Karst Hydrogeology of the Haney Limestone, South-Central Kentucky

Home , Caseyville Formation, Pennyroyal Plateau

Western Kentucky University TopSCHOLAR®

Masters Theses & Specialist Projects Graduate School

5-2013

Karst Hydrogeology of the Haney Limestone, South-Central Kentucky

Sarah Marie Arpin Western Kentucky University, [email protected]

Follow this and additional works at: https://digitalcommons.wku.edu/theses

Part of the Geographic Information Sciences Commons, and the Physical and Environmental Geography Commons

Recommended Citation Arpin, Sarah Marie, "Karst Hydrogeology of the Haney Limestone, South-Central Kentucky" (2013). Masters Theses & Specialist Projects. Paper 1253. https://digitalcommons.wku.edu/theses/1253

This Thesis is brought to you for free and open access by TopSCHOLAR®. It has been accepted for inclusion in Masters Theses & Specialist Projects by an authorized administrator of TopSCHOLAR®. For more information, please contact [email protected].

KARST HYDROGEOLOGY OF THE HANEY LIMESTONE, SOUTH-CENTRAL KENTUCKY

A Thesis Presented to The Faculty of the Department of Geography & Geology Western Kentucky University Bowling Green, Kentucky

In Partial Fulfillment Of the Requirements for the Degree Master of Science

By Sarah Marie Arpin

May 2013

ACKNOWLEDGEMENTS

This research is the product of the help and support of many individuals, without whom the project never would have been possible. There are many who gave me the inspiration to pursue the project in the first place; thank you to all who introduced me to the spectacular world that exists beneath the surface. A whole cadre of individuals gave the moral support necessary to complete the project. Fellow Hoffmanitos Celia Davis,

Nick Lawhon, Ben Miller, Gilman Ouellette, and Sean Vanderhoff offered a tremendous support network throughout the project. A special thanks to my parents who provided their unwavering support throughout, even as the time required to complete the project was more than they thought they were in for. A huge thanks also to the Tallent family, who took me in as their own and continue to do everything possible for the future success and bliss of Jeremy and I. And of course, thank you, Jeremy, for your dedicated love and support. Near or far, you were always right there with me every step of the way.

Support from many others was given through the contribution of information and ideas that helped shape this project. The Kentucky Speleological Survey provided known cave entrance locations and available maps for caves located on privately owned land – special thanks to Jim Currens, Steve Gentry, Howard Kalnitz, and Bill Walden for their help getting all the data together. Mammoth Cave National Park and the Cave Research

Foundation provided the locations of known cave entrances and available maps for caves within the park’s boundaries. Mammoth Cave resource managers Bobby Carson, Lillian

Scoggins, and Rick Toomey were especially helpful in retrieving the data from both the park and CRF archives. Many members of the Cave Research Foundation were eager to discuss the project during expedition weekends; the ideas and insight gained from these

iii conversations were tremendous and are major components of the final project. A special thanks to Roger Brucker who was extremely eager to discuss the research and a joy to interact with. A special thanks also to Pat Kambesis who devoted all her time to the project over the entire week of the CRF 4th of July expedition. Her guidance and support helped to focus the research and bring together the various components of the project into a cohesive final product. Norman Warnell’s extensive knowledge of the land in the

Mammoth Cave area – its history, geology, and hydrology – was personally gathered out of a passion for caves. Norman’s generous contributions to this research included guidance in the field, documents, and cave landowner and location information – all of which are essential to the project. Steve Miller was an invaluable resource for information on Cub Run Cave, providing survey data and volunteering his time to visit the cave for a thorough walk-through of its human and geologic history. Dr. Stan Sides provided survey data for Lulu Mart Cave, additionally offering integral information and documents on Lulu Mart and other caves north of the Green River.

Special thanks to members of the Green River Grotto who donned their wetsuits and volunteered countless hours to survey the cold, wet caves of the Haney Limestone.

Members who contributed to the surveys include: Clint Barber, Josh Brewer, Seth Drake,

Bret Grebe, Brian Ham, Pat Kambesis, Nick Lawhon, Ben Miller, Cody Munday, Gil

Oullette, Dr. Jason Polk, Jeremy Tallent, Sean Vanderhoff, and Virgil Vertrees. Thanks also to the many landowners who graciously allowed us access to their property.

Finally, thank you to my amazing thesis committee members: Dr. Chris Groves,

Dr. Arthur Palmer, Dr. Jason Polk, and Dr. Jun Yan. Every one of you provided great effort, guidance, and your unique expertise to help shape this research project. A very

iv special and heartfelt thanks to my advisor, Dr. Chris Groves, whose tremendous guidance and support extends far beyond this research project.

CONTENTS List of Figures ...... viii List of Tables ...... x Abstract ...... xi Chapter 1: Introduction ...... 1 1.1 Study Area: South-central Kentucky Karst ...... 2 1.2 Research Questions ...... 3 Chapter 2: Literature Review ...... 6 2.1 Karst ...... 6 2.2 Regional Geologic & Geomorphic History ...... 9 2.3 Lithological Characteristics of the Haney Limestone ...... 12 2.4 Karst Development within the Haney Limestone ...... 14 2.5 Hydrologic & Geochemical Investigations of the Haney Limestone ...... 17 Chapter 3: Data & Methodology ...... 19 3.1 Geographic Information Systems ...... 19 3.2 Cave Maps ...... 22 3.3 Field Methods ...... 25 3.4 Geochemical Data ...... 27 Chapter 4: Results ...... 31 4.1 Distribution of Karst Development within the Haney Limestone ...... 31 4.2 Caves and Karst Development in the Haney Limestone ...... 35 4.2.1 Alaska Caverns ...... 36 4.2.2 Barner’s Mill Cave ...... 37 4.2.3 Beaver Dam Creek Cave ...... 39 4.2.4 Chalybeate Cave ...... 40 4.2.5 Cub Run Cave ...... 42 4.2.6 Honaker Cave ...... 45 4.2.7 Lulu Mart Cave ...... 46 4.2.8 Miller (Calyx) Cave ...... 48 4.2.9 Silent Grove Springhouse Cave ...... 49 4.2.10 Regional Cave Trends ...... 51

4.3 Hydrology and Dissolution in the Haney Limestone ...... 53 Chapter 5: Discussion ...... 56 5.1 Recharge Sources ...... 56 5.2 Structural Controls ...... 57 5.3 Hydrologic Surface and Subsurface Relationship ...... 61 5.4 Relative Age ...... 62 Chapter 6: Conclusions & Future Research ...... 64 6.1 Conclusions ...... 64 6.2 Future Research ...... 65 Appendix A: Cave Maps ...... 68 Appendix B: Geochemical Data ...... 93 References ...... 100

vii

LIST OF FIGURES

Figure 1. Map of the study area showing Haney Limestone ...... 3

Figure 2. Geologic formations exposed in the south-central Kentucky region...... 9

Figure 3. Diagram showing the regional landscape ...... 11

Figure 4. Distribution of karst features in the Haney Limestone across a 6 counties ...... 31

Figure 5. Possible spring resurgences not included in the KGS database ...... 33

Figure 6. Distribution of karst features in the Haney Limestone in the study area ...... 34

Figure 7. Alaska Caverns lineplot overlay ...... 37

Figure 8. Barner's Mill Cave lineplot overlay ...... 38

Figure 9. Beaver Dam Creek Cave lineplot overlay ...... 40

Figure 10. Chalybeate Cave lineplot overlay ...... 41

Figure 11. Cub Run Cave lineplot overlay ...... 43

Figure 12. Honaker Cave lineplot overlay ...... 45

Figure 13. Lulu Mart Cave lineplot overlay ...... 47

Figure 14. Miller (Calyx) Cave lineplot overlay ...... 48

Figure 15. Silent Grove Springhouse Cave lineplot overlay ...... 50

Figure 16. Length-weighted rose diagrams ...... 52

Figure 17. Main passage in Lick Creek Cave ...... 58

Figure 18. Length-weighted rose diagram of all Haney caves with Dieke's (1967) findings overlain ...... 59

Figure A1. Description and lineplot of Barnes Smith Cave (Ramsey 1974)...... 69

Figure A2. Davenport Cave map (KSS)...... 70

Figure A3. Honaker Cave map (KSS)...... 71

Figure A4. Spring Trough Caves 1 & 2 (1991 Speleofest Guidebook pg. 49)...... 72

viii

Figure A5. Black Rock Cave map (CRF)...... 73

Figure A6. Bryophyte Cave map (CRF)...... 74

Figure A7. Cade Cave map (CRF)...... 75

Figure A8. Cricket Falls Cave map (CRF)...... 76

Figure A9. Deer Skull Cave map (CRF)...... 77

Figure A10. Feather Cave map (CRF)...... 78

Figure A11. Fish Trap Pits 1 & 2 (CRF)...... 79

Figure A12. Good Spring Cave map (CRF)...... 80

Figure A13. Hearth Pit Cave (CRF)...... 81

Figure A14. Hearth Spring Cave map (CRF)...... 82

Figure A15. Hickory Cabin Cave map (CRF)...... 83

Figure A16. Hideout Spring Cave map (CRF)...... 84

Figure A17. Hornbeam Spring Cave map (CRF)...... 85

Figure A18. Johnson Spring Cave map (CRF)...... 86

Figure A19. Lycopodium Spring Cave map (CRF)...... 87

Figure A20. Salamander Cave map (CRF)...... 88

Figure A21. Squeeze Cave map (CRF)...... 89

Figure A22. Squirrel Hollow Springhouse Cave map (CRF)...... 90

Figure A23. Stillhouse Sink Cave (CRF)...... 91

Figure A24. Two Entrance Cave map (CRF)...... 92

LIST OF TABLES

Table 1. Known caves in the Haney Limestone located within the study area ...... 24

Table 2. Equilibrium constants from White (1988, 121) ...... 29

Table 3. Values for constants k and n from Palmer (1991) ...... 30

Table 4. Summary characteristics of the nine caves over 100 m in length ...... 35

Table 5. Dissolution rates of 6 Haney springs, calculated using the Plummer et al. (1978) equation and water chemistry data collected by Hess (1974) ...... 54

Table 6. Dissolution rates of 6 Haney Springs, calculated using Palmer's (1991) equation and water chemistry data collected by Hess (1974) ...... 55

Table A1. Adwell Spring geochemical data (from Hess 1974)...... 94

Table A2. Blair Spring geochemical data (from Hess 1974)...... 95

Table A3. Bransford Spring geochemical data (from Hess 1974)...... 96

Table A4. Collins Spring geochemical data (from Hess 1974)...... 97

Table A5. Cooper Spring geochemical data (from Hess 1974)...... 98

Table A6. Three Springs geochemical data (from Hess 1974)...... 99

KARST HYDROGEOLOGY OF THE HANEY LIMESTONE, SOUTH-CENTRAL KENTUCKY

Sarah M. Arpin May 2013 103 Pages

Directed by: Chris Groves, Arthur Palmer, Jason Polk, and Jun Yan

Department of Geology & Geography Western Kentucky University

South-central Kentucky has one of the world’s most intensively studied karst areas, with most work focusing on the Mammoth Cave System and related caves and aquifers. However, slightly higher in the stratigraphic section than Mammoth Cave, the

Haney Limestone is a locally important but less well studied carbonate aquifer. This research provides the most comprehensive synthesis to date of the karst hydrogeology of the Haney Limestone of south-central Kentucky, focusing on the distribution and controls on cave and karst features developed within. In contrast to drainage systems within the major limestones below, joints are the most dominant control on passage development in the Haney Limestone within the study area and the orientation of these joints is consistent with that of regional joint sets. Bedding planes and the presence of insoluble rock at the base of the Haney also exert control on conduit development in the Haney Limestone.

Most of the caves of the study area developed in the Haney Limestone are single- conduit caves that receive water through direct, allogenic sources. Cave entrances are frequently perennial spring resurgences and the presence of active streams suggests that the caves function within the contemporary landscape, acting as drains for localized recharge areas. The hydrology of the Haney Limestone plays an important, if localized, role in the regional hydrology of south-central Kentucky, integrated into the current system of surface and subsurface drainage of the regional karst landscape. Evidence supports the idea that caves of the Haney Limestone are, geologically, relatively recent

xi phenomena. A majority of the cave passages in the study area are hydrologically active, the water resurging from the sampled springs is typically undersaturated with respect to limestone, and the caves in some case appear to be developed along potential stress release fractures associated with small, apparently young valleys. This suggests that caves in the Haney Limestone were not directly influenced by the incision of the Green

River over vast periods, like Mammoth Cave, but that cave development is a largely contemporary process.

xii

Chapter 1: Introduction

South-central Kentucky contains one of the world’s most well-known and intensively studied karst areas. Mammoth Cave, the longest cave in the world with a current known length of more than 627 km, lies in the heart of this region and is the primary subject of most scientific literature on south-central Kentucky’s karst. Prolific scientific investigations of Mammoth Cave have allowed geologists to reconstruct and understand major events in the geologic history of the region and the origin of Mammoth

Cave (White et al. 1970; Miotke and Palmer 1972; Hess 1976; Palmer 1981, 1989, 2007;

White 1988; Granger et al. 2001). The Mammoth Cave System extends through three distinct geologic formations of Mississippian age: in ascending order these are the St.

Louis, Ste. Genevieve, and Girkin Limestones (Palmer 1981). Overlying these carbonate rock layers is the Big Clifty Sandstone, an impermeable caprock that protects the limestone from dissolution at the surface and lends itself to the great length of the cave system (Palmer 1981). Above the Big Clifty Sandstone is another carbonate rock unit, the Haney Limestone, which has received much less study. Most research has primarily involved investigations into the lithologic characteristics and depositional history of the

Haney Limestone (McFarlan et al. 1955; Swann 1964; Vincent 1975; Foster 1990), along with some work on the hydrologic characteristics and karst development (Brown 1966;

Hess and White 1993; Ryan and Meiman 1994).

While the hydrogeology is in some places influenced by cave and conduit development within the perched aquifer of the Haney Limestone, the details have not been extensively studied. This study seeks to characterize and describe the hydrogeology of the Haney Limestone within a three county area of south-central Kentucky. This

1 research provides the most comprehensive synthesis to date of the spatial distribution of karst features developed in the Haney Limestone, and the geologic and hydrologic controls on karst development. It provides the first characterization of the patterns of cave origin and passage morphology within the karst aquifer of the Haney Limestone, and compares those with the patterns and controls of development within the principal karst aquifer of the St. Louis, Ste. Genevieve, and Girkin Limestones within the same study area. Knowledge gained from this study provides a more complete understanding of the south-central Kentucky karst and broaden current understanding of the region’s geologic and geomorphic history.

1.1 Study Area: South-central Kentucky Karst

In south-central Kentucky, the Haney Limestone is exposed along the southern and eastern edges of the Illinois Basin. The Haney Limestone outcrops along the

Dripping Springs (Chester) Escarpment and on the higher ridgetops of the Mammoth

Cave Plateau (Chester Cuesta). Due to the slight northwestern dip of the rocks, the

Plateau displays a variation in karst landscape and aquifer development moving from east to west. The eastern portion is in the later stages of development and the landscape is highly fragmented. To the west, karst development is still in the early stages and surface drainage is still prominent. The area of interest includes the region that lies between these two extremes, where cave development is currently most extensive. In addition, this is the region that contains the Mammoth Cave System and much of its recharge area.

This area has been studied extensively and this research has provided a good understanding of the geomorphic history of the region. For these reasons, a three county study area was selected, including Edmonson, Hart, and Warren Counties (Figure 1). A

2 primary goal of this research is to add to current knowledge and understanding of the region’s geomorphic history through the study of another karst aquifer component: the

Haney Limestone.

Figure 1. Map of the study area showing Haney Limestone outcrops at the surface.

1.2 Research Questions

The primary goal of this research is to describe the hydrogeology of the Haney

Limestone in south-central Kentucky, and, in particular, what influence karst development has on the hydrogeology. To describe the karst development and hydrogeology, this study characterizes the controls on cave origin and patterns of passage morphology. What are controls on karst development in the Haney Limestone? The

3 pathways exploited by groundwater moving through the karst aquifer significantly influence the passage morphology. Since water moving through karst aquifers typically follows pathways of secondary porosity, infiltration and solutional enlargement can occur along bedding planes, joints and fractures. Based on the morphology of cave passages in the Haney Limestone, which of these pathways serves as the primary source of infiltration? While a combination of these structural openings is likely to have contributed to passage morphology, one is likely to emerge as dominant and characteristic of the hydrogeology of the Haney Limestone. Is there a consistent pattern of structural origin of caves in the Haney Limestone evident throughout the study area?

The source and nature of the recharge also significantly controls the origin and morphology of caves. As described by Ryan and Meiman (1994) in the case of Lulu

Mart Cave, for example, it is likely that caves in the Haney will receive recharge from a combination of discrete and diffuse sources from both allogenic and autogenic origins, but that the influence of one type may dominate. Is there a type or source of recharge that dominates cave development in the Haney Limestone in the region? In addition, the zone of development strongly influences passage morphology. Development in the epikarst, vadose, and phreatic zones will likely be evident in all caves of the Haney

Limestone; however, given the relative thinness of the rock unit, it is probable that most solutional enlargement of will occur in one of these three zones. Is it the case that passages of caves in the Haney Limestone are characteristic of development within a single zone?

An examination of the distribution of caves and other karst features is necessary to determine whether the level of karst development is consistent throughout the Haney

Limestone or if spatial variation exists. From this, it is possible to proceed by examining what controls the distribution and morphology of caves in the Haney Limestone. Do regional variations in the structure or fault zones exert an influence on karst development and cave distribution? Does the presence and thickness of overburden have an influence on the occurrence of cave and karst features?

Finally, how do Haney caves relate to regional hydrogeology and geomorphology? If incision of the Green River, and coinciding water table levels, controlled passage development in the Haney Limestone, it would follow that these caves may have originated at an earlier time than the Mammoth Cave System. However, it is also possible that caves developed within the Haney Limestone more recently as highly aggressive water flows from the overlying Hardinsburg Sandstone and dissolves away the rock at a more rapid rate. Are caves developed within the Haney Limestone relict features related to past landscape conditions, or, are they more recent features related to current landscape conditions and currently functioning within the regional hydrological regime?

Chapter 2: Literature Review

2.1 Karst

Karst refers to a landscape in which morphology is dominated by the chemical dissolution of soluble rocks (White 1988; Ford and Williams 2007). Although karstification occurs in gypsum, salt, and other rocks, the most common and extensive karst areas occur in carbonate bedrock, most commonly limestone and dolomite. Karst landscapes are characterized by extensively developed underground drainage systems and the level of karst development is dependent upon the capacity of the subsurface to accept water (White 1988).

Water infiltrates the subsurface through existing pathways and zones of weakness within the bedrock. While primary porosity may in some systems be influential in the development of subsurface flow paths, the larger pathways of secondary porosity are the dominant control of karst aquifer evolution (White 1988; Ford and Williams 2007).

Secondary porosity refers to openings in the bedrock formed after deposition; these include joints, fractures and bedding planes (White 1988). The extent of permeability of these types of openings is typically greater than for the primary porosity, and water preferentially follows these pathways. In addition, the larger sizes of these openings allows for the transition from laminar to turbulent flow, which occurs once a conduit reaches a width between approximately 0.5 and one cm (Palmer 2007). A rapid increase in dissolution rates associated with hydrochemical details of the relevant processes that can accompany this threshold ‘breakthrough’ event signal inception of cave development; conduit dissolution and enlargement can occur more rapidly (White 1977; Dreyboldt

1990; Palmer 1991; Groves and Howard 1994). The overall pattern of cave passages is

6 affected by which type of secondary porosity provided the dominant flow paths for the water.

The source of recharge entering a karst aquifer system has a significant impact on the development and characteristics of the system. Precipitation that falls directly on the karst surface is referred to as autogenic recharge. Allogenic recharge refers to precipitation which falls on nonkarstic rock, coalesces to form surface streams, then sinks underground at discrete locations upon reaching the more soluble karst bedrock (Palmer

1984, 2007; White 1988; Ford and Williams 2007). Water can enter underground conduit systems by means of diffuse flow, or, through discrete entry points, such as sinking streams. Flow entering through sinkholes can in some cases be discrete in that water enters the aquifer system at distinct locations, but may also enter through more diffuse pathways in cases where it enters through many distinct locations and converges further downstream in the system. Sinkholes are common sources of autogenic recharge; however, sinkholes may also form in thin layers of overlying insoluble caprock and penetrate to underlying carbonate bedrock below. Sinking streams are common sources of autogenic recharge, as runoff from insoluble bedrock sinks immediately upon reaching soluble carbonate rock. Diffuse flow can be a source of either autogenic or allogenic recharge. Autogenic diffuse flow occurs when water percolates through the soil to the carbonate bedrock surface. Autogenic diffuse flow can also occur when water permeates through porous clastic rock overlying soluble carbonate rocks (White 1988; Ford &

Williams 2007).

The source of recharge has important implications on karst aquifer development due to the effect it has on water chemistry and the volume of water each source is able to

7 accept. Allogenic recharge that has not yet been in contact with carbonate rocks is typically undersaturated with respect to calcite and capable of relatively high rates of dissolution (White 1988; Ford & Williams 2007; Palmer 2007). Discrete sources of recharge concentrate flow and are typically capable of conducting a high volume of water to the subsurface. While many karst aquifer systems receive a combination of inputs from various sources of recharge, the balance between these sources has significant influence on the overall passage patterns of the caves that develop (Palmer 1984; White

1988). Single-conduit caves are a characteristic pattern resulting from the dominance of allogenic recharge; a large volume of aggressive water enters the carbonate bedrock at a single location, flows through and exits the bedrock. Dendritic, branchwork patterns are common in caves where autogenic recharge from multiple inputs dominates; they form as a number of tributary conduits come together to form larger conduits downstream

(Palmer 1984; White 1988).

Water moving through a karst aquifer system flows through three zones of distinct hydrological characteristics. The epikarst zone consists of the dense network of solutionally enlarged conduits at or near the karst surface. The soluble carbonate bedrock may be exposed or buried beneath the soil mantle. In soil covered settings water is most aggressive in this zone, as it has typically been exposed to high CO2 pressures and has yet to interact with carbonate rock. Beneath the epikarst is the vadose zone – a zone of partially or completely air-filled conduits. Water flows downward under the influence of gravity through the vadose zone until it reaches the saturated, or, phreatic, zone. The upper surface of the phreatic zone is the water table; all conduits and pore spaces are water-filled in the phreatic zone (White 1988; Ford & Williams 2007; Palmer 2007). The

8 origin and morphology of cave passages are strongly influenced by these zones and passages exhibit distinct characteristics depending on their zone of development (Palmer

1981, 2007).

2.2 Regional Geologic & Geomorphic History

The sedimentary rocks exposed in the south-central Kentucky region were deposited along coastal margins during the Mississippian and Pennsylvanian Periods,

Figure 2. Geologic formations exposed in the south-central Kentucky region (from Palmer, 1981). The limestone formations in which the Mammoth Cave System is formed are highlighted in blue and the Haney Limestone in pink.

300-350 million years ago (Palmer 1981). Insoluble sandstone and conglomerate rock formations of the area were formed under deltaic conditions, while the soluble carbonate rocks were formed in a shallow marine environment. The types of sediment deposited changed with changes in climate, water depth, and other environmental factors over time, resulting in the formation of different rock layers (Palmer 1981). Erosion reveals the alternating sequence of sedimentary rock types exposed in south-central Kentucky

(Figure 2).

The rock layers exposed in the south-central Kentucky region are located along the outer edge of the Illinois Basin, a synclinal structure which also extends into parts of

Illinois and Indiana. Synclinal structures expose older strata along the raised outer edges and it is along the southern and eastern boundaries of the Illinois Basin that these older strata are exposed in south-central Kentucky. Beds dip inward toward the center of the synclinal structure; in south-central Kentucky, the regional dip of the beds averages around ⅓˚ to ½˚ (6m/km) to the northwest, with local variations due to gentle folding of the beds (Palmer 1981). In this region, the Dripping Springs Escarpment (historically known as the Chester Escarpment) divides the high ridges of the Mammoth Cave Plateau

(Chester Cuesta) from the flat-lying sinkhole plains of the Pennyroyal Plateau to the south. Much of the overlying rock strata have been removed from the Pennyroyal

Plateau and a sinkhole plain has developed in the Ste. Genevieve and St. Louis

Limestones. Precipitation falling on the sinkhole plain flows downdip through underground conduits, beneath the Dripping Springs Escarpment and Mammoth Cave

Plateau, and emerges at spring outlets along the Green River (Deike 1967; White et al.

1970). The plateau consists of high ridges, capped by the insoluble Big Clifty Sandstone

10 and the thinner beds of the Late Mississippian and Early Pennsylvanian System. The upland is dissected by dry karst valleys where the insoluble caprock has been breached

(Figure 3).

Figure 3. Diagram showing the regional landscape (from Cushman, 1965). The limestone formations in which the Mammoth Cave System is formed are highlighted in blue and the Haney Limestone in pink.

The Green River has played a significant role in the evolution of the landscape and regional karst aquifer system. The Green River flows from east to west, cutting a deep valley through the Mammoth Cave Plateau. The elevation of the Green River is the base level for the region. As the Green River downcut, the insoluble Big Clifty

Sandstone was progressively breached from east to west over time, opening up outlets for subsurface flow to exit the karst aquifer (Deike 1967). Surface streams tributary to the

Green River also breached the Big Clifty caprock as the Green River Valley downcut, leaving behind dry karst valleys as flow was diverted to subsurface conduits (Deike 1967;

White et al. 1970). Cycles of glacial advance and retreat during the Pleistocene played a major role in passage development at Mammoth Cave, controlling the height of the Green

River and the level of the water table (Deike 1967; Palmer 1981). The Ohio River was diverted during the Kansan glacial period and resulted in the rapid incision of the Green

River Valley as the Green River quickly adjusted its gradient to the newly routed Ohio

River Valley (Deike 1967). Granger et al. (2001) used cosmogenic dating of cave sediments using 26Al and 10Be ratios to reconstruct the geomorphological evolution of

Mammoth Cave and the regional landscape. Sediment dating enabled the correlation of

Green River incision and major passage level development with regional climatic changes and their consequent, large-scale, hydrologic readjustments. Granger et al.

(2001) estimated the dissolution process that formed the current Mammoth Cave System began during the Miocene or Pliocene Epoch, at least 5 million years ago.

The nearly horizontal passages of the Mammoth Cave System were more recently intersected by deep vertical shafts (White et al. 1970; Brucker et al. 1972). Vertical shaft development in the Mammoth Cave Plateau began after the sandstone caprock was breached and the karst valleys were formed. The vertical shafts are related to the current regional landscape and many but not all are associated with the edges of the insoluble caprock. Aggressive water flowing off the sandstone caprock or out of springs from the perched aquifer of the Haney Limestone formed deep, cylindrical, vadose shafts which integrated into the subsurface conduit systems beneath the plateau (Brucker et al. 1972).

The development of these vertical shafts has also been suggested as a key element in the retreat of the Dripping Springs Escarpment.

2.3 Lithological Characteristics of the Haney Limestone

In 1955, McFarlan et al. established the Golconda Group, which includes the Big

Clifty Sandstone Formation and overlying Haney Limestone Formation. McFarlan et al.

(1955) named the Haney Limestone after the carbonate rocks they observed exposed along Haney Creek in Hardin County, Illinois (Foster 1990). The Haney Limestone and

Big Clifty Sandstone (the Golconda Group) are part of the Chesterian Series of the Upper

Mississippian System. Overlying the Haney Limestone is another siliciclastic formation, the Hardinsburg Sandstone. Regionally, the Haney Limestone occurs throughout the

Illinois Basin of Illinois, Indiana, and Kentucky. It outcrops in some areas along the outer boundaries of the basin, mainly in the southern and eastern areas. The Haney remains deeply buried beneath overlying Pennsylvanian rocks within the inner portion of the Illinois Basin. In the Mammoth Cave area of south-central Kentucky, the Haney

Limestone outcrops along the Dripping Springs Escarpment and on high ridges across the

Mammoth Cave Plateau. Brown (1966) described the Haney Limestone in the Mammoth

Cave area as a pure, crystalline limestone. It is typically yellowish-gray or light-olive gray, with local occurrences of shale and chert (Brown 1966).

Deposition of the thin Haney Limestone occurred in shallow marine environments during a time of transgression and subsidence (Foster 1990). Deltaic conditions characterized the depositional environment of the underlying Big Clifty Sandstone; migrating deltas received sediments transported from the Appalachian Highlands by the

Michigan River. Swan (1964) and Vincent (1975) attributed the shift from the deposition of siliciclastic sediments, to the deposition of carbonate sediments, to transgressive sequences followed by regressive sequences. However, Foster (1990) found no evidence of these shifts in her investigation of the Haney Limestone in Sulphur, Indiana. Rather,

13 shifting climatic conditions in the sediment source areas reduced material transported to the delta. A possible shifting of channels in the Michigan River Delta could have also contributed. These conditions allowed for the deposition of the carbonate sediments that make up the Haney Limestone. Foster (1990) found no evidence of sea level changes during the depositional period of the Haney Limestone, leading her to conclude that the deposition rate must have been nearly equal to the slow rate of basin subsidence occurring simultaneously.

2.4 Karst Development within the Haney Limestone

Brown described the Haney Limestone of the Mammoth Cave region as

“honeycombed with solution channels…much smaller in size than the large caves in lower limestones” (1966, p. 43). In the Mammoth Cave region, the Haney Limestone averages only 12 meters in thickness (Brown 1966). In his investigation of karst hydrology north of the Green River, George (1989) stated that it is typical for caves developed within the Haney Limestone to span the entire vertical extent of this carbonate layer. Lateral extent of theses caves is mostly short, characterized by maze patterns of high, narrow canyon passages (George 1989). Brown (1966, 43) noted that solution channels in the Haney are “especially well developed in the upper part of the formation” and water flows through them to spring outlets at the base of the formation. Water flows laterally upon reaching the shale aquiclude at the base of the Haney Limestone, or the clastic sandstone layer below, resurging from spring outlets at the surface.

The most extensive known cave developed within the Haney Limestone in south- central Kentucky is Cub Run Cave. Cub Run Cave is a commercially operated tourist cave outside Cub Run, Kentucky. Located in Hart County, north of Mammoth Cave

National Park, the entrance was originally a small spring resurgence in the bluffs along

Little Dog Creek, now artificially enlarged to accommodate visitors. The spring emerges atop an erosion-resistant shale layer commonly present at the base of the Haney

Limestone. The current surveyed length of Cub Run Cave is more than two km, with an estimated 500 m known but as yet unsurveyed. The cave also contains the largest known cave passage developed within the Haney; the main trunk remains roughly 8 m high and

10 m wide for about one km. This trunk passage is a large canyon formed from dissolution by the downcutting of a vadose stream that extends through the entire thickness of the Haney Limestone. In the upstream portion of the cave, collapse of the overlying beds exposes the Glen Dean Limestone at the ceiling; the entire thickness of the overlying Hardinsburg Sandstone has collapsed in the passage. The spring resurgence flows out over the shale layer at the base of the carbonate bed. Cub Run Cave appears strongly influenced by the joints and fractures in the rock. The Cub Run Fault, running nearly twenty kilometers in a north-south direction, is roughly three kilometers to the east of the cave entrance (Haynes 1964; Sandberg and Bowles 1965; Deike 1967).

The most significant know cave in the Haney Limestone within the boundaries of

Mammoth Cave National Park is Lulu Mart Cave (Ryan and Meiman 1994). Lulu Mart

Cave is located on the north side of the Green River and has a surveyed length of just over 870 m. The entrance is a spring resurgence which forms the headwaters of the Dry

Prong of Buffalo Creek; the Dry Prong of Buffalo Creek is a perennial surface stream which also receives water from numerous other Haney springs (Ryan and Meiman 1994).

Ryan and Meiman (1994) describe the first 75 m of Lulu Mart Cave as highly joint controlled, noting that the passage formed parallel to joints along two primary axes at

15 azimuths of 13˚ and 290˚. They attribute the joints to stress release fracturing due to overburden removal by the downcutting of Raymond Hollow. Cushman et al. (1965) and

Hess and White (1993) also noted numerous joints and fractures in the Haney Limestone.

Beyond the first 75 m, Lulu Mart Cave becomes a canyon complex with two to three distinct levels. Further upstream is the confluence of two major stream passages, the

Ohio and the Mississippi, both also multi-level canyon complexes. Two smaller streams converge to form the Ohio much farther upstream.

Ryan and Meiman (1994) performed a series of dye traces to delineate the catchment area of Lulu Mart Cave. They found that the Lulu Mart Spring groundwater basin can be divided into two sub-basins, the Fern Hollow sub-basin and the Raymond

Hollow sub-basin, each feeding the two major stream passages in the cave. Water chemistry analysis showed significant differences between the water of the Fern Hollow sub-basin and the Raymond Hollow sub-basin. The headwaters of Fern Hollow emerge from acidic seeps in the Caseyville Formation and are very low in pH. The Glen Dean

Limestone is absent in this area and the Caseyville lies unconformably on the

Hardinsburg Sandstone. However, the headwaters of Raymond Hollow emerge from

Glean Dean springs and the water is much higher in pH. Dye was injected into nine sinking streams within the Lulu Mart Spring groundwater basin and groundwater input was measured at each of these locations. The total groundwater input from these locations only accounted for 65% of the discharge measured at the Lulu Mart Spring.

The remaining 35% of discharge was suspected to be from diffuse infiltration through the

Hardinsburg Sandstone or from stream loss through the Hardinsburg upstream from the gauging stations (Ryan and Meiman 1994).

2.5 Hydrologic & Geochemical Investigations of the Haney Limestone

Hess and White (1993) examined the water chemistry of six small springs

(Adwell, Blair, Bransford, Collins, Cooper, and Three Springs) located within the Pike

Spring Basin, which emerge from the Haney at the Haney-Big Clifty contact. They compared their results against several medium- and high-volume springs in nearby basins to identify any significant differences. Consistent with the concept of hydrochemical facies, Hess and White (1993) found that their study sites were representative of four water types. They classified the springs by their distinct chemical characteristics: surface waters (sinking streams), Haney Limestone springs, medium-volume base level springs fed by water from the plateau and karst valleys, and regional flow fed by water from mixed water sources.

The Haney springs had the smallest recorded temperature fluctuations, which indicated (Hess and White 1993) that water stored in the overlying Hardinsburg

Sandstone entered the Haney slowly through diffuse flow, allowing it time to near a thermal equilibrium. Analyzed individually, the Haney springs showed no variation by season in carbon dioxide (CO2) partial pressure. Averaging the CO2 calculated partial pressure values of each Haney spring revealed seasonal changes, with levels decreasing more rapidly in the fall. The peak concentrations coincided with the growing season while all the calculated values fell above atmospheric levels—evidence that supports the idea that the primary source of CO2 is the soil. The springs remained highly undersaturated with respect to calcite and dolomite and therefore fell under the rapid dissolution regime in terms of reaction kinetics. However, saturation states did vary dramatically throughout the year. Overall, Hess and White (1993) observed low hardness

17 values, with only small seasonal variation. This led them to conclude the residence time in the perched Haney aquifer must be relatively short.

Chapter 3: Data & Methodology

This research provides the first synthesis of information on karst hydrogeology of the Haney Limestone in south-central Kentucky. In addition, the study characterizes patterns of distribution, morphology, and controls on karst development. This synthesis and characterization provides the context needed to relate cave and karst development in the Haney Limestone to the overall regional and geomorphic history of the south-central

Kentucky karst region. A compilation of existing data obtained from multiple sources has been combined with field work.

3.1 Geographic Information Systems

Using the computer software program ESRI ArcGIS, a Geographic Information

Systems (GIS) database was created for examination of the study area and organization of existing data. A basemap was constructed using several component layers that provide cultural reference features as well as information regarding the topographic, geologic, and hydrologic settings. A polygon layer of Kentucky county boundaries was downloaded from the Kentucky Geological Survey (KGS) website and added to the basemap to provide geographic reference and to define the boundaries of the study area.

Raster coverage files of 7.5 minute series (1:24,000 scale) Digital Elevation Model

(DEM) data for the study area were combined using the Mosaic tool and added to the basemap (KGS). The DEM provides computed elevation values for all points in the coverage area; these values can be extracted and applied to specific points, such as cave entrance locations. Shapefiles of the 7.5 minute Digitally Vectorized Geologic

Quadrangles (DVGQs) were also obtained from the KGS. Selected geologic strata relevant to this study were added to the basemap. Structural contour intervals were also

19 included in the DVGQ shapefiles and added to the basemap. All data layers were projected in ArcGIS using the Kentucky South Zone (feet) State Plane coordinate system.

To aid in analysis of the surface-water component of the hydrogeology of the

Haney Limestone, the high resolution (1:24,000) National Hydrography Dataset was obtained from the United States Geological Survey (USGS). The National Hydrography

Dataset contains a Hydrography dataset with directional flow paths and a Hydrographic

Unit dataset containing surface watershed boundaries. A layer of point features showing the locations of springs in Kentucky was downloaded from the KGS. All spring outlets found within or at the edge of the Haney Limestone were selected and a new layer was created containing only these points. A shapefile of digitized sinkholes for the state of

Kentucky was also downloaded from the KGS. A new layer was created by extracting only the sinkholes developed within the Haney Limestone and overlying Hardinsburg

Sandstone from this shapefile.

A data request was submitted to the Kentucky Speleological Survey (KSS) for all known cave entrance locations and corresponding maps located within the Haney

Limestone in the counties of Barren, Butler, Edmonson, Hart, Logan, and Warren, except for those located within Mammoth Cave National Park. A request for this information was submitted separately through the National Park Service’s permit process. Both requests included entrance locations within 100 meters of Haney Limestone outcrops to account for potential contact imprecision during geologic mapping and potential accuracy issues for locations plotted prior to GPS technology. The KSS database contains sixty- one known cave entrances located within the Haney Limestone and 100 meter buffer zone. After eliminating artificial entrances and entrances clearly not within the Haney

Limestone, forty cave entrances remain. Twenty-six of these entrances are located within the amended three-county study area. Resource managers at Mammoth Cave National

Park provided an ArcGIS layer of point features containing 128 caves within a 100m buffer of the Haney Limestone. Following the same process of elimination, forty-nine cave entrances were identified as likely developed within the Haney Limestone. From the two sources, a total of seventy-five cave entrances were identified within the study areas as likely developed in the Haney.

In addition to the seventy-five known cave entrances in the study area, nine previously undocumented caves were identified by speaking with independent cavers and by field investigation. The locations for six of the nine undocumented caves were added to the GIS database, for a total of eighty-one Haney cave entrances within the study area.

The locations of the other three cave entrances are currently unknown and could not be added to the database. A partial map and short description of Spring Trough Caves 1 & 2

(Figure A4) were featured in the 1991 Guidebook to the Kentucky Speleofest (Haun, ed.,

47, 49) and a lineplot and description of Barnes Smith Cave (Figure A1) were featured in the 1974 National Speleological Society’s Speleo Digest (Ramsey, 73) but the entrance locations could not be obtained.

Using this data layer, it is possible to identify localized sinkhole plains in the

Haney Limestone and possible sources of recharge for subsurface conduits. The point locations of cave entrances in the Haney Limestone, digitized sinkhole data, and geologic coverages allow for the interpretation of general recharge sources for the known caves, and, help to identify other areas where caves would be expected. Some indication of the level of aggressiveness of the solvent forming caves within the Haney can be inferred by

21 determining whether the recharge sources are allogenic or autogenic. Cave entrance locations, sinkholes, and surface streams allow for a description of the spatial distribution of caves and karst development within the Haney Limestone.

3.2 Cave Maps

Of the twenty-six Haney caves within the study area on file with the KSS, the archives contained only three rough sketch maps and three drafted cave maps, all for caves located in Warren County. The sketch maps were made from memory and are not to scale; they are therefore unusable. Of the forty-nine known Haney caves in Mammoth

Cave National Park, twenty-two are mapped. Lineplots were created from the twenty- five existing maps by dividing the passages into straight segments and measuring the length and azimuth of each segment (Deike 1969). These measurements were entered into the cave survey data reduction software program COMPASS as survey data and the lineplots generated were georeferenced and exported as shapefiles, readable by ArcGIS.

The cave lineplots were projected and imported into the GIS database. The CRF archives also contained survey data for two additional caves, Silent Grove Springhouse Cave and

Lulu Mart Cave, although the complete maps have not yet been drafted. Independent caver Steve Miller provided up-to-date survey data for Cub Run Cave. Survey data for these three caves was transcribed from field books and other file formats into COMPASS and these lineplots also added to the GIS database (Table 1).

Rose diagrams weighted by length were generated from the lineplots for each individual cave to show the influence of structure on passage development (Deike 1967;

Deike 1969; Palmer 2007). Passage segment lengths were aggregated by their orientation, summing the lengths of each segment within each division of 10˚ azimuths.

In addition, data for all individual caves were aggregated to produce a rose diagram showing the length-weighted passage orientations for the entire study area. This is used for comparison with the orientation of regional joint sets documented by Deike (1967).

7.5’ Location Lineplot Cave Name County Quadrangle Source Source Sulphur Salamander Edmonson Bristow Fieldwork New Survey Cave Alaska Caverns Edmonson Brownsville Norman Warnell Existing Lineplot Barner’s Mill Cave Edmonson Brownsville Norman Warnell Existing Lineplot Chameleon Springs Edmonson Brownsville Norman Warnell None Cave Sulphur Branch Cave Edmonson Brownsville Norman Warnell None Lycopodium Cave Edmonson Mammoth Cave NPS/CRF None Existing Survey A and E Cave Edmonson Rhoda NPS/CRF Data Backbone Cave Edmonson Rhoda NPS/CRF None Bent Tree Sink Edmonson Rhoda NPS/CRF Existing Map Big Hollow Cave Edmonson Rhoda NPS/CRF None Billy Hollow Cave Edmonson Rhoda NPS/CRF None Bird Spring Cave Edmonson Rhoda NPS/CRF None Black Rock Cave Edmonson Rhoda NPS/CRF Existing Map Blossom Cave Edmonson Rhoda NPS/CRF None Bourbon Cave Edmonson Rhoda NPS/CRF New Survey Bryophyte Cave Edmonson Rhoda NPS/CRF Existing Map Cade Cave Edmonson Rhoda NPS/CRF Existing Map Collie Spring Cave Edmonson Rhoda NPS/CRF None Cricket Falls Cave Edmonson Rhoda NPS/CRF Existing Map Deer Skull Cave Edmonson Rhoda NPS/CRF Existing Map Feather Cave Edmonson Rhoda NPS/CRF Existing Map Fishtrap Pit #1 Edmonson Rhoda NPS/CRF Existing Map Fishtrap Pit #2 Edmonson Rhoda NPS/CRF Existing Map Geranium Cave Edmonson Rhoda NPS/CRF None Good Spring Cave Edmonson Rhoda NPS/CRF Existing Map Hearth Pit Cave Edmonson Rhoda NPS/CRF Existing Map Hearth Spring Cave Edmonson Rhoda NPS/CRF Existing Map Hickory Cabin Cave Edmonson Rhoda NPS/CRF Existing Map Hideout Spring Cave Edmonson Rhoda NPS/CRF Existing Map Hornbeam Spring Edmonson Rhoda NPS/CRF Existing Map Cave Johnson Spring Cave Edmonson Rhoda NPS/CRF Existing Map Koepps Sinking Rat Edmonson Rhoda NPS/CRF None Hole Little Hemlock Cave Edmonson Rhoda NPS/CRF None Little Misty Cave Edmonson Rhoda NPS/CRF None Existing Survey Lulu Mart Cave Edmonson Rhoda NPS/CRF Data Luna Cave Edmonson Rhoda NPS/CRF None Misty Hole Cave Edmonson Rhoda NPS/CRF None Mystery Cave Edmonson Rhoda NPS/CRF None

Peger Cave Edmonson Rhoda NPS/CRF None Pip Cave Edmonson Rhoda NPS/CRF None Ridgetop Cave Edmonson Rhoda NPS/CRF None Salamander Cave Edmonson Rhoda NPS/CRF Existing Map Shockley Pit Edmonson Rhoda NPS/CRF None Silent Grove Existing Survey Edmonson Rhoda NPS/CRF Springhouse Cave Data Spined Spider Cave Edmonson Rhoda NPS/CRF New Survey Split Face Cave Edmonson Rhoda NPS/CRF None Squeeze Cave Edmonson Rhoda NPS/CRF Existing Map Squirrel Hollow Edmonson Rhoda NPS/CRF Existing Map Springhouse Cave Steep Bike Cave Edmonson Rhoda NPS/CRF None Stillhouse Sink Cave Edmonson Rhoda NPS/CRF Existing Map Two Entrance Cave Edmonson Rhoda NPS/CRF Existing Map Webb Cave Edmonson Rhoda NPS/CRF None Zigzag Cave Edmonson Rhoda NPS/CRF None Chalybeate Cave Edmonson Smiths Grove KSS New Survey Double Springhouse Edmonson Smiths Grove Fieldwork New Survey Cave Digital Survey Cub Run Cave Hart Cub Run KSS Data Cave 4 Hart Mammoth Cave KSS None Cave 9 Hart Mammoth Cave KSS None Collins Spring Cave Hart Mammoth Cave NPS/CRF None Tater Cave Hart Munfordville KSS Non Bowling Green Indian Creek Cave Warren KSS None North Bowling Green Miller (Calyx) Cave Warren KSS New Survey North Beaver Dam Creek Warren Bristow KSS New Survey Cave Beaver Dam Creek Warren Bristow KSS None Cave 2 T35 Cave Warren Bristow KSS None Clifty Creek Cave Warren Hadley KSS None Crumps Pit Cave Warren Hadley KSS None Davenport Cave Warren Hadley KSS None Daves Dome Cave Warren Hadley KSS None Hadley Cave #1 Warren Hadley KSS None Hadley Cave #2 Warren Hadley KSS None Hadley Cave #4 Warren Hadley KSS None Haneys Cave Warren Hadley KSS New Survey Justice Cave Warren Hadley KSS None Lewis Chapel Cave #2 Warren Hadley KSS None Lewis Chapel Cave #3 Warren Hadley KSS None Lewis Chapel Cave #4 Warren Hadley KSS None Sawdust Cave Warren Hadley KSS None Skillerns Spring Cave Warren Hadley KSS None Indian Creek Warren Reedyville KSS Map (Honaker) Caves Lick Creek Cave Warren Sugar Grove KSS New Survey Table 1. Known caves in the Haney Limestone located within the study area.

Examining the cave maps themselves yields significant information on passage origin and morphology. The plan view of the caves allows for their categorization by general passage patterns following Palmer’s (2007) classification of branchwork and maze patterns in jointed versus bedded rock. Cross-sectional views of passage shape are used to help determine the zone of development in which the passage was formed and provide clues on subsequent enlargement. They are also used to analyze the initial pathways the water followed and enlarged, as well as to identify constrictions on enlargement. Map profiles and survey data are used to determine the average vertical extent of the caves within the Haney Limestone. Although the level of detail of the existing maps varies, water is consistently noted on the maps where it is present. The maps are used to determine the percentage of caves with actively flowing water and entrances that are spring resurgences.

3.3 Field Methods

From the GIS point locations of the cave entrances, landowner names and addresses were obtained from the county assessor’s offices, and permission to access these caves was asked of the landowners. All accessible caves were visited and the locations of all entrances visited were double-checked using a handheld GPS unit at an accuracy of five meters or less. Some entrance locations had been hand plotted on topographic maps and discrepancies of up to 0.5 km were noted. The locations of undocumented entrances found were also obtained using a handheld GPS unit and added to the cave entrance layer in ArcGIS.

Surveys of six previously unmapped caves were conducted. The caves were surveyed and sketched using Suunto compass clinometer instruments and fiberglass tape

(Dasher 2011). Distance from station to station was read to the nearest 0.1 ft. Frontsight and backsight compass and clinometer readings were taken and were required to agree within two degrees; for stations where either a frontsight or backsight reading was not possible, readings were taken twice from the readable station. The precision of distance measurements and variance between compass and clinometer readings is consistent with survey grade 4 established by the Union Internationale de Spéléologie (Häuselmann

2011).

In order to determine the morphology of a passage, all three map views, including plans, profiles, and cross-sections, were included in the survey and maps. The plan shows overall passage patterns, which allows a determination of whether they are dendritic and winding, angular, or maze-like. This provides information on the influence of structural controls on passage development. The profile is necessary to examine passage gradient, which is important when determining the zone in which passage enlargement occurred. Cross-sections show the passage shape, which is key in determining the zone in which enlargement occurred as each zone produces a very characteristic shape. Cross-sections are also the clearest indicator of passage enlargement at one zone with subsequent enlargement in a different zone, or, the intersection of passages formed at different times in different zones. In addition to the survey and sketches done in the field, observations regarding structure and dissolution features were noted. The orientation and passage relationship to joints and fractures, and, the relationship to bedding planes are important in determining the initial pathways the water exploited. Rills and scallops are vadose solution features, and may indicate the passage was formed entirely above the water table, or that subsequent enlargement occurred after

26 the initial passageway was formed. Breakdown occurs once at least part of the passage is above the water table. For example, a phreatic tube may more closely resemble a canyon passage in the event breakdown occurs leaving a flat bedding plane at the ceiling.

Breakdown can enlarge and alter the shape of a passage significantly after initial enlargement and is noted in all map views. In-cave observations of passage shape, gradient, and relationship to structure all help determine the morphology.

3.4 Geochemical Data

In 1974, Jack Hess published his Ph.D. thesis, Hydrochemical Investigations of the Central Kentucky Karst Aquifer System, examining differences in the chemical parameters of various karst waters to interpret the hydrogeology of the region’s carbonate aquifers. Six of the sites sampled by Hess were spring outlets in the Haney Limestone, located on Flint Ridge: Adwell, Blair, Bransford, Collins, Cooper, and Three Springs.

The springs were sampled twice a month for one year (November 1972 to October 1973) to account for seasonal variations. Temperature, pH, and specific conductance (SpC) were measured in the field and laboratory analyses were conducted to measure ion

2+ 2+ + + concentrations of HCO3̄, Ca , Mg , Na , and K (Appendix B). From these parameters,

Hess calculated total hardness (Hd), saturation indices with respect to calcite (SIc) and

dolomite (SId), and CO2 partial pressure (PCO2).

At the time of Hess’s publication, researchers were just beginning to develop kinetic models of calcite dissolution rates (Berner and Morse 1974; Plummer and Wigley

1976; Plummer et al. 1978); therefore, dissolution rates were not included in his analysis.

Plummer, Wigley, and Parkhurst (1978) account for three forward reactions acting to dissolve the limestone surface:

+ 2+ CaCO3 + H ⇄ Ca + HCO3̄

2+ CaCO3 + H2CO3 ⇄ Ca + 2 HCO3̄

2+ CaCO3 + H2O ⇄ Ca + HCO3̄ + OH ̄ each of these occurring simultaneously. Employing Hess’s (1974) geochemical data from the six Haney springs, average dissolution rates were calculated using the Plummer,

Wigley, and Parkhurst (PWP) equation:

$ %$ * Dissolution Rate = k1"# + k2"#%&'( + k3"#%' – k4"&) "#&'( where k1, k2, and k3 are temperature dependent constants and a denotes the activities of the various species. The forward reaction rate constants were calculated as follows:

,,, log k1 = 0.198 − -

/011 log k2 = 2.84 − -

201 log k3 = -5.86 − (T < 298 K) -

0121 log k3 = -1.10 − (T > 298 K) -

The fourth term, k4, describes an observed decrease in rate due to back-reactions and is dependent on all other terms in the equation. This term is calculated as follows:

5% 8 0 k4 = 70 7/ "#%&'( < + 72"#%' < 56 9:$ ; where K2 and Kc are temperature dependent equilibrium constants (Table 2).

Table 2. Equilibrium constants from White (1988, 121).

Dissolution rates were also calculated using the equations developed by Palmer

(1991). The saturation ratio (C/Cs) is the ratio of bulk fluid concentration of calcium, C, divided by the concentration of calcium at saturation, Cs. Based on experimental data,

Palmer (2007) found that the saturation ratio is approximately equal to the saturation index (SI) through the following relationship:

0.35 C/Cs ≈ (IAP/Kc) where IAP is the ion activity product, calculated by multiplying the activity of calcium and the activity of carbonate, and Kc is the calcite dissociation constant. Using this method of approximation, the saturation ratio was then inserted into the following equation to determine the rate of solutional wall retreat (S) in cm/year:

F S = 20.?@ A (0CD D;) GH

The constant 31.56 is used to convert the result into cm/year, k is the reaction coefficient, n is the reaction order, and ρ is the density of the rock (approximately 2.7 g/cm3 for limestone). Values for both constants k and n are dependent upon the temperature and carbon dioxide partial pressure (Table 3).

Table 3. Values for constants k and n from Palmer (1991).

The values for constants k and n are also dependent upon the saturation ratio (C/Cs). The dissolution rate decreases rapidly as the concentration of calcium approaches saturation.

To account for this change in reaction rate, Palmer derived a second regression equation to calculate the values for constants k and n. When the saturation ratio is greater than the saturation ratio at the transition from low-order to high-order reaction rates, C/Cs >

(C/Cs)t , constants k2 and n2 are used.

While Palmer (1991) used the same experimental data gathered and used by

Plummer and Wigley (1976) and Plummer et al. (1978) to develop rate equations for the dissolution of limestone, the two equations are inherently different. While the Plummer et al. equation provides more information each of the individual reactions occurring simultaneously at the limestone surface, Palmer’s equation fits the experimental data better overall. As calcium concentrations near saturation, dissolution rates calculated using the two different equations will differ more due to these inherent differences.

Chapter 4: Results

4.1 Distribution of Karst Development within the Haney Limestone

The general distribution of karst development within the Haney Limestone was examined by a regional overview of two features indicative of karstification – caves and springs. Sinkholes were not included in the regional overview because they could not be meaningfully displayed at this scale. Figure 4 includes the counties of the study area

(Warren, Edmonson, and Hart) as well as several surrounding counties that have significant Haney Limestone outcrops for a regional view of the spatial distribution of

Figure 4. Distribution of karst features in the Haney Limestone across a 6 county area of south-central Kentucky.

these features. The area covered in Figure 4 includes the more mature karst landscape along the eastern edge of the Mammoth Cave Plateau and extends farther to the west than

31 the study area to include the younger karst landscape where surface drainage is still prominent. A total of ninety-three caves and forty-nine springs were identified within this extent. Karst development is relatively homogeneous throughout the region, and karst features are found throughout outcroppings of the Haney Limestone. Though the younger karst landscape in the western area of Logan County does not contain any Haney springs documented in the KGS database, it is clear when viewing the high-resolution data from the National Hydrography Dataset that streams emerge at the base of the Haney

Limestone in many locations (Figure 5). There are eleven known cave entrances developed in the Haney outcrops of Logan County, as well as numerous sinkholes not shown on the map. Therefore, it can be concluded that karstification occurs throughout the Haney Limestone in south-central Kentucky.

Figure 5. Possible spring resurgences not included in the KGS database. A-H indicate possible springs emerging from the Haney-Big Clifty contact; W-Z indicate possible springs emerging from the Haney Limestone above the contact.

Although karst features occur in the Haney Limestone throughout the region and study area, the known features are not evenly distributed. In the three-county study area of Edmonson, Hart, and Warren Counties, excluding Mammoth Cave National Park, there are thirty-one known cave entrances formed in the Haney Limestone. Twenty-one of these are located in Warren County. There are fifty known cave entrances in the

Haney Limestone within the boundaries of Mammoth Cave National Park (Figure 6).

While it may be that the distribution is related to the spatial variation in the maturity of the karst landscape, it is likely that the concentration of known cave entrances in some areas is due to an exploration bias.

Figure 6. Distribution of karst features in the Haney Limestone in the three county study area of south-central Kentucky. Bowling Green, located in Warren County, is home to Western Kentucky

University (WKU). The Green River Grotto is a local chapter of the National

Speleological Society as well as a student organization affiliated with WKU. The Green

River Grotto has a long history of active cave exploration and study in the area and has likely spent the most time in the immediate area – Warren County. The Cave Research

Foundation (CRF) was founded in 1957 by explorers of the Mammoth Cave System. The

CRF was historically located within Mammoth Cave National Park and is now located directly outside the park’s boundary. Both CRF members and park service science and resource managers have been actively documenting, exploring, and researching caves in

Mammoth Cave National Park for over fifty years. Given that the highest concentrations

34 of known cave entrances occur within the areas surrounding two caving and exploration organizations, it is most probable that variations in the distribution of known caves within the Haney Limestone is due to exploration bias. An important implication of this is that there are likely to be more caves in the Haney Limestone to be found in areas that have been less thoroughly examined.

4.2 Caves and Karst Development in the Haney Limestone

Thirty-two lineplots were compiled and added to the GIS database. Of the thirty- two caves in the study area with existing maps and known entrance locations, only nine are over 100 meters in length, including: Alaska Caverns, Barner’s Mill Cave, Beaver

Dam Creek Cave, Chalybeate Cave, Cub Run Cave, Honaker Cave, Lulu Mart Cave,

Miller (Calyx) Cave, and Silent Grove Springhouse Cave. The aggregate sum of the length of all cave passages amounted to 5,369.7 m and these nine caves account for

4,716.8 m, or 88%, of the total. While each of these is subsequently examined in detail,

Surveyed Hydrologic Possible Recharge Cave Name Entrance Setting Length (m) Characteristics Sources Incised Valley Sinkhole; Sinking Cub Run Cave 2105.7 Spring Resurgence Wall Stream; Diffuse Flow Incised Valley Sinking Streams; Lulu Mart Cave 871.2 Spring Resurgence Wall Diffuse Flow Miller (Calyx) 425.6 Sinkhole Spring Resurgence Sinkhole; Diffuse Flow Cave Beaver Dam Incised Valley 292.7 Spring Resurgence Sinkhole; Diffuse Flow Creek Cave Wall Incised Valley Honaker Cave 245.4 Spring Resurgence Sinkhole; Diffuse Flow Wall Incised Valley Sinkhole; Sinking Chalybeate Cave 241.2 Stream Passage Wall Stream; Diffuse Flow Barner’s Mill Incised Valley Sinkhole; 206.3 Spring Resurgence Cave Wall Diffuse Flow Incised Valley Sinkhole; Sinking Alaska Caverns 188.7 Stream Passage Wall Stream; Diffuse Flow Silent Grove Incised Valley Sinkhole; Sinking 133.7 Spring Resurgence Springhouse Cave Wall Stream; Diffuse Flow Table 4. Summary characteristics of the nine caves over 100 m in length.

Table 4 provides a summary of several important attributes of the caves. Maps of each of the nine caves over 100 meters in length were created to examine the relationship between the surface topography and geology and underlying cave passages (Figures 7-

15). The cave lineplot overlays show the relationship between the surface topography and geology; however, they do not reveal much about the structural controls. Rose diagrams were also created for each of the caves with a total length of over 100 m (Figure

16). In addition, a rose diagram for the aggregated cave data was created to look at trends for the entire region.

4.2.1 Alaska Caverns

The total surveyed length of Alaska Caverns (Figure 7) is 188.7 m. The entrance is located in a hillside above the Green River, very close to the Hardinsburg-Haney contact. The entire cave is covered by the insoluble Hardinsburg Sandstone caprock.

The entrance to Alaska Caverns is not currently a spring, but likely was a resurgence site in the past. A majority of the cave is active stream passage; water exits the cave at a spring at the foot of the hillside beneath the current entrance. Alaska Caverns is a single conduit cave, with development concentrated along a single major axis, between 10˚ and

20˚ (190-200˚) azimuths (Figure 16a). There is a smaller axis perpendicular to this at

100-110˚ (280-290˚). A cluster appears between 120˚ and 190˚ (300-10˚) and on the other side of the major axis between 20-50˚ (200-230˚). No passage development occurs between 50˚ and 90˚ (230-270˚). While the results of the rose diagrams may be indicative of strong joint-control, the cave is also oriented downdip. A large sinkhole and a small tributary on the east side of Alexander Creek lie to the southeast of the cave, in the upstream direction. These were not investigated, but water draining into the

Figure 7. Alaska Caverns lineplot overlay showing surface topography and geologic outcrops. depression through the Hardinsburg or sinking into the tributary as it flows across the

Haney are potential sources of recharge for the cave. Both are directly updip from the cave with respect to the local geologic structure. Some water may also reach the cave through the overlying Hardinsburg Sandstone, though the relative proportions of these contributions are not known.

4.2.2 Barner’s Mill Cave

The entrance to Barner’s Mill Cave is a spring resurgence located just above the

Haney-Big Clifty contact (Figure 8). The cave is on the upthrown side of a fault above a drainage channel formed along the fault line. Barner’s Mill Cave is 206.3 m long and

Figure 8. Barner's Mill Cave lineplot overlay showing surface topography and geologic outcrops. does not pass beneath the Hardinsburg caprock. The rose diagram shows a large peak oriented between 10-20˚ (190-200˚) and substantial length between 20-30˚ (200-210˚) azimuths (Figure 16b). Significant passage development also occurs between 40 and 80˚

(220-260˚), with a majority of passage length oriented between 40 and 50˚ (220-230˚). A single minor axis is found between 150˚ and 160˚ (330-340˚) and a minor cluster between

100˚ and 140˚ (280-320˚), each of these roughly perpendicular to the two major peaks.

Barner’s Mill Cave consists of a single, active conduit that exits in the wall of a tributary valley. A spring flows from the entrance. The main source of recharge is likely to be a large sinkhole located to the northeast directly on the mapped fault line. Other sources of recharge may include allogenic water sinking into the Haney Limestone at the

38 edge of the Hardinsburg Sandstone and diffuse flow through the Hardinsburg Sandstone.

Water flows in the updip direction through Barner’s Mill, indicating that fractures and hydraulic gradient, rather than geologic structure, are the dominant influences on conduit development in this instance. Results of the rose diagram support the conclusion that joints and fractures are influential controls, showing major passage development along two principal axes of orientation.

4.2.3 Beaver Dam Creek Cave

A stream flows through the single main passage of Beaver Dam Creek Cave and resurges from an entrance located in the hillside above a creek bed (Figure 9). The cave passages total 292.7 m in length and pass beneath the Hardinsburg Sandstone a short distance into the cave. Results of the rose diagram show a majority of the passage length in Beaver Dam Creek Cave is formed between 20˚ and 40˚ azimuths (200-220˚), with a major orientation between 30-40˚ (Figure 16c). A minor peak is oriented between 60˚ and 70˚.

A small sinkhole formed in the overlying Hardinsburg Sandstone is located just beyond and upstream from the end of the surveyed passage and is likely the main source of recharge for Beaver Dam Creek Cave. Another small sinkhole and a very large sinkhole have formed in the Haney Limestone nearby and may contribute to the water resurging at the cave’s entrance. There are several high domes located off the sides of the main passage that likely conduct water into the cave during heavy precipitation events. The domes have sandstone gravel at their base and are located under the sandstone caprock, suggesting diffuse flow through the Hardinsburg is also a contributing source of recharge. Beaver Dam Creek Cave also contains a section of maze passages

Figure 9. Beaver Dam Creek Cave lineplot overlay showing surface topography and geologic outcrops. beneath a thin layer of the Hardinsburg Sandstone near the entrance which may be the result of diffuse flow concentrated along a fracture zone. Water moving through Beaver

Dam Creek Cave flows downdip; however, prominent jointing is visible in the cave and the results of the rose diagram could be indicative of strong joint control.

4.2.4 Chalybeate Cave

Chalybeate Cave (Figure 10) is currently mapped at 241.2 m; however, a detailed survey is currently still in progress and, according to various reports, the cave is probably at least double this length. The entrance to Chalybeate Cave is at the base of the hillside

Figure 10. Chalybeate Cave lineplot overlay showing surface topography and geologic outcrops. along a creek bed. The entrance itself is not a spring resurgence but the cave does contain a flowing stream. The cave passage follows the surface topography and wraps around the base of the hillside. Spring Trough Caves 1 & 2, located in the bottom of a sinkhole less than a half of a kilometer away, were dye traced to Chalybeate Cave. Water flowing into this sinkhole and through these two caves is certainly one source of recharge; the sinking stream at the head of the stream valley, water from the small sinkhole plain formed in the overlying Hardinsburg Sandstone to the south, and diffuse flow through the Hardinsburg are other possible recharge sources.

Nearly all passage length in Chalybeate Cave is found between 40˚ and 140˚

(220-320˚), with two major peaks at 50-60˚ (230-240˚) and 80-90˚ (260-270˚) azimuths

(Figure 16d). One small cluster between 120˚ and 150˚ (300-330˚) is nearly perpendicular to the major peak between 50˚ and 60˚ (230-240˚). The survey of

Chalybeate is still in progress and a reportedly large portion of the cave remains unsurveyed. At this stage, conclusions cannot be made about the influence of jointing or regional dip.

4.2.5 Cub Run Cave

Cub Run Cave is the longest known cave in the Haney Limestone with a current total surveyed length of 2105.7 meters (Figure 11). The entrance to Cub Run Cave is a spring resurgence perched on top of a shale layer commonly noted at the base of the

Haney. The cave passages are rather concordant with the surface topography and are covered by the Hardinsburg Sandstone and Glen Dean Limestone. Chalybeate Cave spans the entire thickness of the Haney Limestone and in one upstream section of the cave extends through the overlying Hardinsburg Sandstone to reveal the Glen Dean

Limestone on the ceiling. The rose diagram for Cub Run Cave contains segments of sizeable length distributed throughout each division of 10˚ azimuths; however, a few preferential orientations are evident (Figure 16e). A clear peak occurs between 40˚ and

50˚ (220-230˚), part of a larger cluster between 20˚ and 50˚ (200-230˚). Perpendicular to this is a minor peak between 120˚ and 130˚ (200-210˚). The second largest peak occurs between 60˚ and 70˚ (240-250˚) with a small corresponding perpendicular cluster between 150 and 170 (330-350˚). Another cluster representing considerable passage length is found between 80-110˚ (260-290˚).

Figure 11. Cub Run Cave lineplot overlay showing surface topography and geologic outcrops.

Cub Run Cave primarily consists of a single major trunk passage although there are some side passages and smaller meanders of the main stream passage that deviate from the main trunk. The side passages appear to be associated with paleo-springs located at higher elevations along the valley wall. The meandering stream passages that diverge from the massive trunk developed at different levels over time. The sinkhole and stream valley to the northeast are likely the main sources of recharge for the cave. An active vertical shaft, formed beneath the Glen Dean Limestone and Hardinsburg

Sandstone, is located where the cave makes a sharp 90˚ bend. Water flows along the strike until reaching this bend then follows the dip of the beds. While this may be due to

43 the relationship with water table conditions, the orientation of the passages may also be a result of fracturing.

Cub Run Cave is, by far, the longest known cave in the Haney Limestone and contains the largest passage. The great length and passage size may be a result of differences in three conditions: water chemistry, the size of the catchment area and thus volume of discharge, or the length of time the cave has been developing. It is unlikely that the water chemistry is very different from that of other caves in the study area as the geologic conditions of the area are almost identical. In addition, the solutional capacity of the water at the six springs tested by Hess (1974) was near maximum at all locations

(Tables 4 and 5). The apparent catchment area for Cub Run Cave, determined by surface topography based on drawing surface divides along ridges of Hardinsburg Sandstone that surround the upstream reaches of the cave, is approximately 1.65 km2. The catchment area of Lulu Mart Cave was determined, using dye tracer studies, to be 0.72 km2 (Ryan and Meiman 1994). Ryan and Meiman indicated that it is possible the catchment area for

Lulu Mart may have been nearly double in the past, 1.53 km2, but that water was pirated by Big Spring in the recent past. Even if this is the case and it is assumed that the catchment areas were roughly equal in area, Cub Run Cave surpasses Lulu Mart in both length and volume, and particularly the size of the main passages which approach nearly fifteen meters in places. Development of Cub Run Cave thus likely began at an earlier time than other caves in the study area to have produced such extraordinary length and volume compared to other Haney caves. The location of Cub Run Cave is one of the farthest to the east and the farthest north of all caves in the study area. It is located in the most maturely developed karst area of the escarpment, where downcutting of the Green

River first breached the Big Clifty Sandstone. This conclusion is therefore consistent with the geomorphic history of the region, though dependent on future groundwater tracing to confirm the drainage area contributing to Cub Run Cave, and whether additional flow might be passing beneath the Hardinsburg from a larger area.

4.2.6 Honaker Cave

The entrance to Honaker Cave (Figure 12) is located in the bluffs above a creek bed and a spring resurges from breakdown just below the entrance. Honaker Cave is

245.4 m in length and extends to a sinkhole which breaches the overlying Hardinsburg

Sandstone and reveals the Haney Limestone below. Several tall domes beneath the

Hardinsburg Sandstone, up to fifteen meters high, expose large sandstone blocks lodged

Figure 12. Honaker Cave lineplot overlay showing surface topography and geologic outcrops.

45 at the tops. The cave is one main passage with a flowing stream that appears and disappears throughout the cave. One section of the cave contains a long crawlway with standing water. The sinkhole located at the upstream terminus of the passage is likely the main source of recharge, concentrating diffuse flow through the Hardinsburg Sandstone to the Haney Limestone.

Honaker Cave has two major axes along which passage development occurred; one oriented between 10˚ and 20˚ (190-200˚) azimuths and another nearly perpendicular axis oriented between 90˚ and 100˚ (270-280˚) azimuths (Figure 16f). In addition there are two minor axes, also perpendicular to each other, one found between 30˚ and 40˚

(210-220˚) and another between 120-130˚ (300-310˚). The passages show no correlation with the dip of the beds; fracturing and hydraulic gradient dominate the development of

Honaker Cave.

4.2.7 Lulu Mart Cave

Lulu Mart Cave (Figure 13) is the second longest known cave formed in the

Haney Limestone within the study area with a total surveyed length of 871.2 m. The entrance is a spring resurgence located in the side of a creek bed. Lulu Mart is rather unique among Haney caves because it has two major stream passages which converge to form a single main passage leading to the entrance. The two stream passages do not cut back beneath the ridge but rougly parallel the two drainage channels along the side of the ridge. The passages are formed mostly beneath the Glen Dean Limestone and

Hardinsburg Sandstone. Results of the rose diagram show that the most prominent orientation of passage development in Lulu Mart Cave is between 140˚ and 160˚ (320-

340˚) azimuths (Figure 16g). These two peaks are part of a cluster ranging from 120-

170˚ (300-350˚). Perpendicular to this cluster is another cluster with the largest peak between 40˚ and 50˚ (220-230˚). Another major pair of perpendicular peaks is located between azimuths 0-20˚ (180-200˚) and 90-110˚ (270-290˚). Water flowing through the

Figure 13. Lulu Mart Cave lineplot overlay showing surface topography and geologic outcrops. northern branch of the cave and out the entrance flows updip. The southern branch of the cave roughly parallels the strike. Fractures and hydraulic gradient over-ride the dip and dominate passage development in Lulu Mart Cave.

From a series of dye traces, Ryan and Meiman (1994) found that the current catchment area for Lulu Mart Cave is 0.72 km2. The groundwater basin could be divided into two sub-basins, each feeding one of the two major conduits. They determined that the sub-basin feeding the southern branch is 0.26 km2 and the sub-basin feeding the

47 northern branch is 0.46 km2. Dye traces from nine sink points accounted for 65% of the discharge at the cave entrance. They speculated that the remaining 35% of discharge may be from diffuse flow seeping through the overlying rock strata.

4.2.8 Miller (Calyx) Cave

The collapsed sink which contains the entrance to Miller (Calyx) Cave actually contains two cave entrances (Figure 14). Water flowing out of the entrance of Miller flows as a surface stream across the bottom of the sinkhole for approximately twenty- seven meters, and flows into another large cave entrance which becomes impassable after only fifteen meters. The total length of these two caves together is 425.6 m. The rose diagram shows the most preferred orientation of passage development in Miller (Calyx)

Figure 14. Miller (Calyx) Cave lineplot overlay showing surface topography and geologic outcrops.

Cave is between 100˚ and 130˚ (280-310˚), with the largest peak between azimuths 120˚ and 130˚ (Figure 16h). There are two small peaks perpendicular to this orientation between 10˚ and 30˚ (190-210˚). Another pair of perpendicular axes occurs at 80-90˚

(260-270˚) and 170-180˚ (350-360˚). A third perpendicular pair of axes is found from azimuths 50-70˚ (230-250˚) and 140-150˚ (320-330˚). The cave follows the strike; however, a fault is located just south of the cave and the passages may follow related fractures parallel to the fault. Development occurs along prominent joints visible in the passage ceilings.

There is only one major side passage in Miller Cave, which is almost thirty meters in length and ends at the edge of the Hardinsburg caprock. The main passage ends at the edge of another large sinkhole and surface debris is found in this part of the cave.

Recharge most likely comes from the large sinkhole at the end of the main passage.

Allogenic water also sinks as it flows off the edge of the Hardinsburg and onto the

Haney, and diffuse flow is possible through cracks or pore spaces in the Hardinsburg.

4.2.9 Silent Grove Springhouse Cave

Silent Grove Springhouse Cave (Figure 15) totals 133.7 m in surveyed length.

The entrance to the cave is a spring resurgence perched on a relatively thick layer of shale at the base of the Haney Limestone, just above the Big Clifty Sandstone. The cave is a single main stream passage and the floor of the passage is interbedded layers of shale and sandstone. Numerous prominent joints are visible in the ceiling of the passage. Silent

Grove Springhouse is not formed beneath the insoluble Hardinsburg caprock; rather it follows the side of the ridge along a drainage channel. The most preferential orientation

49 of passage development in Silent Grove Springhouse cave is between 160˚ and 170˚

(340-350˚) azimuths (Figure 16i). There is a small length of passage formed

Figure 15. Silent Grove Springhouse Cave lineplot overlay showing surface topography and geologic outcrops. perpendicular to this, between 70˚ and 90˚ (250-270˚). Major passage development occurs along two other axes, one between 20˚ and 30˚ (210-220˚) azimuths and another between 40˚ and 50˚ (230-240˚) azimuths. These are relatively perpendicular to the largest peak. There is no relationship between passage development and regional dipping of the beds; development of Silent Grove Springhouse is dominated by local fracturing.

Silent Grove Springhouse Cave likely receives recharge from the sinkhole located approximately 200 m from the end of the cave. Water may also sink into the Haney

Limestone from the drainage channel and as it flows off the Hardinsburg Sandstone along the edge of the ridge.

4.2.10 Regional Cave Trends

The rose diagram presented in Figure 16j contains the combined lengths of all caves in the study area, divided by 10˚ azimuths. The most prominent orientation of passage development occurs between azimuths 10-50˚ (190-230˚), with a nearly perpendicular group occurring between azimuths 80-110˚ (260-290˚) and 120-130˚ (300-

310˚). One more pair of perpendicular axes is located at azimuths 60-70˚ (240-250˚) and

150-160˚ (330-340˚).

Figure 16. Length-weighted rose diagrams comparing individual cave passage orientations and the regional trend for all caves. 52

Figure 16 (continued).

4.3 Hydrology and Dissolution in the Haney Limestone

There are forty-one Haney caves in the study area with enough information available to determine whether or not they are hydrologically active year-round. Of these forty-one caves, twenty-seven, or 65.9%, are perennial spring resurgences or contain perennial stream passages. All of the major caves over 100 m in length are included in this category. Six (14.6%) of the dry caves are small, single-room collapses with

53 entrances formed in the overlying Hardinsburg Sandstone. The eight remaining caves account for 19.5% of the total and are abandoned conduit systems.

The six springs sampled by Hess (1974) were sampled twice a month for one year. Using the rate laws of both Plummer et al. (1978) and Palmer (1991, 2007), dissolution rates were calculated for each sample taken at each spring, then compiled.

The average, median, and range of dissolution rates calculated using Plummer et al.

(1978) are presented in Table 5. The median was calculated in addition to the mean to account for any skewedness due to outliers. While the ranges for each of the springs are more variable, the mean and median dissolution rates are fairly consistent.

Mean Median Range Haney Spring (mm/yr) (mm/yr) (mm/yr) Adwell Spring 1.0 1.1 0.4 – 1.3 Blair Spring 1.2 1.2 0.8 – 1.3 Bransford Spring 1.0 1.2 -0.2 – 1.4 Collins Spring 0.9 1.0 0.5 – 1.2 Cooper Spring 1.1 1.1 0.9 – 1.3 Three Springs 1.1 1.2 0.6 – 1.3 Table 5. Dissolution rates of 6 Haney springs, calculated using the Plummer et al. (1978) equation and water chemistry data collected by Hess (1974). Negative rates predict over-saturation with respect to calcite.

Dissolution rates calculated using Palmer’s (1991, 2007) equation are presented in

Table 6. Overall, the results of the two different rate calculations are generally in agreement; although those calculated using Palmer’s equation tend to be slightly lower than those calculated using the Plummer et al. (1978) equation. Due to differences inherent in the equations, the closer the water is to equilibrium, the more the rates of dissolution differ. Regardless of which equation is used, water from all the Haney springs sampled emerges from the aquifer undersaturated.

Mean Median Range Haney Spring (mm/yr) (mm/yr) (mm/yr) Adwell Spring 0.4 0.4 0.0 – 0.9 Blair Spring 0.6 0.7 0.1 – 1.0 Bransford Spring 0.6 0.7 0.0 – 0.9 Collins Spring 0.2 0.2 0.0 – 0.5 Cooper Spring 0.4 0.4 0.2 – 0.8 Three Springs 0.6 0.7 0.1 – 0.9 Table 6. Dissolution rates of 6 Haney Springs, calculated using Palmer's (1991) equation and water chemistry data collected by Hess (1974).

Chapter 5: Discussion

5.1 Recharge Sources

In some parts of the Mammoth Cave Plateau within the study area, erosion of the protective Hardinsburg caprock leaves the Haney Limestone exposed and small, localized sinkhole plains have developed. Such sinkholes on the karst land surface provide the direct input sources for autogenic recharge entering the flow system. Allogenic recharge occurs as water flows over the non-karstic Hardinsburg Sandstone and typically dissolves the underlying Haney Limestone on contact as it sinks underground.

The overall pattern of ground water flow and conduit development is closely related to recharge types (Palmer 1984; 1991; 2007). While most of the caves developed in the Haney Limestone consist of only a single major conduit, which Palmer states may be a rudimentary form of any of the branchwork or maze patterns, surface features of the surrounding landscape provide insight into the type of recharge that formed the cave passages. Distinct input locations such as sinkholes and sinking streams in drainage channels are evident in most of the maps of Haney caves over 100 m in length. Palmer also suggested that most of the larger single-passage caves form from sinking streams and are included in the category of branchwork caves. Branchwork caves result from direct, autogenic input sources such as sinkholes and sinking streams. Lulu Mart Cave, the second longest cave formed in the Haney Limestone within the study area, exhibits a clear branchwork pattern and dye tracer results from Ryan and Meiman (1994) confirm that a majority of the recharge comes from sinking streams.

Diffuse flow through pores and fractures in the Hardinsburg Sandstone is likely, in some cases, to contribute some proportion of recharge to caves developed in the Haney

Limestone. As Ryan and Meiman (1994) estimated in their investigation of Lulu Mart

Cave, direct input from autogenic sources accounted for 65% of the total discharge at

Lulu Mart Spring, while 35% was suspected to be from diffuse infiltration through the

Hardinsburg Sandstone or from stream loss through the Hardinsburg. The short side passages that are present in some of the longer Haney caves often lead to relatively tall

(limited by the thickness of the formation) solution domes, some of which have sandstone gravel or breakdown at their base. These are likely located beneath cracks in the overlying Hardinsburg Sandstone or at the edge of the Hardinsburg-Haney contact where allogenic water enters the underground flow system. This is consistent with the characteristics of branchwork caves as these can be considered the “rudiments of tributaries” (Palmer 2007). While direct, allogenic sources of recharge appear to dominate conduit development in the Haney Limestone, the caves are likely a result of a combination of recharge types.

5.2 Structural Controls

All of the caves over 100 meters in length show development along one or two major axes, with major or minor development along corresponding perpendicular axes

(Figure 16). This supports the idea that caves formed in the Haney Limestone have significant joint control. In many of these caves, joints are visible in the ceilings of the passages and enlargement clearly occurred along these joints. Deike (1967) described the joint sets of the region as discontinuous, occurring en-echelon (a grouping of short, parallel or sub-parallel fractures). Deike also found that the joints did not typically span multiple bedding planes. Water flows along openings created by joints, using bedding plane partings to connect the discontinuous joints. The greatest development occurs at

57 the intersection between joints and bedding plane partings (Deike 1967). Joints are a major control on passage development within the Haney Limestone, as are bedding planes which connect discontinuous joints and provide favorable zones of enlargement.

Another significant control on passage development is the presence of insoluble rock at the base which creates the perched carbonate aquifer (Figure 16). Shale is common at the base of the Haney Limestone; where the shale is not present the underlying Big Clifty

Sandstone limits downward solutional enlargement in the same way. Water is perched on these layers until it reaches a resurgence, which is commonly found in the side of a valley wall.

Joint

Bedding Plane

Big Clifty Sandstone

Figure 17. Main passage in Lick Creek Cave showing the three major structural controls on passage development: joints, bedding planes, and insoluble bedrock floors. The Big Clifty Sandstone forms the floor of Lick Creek Cave. Photo by Ben Miller. Data from all thirty-two Haney caves in the study area included in the aggregate rose diagram revealed a regional trend. The highest occurrence of passage length is

58 located between azimuths 10-50˚ (190-230˚), with a nearly perpendicular group occurring between azimuths 80-110˚ (260-290˚) and 120-130˚ (300-310˚). Two other perpendicular peaks occur at azimuths 60-70˚ (240-250˚) and 150-160˚ (330-340˚).

These findings are consistent with Deike’s (1967) analysis of regional joint sets in the

Mammoth Cave area (Figure 18). Deike measured the orientation of joints in outcrops of

Figure 18. Length-weighted rose diagram showing the orientations of all Haney caves within the study area with Dieke's (1967) findings overlain in grey. all strata exposed throughout his study area, which included the low-lying sinkhole plain as well as the uplands of the Mammoth Cave Plateau. He identified two major joint sets between 10-50˚ (190-230˚) and 100-120˚ (280-300˚) azimuths. He also identified two minor sets at 60-70˚ (240-250˚) and 150-160˚ (330-340˚) azimuths. Passage orientations of Haney caves match these regional joint sets very closely. The only variation between the two findings is a lack of development in Haney caves between 110-120˚ azimuths and a larger spread on either side of this sector. This consistency indicates that caves

59 developed in the Haney Limestone are highly joint controlled, in contrast to the major caves below in the Ste. Genevieve and St. Louis Limestones with which bedding planes exert influence (Dieke 1967).

Although the rose diagrams were generated from cave passage data rather than from fractures themselves, it is important that the aggregate results are in agreement with

Deike’s findings. While some diagrams were generated strictly from survey data, others were generated by hand-drawing lines directly along the middle of straight segments of passage for caves with existing maps. Differences in the methods used to determine passage trends appear to be negligible as the results of this rose diagram are consistent with the findings of Deike (1967) in his research on the orientations of regional major and minor joint sets.

Sasowsky & White (1994) described what they term ‘Cumberland style caves’ as caves formed parallel to valley walls along fractures enlarged by stress release from the removal of overlying rock. The main conduits of these caves tend to parallel the walls of reentrant valleys and follow even minor variations in the surface topography. Due to stress release, fractures are more numerous along valley walls and become fewer along ridgetops where less rock has been removed. Importantly, Sasowsky and White noted that the increased size of openings along these fractures is a geologically young phenomenon, contrary to previous ideas regarding caves developed along joints and fractures.

While, based on the data, caves in the Haney Limestone considered in this analysis are not preferentially formed along the downdip side of the valley walls, as in

Sasowsky and White’s (1994) Cumberland style caves, Haney caves do in some cases

60 exhibit other defining characteristics of caves formed along stress release fracture systems. The Haney caves shown in Figures 6-14 are all fairly concordant with the valley walls and topographic surface features. Passages tend to terminate as the overburden thickens, consistent with the reduced frequency of fractures as distance from the valley increases. Miller (Calyx) Cave is the only major cave formed in the Haney

Limestone whose entrance is not located in the side of an incised valley wall; the cave entrance is located in a sinkhole.

5.3 Hydrologic Surface and Subsurface Relationship

The more extensive caves in the Haney Limestone function as more direct drainage routes, at a more favorable gradient, than their surface counterparts. In many cases, surface flow is directed underground through sinking streams in tributary drainage valleys and cuts through the sides of ridges to reemerge along the main stream bed.

The hydrology of the Haney Limestone plays an important role in the regional hydrology of south-central Kentucky and is part of an integrated system of surface and subsurface drainage. As Ryan and Meiman (1994) noted is typical of the Hilly Country of the Mammoth Cave Plateau, surface and subsurface flow often alternate along the course of a single stream valley. Spring water resurges from Lulu Mart Cave to form the headwaters of the Dry Prong of Buffalo Creek. This water then flows along the streambed over the top of the insoluble Big Clifty Sandstone and sinks into cracks in the sandstone and bare Girkin Limestone. The same water is found flowing through Buffalo

Creek Cave and Fort’s Funnel Cave and ultimately resurges at Buffalo Spring along the

Green River (Ryan and Meiman 1994; Sides and Ryan 1996). Water emerging from spring resurgences in the Haney Limestone will in many cases enter the principal

61 carbonate aquifer of the Girkin, Ste. Genevieve, and St. Louis Limestones. The typically undersaturated water from the Haney springs flows off the Big Clifty Sandstone and dissolves the underlying carbonate strata, contributing to the formation of vertical shafts that commonly intersect the passages of the Mammoth Cave System. A Haney spring is perched above the sinkhole collapse entrance to Salts Cave and water from the spring flows directly into the entrance.

5.4 Relative Age

Average dissolution rates calculated using the rate expression of Plummer et al.

(1978) ranged from 0.9-1.2 mm/yr. among the six Haney springs and the median rates ranged from 1.0-1.2 mm/yr. Rates of wall retreat by solution under typical karst groundwater conditions average between 0.1-1 mm/yr (Palmer 1991). The rates calculated for each of the six Haney springs are all near the upper limits of dissolution rates in natural environments. While the geochemistry of the solution indicates the water is capable of rapid dissolution, the higher dissolution rates may be due to a number of different factors. It may be that the water has a very short residence time in the aquifer; however, that these springs are perennial, flowing even during times of drought, seems to contradict this idea. The presence of sediment may impede dissolution by shielding the walls while clean-washed passage walls without sediment interference may also result in higher rates of dissolution (Palmer 1991). Many of the Haney caves are formed at the base of the unit where water is perched on top of the underlying Big Clifty Sandstone or the shale layer found near the bottom of the Haney Limestone. It is likely that the water flows over these insoluble layers without diminishing in its capacity to dissolve, therefore

62 yielding higher dissolution rates. There is no evidence to support the presence of significant sources of acidity other than CO2.

Nearly two-thirds of the caves developed in the Haney Limestone within the study area are spring resurgences or contain actively flowing stream passages. Some of the

Haney caves are abandoned, relict features of the landscape, but a majority of Haney caves are part of the current hydrological configuration of the landscape. The caves are currently functioning to drain water from the landscape through more efficient means than surface drainage. That a majority of caves in the study area are hydrologically active, that the water resurging from the springs is capable of high dissolution rates, and that the caves are developed along potential stress release fractures all supports the idea that caves formed in the Haney Limestone are geologically recent phenomena.

Chapter 6: Conclusions & Future Research

6.1 Conclusions

Most of the caves of the study area developed in the Haney Limestone are single- conduit caves that receive water through direct, allogenic sources. While direct allogenic recharge appears to be the dominant source of conduit development in the Haney

Limestone, the caves likely receive water from a combination of discrete and diffuse recharge types. In all cases, the caves appear to receive water from highly localized sources typically serving in the function of valley drains.

Joints are the most dominant control on passage development in the Haney

Limestone within the study area; when considered together the orientation of these joints is consistent with the orientation of regional joint sets. There is evidence that the spaces along these joints in some cases may have been enlarged by stress release fracturing.

Bedding planes and the presence of insoluble rock at the base of the Haney also exert significant control on conduit development in the Haney Limestone. The strong influence of joints on conduit development in the Haney Limestone contrasts with the major caves of the St. Louis, Ste. Genevieve, and Girkin Limestones below, in which bedding planes are a dominant influence (Deike 1967).

The hydrology of the Haney Limestone plays an important, if localized, role in the regional hydrology of south-central Kentucky; integrated into the current system of surface and subsurface drainage of the regional karst landscape. Evidence supports the idea that caves of the Haney Limestone are, geologically, relatively recent phenomena. A majority of the caves in the study area are hydrologically active, the water resurging from the sampled springs is undersaturated with respect to limestone, and the caves are

64 developed along potential stress release fractures associated with small, relatively young valleys.

6.2 Future Research

While synthesizing and adding to the currently understood nature of the hydrogeology of the Haney Limestone with the study area, this research provides a strong foundation for future studies and identifies many additional questions for study. Very little groundwater tracing work has been down within the Haney, multiple dye tracer studies following the model of Meiman and Ryan (1994) are needed to positively identify the catchment areas and recharge sources for the individual caves. A tracer study for the area surrounding Alaska Caverns, for example, could help identify whether the nearby sinkhole and/or the tributary channel contribute water to the cave. There are two sinkholes formed in the Haney Limestone in the vicinity of Beaver Dam Creek Cave that could potentially conduct water to the cave in addition to the sinkhole immediately upstream of the surveyed passage. Dye tracing has already connected Spring Trough

Caves 1 & 2 to Chalybeate Cave; however, the location of these two caves is not documented. Location of these caves and additional dye traces are needed to delineate the catchment area and understand the complete hydrology of the cave drainage system.

The extent and hydrogeologic characteristics of the Cub Run Cave, in the study area the most significant Haney Cave, and its catchment area are unknown. Important to future research of the Haney Limestone and the anomalous Cub Run Cave is the delineation of the catchment area in order to further understand the nature of this karst flow system. Delineating the size of the catchment area gives evidence useful in determining the relative age of Cub Run Cave. Cub Run Cave is north of the Green

River and the elevation of the spring resurgence entrance is about 200 m. The highest level passages within the Mammoth Cave System are at approximately this same elevation. The origin and development of Cub Run Cave may be tied to the incision of the Green River, like the Mammoth Cave System, and therefore began contemporaneously with the highest levels of Mammoth. However, it is also possible that Cub Run formed at a different time due to differing geomorphological and hydrological conditions. Sequential ordering of these events broadens the current understanding of the region’s hydrogeologic history.

Defining the catchment area for Cub Run Cave may also lead to new research opportunities and provide important information for the landowners. The current owners,

Judy and Terry Schieble, purchased all land overtop the known extent of Cub Run Cave.

The Schiebles commercialized the cave with as little impact to the natural system as possible. They are conscious of the impacts of farming and livestock on the cave’s water quality and chose not to conduct these practices on their land. Such a study at Cub Run

Cave will provide fundamental information for subsequent research on karst water quality and chemistry in an area apparently unaffected by agricultural pollutants.

More quantitative work is also needed in order to better understand the morphology of the individual caves. While many of the caves exhibit apparent floodwater features, such as blind fissures, little is known about fluctuations in discharge and frequency of flooding. A monitoring program set up to include discharge measurements and readings from a pressure transducer could yield the information needed to further describe the hydrologic conditions of caves in the Haney Limestone and determine the influence of floodwater conditions on cave morphology.

The integrated hydrologic relationship between the Haney Limestone underlying carbonate strata also needs further investigation in order to more fully understand the functions of the regional karst landscape. For example, water flowing from springs in the

Haney Limestone very likely contributes to the development of vertical shafts which intersect passages of the Mammoth Cave System. Comparison of the geochemistry of water flowing from the Haney springs and water concentrated by surface channels and the influence each source has on vertical shaft development would identify whether one source is more favorable. Water flowing through vertical shafts may cross apparent drainage basin divides, potentially transporting contaminants to unexpected locations

(Meiman and Groves 2001). Further investigation of these vertical shafts and the potential impacts of concentrated flows coming from Haney Springs would add to the understanding of drainage divides in the region and aid in the ability to protect the resources of Mammoth Cave National Park.

According to Ramsey’s (1974) description in Speleo Digest, Barnes Smith Cave

(Figure A1) is a joint controlled maze cave located in Hart County. With a surveyed length of over 1000 meters, Barnes Smith Cave is a significant Haney cave, second only to Cub Run Cave in length. Efforts to find the location of Barnes Smith should continue, as the cave and information gained from it would be important to any future investigations of cave and karst development in the Haney Limestone.

APPENDIX A: CAVE MAPS

Figure A1. Description and lineplot of Barnes Smith Cave (Ramsey 1974).

Figure A2. Davenport Cave map (KSS).

Figure A3. Honaker Cave map (KSS).

Figure A4. Spring Trough Caves 1 & 2 (1991 Speleofest Guidebook pg. 49).

Figure A5. Black Rock Cave map (CRF).

Figure A6. Bryophyte Cave map (CRF).

Figure A7. Cade Cave map (CRF).

Figure A8. Cricket Falls Cave map (CRF).

Figure A9. Deer Skull Cave map (CRF).

Figure A10. Feather Cave map (CRF).

Figure A11. Fish Trap Pits 1 & 2 (CRF).

Figure A12. Good Spring Cave map (CRF).

Figure A13. Hearth Pit Cave (CRF).

Figure A14. Hearth Spring Cave map (CRF).

Figure A15. Hickory Cabin Cave map (CRF).

Figure A16. Hideout Spring Cave map (CRF).

Figure A17. Hornbeam Spring Cave map (CRF).

Figure A18. Johnson Spring Cave map (CRF).

Figure A19. Lycopodium Spring Cave map (CRF).

Figure A20. Salamander Cave map (CRF).

Figure A21. Squeeze Cave map (CRF).

Figure A22. Squirrel Hollow Springhouse Cave map (CRF).

Figure A23. Stillhouse Sink Cave (CRF).

Figure A24. Two Entrance Cave map (CRF).

APPENDIX B: GEOCHEMICAL DATA

Sample Temp. SpC HCO ̄ Ca2+ Mg2+ Na+ K+ pH 3 Date (˚C) (µmhos) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) 03/11/1972 15.5 147 7.17 139 39.9 3.3 1.3 0.9 15/11/1972 15.0 209 7.25 130 40.9 2.6 1.1 0.6 30/11/1972 13.8 184 7.29 118 35.3 2.1 1.1 0.7 13/12/1972 12.2 62 7.39 30 8.6 1.6 1.1 0.8 26/12/1972 12.3 182 7.22 112 33.0 2.3 1.1 0.7 10/01/1973 12.3 195 7.06 116 32.1 2.0 1.1 0.6 24/01/1973 12.0 159 7.15 110 32.9 2.1 1.2 0.7 07/02/1973 11.1 170 7.11 105 30.8 1.9 1.2 0.6 21/02/1973 10.9 163 7.15 104 30.4 1.9 1.2 0.6 08/03/1973 12.2 65 7.07 30 8.7 1.3 1.2 0.7 19/03/1973 12.2 73 7.19 34 9.9 1.3 1.2 0.7 03/04/1973 11.7 147 7.13 94 28.1 1.7 1.1 0.6 19/04/1973 12.2 69 7.30 35 9.7 1.4 1.3 0.6 01/05/1973 11.7 146 7.12 95 28.1 1.7 1.2 0.7 11/05/1973 11.7 127 7.12 35 9.7 1.3 1.2 0.6 30/05/1973 12.2 63 6.98 35 9.7 1.3 1.2 0.7 12/06/1973 12.8 152 7.12 107 30.8 1.9 1.3 0.7 26/06/1973 13.3 161 7.30 110 31.8 2.0 1.1 0.6 10/07/1973 13.6 153 6.91 110 33.2 1.9 1.1 0.7 23/07/1973 14.4 175 7.14 111 33.5 2.0 1.1 0.5 06/08/1973 14.7 161 7.20 116 34.5 2.1 1.1 0.6 07/09/1973 16.7 182 7.40 122 36.4 2.2 1.1 0.7 17/09/1973 16.6 167 7.48 123 37.0 2.3 1.1 0.7 27/09/1973 16.2 178 7.45 124 37.4 2.3 1.5 0.6 19/10/1973 16.7 184 7.61 129 39.1 2.4 1.1 0.7 Table A1. Adwell Spring geochemical data (from Hess 1974).

Temp. SpC HCO ̄ Ca2+ Mg2+ Na+ K+ Sample Date pH 3 (˚C) (µmhos) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) 03/11/1972 147 101 7.32 61 16.1 3.1 1.5 1.3 15/11/1972 209 81 7.53 38 12.5 1.9 1.5 0.8 30/11/1972 184 69 7.44 33 9.6 1.5 1.3 0.7 13/12/1972 62 70 7.19 35 9.9 1.8 1.3 0.8 26/12/1972 182 85 7.34 39 10.9 1.9 1.4 0.7 10/01/1973 195 82 7.45 42 11.8 1.6 1.3 0.6 24/01/1973 159 75 7.36 38 11.0 1.6 1.6 0.8 07/02/1973 170 71 7.40 35 9.6 1.5 1.3 0.6 21/02/1973 163 69 7.54 38 9.8 1.6 1.3 0.6 08/03/1973 65 66 6.81 28 7.9 1.3 1.2 0.7 19/03/1973 73 62 7.01 29 7.8 1.6 1.2 0.8 03/04/1973 147 68 7.33 37 9.7 1.4 1.3 0.6 19/04/1973 69 69 7.09 40 9.6 1.7 1.3 0.7 01/05/1973 146 70 7.20 39 9.5 1.4 1.4 0.6 11/05/1973 127 66 6.94 34 8.7 1.5 1.2 0.7 30/05/1973 63 63 6.88 37 9.3 1.4 1.4 0.7 12/06/1973 152 98 7.08 62 16.5 2.0 1.5 0.8 26/06/1973 161 118 7.16 77 20.9 2.2 1.3 0.7 10/07/1973 153 105 6.79 69 18.6 2.1 1.3 0.7 23/07/1973 175 80 6.89 47 13.2 1.8 1.3 0.7 06/08/1973 161 115 7.21 75 21.2 2.2 1.4 0.7 07/09/1973 182 173 7.33 85 23.0 2.6 1.4 0.7 17/09/1973 167 134 7.33 88 25.3 2.6 1.7 0.7 27/09/1973 178 148 7.40 90 24.7 2.7 1.4 0.7 19/10/1973 184 142 7.67 96 26.5 2.8 1.4 0.7 Table A2. Blair Spring geochemical data (from Hess 1974).

Temp. SpC HCO ̄ Ca2+ Mg2+ Na+ K+ Sample Date pH 3 (˚C) (µmhos) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) 03/11/1972 13.5 115 7.06 65 18.8 2.7 1.5 0.6 15/11/1972 12.2 88 7.13 46 14.8 1.7 1.3 0.3 30/11/1972 11.4 84 7.50 39 12.4 1.2 1.3 0.4 13/12/1972 11.1 64 7.41 33 9.6 1.5 1.2 0.5 26/12/1972 10.0 88 7.52 45 13.4 1.6 1.2 0.4 10/01/1973 9.7 88 7.41 49 14.5 1.4 1.2 0.6 24/01/1973 10.5 90 7.61 47 13.6 1.5 1.2 0.4 07/02/1973 10.5 86 7.45 45 13.7 1.4 1.2 0.4 21/02/1973 9.7 79 7.69 47 13.5 1.3 1.4 0.4 08/03/1973 11.1 60 6.91 32 9.2 1.0 1.1 0.4 19/03/1973 11.7 62 6.99 36 10.6 1.1 1.1 0.4 03/04/1973 11.7 85 7.15 49 13.7 1.4 1.2 0.4 19/04/1973 11.4 77 7.04 42 11.5 1.3 1.3 0.4 01/05/1973 11.7 78 6.84 46 12.9 1.4 1.3 0.5 11/05/1973 11.7 73 6.97 38 11.0 1.2 1.3 0.4 30/05/1973 11.7 71 6.97 44 11.7 1.2 1.3 0.4 12/06/1973 12.2 95 6.83 57 15.3 1.8 1.4 0.6 26/06/1973 13.3 86 6.89 54 15.0 1.6 1.4 0.4 10/07/1973 14.2 94 6.65 64 15.8 1.6 1.2 0.7 23/07/1973 13.3 77 6.66 42 12.3 1.3 1.2 0.6 06/08/1973 17.2 235 7.32 132 43.4 3.3 1.3 0.8 07/09/1973 17.8 267 7.44 187 59.0 5.1 1.3 4.5 17/09/1973 16.5 286 7.42 187 61.0 4.9 1.8 2.2 27/09/1973 17.6 349 7.35 215 67.5 5.3 1.2 3.7 19/10/1973 11.6 322 7.46 206 70.8 5.7 1.4 3.7 Table A3. Bransford Spring geochemical data (from Hess 1974).

Temp. SpC HCO ̄ Ca2+ Mg2+ Na+ K+ Sample Date pH 3 (˚C) (µmhos) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) 03/11/1972 13.5 166 7.50 120 34.9 3.3 1.0 1.0 15/11/1972 11.1 170 7.70 100 30.3 2.0 0.9 0.6 30/11/1972 11.2 135 7.82 87 26.5 1.6 0.8 0.6 13/12/1972 12.2 106 7.58 66 19.7 1.4 0.9 0.6 26/12/1972 11.3 142 7.89 80 22.7 1.7 0.9 0.6 10/01/1973 12.3 134 7.82 81 24.0 1.4 0.9 0.6 24/01/1973 11.0 132 8.00 83 24.0 1.4 0.9 0.6 07/02/1973 12.0 119 7.90 78 23.4 1.3 0.9 0.6 21/02/1973 11.3 122 7.95 76 22.6 1.3 1.0 0.5 08/03/1973 13.0 82 7.55 49 14.8 1.0 1.0 0.6 19/03/1973 12.8 86 7.63 56 16.4 1.0 0.9 0.6 03/04/1973 12.8 118 7.82 70 20.3 1.2 0.9 0.6 19/04/1973 12.8 108 7.69 68 20.0 1.2 1.0 0.6 01/05/1973 12.8 117 7.62 72 21.2 1.2 1.0 0.6 11/05/1973 12.8 110 7.59 76 21.4 1.2 1.0 0.6 30/05/1973 12.8 104 7.46 75 21.7 1.2 1.0 0.6 12/06/1973 13.3 135 7.30 92 26.3 1.6 1.0 0.6 26/06/1973 13.3 148 7.46 100 27.7 1.8 1.0 0.6 10/07/1973 13.3 135 7.09 101 30.5 1.6 0.9 0.6 23/07/1973 13.9 145 7.32 98 29.7 1.6 0.9 0.6 06/08/1973 13.9 152 7.36 101 30.8 1.7 0.9 0.6 07/09/1973 13.0 168 7.34 74 24.0 1.5 1.0 0.5 17/09/1973 14.2 163 7.54 112 33.2 1.8 0.8 0.5 27/09/1973 14.0 164 7.51 115 34.4 1.9 0.9 0.6 Table A4. Collins Spring geochemical data (from Hess 1974).

Temp. SpC HCO ̄ Ca2+ Mg2+ Na+ K+ Sample Date pH 3 (˚C) (µmhos) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) 03/11/1972 13.5 83 7.31 59 15.7 2.4 1.3 0.7 15/11/1972 13.3 103 7.56 56 17.3 1.8 1.4 0.7 30/11/1972 12.6 85 7.42 43 12.1 1.4 1.2 0.6 13/12/1972 12.0 53 7.43 40 11.1 1.8 1.3 0.8 26/12/1972 8.8 129 7.75 77 23.9 2.2 1.2 0.8 10/01/1973 8.3 108 7.47 64 19.2 1.7 1.2 0.6 24/01/1973 9.7 88 7.48 60 17.7 1.7 1.2 0.6 07/02/1973 9.3 105 7.51 64 18.4 1.7 1.2 0.6 21/02/1973 10.2 1049 7.50 64 18.8 1.7 1.3 0.6 08/03/1973 12.2 65 7.06 33 9.2 1.1 1.2 0.6 19/03/1973 12.8 62 7.13 38 10.2 1.1 1.1 0.6 03/04/1973 14.4 118 7.42 67 18.8 1.7 1.2 0.8 19/04/1973 16.1 108 7.41 68 20.2 1.6 1.3 0.6 01/05/1973 14.4 119 7.30 72 21.2 1.7 1.4 0.6 11/05/1973 14.0 120 7.41 82 23.4 1.8 1.2 0.6 30/05/1973 16.7 107 7.32 77 22.8 1.7 1.3 0.6 12/06/1973 13.9 98 7.24 70 19.2 1.7 1.4 0.7 26/06/1973 13.3 115 7.34 79 21.8 1.8 1.3 0.6 10/07/1973 12.8 110 7.05 72 20.3 1.7 1.2 0.6 23/07/1973 13.3 86 7.08 50 14.1 1.4 1.2 0.7 06/08/1973 13.9 126 7.24 79 23.3 1.9 1.3 0.6 07/09/1973 13.3 167 7.32 88 26.1 2.1 1.4 0.8 17/09/1973 13.3 132 7.40 90 25.7 2.1 1.4 0.6 27/09/1973 13.5 137 7.38 92 26.2 2.3 1.4 0.6 19/10/1973 13.2 135 7.59 97 26.8 2.4 1.4 0.6 Table A5. Cooper Spring geochemical data (from Hess 1974).

Temp. SpC HCO ̄ Ca2+ Mg2+ Na+ K+ Sample Date pH 3 (˚C) (µmhos) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) 03/11/1972 14.5 112 7.20 71 19.8 2.8 1.2 0.8 15/11/1972 12.8 78 6.99 38 11.2 1.7 1.2 0.7 30/11/1972 12.3 69 7.06 33 10.3 1.3 1.2 0.6 13/12/1972 11.8 53 7.43 24 6.3 1.8 1.3 0.8 26/12/1972 9.5 54 7.44 27 7.3 1.4 1.2 0.6 10/01/1973 11.1 63 7.22 43 12.0 1.5 1.1 0.6 24/01/1973 10.0 64 7.74 37 10.7 1.3 1.3 0.6 07/02/1973 11.7 70 7.22 35 10.7 1.3 1.2 0.6 21/02/1973 8.9 60 7.60 35 9.2 1.2 1.2 0.5 08/03/1973 12.2 48 6.94 24 7.2 1.0 1.0 0.6 19/03/1973 12.2 46 7.14 23 6.1 0.9 1.1 0.6 03/04/1973 12.2 49 7.32 26 8.0 1.0 1.1 0.5 19/04/1973 12.2 66 7.17 34 8.6 1.1 1.2 0.5 01/05/1973 12.8 49 7.08 28 6.8 1.0 1.2 0.5 11/05/1973 13.0 52 7.00 27 7.4 1.0 1.3 0.6 30/05/1973 12.8 46 7.08 28 7.6 1.0 1.4 0.6 12/06/1973 16.1 74 7.17 45 10.8 1.5 1.4 0.6 26/06/1973 14.2 115 7.62 78 20.8 2.0 1.2 0.6 10/07/1973 13.3 100 7.37 65 17.5 1.8 1.3 0.7 23/07/1973 13.6 76 6.96 44 13.1 1.4 1.1 0.7 06/08/1973 13.9 133 7.43 76 21.0 2.1 1.2 0.7 07/09/1973 15.3 172 7.71 98 27.4 2.5 1.2 0.6 17/09/1973 14.4 163 7.65 100 28.2 2.5 1.2 0.6 27/09/1973 15.3 163 7.59 106 30.1 2.6 1.2 0.6 19/10/1973 14.5 148 7.73 104 29.0 2.5 1.5 0.7 Table A6. Three Springs geochemical data (from Hess 1974).

REFERENCES

Berner, R.A. and J.W. Morse. 1974. Dissolution kinetics of calcium carbonate in sea water IV. Theory of calcite dissolution. American Journal of Science 274(2): 108-134.

Brown, R.F. 1966. Hydrology of the cavernous limestones of the Mammoth Cave area, Kentucky. U.S. Geologic Survey Water-Supply Paper 1837.

Brucker, R.W., J.W. Hess, and W.B. White. 1972. Role of vertical shafts in the movement of ground water in carbonate aquifers. Ground Water 10(6): 5-13.

Cushman, R.V. 1965. Present and future water supply for Mammoth Cave National park, Kentucky. U.S. Geologic Survey Water-Supply Paper 1475-Q.

Dasher, G.R. 2011. On Station: A Complete Handbook for Surveying and Mapping Caves. 2nd ed. Huntsville, Alabama: National Speleological Society.

Deike, G.H., III. 1967. The Development of Caverns of the Mammoth Cave Region. Ph.D. thesis, Department of Geology, The Pennsylvania State University.

Deike, G.H., III and W.B. White. 1969. Sinuosity in limestone solution conduits. American Journal of Science 267: 230-241.

Deike, R.G. 1969. Relations of jointing to orientation of solution cavities in limestones of Central Pennsylvania. America Journal of Science 267: 1230-1248.

Dreyboldt, W. 1990. The role of dissolution kinetics in the development of karst aquifers in limestone: A model simulation of karst evolution. The Journal of Geology 98(5): 639-655.

Ford, D.C. and P. Williams. 2007. Karst Hydrogeology and Geomorphology. West Sussex, England: Wiley & Sons.

Foster, Z.A. 1990. Depositional and Diagenetic History of the Haney Limestone (Upper Mississippian), Sulphur, Indiana. Master’s thesis, Department of Geological Sciences, Indian University at Bloomington.

George, A.I. 1989. Caves and drainage north of the Green River, in White, W.B., and White, E.L., eds., Karst Hydrogeology: Concepts from the Mammoth Cave Area: New York, Van Nostrand Reinhold: 189-221.

Granger, D.E., D. Fabel, and A.N. Palmer. 2001. Pliocene-Pleistocene incision of the Green River, Kentucky, Determined from Radioactive Decay of Cosmogenic 26Al and 10Be in Mammoth Cave Sediments. GSA Bulletin 113(7): 825-836.

100

Groves, C.G. 1993. Early Development of Karst Systems [Ph.D. Thesis]: University of Virginia.

Groves, C.G. and A.D. Howard. 1994. Minimum hydrochemical conditions allowing limestone cave development. Water Resources Research 30(3): 607-615.

Harris, C.D. and G.H. Fraunfelter. n.d. Depositional aspects of Golconda Group (Chesterian) oolite bodies, Southwestern Illinois Basin. Ch. 9 (?).

Haun, D., ed. 1991. Guidebook to the Kentucky Speleofest: National Speleological Society, Louisville Grotto, Kentucky.

Häuselmann, P. 2011. UIS mapping grades. International Journal of Speleology 40(2): IV-VI.

Haynes, D.D. 1964. Geology of the Mammoth Cave Quadrangle, Kentucky. U.S. Geological Survey.

Hess, J.W. 1974. Hydrochemical Investigations of the Central Kentucky Karst Aquifer System. Ph.D. thesis, Department of Geosciences, The Pennsylvania State University.

Hess, J.W. 1976. A review of the hydrogeology of the central Kentucky karst. NSS Bulletin 38(4): 99-103.

Hess, J.W. and W.B. White. 1993. Groundwater geochemistry of the carbonate karst aquifer, southcentral Kentucky, U.S.A. Applied Geochemistry 8:189-204.

Howard, A.D. 1968. Stratigraphic and structural controls on landform development in the central Kentucky karst. National Speleological Society Bulletin 30(4):95-114.

La Valle, P. 1968. Karst depression morphology in South Central Kentucky. Geografiska Annaler. Series A, Physical Geography 50(2): 94-108.

McFarlan, A.C., D.H. Swann, F.H. Walker, and E. Nosow. 1955. Some old Chester problems – correlations of Lower and Middle Chester Formations of Western Kentucky. Kentucky Geological Survey, Series 9, Bulletin 16.

Miotke, F.-D. and A.N. Palmer. 1972. Genetic Relationship between Caves and Landforms in the Mammoth Cave National Park Area. Würzburg, Germany: Böhler Verlag.

Palmer, A.N. 1981. A Geological Guide to Mammoth Cave National Park. Teaneck, NJ: Zephyrus Press, Inc.

101

Palmer, A.N. 1984. Geomorphic interpretation of karst features, in R.G. LaFleur, ed., Groundwater as a Geomorphic Agent: London, Allen and Unwin: 173-209.

Palmer, A.N. 1989. Geomorphic history of the Mammoth Cave system, in White, W.B., and White, E.L., eds., Karst Hydrogeology: Concepts from the Mammoth Cave Area: New York, Van Nostrand Reinhold: 317–337.

Palmer, A.N. 1991. Origin and morphology of limestone caves: Geological Society of America Bulletin, 103: 1–21.

Palmer, A.N. 1995. Geochemical models for the origin of macroscopic solution porosity in carbonate rocks, In Unconformities in Carbonate Strata: Their Recognition and the Significance of Associated Porosity, eds. Budd, D., P. Harris, and A. Saller, 77-101. : American Association of Petroleum Geologists, Memoir 63.

Palmer, A.N. 2003. Dynamics of cave development by allogenic water. Speleogenesis and Evolution of Karst Aquifers 1(1): 1-11.

Palmer, A.N. 2007. Cave Geology. Trenton, NJ: Cave Books.

Plummer, L.N. and T.M.L. Wigley. 1976. The dissolution of calcite in CO2-saturated solutions at 25 ˚C and 1 atmosphere total pressure. Geochimica et Cosmochimica Acta 40(2): 191-202.

Plummer, L.N., T.M.L. Wigley and D.L. Parkhurst. 1978. The kinetics of calcite dissolution in CO2-water systems at 5 to 60˚C and 0.0 to 1.0 atm CO2. American Journal of Science 278: 179-216.

Quinlan, J.F. 1970. Central Kentucky karst. Mediteranee, estudes et travaux 7: 235- 253.

Quinlan, J.F. and D.R. Rowe. 1977. Hydrology and Water Quality in the Central Kentucky Karst: Phase I. National Technical Information Service. Research Report 101, January 1977.

Ramsey, T. 1974. Barnes Smith Cave, Hart County, Kentucky, In Speleo Digest, ed. B. Mixon, 73: National Speleological Society.

Ryan, M. and J. Meiman. 1994. Lulu Mart Cave and the Hydrogeology of the Chester Uplands Hilly Country, Mammoth Cave National Park, pp 189-201, Proceedings of Mammoth Cave National Park's Third Science Conference, Mammoth Cave National Park, July 5-6, 1994.

Sandberg, C.A. and C.G. Bowles. 1965. Geology of the Cub Run Quadrangle, Kentucky. U.S. Geological Survey.

102

Sasowsky, I.D. and W.B. White. 1994. The role of stress release fracturing in the development of cavernous porosity in carbonate aquifers. Water Resources Research 30(12): 3523-3530.

Sides, S.D. and M.T. Ryan. 1996. Buffalo Creek Cave, pp 157-169, Proceedings of Mammoth Cave National Park’s Fifth Science Conference, Mammoth Cave National Park, August 1-2, 1996.

Swann, D.H. 1964. Late Mississippian rhythmic sediments of Mississippi Valley. Bulletin of the American Association of Petroleum Geologists 48(5): 637-658.

Vincent, J.W. 1975. Lithofacies and biofacies of the Haney Limestone (Mississippian), Illinois, Indiana, and Kentucky. Kentucky Geological Survey, Series X, Thesis Series 4.

White, W.B. 1988. Geomorphology and Hydrology of Karst Terrains. New York: Oxford University Press.

White, W.B., R.A. Watson, E.R. Pohl, and R.W. Brucker. 1970. The central Kentucky karst. Geographical Review 60(1): 88-115.

103