Summer 1999

GEO 589 Research for Earth Science Teachers


The Cliffs at Caumsett State Park

Gloria Mandell

Table of Figures and Plates
fig. 1 Sketch showing position of glacial ice sheet(s) forming Long Island
fig. 2 Sketch showing proglacial lake in Long Island Sound
fig. 3 Sketch showing glacial lacustrine environments
fig. 4
Sketch showing ice cliff depositional environments
fig. 5 Sketch showing restored profile section from Long Island to Connecticut showing terminus of continental glacier in Long Island Sound
fig. 6 Sketch showing three contrasting strata deposited in a Gilbert-type delta
fig. 7  Sketch showing development of strata in cliffs at Caumsett
fig. 8
Sketch showing advance and retreat of glacier forming hummocky terrain at Caumsett
fig. 9 Photo of loess in the wind, Mantanuska Valley, Alaska
Fig. 10
A typical cross section of Long Island
fig.11 Stratigraphic section in Column 1
fig.12 Stratigraphic section in Column 2
fig.13 Stratigraphic section in Column 3

Plate 1 Photo of the westernmost section (Column 1), displays the large foreset beds of a classic Gilbert Delta 
Plate 2 Photo of the intermediate section (Column 2) east of Inspiration point which  does not have the prominent yellow gravel beds. This area is darker with more lacustrine deposits 
Plate 3 Photo of a spectacular Cretaceous outcrop overlain by Pleistocene sands and gravel whichdominates Column 3, near the easternmost section. 1-2 meters of till and a layer of loess as much as 1 meter thick overlie all sections. 
Plate 4 Photo of Cretaceous sediment showing distortion indicating glaciotectonic forces acting upon it subsequent to its deposition. 
Plate 5 Photo of a thin (5-6cm) blue-black layer of partially lithified sediment found at the top of the Cretaceous unit. Perhaps dissolved minerals rich iniron compounds have collected here. Just below this unit In Column 3, there are very bright yellow sands interwoven with the red, white, and purple sand and red and clay. 
Plate 6 Photo The Pleistocene sediments begin with a clay unit 25-30 cm thick. This brown clay layer has 1mm laminations. Above here are thin (2-4cm) layers of fine brown and tan sand laminated with white sand and brown and gray clay. Both the clay and sands show evidence of glaciotectonic disturbance with much folding and faulting with offsets of 3-4 cm.
Plate 7 Photo of foreset beds of a Gilbert-type delta near column 1


Caumsett State Park is located in Lloyd Harbor, New York. The cliffs sit majestically on the north shore of Long Island facing Long Island Sound and Connecticut. At present the Sound is a large body of salt water. However, some 20,000 years ago it was a fresh water glacial lake.


Pleistocene sediments in the cliffs were formed mainly in a Gilbert–type delta in a proglacial lake, whose shores bordered the glacier and the Ronkonkoma moraine to the south. Cretaceous deposits of fine, brightly colored sands and clay underlie the lacustrine deposits of the Pleistocene. The beach sediments and the boulders (erratics) positioned there are the result of erosion and deposition of the Cretaceous and the delta sediments.


The erratics on the beach which are derived from the tills are mostly granites and gniesses with some basalt boulders and numerous cobbles of conglomerate.


Park officials prohibit climbing on and digging into the cliffs. Therefore, the research consists of observations made from site visits and photographs taken of the bluffs and beach. Literary research of texts, journals, and professional papers was done as well.


A self-guided walking tour of the cliffs and beach is included. The more we understand, the greater we can appreciate and enjoy the area.


Gazing at the cliffs from the beach at Caumsett Park, one cannot help but be impressed. The distinctive layers, their different sizes, color, and inclination are certainly intriguing. The rocky beach with its huge boulders resting peacefully on the shore, suggests a time unlike the present. A historical perspective of Long Island can be found at this link.


There is some debate as to whether Long Island formed from the advance and retreat of a single glacier, or by distinctly different glacial advances. Mather (1843) first reported the two prominent curved ridges we know as the Ronkonkoma and Harbor Hill moraines, but he did not realize their glacial origin. Glacial geologist Warren Upham (1879) was the first to recognize them as terminal moraines of a continental glacier. Map patterns revealed the Harbor Hill moraine as the younger (fig. 1). For this reason Upham believed that different glaciers deposited each of the ridges. Chamberlain (1883) argued for one glacier believing the Harbor Hill moraine to be a recessional moraine of the same glacier which formed the Ronkonkoma moraine.


More recently Sirkin (1991) postulates that the two moraines are the result of a single glacier re-advancing over its own deposits.  This glacier then retreats and down wastes with the onset of climactic changes. 


Structural and stratigraphic evidence does not support this.  One glaciation would not have stacked large masses of lower drift and Cretaceous layers in the outwash of the upper drift.  In addition a single event would not account for the distinctly different lobes or the distinctly different sediments found in the two moraines.  The glacial deposits of Long Island are the result of several glacial lobes that formed during both glaciations.  There is evidence for a warm interval separating the two glaciations.  Radiocarbon dating of Harbor Hill outwash sediments in the mid-Wisconsinin range confirms this (Sanders, Merguerian, 1994)


Twenty-two thousand years ago the southern shoreline of Long Island was about 70 miles south of the ice front and near the edge of the continental shelf. The glaciers covering massive areas derived their water from the oceans, as a result sea level was 350 feet lower than at present. Mastodons and Giant Condors roamed the outwash plains.


As the climate warmed, the glacier receded trapping melt water between its face and the Ronkonkoma moraine. This proglacial lake spanned an area from Queens to the Hamptons, and as far north as Connecticut (fig.2). A second glaciation deposited the Harbor Hill Moraine and scraped forward Cretaceous and Pleistocene sediments on the lakebed (fig. 4). The glacier recedes while englacial streams carrying sediments from within it, pour into the lake. These deposits cover the Paleozoic bedrock and Cretaceous sediments below (fig. 5) forming characteristic glaciolacustrine features.


Gilbert deltas form where glacial streams meet the deep proglacial lake. They have horizontal topset beds and south-dipping foreset beds. In front of the foreset beds are shallow horizontal bottomset beds (Gilbert, 1885). The cliffs at Caumsett are the remains of a Gilbert delta (figs. 3, fig. 6).


There is evidence that this glacier advanced overrunning the delta and then stagnated, forming the hummocky terrain just south of the bluffs (Fig. 7 fig. 8). The stratigraphy of the cliffs is consistent with this model.


Initial visits to the site developed an overall impression of cliff development and a selection of the particular areas of the bluffs to be studied. Numerous photographs were taken with a Pentax PZ10 35mm camera with an 80mm zoom lens.


Photographs were developed by Fujifilm and placed on 3-1/2 inch floppy discs. A photomosaic of the cliffs was produced using the 4x6 photos. The discs enabled me enlarge and clarify several images which were also composed into a photo-mosaic.


Photo-discs are made from prints and may be ordered at the time of developing or they may be made subsequent to developing. I have found that the former is much preferred. The quality of the images was much better and the formatting of the disc was much easier to work with.


Videos were taken with a Panasonic Palmcorder with a 1-100x digital zoom. The digital zoom was clear only to 20x.


One hundred-foot ropes marked off in 1m sections were dropped from the top of the cliffs in several areas to enable a better estimate of the size of sedimentary layers. Areas that could be reached by foot were measured with a meter stick and Jacob’s staff. Binoculars (8x) and a small telescope were also used to get a better look at the strata from the beach.


Literary materials and internet sources discussing glaciers and glaciolacustrine environments were reviewed.


Several site visits and assistance by Professor Gilbert Hansen were invaluable in developing an understanding of the area.


A parking permit was necessary to be able to drive down to the Fisherman’s parking lot.

Visits to the site were limited to park hours at the Fisherman’s lot and to regulations prohibiting climbing on and digging into the cliffs.


Approximately 1mile of the cliff face at Caumsett State Park was studied from May through August 1999. Three sections were chosen for stratigraphic columns (fig. 12). They were chosen for their distinctly different geology (fig. 14). The westernmost section (Column 1), displays the large foreset beds of a classic Gilbert Delta (Gilbert, 1885)(fig. 11, Plate 1). The intermediate column (Column 2) is located east of Inspiration point and does not have the prominent yellow gravel beds. This area is darker with more lacustrine deposits (fig.12, Plate 2). A spectacular Cretaceous outcrop overlain by Pleistocene sands and gravel dominates Column 3, near the easternmost section (fig13, Plate 3). 1-2 meters of till and a layer of loess as much as 1meter thick overlie all sections.

Cretaceous Deposits

Cretaceous sediments overlie the Paleozoic bedrock of Long Island. These layers would have been deposited in horizontal beds in quiet waters. Seasonal changes would account for the alternating layers of cross-bedded clay and sand. These sediments form the aquifers from which we derive our fresh water.


Bright white and pink fine grained, cross-bedded sands and clay characteristic of Cretaceous deposits was found at Columns 1, 2 and 3. They reach from the beach to elevations of 18 meters at Column 3. At Columns 1 and 2 Cretaceous deposits are found at elevations of 13and 14˝ meters respectively. There are 6-8 meters of overburden covering the lower elevations of these columns. However, Cretaceous sand and clay can be found not far behind this. The inference must be made that it rests along the length of the bluffs beneath the Pleistocene deposits.


On close inspection the Cretaceous layers contain fine-grained (.01cm) white sand with thin (1cm) layers of bright red, yellow or purple sand and clay. There are areas where the white sand unit is as much as 0.5-1 m thick. Most of the gravel is made up of quartz pebbles (0.5-3cm) and occurs in thin (3-4cm) layers. Thin (2-3cm) layers of clay are laminated within the sand. Clay units are gray and reddish brown. Clay lenses were noted in Columns 2 and 3.


This entire unit shows great distortion indicating glaciotectonic forces acting upon it subsequent to its deposition (Plate 4). The contorted layers show an initial rise and then dip to the SSW. A fault was seen at Column 1 with a 3cm offset.

A thin (5-6cm) blue-black layer of partially lithified sediment is found at the top of the Cretaceous unit (Plate 5). Perhaps dissolved minerals rich in iron compounds have collected here. Just below this unit In Column 3, there are very bright yellow sands interwoven with the red, white, and purple sand and red and clay.

The greater height of the Cretaceous deposits at Column 3 may be the result of differences in the surface at the glacial bed. Or, there was less resistance to glacial forces or less compression with the absence of the delta in this area.

Pleistocene Deposits

Pleistocene deposits are found above the Cretaceous along the length of the bluffs. They contain a wide range of clastic sediment types, from coarse gravels to fine clays to cobbly and pebbly diamicts.

The Pleistocene sediments begin with a clay unit 25-30cm thick (Plate 6). This brown clay layer has 1mm laminations. Above here are thin (2-4cm) layers of fine brown and tan sand laminated with white sand and brown and gray clay. Both the clay and sands show evidence of glaciotectonic disturbance with much folding and faulting with offsets of 3-4 cm.

In Column 1 this unit is 2 ˝ meters thick. Resting above are thin beds (1-2 cm) of fining upward tan and yellowish gravel and sand. Thick unsorted beds of yellowish gravel, cobbles, and sand lie in a 10-meter unit above. The gravel in both these units is mostly quartz, but substantial numbers of highly weathered metamorphic as well as basalt and conglomerate gravel and cobbles are present. This gravel may be an early Pleistocene deposit known as Manetto gravel (Fuller, 1914). The gravel beds dip to the west at 25o and to the south at 10o. They are most likely the foreset beds of a Gilbert-type delta (Ashley, Menzies, 1995, Gilbert, 1885, Merguerian, Sanders, 1996, Sirkin, 1996) (fig. 6, Plate 7).

There is a nearly horizontal layer atop the dipping gravel layers. This may be the topset layer of the delta. The delta beds are not found east of here.

In Column 2, above the fine sand and clay sitting atop the Cretaceous, are 3 meters of brown sand and quartz gravel. Next, 2 ˝ meters of very disturbed brown and tan sand topped with 1 ˝ meters of reddish brown gravel, sand, and clay. A 1 meter unit of very disturbed brown and tan laminated clay follows. Above the clay ˝ meter of brown and tan sand is laminated in 2-3 cm layers. Here again there is evidence of much glaciotectonic disturbance. Small scale folding and faulting, with offsets of 3-4cm is noted. The next 2 meter unit of large beds of unsorted sand and gravel shows no distortion.

Above the brown sand and clay that mark the Cretaceous/Pleistocene contact at Column 3 is 2 meters of sand and gravel with 1meter of cross-bedded sand and clay above. A large (3meter) unit of gravel and sand fining up follows.

Above the delta in Column 1 and the large gravel beds in column 2, is a 1m layer of large cobbles, small boulders, sand, and clay that may be a layer of till. Next we find more sand and gravel, then another larger 2m layer of till which includes much larger boulders.

In Column 3 the lower till layer is missing and there is a ˝-meter thick horizontal bed of sand and clay beneath the large till layer that is also found in column 1.


A layer of loess overlies the till. High winds were generated by air pressure differences where icy glacial air meets warmer temperature near the front. Loess is windblown silt to fine sand with a characteristic yellow color (fig.9). Its fine particles trap much water, making it good environments for rooting plants. It is very thin at the western end of the bluff and thickens to 1 m towards the east. Roots from the soil layer above can be seen intruding the loess in most areas.


The bluffs are currently being eroded at a rate of about 3 ft/year. The soil layer is being undermined roots are exposed and trees have fallen. Slumping and overburden obscures lower layers in most areas.


Park regulations prohibit climbing on or digging into the cliffs. To get a better look at the cliff behind the slump it would be best to come back after a powerful storm. The accompanying winds, rain, and tides may wash away the overburden.


Erratics along the beach have fallen there from the eroding till layers above. They consist mostly of various granites and gneisses. Many of the boulders are augen gneisses. Many are pegmatites with impressive feldspar crystals. Numerous examples of gray Harrison gneiss and many smaller cobble sized conglomerates can be found as well.


There are several basalt boulders, some with quartz veining. One dark basalt erratic has granite attached to it. This erratic is a remnant of the beginning of the breakup of Pangea 200million years ago.


The metamorphic rocks exhibit varying degrees of distortion. Some of the erratics are rounded indicating a long and/or rough glacial journey. More jagged edges speak of a gentler journey. Some are highly weathered and rotted.


The Pleistocene sediments in this sequence were deposited in proglacial environments. Some subsequently underwent glaciotectonic deformation. Looking at the Pleistocene sequence, a glacier north of the cliff face was impounding a glacial lake or pond in which laminated clays were deposited. Sub-glacial and englacial streams formed alluvial deposits as well as the delta deposits found in the western end of the cliffs. At the time, the delta would have been located some distance north of its present location. As the glacier advanced it shoved forward the underlying Cretaceous and the Pleistocene sediments. Interlayering and faulting of the sediments occurred as a result. The glacier then advanced over these sediments depositing the thin till layer. As the glacier retreated lake sediments are once again deposited, this time on top of the till. A subsequent advance moves the glacier over the area at which time it stagnates leaving the larger layer of till (from which most of the erratics found along the beach have fallen), and forms the hummocky area found south of the cliffs (figs. 7&8). The area was then covered with a thin layer of wind blown silt or loess.


Walking west to east along the base of the bluffs, evidence for glacial tectonic deformation is clear. The abrupt changes in stratigraphy near Column 2 may represent a large-scale fault with the Cretaceous and Pleistocene lake sediments being uplifted on the eastern side. The lack of Gilbert delta deposits to the east may reflect deformation or different depositional glacial environments along the glacial face.


Due to tectonic disturbance it is not clear when the Pleistocene sequences were deposited. There is speculation as to whether the lake sequences are part of the Smithtown Clay, a large tract of lake clays underlying the Smithtown region of Long Island east of the Manetto Hills Interlobate Zone and north of the Ronkonkoma Moraine (Sirkin, 1995) or whether they are of separate origin. 




  • Boulton, G.S. (1970a) On the deposition of subglacial and melt-out tills at the margins of certain Svalbard glaciers. Journal of Glaciology, 9, pp 231-245.

  • Chamberlin, T. C. (1883) Preliminary paper on the terminal moraine of the second glacial epoch, p-291-482 in U. S. Geological Survey Annual Report, 3rd (1881-1882), 564pp.

  • Clalyton, L. and Cherry, J. A. (1967) Pleistocene super-glacial and ice-walled lakes of west-central North America.  North Dakota Geological Survey, Miscellaneous Series 30, pp. 47-52.

  • Clemmensen, L.B. and Houmark-Nielsen, M. (1981) Sedimentary features of a Weichselian Glaciolacustrine delta. Boreas, 10, pp. 229-2

  • Cohen, J.M. (1979) Deltaic sedimentation in glacial Lake Blessington, County Wicklow, Ireland.  In:Schlichter, Ch. (ed), Moraines and Varves:Origin/Genesis/Classification,.  A. A. Balkema, Rotterdam. pp. 357-367

  • Dreimanis, A. (1988) Tills:  Their genetic terminology and classification:  In:  Goldthwait, R.P. and Matsch, C.L. (eds), Genetic Classification of Glaciogenic Deposits. A.A. Balkema, Rotterdam. , pp17-83

  • “Earth Science Reference Tables” (1994) Albany: New York State Education Department, 16 pp.

  • Erickson, Jon. (1996) Glacial Geology:  How Ice Shapes the Land.  Facts On File, Inc. NYC, NY 248 pp.

  • Fuller, Myron L. (1914) The Geology of Long Island New York. Professional Paper 82. 231pp

  • G. M. Ashley - Menzies (1995) ch 13, Glaciolacustrine Environments, pp 417-444

  • Gilbert, G. K. (1885) The topographic features of lake shores, p. 69-123 in U. S. Geological Survey, Annual Report, 5th (1883-1884), 469pp.

  • Gustavson, T. S. (1975b) Sedimentation and physical limnology in proglacial Malaspina Lake, southeastern Alaska. In:  Jopling, A. V. and McDonald, B. C. (eds), Glaciofluvial and glaciolacustrine sedimentation. Society of Economic Paleontologists and Mineralogists Special Publication No. 23, pp. 249-263

  • Hambreu, Michael, Alean, Jurg (1992) Glaciers,  Press Syndicate of the University of Cambridge, Cambridge and New York, 208pp

  • Hart, J.K. and Boulton, G.S. (1991) The interrelation of Glaciotectonic and Glaciodepositional processes within the glacial environment.  Quaternary Science Reviews, 10, pp. 335-350.

  • Irwin, J. and Pickrill, R.A. (1982) Water temperature and turbidity in a glacial fed lake:  Lake Tekapo. New Zealand Journal of Freshwater Research, 16, pp.189-200.

  • Jopling, A.V. and Walker, R.G. (1968) Morphology and origin of ripple-drift cross-lamination, with examples from the Pleistocene of Massachusetts.  Journal of Sedimentary Petrology, 38, pp. 971-984

  • Kelly, R.I., I.P. Martini. (1986) Pleistocene glaciolacustrine deltaic deposits of the Scarborough Formation, Ontario, Canada.  Sedimentary Geology, 47, pp. 27-52

  • Martini, I.P. (1990) Pleistocene glacial fan deltas in southern Ontario, Canada. In:  Colella, A. and Prior, D.B. (eds), Coarse-grained Deltas.  International Association of Sedimentologists, Special Publication 10, pp. 281-295

  • Mather, W. W., (1843) Geology of New York. Part I. Comprising (sic) the geology of the First Geological District:  Albany, Carroll & Cook, Printers to the Assembly, 653p., 46pl.

  • Menzies, J. (ed.) (1995) Modern Glacial Environments - Processes, Dynamics and Sediments.  Oxford:  Butterworth-Heineman.

  • Menzies, John, ed. (1996) Past Glacial Environments (Sediments, Forms and Techniques); Butterworh-Heinemann Ltd.; Linacre House, Jordan Hill, Oxford pp. 598

  • Merguerian, C., Sanders, J. (June 1996) Trip NO. 39. Glacial Geology of Long Island 01 & 02, pp132. 

  • Mickelson, D. M., Clayton, L. Fullerton, D. S. and Borns, H. W. Jr. (1983) The late Wisconsin glacial record of the Laurentide ice sheet in the United States.  In The Late Pleistocene.  Late Quaternary Environments of the United States, vol. I (S. C. Porter, ed.), Minneapolis:  University of Minnesota. pp. 3-37

  • Nemec, W. (1990) Aspects of sediment movement on steep delta slopes. In:  Colella, A. and Prior, D.B. (eds), Coarse-Grained Deltas.  International Association of Sedimentologists, Special Publication 10, pp. 29-73.

  • Shaw, J. (1985) Subglacial and ice marginal environments. In: Ashley, G.M., Shaw, J. and Smith, N.D. (eds), Glacial Sedimentary Environments,. Society of Economic Palaeontologists and Mineralogists, Short Course 16, Tulsa, OK. pp. 7-84

  • Sirkin, Les, (1991) Stratigraphy of the Long Island Platform. Journal of Coastal Research, SI #10,. Fort Lauderdale (Florida). ISSN 0749-0208, pp. 217-222

  • Sirkin, Les (1995) Eastern Long Island Geology with field trips. Book and Tackle Shop, Watch Hill, RI 02891,  220pp

  • Sirkin, Les, (1996) Western Long Island Geology with field trips published by Book and Tackle Shop, Watch Hill, RI 02891 179pp.

  • Teller, J. and Clayton, L. (1983). Glacial Lake Agassiz.  Geological Association of Canada Special Paper, 26, 451pp.

  • Thomas, G.S.P. (1984a) A late Devensian Glaciolacustrine fan-delta at Rhosesmor, Clwd, North Wales.  Geological Journal, 19, pp. 125-141.

  • Wateren, D.F.M. van der (1981) A model of glacial tectonics, applied to the ice-pushed ridges in the Central Netherlands.  Bulletin of the Geological Society of Denmark, 34. pp. 55-74