GIS Data Guide Support
Text to Accompany the Quaternary Geologic Map of Connecticut and Long Island Sound Basin
Quaternary Geologic Map of Connecticut and Long Island Sound Basin
This map (sheet 1) portrays the geologic features formed in Connecticut during the Quaternary Period, which includes the Pleistocene (glacial) and Holocene (postglacial) Epochs. The Quaternary Period is the time of development of many details of the landscape and of all the surficial deposits. At least twice in the late Pleistocene, continental ice sheets swept across Connecticut, and their effects are of pervasive importance to the present occupants of the land.
The Quaternary geologic map illustrates the geologic history and the distribution of depositional environments during the emplacement of glacial and postglacial surficial deposits and the landforms resulting from those events. A companion map, the Surficial Materials of Connecticut (Stone and others, 1992) emphasizes the surface and subsurface texture (grain-size distribution) of these materials. The features portrayed on the two maps are very closely related; each contributes to the interpretations on the other.
Connecticut is covered by one-hundred and sixteen 7 1/2' quadrangles, all available as U.S. Geological Survey l:24,000-scale topographic maps with 10-ft contour interval. Of these, sixty-eight quadrangles have U.S. Geological Survey published or open-filed surficial geologic maps, sixteen have Connecticut Geological and Natural History Survey published maps, and fourteen have unpublished surficial geologic data from various authors. We reviewed all these maps, and did reconnaissance mapping in the remaining quadrangles. An index map and a list of references to these quadrangle studies are included in Appendix 1. The maps have been supplemented extensively by subsurface data, also indexed in Stone and others (1992). In the course of compiling this large body of data into the Surficial Materials and the Quaternary Geologic map, based on a consistent interpretive rationale, we have taken considerable liberties with some of the original studies.
The great majority of measurements mentioned, especially in the unit descriptions, are altitudes in feet. These have been kept in U.S. customary units for convenience in finding locations on topographic quadrangle maps, rather than being translated into metric units. Altitudes and depths offshore are given in meters below mean sea level (MSL) for convenience in referencing marine seismic-reflection data which uses metric measurements.
MAP UNITS
The areal units of the map comprise three main groups of deposits: POSTGLACIAL, GLACIAL ICE-LAID, and GLACIAL MELTWATER DEPOSITS. The postglacial deposits, formed by various processes after the recession of the last ice sheet, constitute ten map units; the glacial ice-laid deposits, poorly sorted materials deposited more or less directly by the ice sheets constitute three units. These thirteen units are state-wide in distribution, except for such limitations as are imposed by the geologic processes involved; for example, coastal beach and dune deposits are of course restricted to the coast. The glacial meltwater deposits, laid down by the great volumes of meltwater produced during the shrinkage of the last ice sheet, include six state-wide depositional systems. The six systems of meltwater deposits are further differentiated into 204 units, closely restricted in geographic location and therefore also in age; these units have been given names based on their geographic localities. In the Description of Map Units (Sheet 2), the local map units are organized by their occurrence within seven geographic basins across the State (figure 2, sheet 3) and by type of depositional system within each basin. The rationale for this profusion of units is based on their scientific importance, in that they provide detailed information on the mode of disappearance of the last ice sheet and the depositional processes operating around its margin. A comprehensive understanding of these units has many practical applications in studies of socio-economic importance such as ground water availability and coarse aggregate resources.
Map units beneath Long Island Sound
Glacial and postglacial geologic deposits have long been studied on land in Connecticut. Similar deposits are also present beneath modern marine sediments in Long Island Sound; in fact, it is here that the thickest and most extensive Pleistocene deposits are found (fig. 3, sheet 3). Map units beneath Long Island Sound differ from those on land in two ways. First, offshore geologic units are mapped largely from analysis of seismic-reflection profile data instead of from direct observation as on land. Secondly, due to the third-dimensional aspect of seismic-reflection data, in most places in Long Island Sound map units are superposed and show vertical as well as areal distribution. Nearly all of the offshore geologic units of Long Island Sound occur beneath a generally ubiquitous blanket of Holocene marine mud a few to 10 m in thickness which is not shown on the map.
The distribution of late Quaternary geologic units beneath Long Island Sound was mapped from more than 4,000 line-km of high-resolution, seismic-reflection profiles (see fig. 2 for locations) supplemented by vibracore data. These data were collected as part of an ongoing marine geologic mapping program that began in 1982, conducted by the Connecticut Geological and Natural History Survey (CGNHS) in cooperation with the U.S. Geological Survey. Spacing between seismic lines that were collected perpendicular to the coast is about 1.6 km (locally 0.8 km), and about 4.4 km between tielines paralleling the coast. Vibracores (obtained through CGNHS cooperative programs with the U.S. Minerals Management Service) and submersible and remotely operated vehicle dives (made possible by cooperative programs with NOAA's National Undersea Research Center at the University of Connecticut) provided verification of near-surface geologic interpretations made from seismic-reflection profiles.
Varved lake clays and marine muds in Long Island Sound have distinctive seismic signatures and were easily differentiated with minimal sampling (Lewis and Stone, 1991; Lewis and others, 1993). Other geologic units, including end-moraine deposits, proximal and distal glaciolacustrine fan deposits, glacial lake deltas, channel fill deposits, and marine delta deposits, were mapped by analyzing internal reflection characteristics of seismic units in a synergistic basin-wide context. The offshore mapping was made possible by collaborative interpretation of seismic lines by glacial geologists familiar with the distribution and internal structure of terrestrial Quaternary deposits and marine geologists familiar with the offshore seismic record. Systematic seismic survey coverage of the entire basin (see figure 2) was also an important factor in the offshore mapping effort because knowledge of the areal distribution of seismic units is necessary to their interpretation as specific depositional facies and correlation with on-land geologic units.
As seen in figure 4, the bedrock beneath Connecticut consists of Pre-Cambrian and Paleozoic crystalline rocks in the eastern and western highlands; these rock units extend offshore beneath Long Island Sound. The central lowland of Connecticut is underlain by early Mesozoic arkosic sedimentary rocks locally intercalated with igneous basalt and diabase. Semi-consolidated sands, gravels, and clays of Cretaceous and Tertiary-age coastal plain strata underlie the southern part of the Long Island Sound basin and Long Island. The presence of coastal plain strata beneath Long Island has long been known from extensive water-well and test-boring data (Fuller,1914; Suter and others, 1949); the extent of these units beneath Long Island Sound, terminated on the north side by a fluvially carved, glacially scoured cuesta scarp, has only recently been mapped in detail from systematic seismic reflection profiles in Long Island Sound (Lewis and Needell, 1987; Needell and others, 1987; Lewis and Stone, 1991).
Before glaciation, the crystalline and sedimentary rocks of Connecticut probably were covered by a nearly continuous mantle of weathered rock and saprolite, and unweathered bedrock outcrops were much less abundant than at present. That landscape would have resembled the present northern Piedmont landscape south of the glacial limit. Severely weathered bedrock (shown by map symbol x) occurs at many places beneath the relatively nonweathered glacial deposits, and represents the remnants of the formerly widespread mantle of weathered rock. The removal of most of the weathered rock probably took place before the last or late Wisconsinan ice sheet covered Connecticut, and many differentially weathered fractures were etched in relief; but the main features of the present landscape including the major hills and valleys, uplands and lowlands, were probably present in that preglacial landscape.
The character of the glacial deposits of the State is in part determined by the physical characteristics and mineral composition of the source rocks. This is conspicuously true of glacial deposits, and also of alluvium and stream-terrace deposits, which are largely derived from nearby glacial deposits. The relationship holds not only in areas directly underlain by specific rocks, but also "downstream" in the direction the glacier moved. The directions of movement (shown by the arrows on fig. 4), were south to southeast in most of Connecticut, but southwest to nearly west on the west side of the Connecticut Valley ice lobe. Abundant fragments of a particular rock generally occur within a few miles downstream from the source area, but scattered fragments, particularly of hard rocks, may occur tens of miles away. In the Glastonbury area of central Connecticut, till that is largely derived from Jurassic sedimentary rocks was deposited 0.5-1.5 km downstream on metamorphic rocks (Langer, 1977). Fragments of distinctive rock types downstream from restricted source areas form indicator fans, two of which are shown in figure 4. The indicator fan in western Connecticut is derived from the sedimentary and igneous rocks of the Pomperaug Valley (Pessl, 1970); the fan in southeastern Connecticut is derived from the Preston Gabbro (Goldsmith, 1982).
The following comments about the effects of specific rock types are based mostly on general impressions rather than on detailed studies. These comments apply most strongly to glacial tills; the removal of fine particles from water-washed sediment diminishes some of the contrasts in texture and color. The sedimentary rocks of central Connecticut produced glacial deposits that are commonly reddish brown; siltstones contributed considerable fine-grained material; and conglomerates contributed rounded pebbles and cobbles that have been broken out of matrix. Basalt and dolerite are relatively hard rocks; fragments are abundant near the sources, and occur mainly in the reddish-brown deposits derived from the adjacent sedimentary rocks. Deposits in the Eastern and Western Highlands derived from quartzites contain abundant quartz sand grains and quartzite fragments. Quartzites, because of their hardness, occur as very widespread erratic fragments. Marbles produced fine-grained, light-colored, highly calcareous glacial deposits. Dark schists and phyllites produced fine-grained, dark-colored deposits. Abundant flakes of mucsovite are very conspicuous in deposits derived from muscovite schists. The sulfidic schists and gneisses containing weathered iron sulfides produced very rusty deposits which may become iron-cemented. Granitic rocks that are low in dark minerals produced light-colored deposits, with abundant quartz and feldspar in the sand and coarse silt sizes. Dark mafic and ultramafic rocks produced dark-colored deposits that contain abundant dark iron minerals. The undivided light-gray to medium-gray schists and gneisses underlie more than half of Connecticut; they are the source rocks for the widespread sandy-silty tills of various shades of light and medium gray and yellowish gray; associated sand and gravels are generally yellowish to light brownish gray. These widespread deposits show considerable range of composition, color and texture, but lack the distinctive lithologic effects produced by the other rock types mentioned above.
During the last (late Wisconsinan) glaciation (25-20 ka), a sector of the Laurentide ice sheet of northeastern North America spread across the St. Lawrence Valley and the Green and White Mountains, covered all of Connecticut, and reached its maximum extent on Long Island (Sirkin, l982). The directions of ice movement are indicated by striations and grooves on bedrock, drumlins axes, and inferentially, by the positions of ice margins during retreat (fig. 4). Ice movement across the State was dominantly from NNW to SSE. The principal departure from that general trend was a prominent lobation in and adjacent to the Central Lowland. On the west side of that lobe, which probably became accentuated as the ice thinned during retreat, directions of movement were to the southwest or even west. Weaker lobate patterns occurred in the valleys of the Quinebaug, lower Connecticut, and Housatonic Rivers, and ice movement in westernmost Connecticut was influenced by the large lobe in the Hudson Valley. The glacial meltwater deposits so conspicuous on the map were all deposited during retreat of the late Wisconsinan ice sheet.
Less is known of the earlier (probably Illinoian) glaciation recorded by the presence of the lower till. Drumlins are composed dominantly of lower till, and their directions are probably partly inherited from the earlier glaciation. Glacial meltwater deposits of this earlier glaciation are rare; they evidently were erosionally removed or buried during the late Wisconsinan glaciation. It is likely that still earlier continental glaciations, recorded by deposits in the mid-continent and by oxygen isotope studies in ocean sediments and Greenland ice cores (Imbrie and others, 1984, Mix, 1987, Paterson and Hammer, 1987), also affected Connecticut but no direct evidence has yet been found within the State.
Two glacial tills, distinctive in character and different in age, are present in Connecticut and are described in the Description of Map Units (Sheet 2). These tills are difficult to show as separate map units because the lower, older till occurs almost entirely in the subsurface; generally it is at the surface today only in the floors of artificial excavations that are too small for the scale of the map. Lower till may be at the surface in the upper parts of some drumlins, but this occurrence is known only from local exposures and its extent cannot be predicted; most commonly lower till in drumlins is mantled by thin upper till. Numerous artificial exposures and subsurface well and test-boring data indicate that lower till (also called "drumlin till") comprises the bulk of material within drumlins and other areas mapped as thick till (tt). Lower till exposures are shown by open diamond symbol ( ) on the map and are now known from all parts of the State, in contrast to their limited known extent three decades ago (Schafer and Hartshorn, 1965; Pessl and Schafer, 1968; Pessl, 197l). Upper till, also referred to as "surface till" is generally thin and underlies the areas mapped as thin till (t). The till unit is shown as two separate blocks on the correlation chart to reflect its two different ages.
Recessional end moraines were first mapped in southeastern Connecticut by Goldsmith (1960; 1962; 1964) who defined five belts of segmented moraines including the Clumps, Mystic, Rocky Hollow, Ledyard and Oxoboxo moraines, most of which extend eastward into Rhode Island (Schafer, 1961; Goldsmith, l982). Farther west the Old Saybrook and Madison moraines were mapped by Flint (1971; 1975); however, further field investigation revealed that the distribution some of those moraine segments was exaggerated. We have reduced their areal extent on the map and have mapped separately the Hammonassett moraine in Clinton. Mapping of additional moraines segments between the two areas during State map compilation has allowed linear correlation of the Old Saybrook with the Rocky Hollow moraine (unit oem), the Hammonassett with the Ledyard moraine (unit hlem), and the Madison with the Oxoboxo moraine (unit mom). Offshore mapping of moraine segments and of sub-bottom, continuous, linear ridges of proximal lacustrine fans (unit Cplf), deposited at the grounding line of the ice margin in glacial Lake Connecticut, allows linear correlation of the Old Saybrook moraine along the Lordship lacustrine fan ridge to the Norwalk Islands moraine in southwestern Connecticut.
The moraine belts are relatively linear, but show down-ice topographic deflection where they cross valleys. Accumulations of rock debris were concentrated in the sheer zone where active ice rode up over thin stagnant ice at the margin. When the sheer zone remained in one position for a significant time, concentrations of debris built up within and on top of stagnant ice (Goldsmith, 1982). This material was later deposited on the land surface by ablation processes. The linear trend of the moraine belts reflects the former position of the sheer zone some relatively short distance behind the more ragged margin of stagnant ice. The segmented nature and local boulder-lag character of these moraines is probably due to the action of meltwater in the marginal zone. The moraine segments are most obvious in the upland areas between valleys. In valleys the morainic material may be buried by meltwater deposits; in most places meltwater deposition dominated in the valley and the morainic position is represented by the ice-proximal head of a morphosequence. Locally, moraine segments, more lobate than in upland areas, stand at the proximal heads of meltwater deposits in the valley.
The coastal Connecticut moraines are parallel with a much larger moraine belt that includes the Charlestown and Fishers Island moraines in Rhode Island and New York, and the Harbor Hill moraine on the north shore of Long Island (Sirkin, 1982). Parts of the Charlestown-Fishers Island-Harbor Hill moraine are shown on the Connecticut map because the area is included on the topographic base and this moraine provided basin closure for the containment of glacial Lake Connecticut in the Long Island Sound area.
A few scattered minor moraines have been identified elsewhere in the State to the north; these are much less extensive and reflect more irregular ice margins than those near the coast. One of these, the moraine at Windsorville in the Broad Brook quadrangle, is the only one in the State to have had an exposure of well-developed ice-thrust structures (Stone and others, 1982).
The shrinkage of the late Wisconsinan ice sheet, and the retreat of its margin from south to north across Connecticut, were accomplished by melting of the ice faster than it was resupplied by movement from the north. The meltwater picked up rock debris carried by the ice, and deposited most of it shortly beyond the ice margin as sorted, stratified layers of gravel, sand, silt, and clay, accumulated in streams and lakes, large and small, that were fed by the meltwater. Because meltwater largely flowed in valleys during deglaciation, meltwater deposits are concentrated in those valleys and in many places are more than 30 m in thickness. The drift thickness map (fig. 3) shows that the thickest glacial deposits in Connecticut are in the deepest parts of bedrock valleys and in the Long Island Sound Basin; these thick deposits are largely meltwater sediments that accumulated in glacial lakes.
A major conclusion drawn from the glacial geology of Connecticut is that most of the meltwater sediments were deposited in or graded to glacial lakes, large and small. On R. F. Flint's l930 map of the glacial geology of Connecticut, most of the meltwater deposits are mapped as "sand and gravel deposits in local temporary lakes (dammed by ice and controlled by spillways)" (Flint, l930). Many of the different ideas about the retreat of the last ice sheet in Connecticut were mainly efforts to explain the abundance of deltaic deposits, especially in southerly-sloping valleys (Gulliver, l900; Flint, l930, l932, l934; Lougee, l938, l953; Lougee and Vander Pyl, l95l; Black, l977, l982).
Our Quaternary geologic map of the State reflects our concurrence with Flint's early observation of pervasive deltaic bedding in these deposits, though not with his regional stagnation model for the formation of glacial lakes. Of the two-hundred and four map units of correlated meltwater deposits, one-hundred and eighty-three comprise deposits that were laid down in or graded to glacial lakes. These units are grouped into four types of glaciolacustrine depositional systems. We include within glacial lake units not only lake-bottom sediments and deltas, but also the fluvial deposits laid down in tributary valleys by meltwater streams that fed deltas in the lake. This leaves only twenty-one of the two-hundred and four correlated map units of meltwater deposits as glaciofluvial units grouped into two types of glaciofluvial depositional systems.
The meltwater deposits of Connecticut result mainly from the interaction of three factors: l) the form of the landscape across which the ice was retreating; 2) the form of the margin of the retreating ice; and 3) the locations of the principal meltwater streams emerging from the ice. These factors are not independent of one another; the form of the landscape influenced the other two to a considerable extent. The character of these deposits supports their interpretation through two closely related concepts: morphosequence deposition and stagnation-zone retreat (Currier, l94l; Jahns, l94l; Koteff, 1974; Koteff and Pessl, l981). These concepts have roots more than a century old (Upham, 1878; Emerson, 1898; Woodworth, 1898), were articulated in present form more than five decades ago, and have since been exemplified in many of the quadrangle studies used in compiling this map.
Morphosequences are the basic mappable geographic-chronologic units of glacial meltwater deposits. They are the bodies of sediment formed in particular valleys during particular short periods of time as meltwater streams aggraded their beds, filled proglacial ponds and lakes, and built up to maximum levels, commonly controlled by spillways or by older deposits or remnant ice downstream.
Seven types of morphosequences occur in Connecticut and are described in figure 1. They are defined by the distribution of glaciofluvial, glaciodeltaic, and glacial lake-bottom sedimentary facies, and by whether or not their heads were in contact with the ice. Near-ice-marginal morphosequences were formed short distances in front of the ice margin, separated from it by valley segments that were traversed by meltwater streams without deposition taking place. Ice-marginal morphosequences were deposited in contact with the ice, and their heads are marked by more or less well developed ice-contact scarps, shown by hachured ice-marginal lines on the map.
The depositional heads of many ice-marginal morphosequences probably extended well up onto the edge of the ice or into tunnels within it, but melting of adjacent and subjacent ice generally destroyed such headward parts, or caused them to be collapsed downward and later buried. Most morphosequences are one or two to several kilometers long, and few extend downstream more than l0 km. Meltwater streams either terminated in deltas, or their regimes changed downstream from aggradational to balanced or degradational. Meltwater terraces probably represent the downstream parts of stream systems near the heads of which active aggradation was occurring. The ending of deposition of one morphosequence and the beginning of another probably occurred by diversion of the meltwater flow, because of such events as opening of a new and lower spillway, retreat of the margin of the ice (which was the sediment source), or shifting of position of meltwater flow within the ice. The character of the sediment, lengths and numbers of morphosequences, and regional radio-carbon (14C) dating indicate that deposition of most morphosequences probably occupied periods of time in the range of l0 to l00 years.
Collapse structures-- folds and faults produced by the melting away of adjacent or subjacent ice-- are abundantly exposed in ice-marginal meltwater deposits. In contrast, ice-thrust structures produced by the push or drag of actively moving ice are rare, and have been seen only in a few local exposures. This indicates that for the most part the ice-marginal deposits were laid down in contact with stagnant or dead ice, too thin to transmit forward motion. The distribution of such evidence indicates that the retreating margin of active ice, was fringed by a continuous or nearly continuous zone of dead ice. The outer margin of the dead zone retreated by melting, and thinning of the ice near the margin caused retreat of the shear zone which separated active from stagnant ice. This process, by which the retreating active ice always is fringed with dead ice, is called stagnation-zone retreat. A minimum measure of the width of the fringe is given by the extent of the collapsed deposits headward of the ice contact, especially by the lengths of esker segments (Koteff and Pessl, 1981). Such data, together with inferences made from the topographic setting and texture of deposits about the proximity of continuous high-standing ice, indicate likely widths of 0.5 to 2 km in most places for the fringe of continuous dead ice. The dead ice disappears very irregularly because of differences in such factors as ice thickness, topographic position, thickness of mantling debris, and flow of meltwater through the ice. Therefore, detached ice masses of various sizes and shapes persist well beyond the fringe of continuous dead ice.
Dissent from use of the twin concepts of morphosequence deposition and stagnation-zone retreat was expressed by Black (l977, l979, l982). Using as his main example the Shetucket-Willimantic basin, he urged the alternative of basin-wide regional stagnation. Space does not permit a detailed rebuttal of his arguments here. We disagree with many of his descriptions and his interpretations of field situations, and we believe that many of his conclusions are incorrect. The net result is our conviction that the hypothesis of basin-wide stagnation is in contradiction to the great bulk of the evidence. In fact, the Shetucket-Willimantic basin seems to us to contain one of the best portrayals of morphosequence deposition and stagnation-zone retreat in eastern Connecticut, and our depiction of the area on the map reflects that belief.
The total number of morphosequences in Connecticut is well over a thousand, and they range from several to about twenty-five per quadrangle. For this reason alone, they could not be shown individually on the map. Instead, they have been combined into composite units of two to twelve morphosequences of common depositional setting, formed along the same or closely related paths of meltwater flow within a particular depositional system. These composite map units, of course, represent longer time intervals than do single morphosequences.
The two hundred and four, geographically restricted map units, are classified by six types of depositional systems which formed repeatedly in time and space during late Wisconsinan deglaciation. These depositional systems are discussed and defined in the Description of Map Units (sheet 2); they categorize almost all the meltwater deposits, regardless of the great variety of local detail. Glaciolacustrine systems comprise four of the six types (IL, SL, IP, and SP) and include large, open lakes that formed in broad valleys and basins, and related series of small lakes and ponds that formed in the narrower upland valleys and pocket basins; these large and small lakes formed in northerly sloping valleys and basins where they were ice-dammed and in southerly sloping valleys and basins where they were sediment-dammed. Glaciofluvial systems include meltwater streams in close contact with the ice margin (FP) and meltwater streams that incised and redeposited material distally from the ice margin (FD). The six depositional systems are defined by lithostratigraphic principles; each has typical spatial distribution of various sedimentary facies and types of morphosequences. The depositional systems are distinguished on the map by contrasting groups of colors.
The ice-contact scarps of ice-marginal meltwater deposits mark temporary, local positions of ice margins. Since the deposits generally are controlled by local topographic and hydrologic factors, they are correlatable from one valley to another only locally where lacustrine deposition in one valley required ice-marginal barriers in a nearby valley. However, ice-margin lines can be extended laterally from deposits, on the basis of inferred ice-marginal trends and inferred slopes of the ice surface. General ice-marginal trends are indicated by the positions of deposits against the topography. Slopes of the ice-surface are estimated to be in the vicinity of 100-l50 ft/mi (20-30 m/km), substantially less than the known slopes of the margin of active ice; such slopes are plausible given the extent, altitudes, and requisite ice-margin positions of the glacial lake deposits.
The method by which ice-margin lines were extended is not without error, and the accuracy of a particular position may be explained by the following rationale. The more extensive the meltwater deposits in an area, the more reliably one may extend an ice-margin line across it. The farther one extends a line across an upland with few or no meltwater deposits, the greater the likely error. Incorrect estimates of general trends will produce errors in correlation from east to west. Incorrect estimates of ice-surface slopes will produce errors in the degree of topographic inflection of ice margins. Nonetheless the extended ice-margin lines, though probably incorrect in detail, are believed to give a reasonable picture of the forms of successive ice margins during ice retreat: the general east-northeast to west-southwest trend; the great lobe in the Central Lowland and the sub-lobes within it; the lesser lobes in the Quinebaug, lower Connecticut, and other valleys; and the zigzag patterns across uplands. We emphasize that these lines represent the outer (down-ice) margins of continuous dead-ice fringes. The inner margin of the stagnant zone, although not shown would have much less topographic inflection.
Figure 5 shows selected ice margin retreat positions across Connecticut highlighted from those shown on the map (Sheet 1); approximate dates (in 14C years) associated with four recessional ice-margin positions are estimated from a regional array of deglaciation dates from outside of the area as well as within it (correlation diagram, map sheet 3) (Stone and Borns, 1986; Ridge and Larson, 1990; Stone and Ashley, 1992). 14C dates in Connecticut which are relevant to the deglacial chronolgy are discussed and referenced in the section entitled Radiocarbon-dated Localities on sheet 2; their locations are shown on sheet 1. Positions of major glacial lakes which dominated the deglacial history of the State are shown schematically in Figure 5. Deposits of each of these lakes as well as the multitude of smaller glacial lakes and glaciofluvial systems shown on the map record a detailed history of ice retreat across Connecticut; however space here does not permit a detailed discussion of them all. Glacial Lakes Connecticut, Middletown, and Hitchcock were the three largest, longest-lived, and most important lakes. Deposits of these lakes are extensive and are shown on the map as multiple units; unit descriptions (Sheet 2) refer only to the individual deposits and not to the lake as a whole; therefore the details of lake formation, history of deposition and drainage are discussed here.
Deglaciation of the Long Island Sound Basin was entirely dominated by the presence of Glacial Lake Connecticut (C, in fig. 5) which was impounded in the Long Island Sound Basin behind the Harbor Hill-Roanoke Point-Fishers Island-Charlestown end moraine. Formation of the lake began at about 19 ka when according to Stone and Borns (1986) the ice front began to recede from the end moraine position. Meltwater was impounded in the expanding long narrow basin between the moraine to the south and the retreating ice margin to the north. Initially, when lake-levels were highest, the impounded water was probably coextensive with a lake in Block Island Sound, and the whole system spilled across a notch in the terminal moraine at the head of Block Channel (Lewis and Stone, 1991). As the Block Channel spillway was eroded, lake-levels lowered and eventually the surface of the inner moraine in the relatively low area between Fishers Island and Orient Point emerged; this resulted in the formation of a separate body of water in the Long Island Sound Basin defined by Stone et al (1985) as glacial Lake Connecticut. Lake Connecticut water-levels were controlled by a spillway across the lowest point on the inner moraine at the place today called The Race. From a relative initial spillway altitude of about -10 m (33 ft), the lake gradually lowered by erosion at the Race to a final spillway altitude of about -60 m (197 ft); the rate of this erosion was controlled by the rate of lowering of the lake in Block Island Sound to the south.
Systematic northward retreat of the ice margin through the Long Island Sound Basin is recorded by sequential ice-marginal lacustrine-fan deposits (map unit Clf) built on the lake bottom by meltwater that issued from tunnels at the grounding line of the ice. The proximal parts of the largest of these lacustrine fan deposits are linear, ice-margin-parallel features deposited in deeper water areas along the trends of submerged extensions of the recessional moraines of coastal Connecticut.
Deltas deposited in glacial Lake Connecticut (map units Cmy, Cp, Cjt, Cni, Cwo, Cmi, Ch, Cew, Ce, Cn, Cdm, Cl, Cb, Cs) exist along the Connecticut shoreline near the mouths of most rivers entering the Sound. Delta topset-foreset contacts are paleo-water-level indicators of the lake; they occur at altitudes of 0 to 10 m above present sea level in the coastal deltas and indicate that, at its highest levels (when the ice margin stood at positions near the present Connecticut shoreline), glacial Lake Connecticut occupied all of the Long Island Sound Basin and extended into the present river estuaries. When their present altitudes are adjusted to the regionally established glacio-isostatic rebound slope of 0.9 m/km (4.74 ft/mi) to the N21oW (Koteff et al., 1988), coastal deltas record slowly lowering lake levels during their sequential construction. Because of its east-northeast trend, the ice margin first retreated onto higher ground out of the lake basin in the southeast corner of Connecticut (east of the mouth of the Thames River). Meltwater flowed from on-land ice-margin positions down bedrock valleys and built deltas into the lake near the present coastline; these deltas record the highest lake levels [0 to about 2 m (6 ft) below present sea level which project to the Race spillway at about -9 to -10 m (31 ft)].
As estimated by correlation of the Norwalk Islands moraine with morainal positions in New York and New Jersey, by about l7.6 ka (Stone and Borns, 1986), the ice sheet had established a recessional position just offshore of the present western Connecticut coast, along the Norwalk Islands moraine and its submerged eastward extension. Eastward along this trend, the ice margin stood at the head of an emergent delta at Lordship which marks an interlobate angle. Lobation of the ice front southeastward from Lordship, across the deep offshore extension of the Hartford Basin is marked by an extensive lacustrine fan deposit (Lewis and others, 1988); this fan deposit appears to correlate with a line of less extensive fans and submerged end moraine that trends northeast and passes onshore at Old Saybrook. Deltas at the mouth of the Connecticut River (unit Cwo) built in front of this ice margin position indicate that lake levels had lowered only a few meters from the initial spillway altitude of -10 m (33 ft) by this time.
Further ice retreat progressively deglaciated shoreline areas west of Old Saybrook. A recessional position marked by the Hammonassett-Ledyard moraine line (unit hlem) that passes offshore at Hammonassett Point near Clinton appears to correlate with a lacustrine fan line south of New Haven which mimics the lobate shape of the Lordship lacustrine fan deposit and lies about 3 mi (5 km) to the north of it. West of Milford, the ice front had also retreated out of the lake. Deltas were built into the lake directly in front of the Hammonassett moraine and the slightly younger, Madison moraine (unit mom) to the northwest. At this time, west of Milford, meltwater streams flowed southward down bedrock valleys and built deltas into the lake at altitudes indicating that the lake had lowered about 5 - 7 m (16 - 23 ft).
As deglaciation progressed, the topography of the Central Lowland produced a lobe of ice extending southward from New Haven that lingered the longest in the lake. As this ice lobe retreated out of the lake, a complex of deltas was built northeast of Milford, in West Haven, New Haven and East Haven; the lake had lowered about 10 m (33 ft) by this time. These deltas are extensive both on land and offshore. It is clear from continuous internal reflectors indicating stratigraphic equivalence that in the deep central basin much of the varved lake-clay section [here commonly 100 m (330 ft) thick] settled out in the lake concurrently with the Milford-New Haven area delta building. Onland evidence indicates that this delta building took place during the time of ice retreat from the Lordship position to about 6 km north of New Haven (see 16.5 ka ice position on fig. 5). Current estimates on deglacial chronology suggest that this period was perhaps on the order of 1000 years.
A much thinner section of varved lake clay overlies the section that is stratigraphically equivalent to the New Haven deltas and provides evidence that the lake continued to exist as the ice margin retreated northward; however, coarse-grained sediment supply to the lake was largely cut off when the ice margin left the shorter drainage basins feeding the lake, and when new glacial lakes trapped much of the sediment in larger drainage basins to the north. During this time, the level of glacial Lake Connecticut continued to lower due to erosion at the spillway. As shoreward portions of the lakebed were subaerially exposed, streams locally entrenched older, higher-level delta deposits and redeposited coarser material farther out into the lake basin; this material is seen in places at the very top of the varved lake-clay section. The gradually shrinking glacial lake may have lasted another 1000 years; this seems a reasonable estimate for duration of varve deposition in the upper lake section. If so, by about 15.5 ka, the lake was completely drained and the lakebed was subaerially exposed.
The Eastern Highlands include the Quinebaug-Southeast Coastal Basin, the Thames Basin, and the Lower Connecticut Basin (fig. 2). During early deglaciation of the southeast-coastal area, the ice margin emerged from glacial Lake Connecticut and retreated onto land. The resultant change in ice-flow regimen and slow ice-margin retreat rate in the coastal upland area resulted in deposition of three recessional moraines, the Mystic, Old Saybrook-Rocky Hollow and Hammonassett-Ledyard moraines. Concurrent with early morainic deposition, fluviodeltaic deposits were laid down, graded to glacial Lake Connecticut in south-draining coastal valleys. These early deposits served as initial sediment dams for series of ponds (SP depositional system) that developed with continued northward ice retreat. Once the ice margin had retreated to north of the east-west trending Quinebaug River drainage divide, series of small ice-dammed ponds (IP) and three major ice-dammed glacial lakes (IL), Voluntown, Pachaug, and Oneco (units lvo, lp, lon) were dammed by the edge of the ice. With further retreat into the broad Quinebaug River valley sediment dammed lake deposition (SL) began with the formation of glacial Lake Quinebaug (unit lqb) (first described by Stone and Randall, 1978) which controlled meltwater deposition for a considerable time during ice retreat in the valley northward to the vicinity of Danielson. This lake was impounded by a glacial drift dam that filled the bedrock gorge of the present Quinebaug River south of Jewett City. The water level of the lake was controlled by a spillway over bedrock adjacent to the gorge with an altitude of 134 ft (41 m). The glacial lake lengthened northward in the Quinebaug lowland as the glacier margin retreated; deltaic deposition controlled by the Lake Quinebaug water plane was interrupted for a time by SP depositional processes of unit qb as small, sediment-dammed ponds temporarily controlled deltaic deposition at levels slightly higher than Lake Quinebaug; however after this brief interval, erosion of sediment dams farther south returned the controlling base level to the Lake Quinebaug water plane. During ice retreat in the Quinebaug valley northward from Danielson, SP depostional systems controlled the levels of predominately ice-marginal deltaic and fluviodeltaic deposits including units da, pt, ew, wt and fr. In small north-draining tributary valleys series of small ice-dammed ponds (IP) controlled deltaic deposition for five map units (mb, wr, wb, epg and nth).
In the Thames basin, early meltwater deposits in south-draining coastal streams were graded to glacial Lake Connecticut; In the lower Thames River valley these deposits were largely removed by later meltwater erosion. Moraine segments also locally cross the valley, and glacial Lake Uncasville was impounded as a sediment-dammed lake probably behind a later-eroded segment of the Ledyard moraine in the Thames River valley. Ice-marginal deltas and fluviodeltas built into this lake during the time of ice retreat northward to the Norwich area. Farther north in the basin, SP depositional proccesses prevailed in the relatively narrow Willimantic, Yantic, and Shetucket River valleys where sixteen map units have been differetiated. The most laterally extensive SP deposits (unit wil) were laid down in a series of small sediment dammed ponds associated with 12 successive ice-marginal positions in the Shetucket River valley. In the small northerly-draining tributary valleys to the larger southerly-draining valleys, IP depositional processes controlled deposition of six map units.
South-draining valleys in the lower Connecticut basin were initially blocked by meltwater deposits graded to glacial Lake Connecticut and locally by moraine segments. This process was succeeded by sequential sediment-dammed ponding (SP) behind the coastal-area deposits; twelve units of sediment-dammed pond deposits were differentiated in the lower Connecticut basin. Of these units lcc, lce, and lct in the Connecticut River valley were the most extensive, completely filling the valley from side to side. The deposits formed the lengthy sediment dam for glacial Lake Middletown. Major ice-dammed glacial lakes (IL) formed in two northerly draining valleys in the lower Connecticut basin, glacial lakes Essex (unit lex) and Colchester (unit lc). Five series of small ice-dammed ponds (IP) controlled deposition in smaller north-draining valleys.
Because of the overall east-northeast trend of the retreating ice margin, the Eastern Highlands were entirely deglaciated while ice still remained in the Western Highlands and Central Lowland. Ice retreat across this area probably occurred between 18.0 and 16.0 ka based on regional correlation of ice margins (Stone and Borns, 1986). The earliest 14C-dated organic horizons in the area, 15.0 ka at Rodgers Lake (Davis and others, 1980) and 15.2 ka at Cedar Swamp (McWeeney, 1995), both record tundra vegetation at that time, but at both sites cores did not penetrate to the base of the organic section.
The Western Highlands include the Housatonic-Southwest Coastal Basin and the Naugatuck Basin (Fig. 2), and are dominated by narrow bedrock valleys within extensive upland areas. Because of the narrowness of most valley segments in this area, meltwater deposits are predominantly units of the SP and IP depositional systems. Several valleys contain proximal fluvial deposits (FP) in their steeper reaches. Ten major glacial lakes developed in the broader valley segments as the ice margin retreated northwestward through the area. Once the ice margin had retreated across the east-west trending divide that separates the southwest coastal rivers from the Housatonic drainage basin, Glacial Lakes Danbury and Pootatuck formed as ice dammed lakes (IL) in north-draining tributary valleys following initial deposition in a series of ice-dammed ponds (IP system). Glacial Lake Danbury, first described by Thompson (1976) was the most extensive lake in the Western Highlands; it existed in the Still River valley and is recorded by deposits (units lds, ldp and ldh) graded to three successively lower stages and by extensive lake-bottom deposits. Glacial Lake Pomperaug (unit lpt) formed as a sediment dammed lake (SL) in the Pomperaug valley, a southerly draining tributary valley to the Housatonic River; it was dammed by sediment in the Housatonic valley and coexisted with the Pumpkin Hill stage of glacial Lake Danbury. Northerly draining tributary valleys in the Western Highlands contain deposits of ten series of ice-dammed ponds (IP depositional system). Ice retreat in the relatively narrow Housatonic and Naugatuck River valleys was dominated by sedimentation in series of small sediment-dammed ponds (SP depositional system). Locally, in the more steeply sloping valley reaches, proximal glaciofluvial (FP) deposition took place. Continued flow of distal meltwater, especially in the Housatonic valley, eroded large parts of the ice marginal deposits leaving only remnant terraces on valley sides.
With retreat into the northwest corner of Connecticut, the ice margin dammed glacial Lake Hollenbeck in the north-draining Hollenbeck River valley and glacial Lake Norfolk in the Blackberry River valley. Deposition in these north-draining valleys largely preceded development of glacial Lake Great Falls. The lake developed in the broad, over-deepened basin of the Housatonic River valley, which is underlain by easily erodable marble. Ice-marginal deltas built from the last upland ice-margin positions in Connecticut were constructed in glacial Lake Great Falls deposits of which are more extensive in Massachusetts.
The Central Lowland includes the Farmington-Quinnipiac Basin and the Upper Connecticut Basin (fig. 2). Deglaciation of this area was characterized by major lobation of the ice margin, strongly influenced by the topographic relief between the Eastern and Western Highlands and the Central Lowland; the traprock ridges which separate the Farmington-Quinnipiac basin from the Upper Connecticiut basin also produced a topographic inflection in the lobate ice margin. Ice retreat was northwesterly on the east side of the Lowland and northeasterly on the west side (see fig. 4). Meltwater deposition in the Central Lowland was dominantly controlled by major glacial lakes. In the Farmington-Quinnipiac Basin, initial damming of the Quinnipiac valley was provided by part of the New Haven delta deposits (unit Cn) graded to glacial Lake Connecticut; these deposits filled the valley at Fair Haven and impounded glacial Lake Quinnipiac (unit lq). This lake recieved only local ice-marginal deltaic deposition, but extensive lake-bottom deposits (New Haven Clay) filled the Quinnipiac valley to just south of the bedrock gorge at South Meriden. North of the gorge, glacial lakes Southington (ls), Farmington (lf), and Tariffville (lt), all major sediment-dammed lakes (SL depositional system) controlled meltwater deposition in the basin. In narrower, steeper parts of the valley, SP depositional processes prevailed during northward ice retreat and deposits of units sb, un, ws. sl, and ss were laid down also in the valley. In northerly-draining tributary valleys to the main valley the ice margin dammed two major glacial lakes (IL), Lake Bristol and Lake Nepaug and six series of small, ice-dammed ponds (IP).
Following drainage of glacial lakes in the Farmington-Quinnipiac valley, an extensive glaciofluvial deposit (unit qt) of the FD depositional system was deposited, inset into the older lake deposits. The Quinnipiac valley terrace (Stone and others, 1985) is the most extensive glaciofluvial deposit in the State. It was deposited by distal meltwater that flowed from the Western Highlands southward down the Quinnipiac valley and carried mica-rich, quartzo-feldspathic light-colored sands derived from crystalline rocks. The sands contrasts in color with the red brown color of the ice-marginal glacial lake sediments derived from local Mesozoic rocks in the Valley (Krynine, 1937; Lougee, 1938). When the Tariffville gap, where the present Farmington River flows through the trap ridge into the upper Connecticut Basin, was deglaciated meltwater drainage was diverted northward and deposition of unit qt ceased.
Deglaciation of the upper Connecticut Valley was dominated by sedimentation in glacial Lakes Hitchcock and Middletown. Preceding these two lakes were several ice-dammed lakes in north-draining tributary valleys of the Connecticut River valley. Glacial Lake Coginchaug was impounded in the north-draining Coginchaug River valley. The early stage of the lake (unit lcd) spilled across the main drainage divide to the south; the later stage of the lake (unit lcm) spilled eastward into the Connecticut valley. IP deposition of unit hh also preceded formation of glacial Lake Middletown in Meriden in small northeast sloping valleys which drain the dip slope of the Hanging Hills (trap ridges) and are tributary to the Mattebessett River.
On the east side of the upper Connecticut valley, north-draining valleys contained a series of ice-dammed ponds (unit cd) which spilled across the upper Connecticut drainage divide. These were followed by a series of long, narrow ice-dammed glacial lakes that formed in valleys oblique to the trend of the ice margin: lakes Roaring Brook (unit lrb), Salmon Brook (unit lsb), and Manchester (unit lma). These ice-dammed depositional systems formed in northerly sloping tributary valleys to the upper Connecticut basin and preceded the development of major sediment-dammed lakes in the main basin.
Glacial Lake Middletown first developed along the Connecticut River and in the Mattabessett River basin. It was impounded by a long mass of earlier deposits (unit lct) in the lower Connecticut River valley at and south of The Straits; the spillway, with an initial altitude of about 130 ft (40 m) was over these deposits. Successive Ice-marginal deltaic deposits were built into the lake as the ice retreated northward. When adjusted for the regionally established postglacial tilt of 4.74 ft/mi (0.9 m/km) to the N21oW, delta topset-foreset contacts indicate that the lake slowly lowered due to erosion of its sediment dam. Lake Middletown occupied the Middletown basin in the lower Mattabessett valley and extended into the Berlin basin in the upper Mattabessett valley, as is indicated by accordant delta levels, by basin geometry resulting in ice-margin positions that trend NW-SE, and by the extent of the Berlin clays. Deltas in Cromwell (unit Mc), Newington (unit Mn), and New Britain (unit Mw) were built contemporaneously and record lake levels at the spillway of about 110-115 ft (34-35 m).
Just north of the Cromwell deltas, deltas of unit db were laid down in waters that were temporarily ponded to a higher level than Lake Middletown, controlled by the Dividend Brook spillway over deposits of unit Mc; this spillway was not eroded lower than its present level of 129 ft (39 m) because of the presence of Lake Middletown at its mouth.
When the ice uncovered the lower part of the divide between the Hartford basin and the Middletown/Berlin/New Britain basin where the New Britain spillway of Lake Hitchcock would later exist, Lake Middletown persisted at a level high enough to spread across the divide into the Hartford basin. When the ice retreated from the north end of Cedar Mountain (Newington-Hartford town line), the Dividend Brook spillway was abandoned and lake Middletown spread eastward into the south end of the basin later occupied by Lake Hitchcock. Deposits of units Mw, Me, Mg, and Mwv as well as lake-bottom deposits (unit Mb) in the Hartford basin all occur at altitudes accordant with Lake Middletown, but too high to have been controlled by any possible early level of the New Britain spillway. Not until Lake Middletown had lowered to below 110-115 ft (34-35 m) at the divide [about 65 ft (19 m) at the Straits spillway] could the New Britain spillway come into use as the outlet for Lake Hitchcock; this did not happen until the ice margin had retreated to north of Windsor and Windsorville.
Glacial Lake Hitchcock existed in the upper Connecticut River basin in Connecticut, Massachusetts, Vermont, and New Hampshire, lengthening to at least 185 mi (300 km) as the ice retreated northward to the vicinity of Burke, Vermont. The Connecticut River valley was dammed to an altitude of l50-l60 ft (46-49 m) at Rocky Hill - Glastonbury by deposits of glacial Lake Middletown (units Mc and db on Sheet 1); this mass of stratified drift is often referred to as "the Rocky Hill dam". The spillway for Lake Hitchcock was not over the dam however, but at the lowest place across the Mattabassett River drainage divide between the Hartford Basin and the Middletown-Berlin Basin in New Britain. When the ice margin first retreated into the Hartford basin, north of that divide, Lake Middletown water covered the later New Britain spillway location and early ice-marginal deltas in the Hartford basin were controlled by Lake Middletown. Not until Lake Middletown had dropped to below ll5 ft (35 m) could the New Britain spillway area emerge and Lake Hitchcock exist as a separate water body; this occurred at about the time that the ice margin was at Windsor and East Windsor.
During the early life of the lake, the New Britain spillway was eroded into till and older stratified drift so that water levels at the spillway dropped from about 115 ft (35 m) down to 82 ft (25 m) in altitude (Langer, 1977; Langer and London, 1979). In Connecticut, all ice-marginal and distal meltwater-fed deltas as well as one small delta built by meteoric water record lake levels higher than the longer-lived stable level. These deltas show a gradual lowering of lake level as the ice retreated northward and the New Britain spillway was incised down to bedrock. Ice-marginal deltas in Windsor (unit Hhw) and East Windsor (unit Hhe) record 110-115-ft (33-35-m-) levels at the spillway. To the north ice-marginal deltas in Suffield (unit Hhr) and Enfield (unit Hhs) indicate 105-110-ft (31-33-m-) levels at the spillway; still farther north in Suffield and Enfield, deltas of units Hhc and Hhn record levels just below 100 ft (30 m) at the New Britain spillway. This early phase of glacial Lake Hitchcock is recorded by ice-marginal deltas that are found well into southern Massachusetts and that were built to lake levels between 85 and 95 ft (26 and 29 m) at the spillway. This higher-than-stable-level phase of the lake is referred to as the "Connecticut Phase" (Koteff and others, 1988). It is important to note that deepening of the spillway channel was controlled by conditions 31-37 mi (50-60 km) to the south. Base level for waters exiting the spillway was controlled by down-cutting in the lower Connecticut River and lowering levels of glacial Lake Connecticut in the Long Island Sound Basin. The Rocky Hill dam area was glacio-isostatically depressed 144 ft (44 m) and the New Britain spillway area was depressed 164 ft (50 m) more than the area at the mouth of the Connecticut River. In order for the New Britain spillway to lower by 10 m (33 ft) during the early phase of the lake, Lake Connecticut had to have already lowered to below -82 ft (-25 m) in altitude.
Delta levels in Massachusetts indicate that a stable lake level, 82 ft (25 m) in altitude, had been reached by the time the ice margin had retreated to just north of the Chicopee River Valley; regional correlation of 14C dates (Stone and Borns, 1986) place the ice front in this position at about 15 ka. The 82-ft- (25-m-) level indicates that the water flowing through spillway was about 24 ft (7.3 m) deep since its bedrock floor today is at about 58 ft (17.7 m) in altitude. Altitudes of topset-foreset contacts of ice-marginal deltas, from southern Massachusetts to the lake's northernmost extent, project to the stable level [82 ft (25 m) at the New Britain spillway] on a straight line which is tilted up to the north-northwest at a slope of 4.74 ft/mi (.9 m/km). The linearity of these projected delta altitudes indicates that the lake level was stable during the time of ice retreat from Chicopee, MA to Lyme, NH and that postglacial rebound of the land surface did not begin until after all ice-marginal deltas had been built, probably between l4 and 13.5 ka (Koteff and Larsen, 1989). Deltas that were not associated with the ice margin, but rather were built by meteoric water in most river valleys that entered the lake, also project to the stable lake level. In Connecticut, these include unit Hlh associated with the Hockanum River, unit Hls associated with the Scantic River, and unit Hlb where the Farmington River constructed a large delta northeastward into the lake in the area now surrounding Bradley International Airport. The Bradley Airport delta covers about 20 mi2 (50 km2) and the fact that its entire surface [which is tilted up to the N21oW in the amount of 4.74 ft/mi (0.9 m/km)] is graded to the stable 82-ft (25-m-) level provides evidence for the long duration of the stable level and also indicates that the lake was not affected by glacio-isostatic tilting until after nearly all of its deltas had been constructed.
It is important also to note that the New Britain spillway could not have lowered further than the 82-ft- (25-m-) level; a +82-ft (+25-m) altitude at the New Britain spillway is equivalent to a -82-ft (-25-m) altitude at the mouth of the Connecticut River when the (164 ft (50 m) of differential depression between the two localities is taken into account. The base of the channel through which the paleo Connecticut River carrying water that spilled from Lake Hitchcock was imposed on bedrock at -27 m (-88.5 ft) in altitude at the mouth of the present Connecticut River east of Saybrook Point; this point was the actual control for the "Stable Phase" of glacial Lake Hitchcock. The Stable Phase of Lake Hitchcock lasted from about 15 ka until about 13.7 ka and during this time, the southern part of the basin (south of the Holyoke Range in Massachusetts) was largely filled with deltaic and lake-bottom sediments. Preserved lake-bottom surfaces in Connecticut are at about 45 ft (14 m) in the south and 145 ft (44 m) in the north; the tilted stable-level paleo-waterplane over this area is at 63 ft (19 m) in altitude at the north edge of the Rocky Hill dam and 172 ft (39 m) at the Massachusetts border; thus, toward the end of the stable phase before the dam was breached, water depths in the lake were only 20-25 ft (6-7 m). Due to the fact that the bedrock basin which contained the lake north of the Holyoke Range in Massachusetts is deeper, the lake was not filled with sediment to the extent that it was in the southern basin. North of the Holyoke Range in Massachusetts, preserved lake-bottom surfaces are at 150 ft (46 m) in altitude, and at the end of the Stable Phase water depth was about 150 ft (46 m).
Fluviodeltaic deposits built southeastward into the lake by the Farmington River (units ft and Hf) record a "Post-stable Phase" of the lake during which levels were lower than the 82-ft (25 m) level at the New Britain spillway. A topset-foreset contact in the Hf deltaic deposits north of the Farmington River is at 127 ft (39 m); delta-surface altitudes in the same unit to the south of the river indicate slightly lower water levels. These levels project southward below the New Britain spillway-level to 50-60 ft (15-18 m) in altitude at the Rocky Hill dam and record lowering of lake levels as the dam was entrenched. A preserved 55-ft (17 m) terrace inset into the drift-dam sediments on both sides of the present Connecticut River in Rocky Hill and Glastonbury records this post-stable phase which was relatively brief in Connecticut. A 14C date [13,540+90 (Beta-59094, CAMS-4875)] on plant debris in lacustrine sands at the top of the lake-bottom section (locality #9 on the map and on sheet 2) associated with the Hf delta establish that the time of dam breech was at about 13.5 ka.
The dam was breached most likely by headward erosion of streams on its south side, possibly by ground-water sapping and possibly aided by earthquakes generated due to the initiation of postglacial rebound. Regardless of the mechanism by which the dam was breached, Lake Hitchcock could not lower below stable level, much less drain, until its bed was raised by glacio-isostatic tilting. Dam breaching and initiation of isostatic rebound was requisite in order to establish the lower water-level altitudes recorded in the post-stable Farmington River (unit Hf) deltaic deposits. Once this process began, it proceeded rapidly as the drift-dam was incised from just above 60 ft (18 m) in altitude (the stable level at the dam) down to just above 40 ft (12 m); once this 20 ft (6 m) of lowering was accomplished, Lake Hitchcock, south of the Holyoke Range, was entirely drained and the newly formed Connecticut River began to incise the lake floor (long the terraces of unit st) over the 50-mi (80-km) stretch between the Holyoke Range and the breached dam. Lake Hitchcock continued to exist north of the Holyoke Range with initial water depths [(lowered from stable level by only 20 ft (6 m)] of about 130 ft (40 m); continued lowering of the lake was controlled by the rate of rebound which made it possible for the lake bed south of the Holyoke Range to be incised.
An approximate 4,000-yr life span for Lake Hitchcock was indicated by Antevs (l922) through a method of varve-correlation in clay pits from Hartford, CT to the north end of the lake basin in St. Johnsbury, VT. This method assumes that the silt/clay varve-couplets are annual summer/winter deposits and that regional seasonal fluctuations affected the thickness of individual varves over the entire lake basin. Varved silts and clays of glacial Lake Hitchcock were used to construct Antevs' New England varve chronology between varve-yr 3001 and varve-yr 7000. Recently, Ridge and Larsen (1990) fit a 533-yr varve section from Canoe Brook in southern Vermont into the relative varve chronology of Antevs; they also placed the chronology in an absolute time frame with a 12.4 ka C14 date on plant debris in the Canoe Brook section at the position of varve 463 (varve 6150 in the Antevs chronology). Using this calibration of the varve chronology, lacustrine deposition at the south end of Lake Hitchcock (varve 3001) began at about 15.5 ka. The early Connecticut phase was followed by the longer Stable phase of the lake which lasted until about 13.5 ka (varve 5050). The Post-stable phase of the lake, which lasted only briefly in Connecticut, continued for another 2000 years north of the Holyoke Range until about 11.5 ka (varve 7000) (Stone and Ashley, 1992; 1995).
Postglacial deposits in Connecticut include stream terrace deposits (unit st), talus (ta), dunes (d), floodplain alluvium (a), swamp (sw) and salt-marsh (sm) deposits, beach deposits (b), fluvial-estuarine channel fill deposits (ch) and marine delta deposits (md); the onset of postglacial conditions was time-transgressive and began several thousand years earlier in the southern part of the State than in northern parts.
In the Long Island Sound Basin, significant postglacial events were drainage of the glacial lake and subsequent sea-level rise. The remnant glacial lake was probably completely drained by 15.5 ka and a fluvial channel system (linear scarp symbol on map) was being carved on the lake floor by meteoric streams flowing southerly in coastal Connecticut, and northerly on the north shore of Long Island; these tributary channels joined a major east-west trending trunk channel which also received distal meltwater drainage from the Hudson valley to the west (Stanford and Harper, 1991). The channel system exited the Basin through the lake-spillway notch in the terminal moraine at The Race and provided a path through the moraine for transgression of the sea from the south (Lewis and Stone, 1991). Minor fluvial sediments were deposited in the bottoms of the channels, but predominantly the channels are filled with estuarine sediment (unit ch) that was deposited as the early postglacial sea flooded these low-lying areas of the drained lake basin as eustatic sea-level began to rise significantly between 15 and 16 ka (Fairbanks, 1989; Bard and others, 1990) and before glacio-isostatic rebound began.
A major, wave-cut marine unconformity (mu on seimic section D-D'on sheet 3) was cut across the top of the estuarine channel fill and over higher lying lake deposits as sea level rose. The marine unconformity is present in seismic sections up to altitudes of about -25 m (-82 ft) indicating that sea level probably rose to this height in central Long Island Sound before crustal rebound began.
The diagram below shows a conceptual relative sea-level curve for central Long Island Sound (highlighted line) which was derived by combining the glacio-eustatic sea-level curve (blue dashed line) from Barbados (Fairbanks, 1989; Bard and others, 1990) and a curve representing the timing and total depth of glacio-isostatic depression in central Long Island Sound (dashed red line). The uplift curve is based on several assumptions (listed on diagram) which are indicated from regional evidence some of which is presented in this report and in others (Koteff and Larsen, 1991; Stone and Ashley, 1995). The presence of the extensive marine delta (unit md) that records a -40-m (-130 ft) relative sea-level in central Long Island Sound (Stone and Lewis, 1991: Lewis and Stone, 1991) provides good evidence for the conceptualized relative sea-level curve. The large volume of delta sediment required a significant length of time for construction and the constant -40 m (-130 ft) depth of the topset-foreset contact indicates that relative sea level was stable during the deposition of the delta. The only possible source of the great volume of sediment contained within the marine delta was the drained lakebed of glacial Lake Hitchcock in the Connecticut valley to the north. This sediment supply became available only when the stable phase of Lake Hitchcock ended at about 13.5 ka; as previously discussed glacioisostatic uplift had to occur in order for Lake Hitchcock to drain. Regional evidence from northern New England (Barnhardt and others, 1995; Koteff and others, 1993; Koteff, 1995) also indicates that isostatic rebound began around this time. Thus, the early rapid rate of uplift was balanced with the equally rapid rate of eustatic sea-level rise resulting in a sea-level stand in Long Island Sound at about -40 m (-130 ft) for several thousand years between 13.0 and 9.5 ka. during which time the marine delta was built and the Connecticut River terrace and floodplain surfaces were incised. Recently obtained 14C dates (9370+100 Beta-52257, 8530+80 Beta-52256) on basal organics beneath lowest terrace surfaces along the Connecticut River in Massachusetts indicate that most of the postlake incision into the lakebed had been accomplished by ~9.0 ka (Stone and Ashley, 1992). The volume of eroded lakebed sediment, as calculated from the area and depth of incised terraces, is 12 billion m3; this material now composes the marine delta, the calculated volume of which is 11.5 billion m3.
As eustatic rise overtook the rate of isostatic rebound, relative sea level in central Long Island Sound rose continuously; the transgression submerged the marine delta and a blanket of marine mud (not shown on map, seismic unit mm in section DD') accumulated over the entire basin. As marine waters deepened, intense tidal-scour conditions developed in eastern Long Island Sound, resulting in local reworking of marine delta sediments and the development of a very large sand-wave field in eastern Long Island Sound (Fenster and others, 1990). A record of 4 to 5 m (13-16 ft) of sea-level rise during the last 4,000 to 5,000 years is preserved in coastal salt-marsh deposits (Bloom and Stuiver, 1963; van de Plassche and others, 1989; van de Plassche, 1991; Patton and Horne, 1991).
In most of mainland Connecticut, postglacial conditions consisted predominantly of downcutting of glacial deposits by meteoric streams along stream terrace surfaces, followed by the establishment of floodplains at modern levels. Streams had eroded to modern floodplain levels relatively early, in some cases before 12.0 ka (O'Leary, 1975, Stone and Randall, 1978). Postglacial winds were intense and widespread as evidenced by the ubiquitous blanket of eolian sand and silt that overlies glacial sediments throughout the State and in which the modern soil is developed. The postglacial climate was severely cold for several thousand years following deglaciation. Paleobotanical studies reveal that treeless, tundra vegetation dominated by dwarf willow (salix herbacia) sedges (Cyprus and Carix), herbs and shrubs (dryas, artemesia) dated from earlier than 15 ka to about 13 ka was present in the area (Davis, 1980, Gaudreau and Webb, 1985, Jacobson and others, 1987, Thorson and Webb, 1991). Also, features interpreted as ice-wedge casts with polygonal-ground pattern deform eolian-capped glacial sediments in numerous localities in Connecticut (Schafer and Hartshorn, 1965; Schafer, 1968; O'Leary, 1975; Stone and Ashley, 1992, ). These features indicate that permafrost existed locally in areas where substrate conditions were favorable to its formation. The presence of permafrost structures indicates that mean annual temperatures were below 0o C during the early postglacial time interval.
In the upper Connecticut basin postglacial conditions were dominated by the continued existence of lake Hitchcock several thousand years after the ice-margin retreated from the area. Extensive fields of eolian sand dunes formed in the treeless environment, indicating the continued effects of strong winds. Dunes are present on deltaic and lake-bottom surfaces of glacial Lake Hitchcock. Dunes on deltaic and high-level lake-bottom surfaces were formed by north to north-northeastly paleo winds; these surfaces were available as early as 15.5 ka. Dunes on stable-level lake-bottom surfaces were formed by northwesterly paleo winds; these surfaces became available at about 13.5 ka as Lake Hitchcock drained. Evidence that severely cold temperatures persisted until the time of Lake Hitchcock drainage, exists due to the presence of hundreds of circular to sub-circular, rimmed depressions interpreted as pingo scars developed in the drained lakebed sediments (Stone and Ashley, 1989; Stone and others, 1991; Stone and Ashley, 1992). Paleobotanical records indicate warming of the postglacial climate about 12.5, accompanied by reforestation of the landscape by successive spruce, pine and hardwood forest from 12.5 to 9 ka (Davis, 1980, Gaudreau and Webb, 1985, Jacobson and others, 1987).
Antevs, Ernst, 1922, The recession of the last ice sheet in New England: American Geographical Society Research Series 10, 120 p.
Barnhardt, W.A., Gehrels, W.R., Belknap, D.F., and Kelley, J.T., 1995, Late Quaternary relative sea-level change in the western Gulf of Maine: Evidence for a migrating glacial forebulge: Geology, v. 23; no. 4, p. 317-320.
Bard, E., Hamelin, B., Fairbanks, R.G. and Zindler, A., 1990, Calibration ofthe 14C timescale over the past 30,000 years using mass spectrometric U-Th ages from Barbados corals: Nature v. 345, p. 405-410.
Beetham, N. and Niering, W.A., 1961, A pollen diagram from southeastern Connecticut: American Journal of Science, v. 259, p. 69-75.
Bender, M.M., Baerreis, D.A., Bryson, R.A., and Steventon, R.L., 1981, University of Wisconsin radiocarbon dates XVIII: Radiocarbon, v. 23, p. 145-161.
Black, R.F., 1977, Regional stagnation of ice in northeastern Connecticut: An alternative model of deglaciation for part of New England; Geological Society of America Bulletin, v. 88, p. 1331-1336.
______ 1979, Regional stagnation of ice in northeastern Connecticut: An alternative model of deglaciation for part of Nw England: Reply; Geological Society of America Bulletin, Part 1, v. 90, p. 118-120.
______ 1982, Modes of deglaciation of Connecticut: a review, in Larson, G.J., and Stone, B.D., eds., Late wisconsinan glaciation of New England: Dubuque, Iowa, Kendall/Hunt , p. 77-100.
Bloom, A.L., and Ellis, C.W., Jr., 1965, Postglacial stratigraphy and morphology of coastal Connecticut: Connecticut Geological and Natural History Survey Guidebook 1, 10 p.
Bloom, A.L., and Stuiver, M., 1963, Submergence of the Connecticut coast: Science, v. 139, p. 332-334.
Colton, R.M., 1960, Surficial geology of the Windsor Locks quadrangle, Hartford, Co., Connecticut: U.S. Geological Survey Geological Quadrangle Map GQ-137, scale 1:24,000.
Currier, L.W., 1941, Disappearance of the last ice sheet in Massachusetts by stagnation zone retreat [abs.]: Geological Society of America Bulletin, v. 52, p. 1895.
Dana, J.D., 1883-1884, Phenomena of the Glacial and Champlain periods about the mouth of the Connecticut valley-that is, in the New Haven region: American Journal of Science, 3rd series, v. 26, 1883, p. 341-361; v. 27, 1884, p. 113-130.
Davis, M.B., 1965, Phytogeography and palynology of northeastern United States, in Wright, H. E., Jr., and Frey, D. G., eds., The Quaternary of the United States: Princeton, N.J., Princeton University Press, p. 377-401.
______ 1969, Climatic changes in southern Connecticut recorded by pollen deposition at Rogers Lake: Ecology, v. 50, p. 409-422.
Davis, M.B., Spear, R.W., and Shane, L.C.K., 1980, Holocene climate of New England: Quaternary Research, v. 14, p. 240-250.
Davis, M.B. and Deevey, E.S., Jr., 1964, Pollen accumulation rates: estimates from later-glacial sediment of Rodgers Lake: Science, v. 45, p. 1293-1295.
Deane, R.E., 1967, The surficial geology of the Hartford South quadrangle, with map: Connecticut Geological and Natural History Survey Quadrangle Report 20, 43 p.
Deevey, E.S., 1958, Radiocarbon-dated sequences in eastern North America; Geobotanisches Institut Rubel veroffentlicheng, v. 34, p. 30-37.
Denny, C.S., 1936, Periglacial phenomena in southern Connecticut: American Journal of Science, 5th series, v. 32, p. 322-342.
Emerson, B.K., 1898, Geology of Old Hampshire County, Massachusetts: U.S. Geological Survey Monograph 29, 790 p.
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
Fenster, M.S., Fitzgerald, D.M., Bohlen, W.F., Lewis, R.S., and Baldwin, C.T., 1990, Stability of giant sand waves in eastern Long Island Sound, U.S.A. Mar. Geol., v. 91, p. 207-225.
Flint, R.F., 1930, The glacial geology of Connecticut: Connecticut Geological and Natural History Survey Bulletin 47, 294 p.
______ 1932, Deglaciation of the Connecticut Valley: American Journal of Science, 5th series, v. 24, p. 152-156.
______ 1933, Late-Pleistocene sequence in the Connecticut Valley: Geological Society of America Bulletin, v. 44, p. 965-988.
______ 1934, Late-glacial features of the Quinnipiac-Farmington lowland in Connecticut: American Journal of Science, 5th series, v. 27, p. 81-91.
______ 1953, Probable Wisconsin substages and late-Wisconsin events in northeastern United States and southeastern Canada: Geological Society of America Bulletin, v. 64, p. 897-919.
______ 1956, New radiocarbon dates and late-Pleistocene stratigraphy: American Journal of Science, v. 254, p. 265-287.
______ 1963, Altitude, lithology, and the Fall Zone in Connecticut: Journal of Geology, v. 71, p. 683-697.
______ 1971, The surficial geology of the Guilford and Clinton quadrangles, Connecticut (with maps): Connecticut Geolgical and Natural History Survey Quadrangle Report No. 28, scale 1:24,000, 33 p.
______ 1975, The surficial geology of the Essex and Old Lyme quadrangles, Connecticut (with maps): Connecticut Geological and Natural History Survey Quadrangle Report No. 31, scale 1:24,000, 41 p.
Flint, R.F., and Gebert, J.A., 1976, Latest Laurentide ice sheet: new evidence from southern New England: Geological Society of America Bulletin, v. 87, p. l82-l88.
Fuller, M.L., 1914, The geology of Long Island, New York: U.S. Geological Survey Professional Paper 82, 231 p.
Gaudreau, D.C. and Webb, T., III, 1985, Late-Quaternary pollen stratigraphy and isochrone maps for th northeastern United States, in Bryant, V.M., Jr. and Holloway, R.G., eds., Pollen records of the Late-Quaternary North American Sediments: American Association of Stratigraphic Palynologists, p. 247-280.
Goldsmith, Richard, 1960, A post-Harbor Hill-Charlestown moraine in southeastern Connecticut: American Journal of Science, v. 258, p. 740-743.
______ 1962, Surficial geology of the Montville quadrangle, Connecticut: U.S. Geological Survey Geological Quadrangle Map GQ-148, scale 1:24,000.
_____ 1964, Surficial geology of the Niantic quadrangle, Connecticut: U.S. Geological Survey Geological Quadrangle Map GQ-329, scale 1:24,000.
______ 1982, Recessional moraines and ice retreat in southeastern Connecticut, in Larson, G. J., and Stone, B. D., eds., Late Wisconsinan glaciation of New England: Dubuque, Iowa, Kendall/Hunt, p. 6l-76.
Gulliver, F.P., 1900, Thames River terraces in Connecticut: Geological Society of America Bulletin, v. 10, p. 492-495.
Hanshaw, P.M., 1962, Surficial geology of the Meriden quadrangle, Connecticut: U.S. Geolgoical Survey Geological Quadrangle Map GQ-150, scale 1:24,000.
Hill, D.C., and Shearin, A.E., 1970, Tidal marshes of Connecticut and Rhode Island: Connecticut Agricultural Experiment Station Bulletin 709, 34 p.
Hokans, D. H., 1952, Glacial water levels in the Housatonic-Naugatuck River valleys of western Connecticut: Unpublished M.A. thesis, Clark University, Worcester, Massachusetts, l06 p.
Imbrie, J., Hays, J.D., Martinson, D.G., McIntyre, A., Mix, A.C., Morley, J.J., Pisias, N.G., Prell, W.L., and Shackleton, N.J., 1984, The orbital theory of Pleistocene climate: support from a revised chronology of the marine 18O record, in Berger, A., Imbrie, J., Hays, J., Kukla, G. and Saltzman, B., eds. Milankovitch and Climate, Part 1: Dordrecht, Reidel, p. 269-305.
Jacobson, G.L., Jr., Webb, Thompson III, and Grimm, E.C., 1987, Patterns and rates of vegetation change during the deglaciation of eastern North America, in Ruddiman, W.F. and Wright, H.E., Jr., eds., North America and Adjacent Oceans During the Last Deglaciation: Boulder Colorado, Geological Society of America, The Geology of North America, v. K-3, p. 277-288.
Jahns, R.H., 1941, Outwash chronology in northeastern Massachusetts [abs]; Geological Society of America Bulletin, v. 52, p. 1910.
Jahns, R.H., 1953, Surficial geology of the Ayer quadrangle, Massachusetts: U.S. Geological Survey Geologic Quadrangle Map GQ-21, scale 1:31,680.
Kaye, C.A., 1964, Outline of Pleistocene geology of Martha's Vineyard, Massachusetts: U.S. Geological Survey Professional Paper 50l-C, p. 134-139.
Kelley, G.C., 1975, Surficial Geologic map of the Kent and Ellsworth quadrangles, Connecticut; U.S. Geological Survey Open File Report OF-75-171.
Koteff, Carl, 1974, The morphologic sequence concept and deglaciation of southern New England, in Coates, D. R., ed., Glacial geomorphology: Binghamton, N. Y., State University of New York, Publications in Geomorphology, p. 121-144.
Koteff, Carl, and Pessl, Fred, Jr., 1981, Systematic ice retreat in New England: U.S. Geological Survey Professional Paper 1179, 20 p.
Koteff, Carl, and Larsen, F.D., 1989, Postglacial uplift in western New England: Geologic evidence for delayed rebound: in Gregersen, S., and Basham, P.W., eds., Earthquakes at North Atlantic passive margins: Neotectonics and postglacial rebound: Norwell, Massachusetts, Kluwer Academic Publishers, p. 105-123.
Koteff, Carl, Robinson, G.R., Goldsmith, Richard, and Thompson, W.B., 1993, Delayed postglacial uplift and synglacial sea levels in coastal central New England: Quaternary Research, v. 40, p. 46-54.
Koteff, Carl, Stone, J.R., Larsen, F.D., Ashley, G.M., Boothroyd, J.C., and Dincauze, D.F., 1988, Glacial Lake Hitchcock, postglacial uplift and postlake archeology: in J. Brigham-Grette, ed, Fieldtrip Guidebook AMQUA 1988, University of Massachusetts, Department of Geology and Geography, Contribution 63, p. 169-208.
Koteff, Carl, Thompson, W.B., Goldsmith, Richard, and Larsen, F.D., 1995, Correlation of deglacial events in western and coastal New England; Geological Society of America, Abstracts with Programs, v. 27, no. 1, p. 61.
Krynine, P.D., 1937, Glacial sedimentology of the Quinnipiac-Pequabuck lowland in southern Connecticut: American Journal of Science, 5th series, v. 33, p. 111-139.
Langer, W.H., 1977, Surficial geologic map of the Glastonbury quadrangle, Hartford and Middlesex Counties, Connecticut: U.S. Geological Survey Geologic Quadrangle Map GQ-l354, scale 1:24,000.
Langer, W.H. and London, H.H., 1979, Surficial geologic map of the upper Connecticut River basin, Connecticut: U. S. Geological Survey Open-file Report OFR-79-232, scale 1:125,000.
Leopold, E.B., 1955, Climate and vegetation changes during an interstadial period in southern New England: unpublished doctoral dissertation, Yale University, New Haven, Connecticut, 99p.
Leopold, Estella, 1956, Two late-glacial deposits in southern Connecticut: National Academy of Science Proceedings, v. 52, no. 11, p. 863-867.
Lewis, R.S. and Needell, S.M., 1987, Maps showing stratigrqphic framework and Quaternary geologic history of eastern Long Island Sound; U.S. Geological Survey Miscellaneous Field Studies Map MF-1939A, scale 1:125,000.
Lewis, R.S., Oldale R.N., and Harris, S.L., 1988, A seaward extension of the Lordship deposit: evidence for a late-Wisconsinan ice position in Long Island Sound; Geological Society of America Abstracts with Programs, v. 20, no. 1, p. 32.
Lewis, R.S. and Stone, J.R., 1991, Late Quaternary stratigraphy and depositional history of the Long Island Sound Basin: Connecticut and New York: Journal of Coastal Research, Special Issue No. 11, p. 1-23.
Lewis, R.S., Stone, J.R., and DiGiacomo-Cohen, M.L., 1993, Interpretation of glaciolacustrine and postglacial marine facies in Long Island Sound; Geological Society of America Abstracts with Programs, v. 25, no. 2, p. 33.
Lougee, R.J., 1938, Physiography of the Quinnipiac-Farmington lowland in Connecticut: Colby College Monograph 7, 64 p.
Lougee, R.J., 1953, A chronology of postglacial time in eastern North America: Scientific Monthly, v. 76, p. 259-276.
Lougee, R.J., and Vander Pyl, A.W., 1951, Glacial water levels in the Thames-Willimantic river valley: Scientific Monthly, v. 73, p. 275-283.
McWeeney, L.J., 1995, Revised revegetation history for the post-glacial period (15,200 - 10,000 14C yrs BP) in southern New England, Geological Society of America Abstracts with Programs, v. 27, no. 1, p. 68.
McWeeney, L.J., 1994, Archaeological settlement patterns and vegetation dynamics in southern New England in the late Quaternary: unpublished doctoral dissertation, Yale University, 258p.
Mix, A.C., 1987, The oxygen-isotope record of glaciation, in Ruddiman, W.F. and Wright, H.E., Jr., eds., North America and Adjacent Oceans During the Last Deglaciation: Boulder Colorado, Geological Society of America, The Geology of North America, v. K-3, p. 111-135.
Moeller, R.W., 1980, 6LF21: a paleo-Indian site in western Connecticut; American Indian Archeological Institute, Occasional Paper Number 2, p.
Needell, S.W., Lewis, R.S., and Colman, S.M., 1987, Maps showing the Quaternary geology of east-central Long Island Sound; U.S Geological Survey Miscellaneous Field Studies Map MF-1939B, scale 1:125,000.
O'Leary, D.M., 1975, Surficial Geologic Map of the Moodus quadrangle, Connecticut, U.S. Geological Survey Geologic Quadrangle Map GQ-1205, scale 1:24,000.
Paterson, W.S.B., and Hammer, C.U., 1987, Ice core and other glaciological data, in Ruddiman, W.F. and Wright, H.E., Jr., eds., North America and Adjacent Oceans During the Last Deglaciation: Boulder Colorado, Geological Society of America, The Geology of North America, v. K-3, p. 91-109.
Patton, P.C., 1988, The geomorphic response of streams to floods in the glaciated terrain of southern New England: in Baker, V.R., Kochel, R.C. and Patton, P.C, Flood Geomorphology, J. Wiley and Sons, p. 261-278.
Patton, P.C. and Horne, G.S., 1991, A submergence curve for the Connecticut River estuary; Journal of Coastal Research Special Issue Number 11, p. 181-196.
Pessl, Fred, Jr., 1970, Surficial Geologic Map of the Woodbury Quadrangle, Connecticut. U.S. Geological Survey Geol. Quad. Map GQ-896, scale 1:24,000.
Pessl, Fred, Jr., 1971, Till fabrics and till stratigraphy in western Connecticut, in Goldthwait, R.P., ed., Till: a symposium: Columbus, Ohio, Ohio State University Press, p. 92-l05.
Pessl, Fred, Jr., and Schafer, J.P., 1968, Two-till problem in Naugatuck-Torrington area, western Connecticut, in New England Intercollegiate Geological Conference 60th Annual Meeting, New Haven, Conn., Oct. 25-27, 1968, Guidebook for fieldtrips in Connecticut: Connecticut Geological and Natural History Survey Guidebook 2, Trip B-l, 25 p.
Ridge, J.C. and Larsen, F.D., 1990, Re-evaluation of Antevs' varve chronology and new radiocarbon dates of sediments from glacial Lake Hitchcock: Geological Society of America, v. 102, p. 889-899.
Rodgers, John, (compiler), 1985, Bedrock geological map of Connecticut: Connecticut Natural Resources Atlas Series, Connecticut Geological and Natural History Survey, Hartford, Conneticut, scale 125,000.
Schafer, J.P., 1961, Correlation of end moraines in southern Rhode Island: U.S. Geological Survey Professional Paper 424-D, p. D68-D70.
Schafer, J.P., 1965, Surficial geologic map of the Watch Hill quadrangle, Rhode Island, Connecticut; U.S. Geological Survey Geol. Quad. Map GQ-712, scale 1:24,000.
Schafer, J.P., and Hartshorn, J.H., 1965, The Quaternary of New England, in Wright, H.E., Jr., and Frey, D.G., eds., The Quaternary of the United States: Princeton, N. J., Princeton University Press, p. 113-128.
Sirkin, Les, 1982, Wisconsinan glaciation of Long Island, New York, to Block Island, Rhode Island, in Larson, G. J. and Stone, B. D., eds., Late Wisconsinan glaciation of New England: Dubuque, Iowa, Kendall/Hunt, p. 35-59.
Stanford, S.D., and Harper, D.P., 1991, Glacial lakes of the lower Passaic, Hackensack, and lower Hudson Valleys, New Jersey and New York, Northeastern Geology, v. 13, no. 4, p. 271-286.
Stone, B.D. and Randall, A.M., 1978, Surficial geologic map of the Plainfield Quadrangle, Connecticut; U.S. Geological Survey Geologic Quadranlge Map GQ-1422, scale 1:24,000.
Stone, B.D., and Borns, H.W., Jr., 1986, Pleistocene glacial and interglacial stratigraphy of New England, Long Island, and adjacent Georges Bank and Gulf of Maine: in, Sibrava, V., Bowen, D.Q., and Richmond, G.M., (eds.), Quaternary Glaciations in the Northern Hemisphere: Oxford, Pergamon Press, p. 39-52.
Stone, J.R. and Ashley, G.M., 1989, Fossil Pingos?: New evidence for active permafrost in southern New England during late-Wisconsinan deglaciation; Geological Society of America, Abstracts with Programs, v. 21, no. 2, p. 69.
Stone, J.R. and Ashley, G.M., 1992, Ice-wedge casts, pingo scars and the drainage of glacial Lake Hitchcock, in Robinson, P. and Brady, (eds.) New England Intercollegiate Geological Conference 84th Annual Meeting, Amherst, Mass., Oct. 9-11, 1992, Guidebook for fieldtrips in the Connecticut Valley region of Massachusetts and adjacent States: University of Mass., Dept. of geology and geography Contribution no. 66, Trip A-7, p. 305-331.
Stone, J.R., and Ashley, G.M., 1995, Timing and mechanisms of glacial Lake Hitchcock drainage, Geological Society of America Abstracts with Programs, v. 27, no. 1, p. 85.
Stone, J.R., Ashley, G.M., and Peteet, D.M., 1991, Cross-section through a post-Lake Hitchcock surface depression; Geological Society of America, Abstracts with Programs, v. 23, no.1, p. 135.
Stone, J.R. and Lewis, R.S., 1991, A drowned marine delta in east-central Long Island Sound: evidence for a -40-m-relative sea level at > 12.3 ka; Geological Society of America Abstracts with Programs, v. 23, no. 1, p. 135.
Stone, J.R., Schafer, J.P., London, E.H., and Thompson, W.B., 1992, Surficial materials map of Connecticut: U.S. Geological Survey Special Map, scale 1:125,000.
Stone, J.R., Stone, B.D., and Lewis, R.S., 1985, Late Quaternary deposits of the southern Quinnipiac-Farmington lowland and Long Island Sound basin: their place in a regional stratigraphic framework, in R.J. Tracey, ed, New England Intercollegiate Geological Conference 77th Annual Meeting, New Haven, Conn., Oct. 4,5, and 6, 1985, Guidebook for fieldtrips in Connecticut and adjacent areas of New York and Rhode Island: Connecticut Geological and Natural History Survey Guidebook No. 6, Trip C6, p. 535-576.
Suter, R., de Laguna, W., Perlmutter, N.M., and Brashears, M.L., Jr., 1949, Mapping of geologic formations and aquifers of Long Island, New York: New York Water Power and Control Commission Bulletin, GW-18, 212 p.
Thompson, W.B., 1975, Surficial geologic map of the Danbury quadrangle, Connecticut, U.S. Geological Survey Open File Report, OF-75-547. scale 1:24,000.
Thompson, W.B., 1975, Surficial geologic map of the New Milford quadrangle, Connecticut, U.S. Geological Survey Open File Report, OF-75-548. scale 1:24,000.
Thorson, R.M. amd Webb, R.S., 1991, Postglacial history of a cedar swamp in southeastern Connecticut; Journal of Paleolimnology, v. 6, p. 17-35.
Upham, W., 1878, Modified drift in New Hampshire, in Hitchcock, C.H., 1878, The Geology of New Hampshire; Pt. 3, Surface geology: Concord, N.H., v.3, p. 3-176.
Upson, J.E., 1971, Surficial geologic map of the Mystic quadrangle, Connecticut, New York, Rhode Island; U.S. Geological Survey Geologic Quadrangle Map GQ-940, scale 1:24,000.
Upson, J.E., and Spencer, C.W., 1964, Bedrock valleys of the New England coast as related to fluctuations of sea level: U.S. Geological Survey Professional Paper 454-M, 44 p.
Van de Plassche, Orson, Mook, W.G., anbd Bloom, A.L., 1989, Submergence of coastal Connecticut 6000-3000 14C-years B.P.; Marine Geology, v. 86, p. 349-354.
Van de Plassche, Orson, 1991, Late Holocene sea-level fluctuations on the shore of Connecticut inferred from transgressive and regressive overlap boundaries in salt-marsh deposits; Journal of Coastal Research, Special Issue Number 11, p. 159-179.
Woodworth, J.B., 1898, Some glacial wash-plains of southern New England: Essex Institute Bulletin, v. 29 (1897), p. 71-119.
Contents | Credits
Description of Map
Introduction | Map Units | Preglacial Landscape and Bedrock Source Areas | Glaciation | Glacial Ice-laid deposits | Deglaciation | Glacial Meltwater Deposits | Chronology of Ice Retreat and Major Glacial Lakes | Postglacial Conditions | References
Description of Map Units
Postglacial Deposits | Early Postglacial Deposits | Glacial ice-laid deposits | Glacial meltwater deposits | Housatonic - Southwest Coastal Basin | Naugatuck Basin | Farmington - Quinnipiac Basin | Upper Connecticut Basin | Lower Connecticut Basin | Thames Basin | Quinebaug - Southeast Coastal Basin | Long Island Sound Basin | Radiocarbon-dated Localities
