, 1995; Barnhardt, 1992, 1994; Dolby, 1994; Friez, 1993; Hannum, 1997; Hay, 1988; Malone, 1997; Robbins, 1992; Shipp, 1989). This report is written to accompany the map series and to explain the field techniques used to collect data. The nature of the seafloor, as well as the late Quaternary geologic history of the area, is also described.

Previous Work

Terrestrial

Investigations of terrestrial glacial and late Quaternary geology were fundamental to establishing the nature of the deposits found offshore and the timing of their deposition. A synthesis of the early terrestrial work by Borns (1989) describes the introduction of the theory of glaciation to Maine, and the evolution of hypotheses regarding the timing and geography of deglaciation.

Till of Wisconsin age forms the base of the Quaternary stratigraphic section in coastal Maine and New Hampshire. Although scattered deposits of pre-Wisconsin material have been reported from the region (Borns and Allen, 1963), they are few in number and poorly preserved on land. To the south of the study area, multiple till deposits exist, but till described from Maine is of Wisconsin age (Weddle, 1992).

Till exists as a laterally extensive and variably thick deposit of mud, sand, and gravel (ground moraine) resting unconformably on bedrock throughout the region. It also exists in the form of drumlins and recessional moraines ("end moraines," Thompson and Borns, 1985). Till may be absent in many locations, and deposits of till commonly erode along the present coastline (Figure 2).

Drumlins are teardrop-shaped hills of till concentrated along the Maine-New Hampshire border, in Boston Harbor, and in many locations on the Nova Scotia coast. They have not been recognized in the study area, however, and will not be further discussed.

Recessional moraines are widely recognized along the coast and have been reported from throughout the Gulf of Maine (Oldale and O'Hara, 1984; Oldale, 1985; King and Fader, 1986). Moraines vary from a few meters to tens of meters in height and width. Although frequently discontinuous in length, morainal segments may extend for up to 10 kilometers (Thompson and Borns, 1985; Figure 3a, Figure 3b). The surface of moraines is highly irregular and often littered with boulders.

Concentrations of moraines stretch from New Hampshire into coastal Maine (Katz and Keith, 1917), occur in local abundance in central Maine's coastal zone (Smith, 1982), and are widespread throughout eastern Maine (Ashley and others, 1991). These moraines often contain interfingering deposits of lodgement till, sand and gravel outwash, and fossiliferous glacial-marine mud (Figure 4; Smith, 1980, 1982, 1985; Smith and Hunter, 1989; Ashley and others, 1991). Radiocarbon dates from marine shells in the Great Hill moraine in Kennebunk first established that melting glaciers were on the southwestern Maine coast by at least 13,830 yr B.P. (thousands of years before present; Stuiver and Borns, 1975). More recent work from similar stratified moraines in Machias and Lubec suggests that receding ice was in coastal Maine even earlier, possibly by 14,000 yr B.P. (Dorion, 1997).

The most widespread late Quaternary deposit occurring in the coastal zone and offshore regions of the study area is muddy glacial-marine sediment, locally known as the Presumpscot Formation (Bloom, 1963). Although recognized since the early 19th century (Borns, 1989), the deposit was first carefully mapped by Bloom (1960), who also recognized its late glacial-early postglacial time of deposition. The Presumpscot Formation ranges in texture from massive deposits of mud to well layered beds of sand, silt, and clay (Figure 5; Stuiver and Borns, 1975; Ashley and others, 1991; Kelley, 1989). Its surface is relatively flat to rolling, and it is thickest in valleys and thin over topographic high points. It probably entered the sea as rock flour from glacial tunnels (Kelley, 1989) and accumulated rapidly near its source between 13,000 and 14,000 yr B.P. (Dorion, 1997). After ice had retreated into the western mountains of the region, marine mud continued to accumulate more slowly on the isostatically depressed landscape. Numerous fossils occur in the Presumpscot Formation and have been most frequently radiocarbon-dated to the time between 12,000 and 13,000 yr B.P. (Smith, 1985).

Isostatic loading of the crust by ice led to the drowning of the present land surface by the sea between about 14,500 and 11,500 yr B.P. (Belknap and others, 1987a; Stuiver and Borns, 1975; Dorion, 1997). At the landward edge of marine inundation is the marine limit where glacial-marine deltas mark the highstand of the sea (Crossen, 1991; Thompson and others, 1989; Koteff and others, 1993). As sea level began to fall, estuarine conditions prevailed in some river valleys and the sandy Embden Formation accumulated in the Kennebec River valley (Borns and Hagar, 1965). Sand and gravel outwash deposits (sand plains) followed the retreat of the sea across Maine's coastal lowland, possibly even seaward of the present coast.

Marine

Ostericher (1965), in Penobscot Bay, first recognized the thick deposits of glacigenic sediment offshore through seismic reflection methods. He obtained a radiocarbon date from above the unconformity at the top of the glacial-marine sediment, and established that sea level was at about - 18 m by 7,390 ± 500 yr B.P. Schnitker (1972) found similar glacial-marine sediment in Sheepscot Bay, as did Folger and others (1975) off southwestern Maine and New Hampshire. Borns and Hagar (1965) had focused attention on the role of rivers in delivering sediment from the newly deglaciated landscape to the falling level of the sea, and Schnitker (1974) recognized a large accumulation of sediment at the mouth of the Kennebec River as a deltaic feature. He argued for a -65 m lowstand of sea level off the Kennebec River mouth on the basis of the morphology of the lowstand delta and submerged berm on its seaward margin. Belknap and others (1987a) incorporated Schnitker's (1974) sea-level lowstand estimate into a sea-level record for the region, but noted the uncertainty of the offshore data. Shipp and others (1991) provided regional evidence for a lowstand of the sea between 55 m and 65 m depth based on seismic reflection data between Wells and Machias, Maine. Through many offshore vibracores, Kelley and others (1992) and Barnhardt and others (1995) established the complex rate of change of early Holocene sea level as an effect of long-term isostatic adjustment coupled with eustatic sea-level rise (Figure 6).

Understanding sea-level change is important in the western Gulf of Maine because of its profound effect on the locations of sediment deposition and sediment reworking (Belknap and others, 1987a; Kelley and others, 1992). The regression (relative sea-level fall) and transgression (relative sea-level rise) of the sea moved the shoreline back and forth across much of the inner continental shelf and generally stripped glacial-marine sediment from bathymetric high points and transferred the material to lower, more seaward regions. Shipp and others (1991) noted that areas shallower than the lowstand of the sea (55 m to 65 m) were rockier and had lost some of their glacial-sediment cover through wave reworking during both the late Pleistocene fall in sea level and the early Holocene rise of the sea. A marked unconformity on the surface of the glacial-marine sediment truncates acoustic reflectors in that material (Figure 7) at depths shallower than 40 m in Penobscot Bay (Knebel, 1986; Knebel and Scanlon, 1985), and even deeper elsewhere (Kelley and Belknap, 1991, Barnhardt and others, 1997). By constantly shifting the position of the shoreline and locations of sediment erosion and deposition, sea-level changes have been the single most important factor shaping the surficial sediments of the inner continental shelf since the end of the Ice Age.

Despite reworking during sea-level changes, many of the glacial deposits known from land are recognized offshore. Till has been imaged by seismic reflection methods offshore as a patchy deposit overlying bedrock, as well as in the form of moraines (Figure 8; Belknap and others, 1986, 1987b; Kelley and others, 1989c, Kelley and Belknap, 1991; Knebel and Scanlon, 1985, Miller, 1997; Barnhardt and Kelley, 1995). Till is recognized on acoustic records by its stratigraphic position over bedrock, its general lack of internal stratification and its strong acoustic surface return (Belknap and others, 1989). On side-scan sonar records till is recognized by its strong acoustic return and boulder-littered surface, as well as by its geometry (moraines, Figure 3a, Figure 9). In shallow water, morainal deposits can be traced directly from eroding outcrops on land (Figure 4; Belknap and others, 1987b), but in other locations moraines are far removed from land (Kelley and Belknap, 1991).

The dominance of glacial-marine sediment as an offshore deposit was recognized by Belknap and others (1986, 1987b), and Kelley and others (1986, 1989a, b, c). It typically fills depressions in bedrock and is often covered by modern mud (Figure 7) or sand. In areas of currents, however, glacial-marine sediment may be exposed on the seafloor (Figure 10). Glacial-marine material has been subdivided into several acoustic facies on the basis of the geometry of the deposit and its internal structure (Belknap and others, 1989), but it is often difficult to differentiate from modern mud on side-scan sonar profiles.

Deltaic and estuarine materials are recognized from seismic profiles and vibracores off major river mouths (Figure 11; Barber, 1995, Barnhardt, 1994; Barnhardt and others, 1997). The abundance of sand near rivers has been re-evaluated through coring, and contemporary volume evaluations are reduced from earlier estimates (Barber, 1995; Barnhardt, 1994; Kelley and others, 1995a, b).

Some deposits of mud offshore possess natural gas that imparts properties to the material that are not common in terrestrial sediment. Gas-charged (methane) sediment is generally found in areas with thick deposits of Holocene mud, although gas may also occur in glacial-marine mud (Shipp, 1989). Gas reduces the sediment shear strength and abets submarine slumping (Kelley and others, 1989c). In some locations, like Belfast Bay, Blue Hill Bay, and Passamaquoddy Bay, gas has erupted from the seabed and excavated large pockmarks on the seafloor (Figure 12; Scanlon and Knebel, 1988; Kelley and others, 1994, Barnhardt and Kelley, 1995). The exact origin of the gas and the triggering mechanism(s) are unknown (Kelley and others, 1994), although some gas-escape pockmark fields may overlie lakes and bogs drowned by rising sea level. Bottom sediment mapping in the study area began with the collection of sediment associated with early bathymetric soundings in the late 19th century (see Trumbull, 1972 for summary of early work). The results of systematic bottom sampling were presented in a series of U.S. Geological Survey publications in the early 1970's (Folger and others, 1975; Schlee and Pratt, 1970; Schlee, 1973; Schlee and others, 1973, Trumbull, 1972). This work was based on a large number of widely spaced samples that were analyzed for both composition and texture. The textural data were presented in the format of ternary diagrams which depict map units in terms of % mud, % gravel, and % sand, or % sand, % silt, and % clay. Owing to the inherent variability of the seafloor and the wide spacing of the grab samples, the resulting surficial maps were of small scale and lacked detail. When bathymetric information and seismic reflection observations were added, a larger scale, more detailed map was produced, but for a restricted region (Folger and others, 1975). Even when more recent compilations were produced, extensive regions (>25 km2) within our study area were represented by only a single bottom sample (Poppe and others, 1989).

Beginning in the late 1980's, the Maine Geological Survey, University of Maine, and University of New Hampshire (Birch, 1984a, b, 1990) began offshore mapping programs. Although early maps used geophysical tools extensively in addition to bottom samples, the resulting maps employed conventional, ternary (three axis, sand, mud, gravel) diagrams for textural map units (Kelley and others, 1987a, b). These reports also defined physiographic regions (Kelley and others, 1989a, Kelley and Belknap, 1991), however, by using the provisional bathymetric charts of NOAA. More recent work recognized the mappng advantages of side-scan sonar, as well as its limitations, and defined map units that were recognizable by acoustic imagery alone (Barnhardt and others, 1998). This report and accompanying maps are the culmination of that mapping style.

Contents Introduction Methods Physiography Surficial Geology Geological History Summary Glossary References

Last updated on October 6, 2005