The Pleistocene Ice Age

The central Indiana landscape is primarily a product of the Pleistocene Epoch of the Quaternary Period, or Ice Age, a period of widespread continental glaciation in which the temperate northern latitudes were repeatedly invaded by large ice sheets. The Pleistocene began about 2.6 million years ago and was characterized by a cooler, wetter, and presumably cloudier climate than today. These conditions led initially to the development of ice caps in the vicinity of Hudson Bay and Labrador. Over tens or hundreds of thousands of years the ice caps gradually expanded, eventually coalescing into the massive Laurentide Ice Sheet that flowed southward into the temperate latitudes of North America. During the course of the Pleistocene, the glaciers carved out the basins of the Great Lakes, which helped direct the flow into Indiana from three principal directions: (1) the northwest (Lake Michigan Lobe); (2) the north-northeast (Saginaw Lobe); and (3) the east-northeast (Huron-Erie Lobe). The deposits of each lobe are characterized by a distinctive suite of rocks and minerals derived from the particular bedrock each lobe flowed over. As far as is known, all of the glacial deposits in Marion County are from the Huron-Erie and Saginaw Lobes.

The Ice Age was punctuated by several prolonged warm periods during which the glaciers disappeared entirely from the temperate latitudes and a climate similar to today or even warmer prevailed. The interglacial landscape was modified by typical terrestrial processes, such as erosion by streams to produce hills and valleys, and deep soil development; it also supported abundant vegetation, including hardwood forests and wetlands similar to those present today. These warmer periods are known as interglacial stages (fig. 1), and they separate several major stages of glacial advance that are recorded by the Pleistocene deposits in the Midwest.

Indirect geophysical evidence suggests that the core of the Laurentide Ice Sheet in the vicinity of southern Canada and the Great Lakes was up to 2 miles (3.2 km) thick, similar to the modern Greenland ice cap. The thickness of the glacial lobes that affected central Indiana is less clear, but was probably several thousand feet at times, with progressively thinner ice toward the margin of the glacier. In general, deposition occurs near the margin of glaciers, whereas erosion is the dominant process further back up-ice — especially beneath the cores of large ice sheets, where the substrate beneath the glacier may be severely scoured, completely removing any earlier glacial deposits that may have been present, and deeply eroding the underlying bedrock. Topographic characteristics of the substrate also greatly affect where erosion and deposition occur, with hills and other obstructions often experiencing intense erosion, while valleys and lowlands typically are sites of deposition where earlier deposits tend to be well preserved (see How Glaciers Work).

Unconsolidated Deposits

The modern landscape of Marion County, and the unconsolidated deposits within about 30 to 50 ft (9 to 15 m) of the surface, are chiefly the product of the most recent stage of glaciation, called the late Wisconsin, which affected the county between about 22,000 and 17,000 radiocarbon years B.P. At many places, however, the late Wisconsin deposits comprise a relatively thin veneer that mantles a thick series of deposits from earlier ice advances, the oldest of which may approach one million years in age. Marion County lay near the southern terminus of ice sheets throughout the Pleistocene, a position that helped protect older deposits from erosion during younger ice advances and contributed to the preservation of a fairly robust, though complex and locally incomplete, record of glacial events (fig. 2). The total thickness of unconsolidated deposits in the county is commonly between 100 and 200 ft (30.5 and 61 m), and locally exceeds 300 ft (91 m) in Lawrence and Franklin Townships, where a thick sequence of ancient pre-Illinoian deposits is preserved in a series of deep bedrock valleys and lowlands. In contrast, the glacial deposits are extremely thin in parts of Decatur Township, where large bedrock hills associated with the buried northern extension of the Knobstone Escarpment obstructed ice flow and stand within a few feet of the modern land surface.

Figure 2.
Simplified west-to-east geologic cross section illustrating the relationship of the topography of the bedrock surface and the modern land surface to the total thickness of unconsolidated sediments. The diagram also illustrates the complicated cross-cutting relations between several major and minor paleosurfaces (ancient landscapes typically represented by weathering horizons and changes in geologic properties) that bound the different glacial sequences beneath the county. (Adapted from cross sections in Brown and Laudick, 2003).

These examples demonstrate the close relationship between the thickness of glacial deposits and the underlying bedrock topography: glacial deposits are almost invariably thicker and better preserved in low areas on the bedrock surface, such as buried valleys, whereas they are usually much thinner over bedrock highs, such as that in the southwestern part of the county (fig. 2). Certain kinds of glacial deposits, such as end moraines and kames, are associated with elevated, irregular topography. Crown Hill (fig. 3) and the large ridges at Glenns Valley are good examples, where glacial action has produced conspicuous topographic high points underlain by very thick glacial deposits. On the other hand, erosion by streams both during and after glaciation has produced low-lying valleys, below which the glacial deposits are generally thinner than beneath adjacent uplands.

Pleistocene History and Glacial Terrains of Marion County

The landscape of Marion County is made up of a series of glacial terrains, each characterized by a specific set of landforms and underlying sedimentary sequences that reflect a particular geologic history and set of depositional (or erosional) processes local to that region of the landscape. These terrains are primarily the result of late Wisconsin glaciers active in the county from about 22,000 to 17,000 radiocarbon years B.P., and the meltwater they produced (fig. 4). Late Wisconsin sequences average about 50 to 75 ft (1.5 to 22.9 m) thick in the county, but are locally much thicker or thinner, depending on the type of terrain they form and the relief on underlying deposits and bedrock. These deposits are widely exposed at the surface, cropping out in bluffs, along stream banks, and in many excavations almost anywhere in the county, thus they are readily observed and relatively well known. Among other things, they form the parent materials for the surface soil, act as the foundation for most infrastructure, and contain major groundwater resources, hence their character is of immediate relevance to everyday life. Therefore, most of this section focuses on the late Wisconsin glacial history and depositional sequences of the county.

At many places, however, the Late Wisconsin sequences make up only a fraction of the total glacial deposits present above bedrock, and are draped over thick sequences of Illinoian and pre-Illinoian age sediments that collectively are referred to as "pre-Wisconsin deposits." By virtue of their greater depth, the pre-Wisconsin deposits are poorly exposed at the modern land surface, cropping out only along the sides of several deeply entrenched valleys, such as Eagle Creek, Fall Creek, White River, and their major tributaries. Hence, most of our understanding of the pre-Wisconsin sequences comes from water wells and other boreholes that penetrate below the late Wisconsin deposits. But even though they are not typically close to the surface, the pre-Wisconsin deposits are of immense scientific and practical importance. They contain vast groundwater resources that are used at many places in the county, as well as evidence crucial to understanding the history of the Ice Age and the behavior of climate over the past million years or so.

Pre-Wisconsin History and Sequences

The oldest glacial deposits in the county (fig. 5) consist of a series of reddish-brown, clay-rich lake sediments, tills, and minor sand and gravel deposits that fill the deep bedrock valleys in the northeastern part of the county. These deposits contain clasts of Jurassic mudstone ("redbeds," which impart the distinctive color) and other rock types from the Michigan Basin, and are thus thought to have been deposited by an early version of the "Saginaw Lobe" that advanced into Indiana from the north. They closely resemble deposits found in large buried valleys north of Marion County, which were named the "West Lebanon" Member by Bleuer (1991) and mark the first incursion of Pleistocene ice into Indiana. Lake sediments of the West Lebanon Member are magnetically reversed, meaning they were deposited during the last polar reversal and are older than 780,000 years. The West Lebanon glaciation disrupted the pre-glacial drainage pattern throughout the northern half of the state by damming up major drainages and creating large glacial lakes that became filled with sediment. Its diagnostic red-brown color makes the West Lebanon Member the most readily recognizable pre-Illinoian unit in the subsurface of Marion County.

At least two other major pre-Illinoian glaciations are recognized in the deposits beneath Marion County. One, represented by a series of pinkish tills and associated sand and gravel, was deposited by ice flowing into Indiana from the northeast and probably correlates with the so-called "Hillery Till Member" of eastern Illinois (Johnson and others, 1972; Bleuer, 1991). In contrast, the top of the pre-Illinoian section consists of a thick sequence of weathered sand and gravel deposits interbedded with olive-gray sandy till. This interval is an important deep aquifer at many places in the county and is commonly recognizable by the strong interglacial weathering profile formed on its surface (fig. 6). Although evidence of weathering profiles locally occurs within and between all of the pre-Illinoian sequences, none are as pronounced or as widely recognizable as the one at the top of the section, which is marked by a reddish or olive-colored buried soil profile more than 20 ft (6.1 m) thick in some boreholes. This weathering horizon is believed to have formed mainly during the Yarmouth interglacial stage, prior to 200,000 years ago, and represents the interglacial landscape that existed just prior to the Illinoian glaciation.

Much of the pre-Wisconsin section beneath Marion County consists of a series of at least four, hard, gray-brown, loam-textured Illinoian tills deposited between 200,000 and 130,000 years ago (fig. 7). The tills are mostly similar in appearance and difficult to distinguish without detailed chemical and mineralogical analyses, which suggest that at least two of the tills were deposited by glaciers that came from the northeast, while another was deposited by ice that advanced out of Michigan. These till sheets are, however, locally separated from one another by variably eroded weathering horizons that exhibit loss of carbonate minerals, strong jointing, and olive-brown paleosols a few feet thick. Illinoian glaciers advanced as far as the Ohio River valley and northern Kentucky — further south than any other Pleistocene glaciation in Indiana — resulting in significant erosion of earlier deposits by each successive ice advance that came over Marion County. The Illinoian tills are also separated at places by thin, discontinuous lenses and some larger bodies of sand and gravel, the largest of which form extensive sheet-like bodies in the southern part of the county, where they serve as important groundwater sources.

The tendency of later ice advances to modify and erode the older deposits they advanced over has produced complex subsurface relations among the various pre-Wisconsin sequences, characterized by numerous cutouts of older sequences and paleosurfaces by younger ones. Large, outwash-filled buried valleys localized within the glacial section are fairly common, and may or may not exhibit any relationship to buried valleys or other topographic features associated with the bedrock surface. These relations, along with the limited number of surface exposure, hinder systematically sorting out the character, continuity, and history of these ancient deposits at any except the most local scale. More extensive exposures of pre-Wisconsin deposits elsewhere in the state (for example, fig. 8) offer additional clues as to their character and complex history, and serve as a useful analogue for the subsurface of Marion County. A more comprehensive treatment of the pre-Wisconsin deposits of Marion County can be found here.

Figure 8.
A complicated series of old glacial deposits underlies the pre-Wisconsin surface in this exposure along Wildcat Creek in Clinton County. Well-jointed, pink, crudely layered pre-Wisconsin till (1) is separated from overlying massive greenish-gray till (2) by a thin layer of gravel and dark silt. Both till units have strong fabrics, oriented in sharply different directions, indicating the tills came from different sources. Unit 3 is a sheared and folded, discontinuous body of sandy silt that appears to fill depressions in unit 2. It is overlain unconformably by gray, slightly weathered, well-layered diamicton (4), which is in turn overlain along a sharp erosional contact by brownish-orange, weathered till (5) with a thin layer of muddy gravel (5G) at its base. The till has a well defined fabric that is oriented in a direction different from either of the two tills (1 and 2) below it. The gravel is seeping groundwater, which gives it a dark color. The pre-Wisconsin surface (dashed line) is underlain by a unit of organic silt (6) having a radiocarbon age of more than 50,000 years. Late Wisconsin till and gravel (7) of the Trafalgar Formation are above that, and are capped by windblown silt (8) in the highest part of the outcrop. The exposure is about 26 ft (8 m) tall at the highest point on the far left. Photo by A. H. Fleming.

The Pre-Wisconsin Surface

Despite these complications, or perhaps because of them, the most readily recognizable horizon associated with the pre-Wisconsin deposits below Marion County is the paleosurface (fig. 9) that developed during the Sangamon interglacial stage between 130,000 and 22,000 years ago and was subsequently modified by late Wisconsin glaciers and their meltwaters. The pre-Wisconsin surface represents the landscape initially encountered by late Wisconsin glaciers advancing into the county, and upon which their sediments were deposited. This horizon appears to have contained a considerable amount of topographic relief (fig. 10), which exerted a major influence on the behavior of late Wisconsin ice sheets and served to focus meltwater streams and outwash deposition in valleys and other low areas along the paleosurface. Although the pre-Wisconsin surface experienced a considerable amount of erosion and valley incision at places during late Wisconsin glaciation, it nevertheless preserves a variety of features that make it a fairly recognizable horizon that can be mapped in the subsurface throughout the county. Among the most interesting of these features (fig. 9) are the presence of buried wood, organic silt deposited in interglacial wetlands, and other evidence of the terrestrial habitats that existed about 22,000 years ago, just before they were obliterated by the late Wisconsin glaciation.

Figure 10 shows a simplified map of the major physiographic features of the pre-Wisconsin surface beneath Marion County. The configuration of the surface was identified by analyzing thousands of water well records, along with hundreds of samples and gamma-ray logs collected from boreholes, and numerous exposures along streams and in deep excavations. This key buried horizon is a composite feature having as much as 300 ft (91.4 m) of relief, and it truncates many pre-Wisconsin sequences whose ages span the entire range of Illinoian and older glacial events that affected this part of the state. The respective terrain regions have experienced differing geologic histories, consequently the age of the surface varies from place to place. In general, the pattern of terrains and distribution of underlying sequences and paleosols suggest that the pre-Wisconsin surface in eastern and northwestern Marion County is, in fact, representative of the Sangamon paleosurface. In contrast, the pre-Wisconsin surface in the central part of the county near the White River valley appears to have been extensively modified by late Wisconsin meltwater, and most buried soils or other evidence of a paleosurface present along the pre-Wisconsin surface in that area are likely to be exhumed pre-Sangamon paleosurfaces. (Diagram from Brown and Laudick, 2003, fig. 5, adapted from Fleming and others, 1993.)

Late Wisconsin History and Terrains: Origin of the Modern Landscape

The late Wisconsin Laurentide Ice Sheet began to affect Marion County and surrounding areas approximately 22,000 years ago with the arrival of large volumes of meltwater that were carried down the already well-established White River valley from the ice front, which lay north of the county. Over the next 1,000 years, the ice sheet advanced into and eventually covered all of Marion County as it established a terminal position well to the south in Morgan and Johnson Counties. Radiocarbon ages from wood at the base of the Wisconsin section in northwestern Marion County indicate that glacial conditions were well established in that part of the county no later than 21,000 radiocarbon years B.P.

All of the late Wisconsin sequences in Marion County are members of the Trafalgar Formation (Wayne, 1963), which is named for the small Johnson County village where these deposits were first described. The Trafalgar Formation is the principal surface unit throughout the central till plain and was deposited by ice flowing out of the Huron-Erie Basin. It has a distinctive eastern source mineralogy, characterized by a high concentration of calcite (limestone) in the silt and sand fraction, as well as abundant fragments of Paleozoic limestone and dolostone, and dark colored shale (fig. 11), which crop out along the path of the Erie Lobe and were incorporated into the glacier as it advanced into Indiana.

The Trafalgar Formation, as currently defined, encompasses several different Huron-Erie Lobe events and till sheets, which are difficult to correlate with one another, in various parts of northern and south-central Indiana. All these eastern-source tills are strikingly similar and frequently give way to bodies of outwash and lake sediment, making it highly problematic to trace individual events or deposits across the major region affected by the lobe. Thus, interpretations regarding the sequence of events as well as the character and continuity of the deposits they left, are best made at a more local scale, such as a county, and are often based as much on differences between landscapes affected by various events as they are on sediment properties. The following discussion simply highlights the major late Wisconsin events in Marion County, with a particular emphasis on natural and historical areas in the county where the impact of these events can readily be seen in the modern landscape. Additional details about the late Wisconsin deposits and events in the county, how they are manifested and identified, and the methods used to map them, can be found here.

Beginning between 21,000 and 22,000 years ago, late Wisconsin glaciers appear to have been active in or near Marion County almost continuously for the next several thousand years, producing three major depositional sequences of somewhat different character. The initial incursion of the ice sheet produced a robust till sheet, referred to herein as "T1," which can be traced in the subsurface throughout the county, along with large outwash fans in and adjacent to the White River valley. This till sheet, and its associated outwash, overlie the pre-Wisconsin surface at most places in the county, and are distinguished from the older deposits at and below that surface by their fresh, calcareous character and gray color, which typically contrast with the brown or olive, carbonate-depleted, weathering features associated with deposits immediately below the pre-Wisconsin surface. In general, all the till deposited by late Wisconsin ice sheets has a loam texture, in which the matrix consists of roughly two parts each of sand and silt, and one part clay. Local textural variations are fairly common, however. This is particularly true near the base of the T1 sequence (fig. 12), where the till is commonly more silty or sandy at places where the glacier overrode and incorporated silt and clay from older interglacial silt units, or sand and gravel from its own outwash.

Figure 12.
Left-Compact, dark brown T1 till overlies gray pebbly sand (outwash) in this exposure of late Wisconsin deposits along Buck Creek. The lowest several inches of the till are sandier than the main mass because of incorporation of granular sediment from the underlying outwash. The till has a knife-sharp basal contact and a strong fabric produced by the alignment of flatiron-shaped stones and thin silt lenses, indicating ice flow was out of the northeast, parallel to the arrow. Knife is 3.15 inches (8 cm) long. Right-Gray T1 till makes up most of this 65-ft (20-m) tall bluff near the confluence of Fishback and Eagle Creeks. Photos by A. H. Fleming.

The T1 till sheet and its associated outwash are generally buried by younger deposits at most places, but they are well exposed in numerous bluffs along major streams (fig. 12). Excellent exposures of thick T1 till (and its subjacent outwash sheet, at some places) occur at many places along Eagle Creek and its tributaries, notably at Eagle Creek Park and Reservoir, where it can be seen overlying variously eroded weathering horizons along the pre-Wisconsin surface. Good exposures also occur along Buck Creek at Southeastway Park, the White River at Holliday and Marott Parks, and along Indian Creek in Lawrence.

The ice sheet appears to have withdrawn briefly to the north, before readvancing and again covering the entire county. This "T2" event also produced a relatively widespread till sheet (fig. 13) that is somewhat less robust and more irregular than the previous one, and typically has a larger number of sand and gravel bodies associated with it. Throughout the county, thin to very thick sheets of outwash locally separate the T1 and T2 sequences in the subsurface, but at some places, the two till sheets are stacked without any intervening sediments, forming a robust block of till within which it is impossible to discern any obvious break between the two events. In any case, the comparatively thick and widespread nature of these first two sequences suggests that the ice that produced them was active in the county for a substantial period of time.

The T2 event also produced two end moraines and associated outwash fans with markedly greater local relief and a rolling to hummocky character. These ridges flank the White River in the southern part of the county and appear to be related to the opening of a major reentrant in the ice sheet where the glacial White River disgorged. Prominent among these are the tall, hummocky ridges in the Glenns Valley area (fig. 14), which stand as much as 150 ft (45.7 m) above the floor of the adjacent White River valley. They are part of a feature known as the Greenwood Moraine, most of which lies just south of the county line in northern Johnson County, but is nevertheless one of the most conspicuous topographic features in and adjacent to the county. A second T2 end moraine comprises a broad, rolling upland that extends southward from Eagle Creek through Wayne and Decatur Townships, and is composed of T2 till and small- to medium-sized sand and gravel bodies. This upland is prominent in the vicinity of Indianapolis International Airport and the nearby I-70 corridor, as well as in western Decatur Township along the Hendricks County line.

Figure 14.
Left-The rolling, hummocky topography of the Greenwood moraine on display at Glenns Valley Park. This end moraine was constructed where the margin of the T2 ice became stationary for an extended period of time and the sediment delivered by the glacier accumulated in a ridge. Right — In the Glenns Valley area, the moraine has a gravelly surface, producing a well drained landscape that is somewhat dry compared to T2 till plains and supports oak-hickory forests. The summit of the ridge, just west of Bluff Road, is underlain by some 250 ft (76 m) of sand and gravel, which extends to bedrock. The abundance of gray Paleozoic carbonate pebbles in the photo is striking. Photos by A. H. Fleming.

The ridges at Glenns Valley are composed chiefly of sand and gravel, and their up-ice (northeast) sides are capped by a thin veneer of T2 till. They represent the heads of one or more large, coalescing ice-contact fans (fig. 15) formed along this part of the moraine, and were evidently deposited where one or more major meltwater conduits exited the glacier. The body of the fan(s) formerly occupied what is now the modern valley of the White River, but was subsequently eroded away by massive meltwater discharges during later events. The meltwater erosion left only a few small, streamlined mounds of sand and gravel projecting above the central floor of the valley, along with the isolated ridges near Glenns Valley to mark the heads of the fan(s). Interestingly, the river channel in this immediate area follows an unusually circuitous route that appears to have a vaguely radial symmetry about the ridges at Glenns Valley. This pattern strongly suggests that the river channel migrated to the west side of the sluiceway along the toe of the growing fan during T2 time, and has largely remained there ever since.

In contrast, the third late Wisconsin event, called "T3," appears to have resulted from a rather short-lived surge of thin ice, which occurred between 17,000 and 18,000 years ago and covered only the northern two-thirds of the county before becoming largely stagnant. This final pulse of late Wisconsin ice did not produce a persistent till sheet, but instead left a heterogeneous assemblage of generally thin, hummocky ablation sediments, composed of small silt and sand units interbedded with irregular, discontinuous, and texturally variable diamictons that were mostly the result of mud flows and other reworking of existing sediments on top of a decaying mass of ice.

Figure 16.
Left-Processes visible on the surface of a stagnant modern glacier in Alaska illustrate the reworking of freshly melted-out sediment by small streams and numerous types of mass movement, such as the debris flow in the foreground (arrow), triggered as supporting ice melts from around the sediment. Sediment ultimately accumulates in holes and depressions on the ice surface, leaving a hummocky deposit of texturally variable ablation sediment. Person (dark) in left center of photo for scale. Right — The hummocky T3 landscape, seen north of Southeastway Park, is envisioned to have formed in the same manner, as a thin T3 ice sheet stagnated and decayed in place. Photos by A. H. Fleming.

The most prominent features associated with this event are the hummocky ridge known as the Bunker Hill moraine, which marks the terminal position of T3 ice and is more or less paralleled by Shelbyville Road, and the hummocky landscape north of that ridge (fig. 16), which dominates the northeastern third of the county. The contrast between the smooth, poorly drained T2 till plain immediately in front of the moraine, and the hummocky, irregular T3 landscape on and behind the moraine, is stark and can be observed from numerous roads that cross this fundamental terrain boundary in the southeastern quadrant of the county (fig. 17).

Figure 17.
Left-The topography shown on this portion of the Beech Grove 7.5-minute quadrangle illustrates the sharp contrast between the smooth T2 till plain to the southwest and the characteristic hummocky disintegration topography associated with T3 ice to the northeast, with the two being separated by the high, irregular ridge known as the Bunker Hill moraine. (Diagram from Brown and Laudick, 2003). Right-View northwestward along the toe of the Bunker Hill moraine from near Stop 11 and Shelbyville Roads. The poorly drained T2 till plain (left of dashed line) was washed by sheet flow of meltwater (arrows) emanating from gaps in the moraine, whose toe lies to the right of the dashed line. The till plain contains numerous wet depressions (w) that mark former ephemeral swamps prior to artificial drainage, whereas the high-relief moraine is much better drained. Photo by A. H. Fleming.

Crown Hill and the rolling, elevated landscape immediately around it (fig. 18) are also thought to have formed along the terminal position of T3 ice. This somewhat enigmatic terrain fragment is truncated on both sides by the younger valleys of the White River and Fall Creek, which isolate it from other T3 uplands. It is aligned with the northwestward projection of the Bunker Hill Moraine, however, from which it is separated only by the valley of Fall Creek. Fall Creek valley probably was cut out somewhat later during the T3 event, hence it seems plausible that Crown Hill and the Bunker Hill moraine may have been one contiguous feature prior to that.

West of the White River, the T3 topography loses its distinctiveness at many places, blending in with the T2 landscape beneath it. At most places west of the river, T3 is recognizable only as a very thin veneer of low-relief, hummocky ablation deposits, typically a few feet thick, plastered over T2 features. No obvious moraine or other T3 terminal position can be recognized west of the river, although radiocarbon ages from Hendricks County clearly show that this event affected that area. The basic contrast in T3 morphology across the river valley suggests that there were fundamental differences in the behavior of T3 ice on either side of the central sluiceway defined by the White River and Fall Creek. It seems possible that the meltwater outbursts that led to the development of these gorges during T3 time may have sapped the ice sheet west of the river, perhaps partially detaching it from a thicker and more active part of the ice sheet east of the central sluiceway. The scale of these outbursts may signify some fundamental instability or discontinuity in the T3 ice sheet, perhaps related to the storage of massive amounts of meltwater in and below the glacier. Inasmuch as the evidence for large meltwater outbursts appears concentrated in the central and eastern parts of the county, it may be that the eastern part of the ice sheet surged southward repeatedly as it was buoyed by subglacial meltwater, while the western part held less meltwater and simply stagnated in place after its initial advance.

Regardless of whether the landscape was produced by T2 or T3 events, all of the late Wisconsin deposits, and especially the tills, produce extremely fertile soils. The Laurentide Ice Sheet advanced over dozens of different kinds of bedrock lithologies on its thousand-mile journey into central Indiana, including large areas of Paleozoic limestones and dolostones. In this process, the glacier incorporated silt and sand containing a wide range of major and minor elements, redepositing them as a homogenized mixture that resembles an ideal natural fertilizer. Late Wisconsin tills are referred to as "calcareous" for their plentiful content of the essential nutrient calcium, whose abundance facilitates the uptake of several other essential nutrients and leads to the presence of natural communities dominated by "calciphiles" — plants endemic to high-calcium soils. This assemblage, broadly known as the "till plain forest" (fig. 19), once covered the central Indiana landscape prior to the advent of agriculture, which now dominates this landscape. Subtle differences in topography, drainage, and moisture content within the till plain forest created almost endless variations in natural community structure and composition, ranging from swamps and wet woods in swales and floodplains, to rich mesic cove forests in deep ravines, to knolls dominated by oaks and hickories. Where preserved today, till plain forests host some of the showiest displays of wildflowers found anywhere in the midcontinent region, if not North America.

Figure 20.
Cultural history and ice-age topography combine at the Indianapolis Museum of Art (left) and the central canal (center), while bluff-side parks such as Holliday Park (right) exemplify the tight confines of the Rocky Ripple segment of the river, here seen in flood following heavy rain and snowmelt. Photos by A. H. Fleming.

Figure 19.
The calcareous till plain forest at Southeastway Park. Sugar maple (left) is dominant in the canopy, and is tapped for maple syrup in the late winter. Some of the showy, calcium-loving wildflowers include (left to right): Virginia bluebells and blue-eyed Mary; trout lily; Dutchman's breeches; and sessile trillium. Photos by A. H. Fleming.

As alluded to previously, the primary effect of the T3 glaciation was the production of copious quantities of meltwater, much of which appears to have been stored beneath the ice sheet and then released abruptly in one or more sizable outbursts. These massive meltwater outbursts are singularly responsible for shaping the more dramatic aspects of the county's topography, including the deep gorges of Fall Creek, the central and northern White River valley (fig. 20), Eagle Creek, and several other deeply entrenched tributaries. The meltwater outbursts appear to have occurred while T3 ice was still active in the area, because the gorges they cut have several medium and large tributaries with markedly rectilinear drainage patterns that clearly initiated in intersecting crevasses in and below the ice sheet. Examples include Fishback Creek, Indian Creek, and Williams Creek, among others. Such tributaries likely represent conduits through which meltwater stored in the glacier were discharged to the larger sluiceways, and suggest the great magnitude of the internal "plumbing" within the ice sheet that led to the meltwater outbursts. In contrast to these larger tributaries, whose lengths are typically measured in miles, the bluffs that flank most of the main gorges are also dissected by numerous short but very steep ravines (fig. 21) which appear to largely be post-glacial features. Seldom more than a quarter-mile long, such ravines are usually ephemeral and typically descend precipitously to their receiving streams. While some of these ravines may have originated in cracks in melting ice, they lack the rectilinear pattern and have no continuity beyond the immediate bluff walls.

Particularly good places to see the gorgelike topography associated with these late meltwater outbursts include Fort Benjamin Harrison State Park along Fall Creek, and several places along the bluffs of the White River (fig. 20), such as Marott, Broad Ripple, and Holliday Parks and the Indianapolis Museum of Art grounds. The central canal towpath also provides striking views of the deeply-entrenched gorge section of the river.

The section of Fall Creek above 56th Street appears to be a special case, in that its gorge probably originated entirely beneath active T3 ice as a feature known as a tunnel valley (fig. 22). In this setting, meltwater was under high hydrostatic pressure from a combination of overlying ice and hydraulic head produced by a tall column of meltwater stored in the ice above. Under such high head, the meltwater excavated a deep, flat-bottomed valley in the substrate below the glacier. At the time Fall Creek Gorge was cut, the base of the ice sheet sat on the upland surfaces that today stand some 100 to 150 ft (30 to 46 m) above the floor of the modern valley, which gives some visual idea of the depth of incision.

The current valley floor, however, is not the ultimate floor of the tunnel valley. Subsurface data indicate that the meltwater excavated a canyon some 75 ft (23 m) or more below the modern floor of the valley, which then filled back up to its current level with outwash, once the ice melted back from the sluiceway and the confining pressure was reduced. In doing so, the walls of the tunnel cut completely through not only the entire section of late Wisconsin glacial sediments, but also through a thick series of older Illinoian tills, ultimately bottoming out in pre-Illinoian deposits. The Rocky Ripple segment of the White River as well as portions of Eagle Creek may have also been cut as tunnel valleys during T3 time, though their subsurface geometries are not as well defined as Fall Creek.

Figure 22.
Composite image (top) of the steep sided Mud Creek valley, looking downstream from Fall Creek Road. Mud Creek is the major tributary of Fall Creek in Marion County and also originated as part of the Fall Creek tunnel valley system. The wide, flat bottom of the valley is deceptive: as shown in the diagrams on the left, the true bottom of the tunnel valley is up to 50 to 75 ft (15 to 23 m) below the modern valley bottom, which is floored by coarse late Wisconsin gravel (inset) that partially filled the tunnel valley once the ice had melted back and unconfined conditions were reestablished. The structure in the distant background on the right side of the photo above is one of the wells in the Mud Creek municipal wellfield, which taps a large body of pre-Illinoian age sand and gravel into which the tunnel valley was excavated. Photos by A. H. Fleming.

Throughout the late Wisconsin glaciation, the White River acted as the major sluiceway for meltwater generated by glaciers not only in Marion County, but elsewhere in central Indiana. The glacial White River changed courses several times during the late Wisconsin and actively carried large volumes of meltwater and sediment even while beneath or flanked by glacial ice that may have been thousands of feet thick. Despite being at temperatures just barely above freezing, the continuous flow of such a large volume of meltwater generated sufficient heat to melt the ice adjacent to the sluiceway, producing a major ice-walled channel that frequently was free of ice while the surrounding uplands were being glaciated. This battle between relatively cold ice and relatively warm meltwater played out throughout the late Wisconsin and resulted in complex interactions between ice margins and meltwater as the glacier periodically surged into the valley, only to melt back when the resulting constriction of the sluiceway intensified the heat output from the meltwater. This cycle repeated itself several times, leading to a wide variety of landforms and abrupt changes in the style of deposition along the margins of the valley. The changing behavior and position of the sluiceway during glaciation created several distinct segments of the river valley (figs. 23 and 24) that are identified based on differing morphology, hydrology, and subsurface geology.

The modern appearance of the White River valley and its tributaries may be the result of late Wisconsin glaciation, but there is ample evidence that the river, or a precursor stream, was already well established in this general location when the late Wisconsin glaciers arrived in Marion County. In fact, the subsurface record in Marion County provides compelling evidence that the White River valley functioned as a major glacial sluiceway and important interglacial drainage throughout virtually the entire Pleistocene glaciation, and probably existed as an important pre-glacial drainage as well, albeit in a somewhat different form than its current one. Indeed, the recurring role of the White River and its precursor streams is a common thread that links together all of the different glacial episodes that have affected the county. The history of the valley is, in its own right, a fascinating, if not tantalizing, story of geologic change.

Figure 24.
Left-The horseshoe bend just above Broad Ripple. Here, the river is bending away from its former glacial course, and leaving the main sluiceway which continues straight south to the Fairgrounds. Beyond the bend, at the west end of Broad Ripple, the river enters a narrow, straight, gorgelike valley known as the "Rocky Ripple segment," which is also inferred to have initiated as a tunnel valley at the same time as Fall Creek. At the top of that segment, the river flows across Devonian Limestone, one of only two bedrock outcrops in the county. Right-Cross-sectional view of pitted late Wisconsin outwash in a sand and gravel pit along the White River valley. The deformed beds behind the person underlie a depression that developed where a sizable block of ice was buried in the outwash and subsequently melted, causing the overlying beds to collapse. Photos by A. H. Fleming.


Bleuer, N. K., 1991, The Lafayette Bedrock Valley System of Indiana; Concept, form, and fill stratigraphy, in Melhorn, W. N., and Kempton, J. P., eds., Geology and hydrogeology of the Teays-Mahomet Bedrock Valley System: Boulder, Colorado, Geological Society of America Special Paper 258, p. 51-77.

Brown, S. E., and Laudick, A. J., eds., 2003, Hydrogeologic framework of Marion County, Indiana — a digital atlas illustrating hydrogeologic terrain and sequence: Indiana Geological Survey Open-File Study 00-14, CD-ROM.

Johnson, W. H., Follmer, L. R., Gross, D. L., and Jacobs, A. M., 1972, Pleistocene stratigraphy of east-central Illinois: Illinois State Geological Survey Guidebook Series 9, 97 p.

Fleming, A. H., Brown, S. E., and Ferguson, V. R., 1993, Hydrogeologic framework of Marion County, Indiana: Indiana Geological Survey Open-File Study 93-05, 67 p.

Wayne, W. J., 1963, Pleistocene formations in Indiana: Indiana Geological Survey Bulletin 25, 85 p.