DEEP: Geology of Mansfield Hollow State Park

The Geology of Mansfield Hollow State Park
Trail Map
Rock Types
  • Igneous
    • None 
  • Metamorphic
    • Gneiss 
  • Sedimentary
    • None
  {Photograph of sign in Mansfield Hollow State Park}
Rock Units
  • Hope Valley Alaskite Gneiss: medium grained light-gray to cream colored granitic gneiss composed of potassium-feldspar and mica with minor quartz. Tatnic Hill Formation: gray and dark gray gneiss and schist.
Minerals of Interest
  • No outstanding mineral deposits
Interesting Geologic Features
  • Stratified sand and gravel deposits
    • Kettles
    • Eskers
  • Peat deposits

INTRODUCTION. Outdoor recreational enthusiasts come to Mansfield Hollow State Park during all four seasons for activities that range from picnicking, fishing and ice fishing, boating, hiking, and skiing (cross-country). It is also a good place to appreciate the processes of inferential reasoning and geological interpretation. The park can be accessed by numerous trails from several different parking areas. The trail map (see above) will be useful and we recommend that you print it out before starting your exploration of the park.

The Mansfield Hollow reservoir was created by construction of a flood-control dam by the Army Corps of Engineers in the early 1950’s. Water backed up behind the dam and flooded "Turnip Meadows", a swamp located at the confluence of the Fenton, Mount Hope, and Natchaug rivers. The reservoir so created is referred to both as the Mansfield Hollow Lake and Lake Naubesatuck. When the flood gates restrict drainage of the Natchaug River, water may back up a couple of miles into the three valleys. Normal reservoir level is maintained at an elevation of about 210 feet above sea level. The spillway elevation of the dam is 47 feet higher. Joseph Leary (2004, p. 118-119) provides a nice description of the park and its facilities.

From Hartford: take Route 84 east, to Exit 68. Follow Route 195 south about 10 miles to Bassett Bridge Road. (You will cross Route 44, go through the UCONN campus and past the junction of Route 89 before coming to Bassett Bridge Road. Bassett Bridge Road is the first traffic signal about half a mile past Route 89.) Turn left onto Bassett Bridge Road. The Park is about a mile down the road on the left side of Bassett Bridge Road.

From Willimantic: Route 195 north to Bassett Bridge Road. Take a right onto Bassett Bridge Road. Park is on your left, on Bassett Bridge Road.


Introduction. The geologic story of Mansfield Hollow State Park is mainly about sediments that were deposited at the end of the last Ice Age (about 16,500 years ago) when the glaciers were melting. For the most part, these younger unconsolidated deposits cover the local bedrock.

Bedrock is exposed in two different locations within the park. The largest exposure is at the base of the dam and just southward. There, Hope Valley Alaskite Gneiss (pronounced "nice") crops out within the bed and along the banks of the Natchaug River (Figure 1).

{Outcrops of alaskite gneiss} {Waterfall just below the Mansfield Holllow Dam Spillway} {Rapids downstream of the waterfall.}
  A. B. C.

Figure 1. Outcrops of alaskite gneiss on Natchaug River below Mansfield Hollow Dam. A. Light gray phase of gneiss. Foliation does not show well on this surface. Location of picture just to left of waterfall shown in center photo. B. Waterfall over an old mill dam just below the Mansfield Hollow Dam spillway, which is seen in background. The mill is located just north (to the left) and downstream from this waterfall. A water race goes from upstream of the waterfall toward the mill. C. Rapids downstream of the waterfall. Outcrops on far shore are composed of cream-colored alaskite gneiss.

The river forms a fall and rapids where it crosses the outcrop. Extreme caution is advised when visiting this area because a misstep could be tragic. Alaskite2 gneiss is composed of mineral grains of potassium feldspar, quartz and minor amounts of dark minerals. It has a granitic look to it. The rocks that crop out are light gray to pinkish colored and are clearly foliated. Outcrops of Tatnic Hill gray gneiss are found along the Mount Hope River near the Atwoodville Bridge. There the river flows rapidly and a large pothole-like pool is found at a bend in the river just upstream of the bridge.

2 Alaskite is similar to granite. Alaskite is composed of potassium-feldspar, quartz and mica. Granite contains potassium-feldspar and sodium-feldspar, quartz and mica.

20,000 years ago Connecticut was in the grips of the last Ice Age. Glacial ice about a mile thick covered the state; indeed, it covered most of the northern third of North America. The glaciers pushed as far south as Long Island, Block Island, and the Islands south of Cape Cod. Glacial ice is a powerful agent of erosion. Glaciers create enormous amounts of sand, gravel and mud that is left behind as glacial soil (till) at the end of an ice age.

When all the ice begins to melt, melt-water accumulates on top of the glacier in streams that flow down hill. Many may flow on top of the ice to the edge of the glacier. The streams then flow away from the glacier in coalescing channels that may form a broad plain of "out-wash". Many streams on top of the glacier get swallowed into cracks and crevasses and flow in tunnels beneath the ice to an edge. There they may join other streams on a large out-wash plain. Outwash sediments, deposited by melt-water streams, are composed of rounded rocks (river-rocks) and sand grains that are deposited in beds (i.e. they are stratified). The mud usually gets carried to the sea or is deposited in nearby ponds.

Melt-water stream deposits.

Notice the shape of the land when you turn east onto Bassett Bridge Road. In general it is a flat plain that has a generally hummocky surface (containing small hills and depressions; see Figure 2, A and B).

{Bassett Bridge Road looking northwest} {Bassett Bridge Road looking southeast} {Stratified sand and gravel deposit}
A B. C.

Figure 2. Pictures showing plain-like surface at Bassett Bridge road. A. Along Bassett Bridge Road looking north-west: B. looking southeast. Note slight irregularities of topography. C. Stratified sand and gravel deposit in a working gravel-pit just south of Bassett Bridge Road, confirming that plain is composed of river-deposits.

{Topographic map }

Figure 3. Topographic map showing part of the plain over which Bassett Bridge Road travels. The parking area is near the area on the map designated "Picnic Area". The plain top (far west of map) has an elevation in excess of 270 feet. Notice two small ponds filling the bottoms of depressions (holes) just to east of Picnic Area. Contour interval 10 feet.

The map, Figure 3, shows that the plain surface has an elevation about 275+ feet. This is about 65 feet above the normal reservoir level and up to 85 feet higher than the elevation of the river channels that were flooded by the impoundment. As you approach the park you will notice several depressions on the surface. Some have a small pond in their bottom; they are deep enough to intersect the water table. If you look at the map you will notice there are several "holes’ with small ponds on the plain surface. We will talk about the depressions later in this discussion.

What underlies the plain? Stratified sand and gravel are being removed from a gravel pit just south of Bassett Bridge Road (Figure 2C.). The grains of gravel are all rounded like river-rock. Some of the sand and gravel layers contain cross-beds indicating deposition from flowing water. Mud layers are lacking. It seems clear that the Bassett Bridge area was formed by deposits of sand and gravel from some ancient stream or river that flowed across this surface. The stream was likely filled with glacial melt-water. Glacial melt-water streams flowing at the end of the last ice age carried an abundant sediment load and could deposit much of the sand and gravel that they carried. The problem with this interpretation is why the melt-water streams would flow at an elevation 65-85 feet higher than the modern stream valley. We can offer two interpretations: (1) Rahn (1971) inferred that the melt-water streams did not flow along the bottom of the modern valley because left over masses of ice partially filled the valley bottom.

Thus, sand and gravel were deposited next to the left over ice that filled the middle of the valley. Such deposits are called ice-contact deposits and, in the older literature, kames and kame terraces. (2). Alternatively, sand and gravel deposits may have completely filled the valley, but the current valley bottom was eroded into the gravel deposits by slightly younger melt-water streams (torrents) as the ice melted northward during warming climates about 15,000 years ago.  Such sand and gravel deposits are found in many Connecticut valleys. Many are mined as a source of sand and gravel for construction. The surface of these deposits is usually 20-60 or more feet above the valley floor. Sand and gravel terraces line the sides of the Fenton, Mount Hope and Natchaug River valleys. All have a rather hummocky surface that may be pitted with kettles (see below).


Along the red trail close to the picnic area (about 50 feet east of where the trail makes a sharp bent to the south) or along the blue trail ("The Kettle Hole Trail") at station 7 you will be able to look into a large depression, a 50-60’ deep hole in the ground (Figure 4A, B). Both have steep sides (about a 30-40 o slope) and both have water in their bottom. Perhaps you can see on the map other depressions nearby (some will have water in their bottom if they are deep enough to intersect the water table). They are holes that exist on the plain of river-deposited sand and gravel. We can confirm this by looking around the edges and noticing that erosion has occurred in low areas where flood waters have gained access to the depressions. In the past when the flood gates of the dam have been closed water has backed up sufficiently to flow into the depressions and has eroded the adjacent low area.

{Looking into depression north of red trail adjacent to picnic area.} {Depression at station 7 along the Kettle Hole Trail} {Eroded area where flood waters entered depression.}
A B. C.

Figure 4. A. Looking into depression north of red trail adjacent to picnic area. Depression is about 60 feet deep and has a small pond and swamp in its base. This depression may be hard to find when leaves are on the trees. B. Depression at station 7 along the "Kettle Hole Trail (blue blazed trail) just south of boat launch parking area. C. Eroded area where flood waters entered the depression adjacent to boat-launch parking. Shiny-black instrument is 6" long GPS receiver. Notice rounded ("river rock") gravel-sized grains.

This allows us to see that also, the land here is underlain by stratified sand and gravel composed of rounded grains (Figure 4C). This helps confirm our interpretation that the area is underlain by river deposits. But, we need to answer another question: How were the holes formed?" Modern rivers can erode pools that are usually elongate and sinuous. They usually fill in with sediments shortly after they form (rivers are dynamic and continually changing). They are only a fraction as deep as they are wide. The holes we see are roughly circular and have a much greater depth to width ratio than pools in a modern river. They were not formed by river erosion!

If a large block(s) of left-over glacial ice remained in the valley bottom and along the sides of the valley, the melt-water streams would flow around the left-over blocks and deposit their load of sand and gravel surrounding the ice block, perhaps even covering it. Later when the ice melted, the sand and gravel would collapse into the void creating the hole. Interestingly, when gravel collapses, or is piled up, it naturally assumes a 30-40 o slope. Such depressions are called "kettles". There are several kettles on the sand and gravel plain in the area.

Esker site.

The last feature we will consider is located about two miles to the south of the boat launch area. You can follow the blue-dot trail across Bassett Bridge and hike about 2 miles southeast to the feature. Alternatively, you may get back in your car and drive (left out of parking area) on Bassett Bridge Road about 2 miles to North Windham Road. Go right on North Windham Road just before the bridge across the Natchaug River. A parking area is at the end of the road (about a quarter mile down the road). This is in a permitted hunting area and appropriate high-visibility clothing (orange) should be worn during the hunting seasons. There follow the abandoned continuation of North Windham road north to the site until you are on or next to a low ridge, perhaps 30 feet high and several hundreds of feet long (Figure 5).

{Eskers along abandoned portion of North Windham Road} {Walkers nearing cut into side of esker} {Looking at side of esker norht of road}
A B. C.

Figure 5. Eskers along abandoned portion of North Windham Road. A. Road follows top of ridge. Rounded boulders litter sides of ridge (see Fig. 6A). B. Walkers nearing cut into side of esker. Figure 6B shows that cut. C. Looking at side of esker north of road.

About a quarter of a mile north, the side of the ridge was cut into to form the road (Figure 5B). Here we can see that the ridge is composed of sand and gravel (Figure 6B). You can tell by the rounded cobbles and boulders (i.e., river rocks) that the ridge is made of river deposits. The largest clasts3 are boulders foot or two in diameter. They must have been deposited by a high velocity stream. This landform is called an esker. The melt-water stream that deposited the gravel did not flow on top of the ridge. Ice must have constrained the stream to flow in a long narrow space. Today we know that melt-water streams can flow on top of glaciers, in cracks and crevasses within a glacier, or within a tunnel beneath a glacier. The size of the largest boulders suggest that the velocity of the stream was forced by hydraulic pressure in a tunnel beneath rather thick glacial ice rather than gravity in an open crack or channel on top of the ice. The ice must have been thicker and filled much more of the valley when the esker formed than when the kame plain and kettles formed. However, at the very top of the esker sharp sided, angular boulders are found (Figure 6B). They probably were not transported by water (otherwise they would have been rounded). After stream-flow in the tunnel ceased, the tunnel collapsed, allowing angular blocks of rock on the surface of the ice to drop onto the esker.

3A clast is a broken rock fragment, no matter how large or small. The term refers to grains boulders that are freshly broken from an outcrop as well as sand grains that are broken up rock fragments.

{Rounded clasts on surface of esker} {Road cut into side of esker exposing sand and gravel}
A B.

Figure 6. A. Rounded clasts on surface of esker. This one is about 2 feet in diameter. B. Road cut into side of esker exposing sand and gravel. Note most clasts are cobble sized and rounded. Several large angular fragments are found neat the top of the esker. They probably dropped into a collapsed ice tunnel and onto the esker surface after the deposition of the esker was completed.

A question for you to ponder is: did the esker form at the same time, before or after the sand and gravel. Did the kettle form at the same time, before or after the esker? To answer the question try to infer how much ice was left in the valley when each feature was formed. When the esker formed the ice must have been thick enough for a tunnel that connected to a surface several tens to hundreds of feet above it. That height provided water pressure to force the large boulders through the tunnel and deposit them within the esker. When the sand and gravel formed the ice had shrunk to at most the size of the modern lake and only a few tens of feet higher than the top of the gravel deposit. That is much thinner than the ice that was when the esker formed. Finally the kettle formed when buried ice melted. But it had to have melted after the ice in the center of the valley melted allowing the streams to flow on the valley floor. That is because the depression did not get filled in as soon as it formed. With streams still flowing across the gravel plain, incipient kettles would be filled as fast as they formed. It is likely that there were several kettles that formed before enough ice melted in the valley center to allow the melt-water streams to flow at that lower elevation. In summary, the esker formed first, the sand and gravel was deposited later and the kettles formed last.

Finally it should be noted that the post glacial landscape has not been greatly altered by natural modern day processes.

A geological map, showing the various glacial deposits in part of Mansfield Hollow State Park, is shown below as Figure 7. It was compiled by Stone and others, 2005.

{Map showing surface (glacial) deposits}

Figure 7. Map showing surface (glacial) deposits surrounding the south end of the Mansfield Hollow Reservoir (after Stone et al, 2005).


Leary, Joseph, 2004, A Shared Landscape: A Guide and History of Connecticut’s State Parks and Forests. Friends of CT State Parks, Inc, 240p.

Rahn, P.H., 1971, The surficial Geology of the Spring Hill Quadrangle with map. State
Geol. Nat. Hist. Surv. of Connecticut Quad. Rpt. 26, 33p.

Stone, J.R., Schafer, J.P., London, E.H., DiGiacomo-Cohen, M., Lewis, R.S., and Thompson, W.D., 2005, Quaternary Geologic Map of Connecticut and Long Island Sound Basin, U. S. Geol. Survey, Sci. Inv. Map 2784, 2 sheets. Also see U.S. Geological Survey Open File Rept. 98-371.

Virtual field trips:

Hyatt, J. A. and others, 2002, Accessing Connecticut Landscapes On-Line. Connecticut Teaching and Learning Enhanced by Technology Program,

Esker in Mansfield hollow

Kettles in Mansfield hollow

Mansfield Hollow State Park  |  Geology of Connecticut State Parks
{Link to Top of Page}