Contamination and Collapse:

Human Settlement and Karst.










Kristina Perry

Geography 810

May 21, 1999

Environmental change and geomorphic response are a part of any natural system. However, in contrast to natural environmental processes, the impact of human activity on karst terrains is characterized by extreme rates of change and an exaggerated scale of the types of changes that occur. (Williams, 1993.). The impact of human activity on karst terrain is amplified by the fact that karst drainage and aquifers are unlike surface drainage basins and "regular" aquifers in three ways: first they are underground, second the flow is complex –both laminar and turbulent – and finally, they are dynamic and evolving. It is, in fact, their very nature is what makes them fragile and susceptible to pollution.

The study of karst solution processes and karst hydrology (besides being interesting) is increasingly important as populations in areas dominated by karst features grow. These populations not only utilize karst aquifers for drinking water, but also must be careful about waste generation and disposal so as not to contaminate that very source of water upon which they depend. It appears, though, that the stuff of human settlement is the stuff of pollution and systems disruption: agriculture and animal husbandry; chemical manufacture, storage and use; and automobile use. If people are going to settle in karst terrains, they need to be as informed as possible about their home hydrology, in order that they might protect it. In addition to the threat of groundwater depletion and contamination in areas of human settlement on karst, there is also the very real threat of ground collapse. Water use and improper building practices can accelerate natural sinkhole development in such a way as to endanger both people and structures.

My interest in the interface of karst and human settlement began at the 1998 annual meeting of the AAG in Boston. There, I heard a geographer from the University of Kentucky talk about the threat to karst aquifers posed by waste lagoons in large scale confinement-style hog farming. Already interested in the environmental issues posed by industrial agriculture, I was particularly intrigued by this new dimension. I spent the semester looking at the pollution problems associated with agriculture and the pollution and the structural failure caused urbanization. To date I have yet to find an article in a scholarly journal documenting the problem of hog waste and karst. (But, I haven’t given up – and when I find it I will send it your way!) Looking at pollution and karst I realized that it was necessary to understand the basic mechanisms and morphology of karst itself. Therefore, it was important to review some "karst basics" before talking about specific problems, remediation and management.

Accordingly, this literature review will first examine karst formation and karst drainage, with a brief comparison to surface drainage basins for illustration purposes. That will be followed by a discussion of karst hydrology and susceptibility to pollution with particular attention paid to the range of human impact on karst. I will then review representative articles of those I read, to illustrate pertinent issues in pollution and contaminant transport, and sinkhole formation in karst aquifers and landscapes. Examples of monitoring and remediation techniques and some recommendations for management will also be discussed. In retrospect, I wish I had focussed more closely on groundwater contamination by agriculture and learned a little about the transport chemistry and more about cleanup strategies.


Karst Formation And Karst Drainage

karst formation

Karst landscapes are unique in that the principal geomorphic agent is chemical solution and solution transport. Karst features occur in areas where bedrock is made up of carbonate, gypsum or other evaporite rock. This discussion will focus on carbonate bedrock. The chemical that dissolves limestone is carbonic acid – formed from the mixing of CO2 and water. (H2CO3) In a gross simplification of the process: gaseous CO2 mixes with water to become aqueous CO2 which in turn mixes with water to become a neutral (in charge) carbonic acid. The carbonic acid then dissolves the Calcite / limestone / CaCO3 and transports it in solution as the charged Ca2+ and 2HCO3-. (White, 1988.)


karst drainage

Accordingly, fluvial systems in carbonate areas differ from surface drainage systems. In non-karst areas drainage basins are found on the land’s surface. For example, water from precipitation becomes first order tributaries, second order streams, third order creeks and then rivers that wash to the sea. In contrast, karst drainage is characterized by its subterranean nature and its dynamic evolution. Underground Karst aquifers often begin as surface drainage. Rivers may downcut until carbonate rock is exposed. At that point, solution processes begin to dissolve the limestone and tributary water may disappear into the ground in a swallow hole or a sinking stream only to emerge kilometers away as a full-blown (high order) river from the mouth of a cave or from an artesian spring. The evolution of this underground drainage is controlled by the amount of water, composition of the bedrock (is it largely limestone or is their a lot of clay and/or chert mixed into it), presence of existing joints or fractures, porosity of alluvium, and CO2 content of the soil. (White, 1988.)

At some point, underground drainage may replace surface flow as subterranean conduits and caves continue to enlarge. The drainage basin itself becomes completely underdrained except at the time of large storm events when underground drainage is overwhelmed. This process of replacing surface with underground drainage produces a dry valley.

Ultimately, water re-emerges at the surface. Resurgence most often takes the form of spring which come in three types: gravity springs which flow from the openings of cave without much pressure (also called hydrostatic head), alluviated springs which water re-emerges under some pressure against surface alluvium, glacial outwash or rockfalls, and artesian springs, which emerge under the greatest hydrostatic head. In fact, at times of high discharge they may even show a pronounced fountain effect. (White, 1988)


Karst Hydrology

groundwater review

Almost all rocks are permeable to some extent and they retain and transmit water through their pore spaces. If a rock-type is able to store large quantities of water it is called an aquifer. Aquifers are supplied by water that percolates from the surface until it reaches the phreatic zone (the water saturated zone). The phreatic zone is below the vadose zone (the unsaturated zone) and the two are divided by the water table. In non-karst areas there is a distinction between surface water and groundwater. Surface water is drained and directed according to local topography and surface water is loosely connected to groundwater through infiltration. Aquifer boundaries, and in fact aquifer flows, are quite independent of surface drainage basin boundaries. (Heath, 1998)

In contrast, karst fluvial systems and groundwater are directly connected through subsurface conduits. Groundwater basins are flow-paths through which previously surface drainage moves to its point of resurgence and discharge. The boundaries and direction of the groundwater basin and the relic surface basin can be the same where the underground drainage has merely replaced surface flow. However, underground conduit systems are often more complex. They may have more than one channel carrying water and they may have branching such that water from one drainage basin discharges at more than one spring. (Williams, 1993)

For these reasons karst aquifers are not well understood. It is thought that water flows through porous surface media to the vadose zone or enters the vadose zone through swallow holes and conduits. Water is stored below these conduits in the phreatic zone and beside the conduits in non-flow conduits and caves. Phreatic storage further

dissolves limestone, enlarging conduits and forming caves. Unlike non-karst areas there exists a great potential for mixing of drainage and storage waters as flow is sometimes flashy and turbulent. (Above is a diagram from White, 1988) With this potential for mixing comes greater possibilities for the transmission of contaminants (White, 1988.). In many carbonate aquifers the flow type varies between diffuse and conduit. Diffuse flow occurs in tight fractures, joints, and bedding planes. Such flow, where velocities are low, is said to be laminar (not turbulent and where no mixing occurs). (Williams, 1983) Storage can even take place in this "subcutaneous" zone when a perched water body develops because of seasonal saturation of the lower permeability underlying rocks.

human impact and karst

With such rapid subsurface water entry and all this connectivity and mixing its easy to see why groundwater contamination is a problem. But, groundwater pollution is not the only degradation of karst systems. In an introductory article in a special supplement to Catena Paul Williams classifies human-induced impacts on karst terrains in the following categories:

(The current conflict in Yugoslavia is occurring in a region that is highly karstified. At one point on NPR I heard some military talking head mention some of the obstacles to troop movement in the area and he mentioned the "difficult terrain." What he was referring to is karst terrain riddled with pinnacles, sinkholes and poljes that make driving a tank just a little bit difficult. Surely, bombing the hell out of an area can’t be very good for the landscape. But, I digress…)

Some examples of impact: (all examples the special issue of, CATENA devoted to karst and human impact.)

Caves provide natural shelter and cave occupation, though dating back to mid Pleistocene, is still practiced today in areas of Southern Europe and of china

Deforestation has a profound indirect impact and dates back to 6000 years ago in the northern Mediterranean basin. In the Greek and Roman eras deforestation stimulated the erosion of hillsides and sedimentation of the Mediterranean valleys. The loss of trees increases runoff. In many karst areas soils are poor and thin and runoff can expose bedrock outcroppings – leaving little left for cultivation.

Karst areas that are cleared for agriculture are prone to soil loss in sinkholes and swalletts. Terracing of hillsides (as is done in China) sometimes helps stem the tide, but more often than not agriculture results in desertification.

The over-tapping of groundwater supplies results in a lowing of the water table and facilitates sinkhole collapse and other surface collapse.

Limestone and marble have been mined for building and for sculpture "since the minoan age of Crete"! (p.13) Mineral quarrying degrades karst systems and provides more ready access to groundwater for contaminants.

The most common problems associated with urbanization in karst area are flooding, pollution and groundwater collapse. Increased demand for water depletes aquifers, impervious surfaces concentrate runoff, and construction of large buildings on unstable ground is hazardous to people and to karst.

Recreation in cave and karst cause wear and tear on caves. In addition, the waste

generated by visitors is another potential contaminant source.

In the remaining portion of this paper, I want to focus on two specific issues: groundwater contamination (and to a degree, agricultural activities as a source of groundwater impacts) and sinkhole collapse.

Groundwater contamination

According to geomorphologists and karst hydrologists, like Nicholas Crawford, shallow aquifers in karst areas are probably the most vulnerable in the world to groundwater contamination. Karst aquifers receive both diffuse recharge from percolation through the soil and concentrated recharge from surface runoff that flows directly into the aquifer at stream sinks and sinkhole drains. (Crawford and Whallon, 1985.) Because of the rapid velocities of these underground streams, contaminants may travel several miles through the aquifer in only a few hours.

Contaminants from agricultural activities, such as nitrates, bacteria from livestock waste, and pesticides, are potential problems in karst terrain. Contaminants found in urban storm water runoff such as: lead, chromium, oil and grease, petroleum products, solvents, and bacteria from pet and animal wastes may be a threat to people using karst water supplies and to cave life. Additionally, it has been shown that septic tank effluent can travel through the thin soils which are characteristic of most karst areas into the aquifer and then to a spring in only a few hours. Alternately, percolating soil water can push effluent from septic tanks down into the phreatic zone resulting in high fecal coliform counts and high fecal coliform/fecal streptococcus ratios at some springs following storm events. (Crawford and Whallon, 1985)

Transport in Karst Aquifers

John Hoke and Carol Wicks investigate transport in karst aquifers in The Engineering Geology and Hydrogeology of Karst Terranes (Hoke & Wicks, 1997). They assert that predicting flow response is not possible on the individual conduit level. Rather, they propose that modeling transport is much more accurate if it is done on a "basin scale" perspective, because the hydraulic geometry is exceedingly complex and transport depends on volume, shape, location of conduits, position of conduits relative to the water table and the degree of development of other karst features, like sinkholes and losing streams. Hoke and Wicks suggest that using a linear systems approach is useful because it describes basin transport in terms of the distribution of travel or as a residence time -- and without having to have detailed knowledge of internal structure. In the linear systems model each basin has a unique "kernal function" which relates recharge over a spatial area. Kernal functions can be derived through dye tracing methods and takes into account the method by which contaminants enter the cost system. For example, a spatially and uniformly distributed non-point source contaminant to the basin will exhibit a different response than a discrete point source contaminant. Their model used precipitation, spring discharge, and recharge over a spatial area as the variables. Testing of the model yielded predictable results. Transport models are useful in that, if accurate, they can be generalized to basins of similar character.

Barner and Uhlman wrote an article, "Contaminant Transport Mechanisms In Karst Terrains And Implications On Remediation" for the Fifth Multidisciplinary Conference On Sinkholes And The Engineering And Environmental Impacts Of Karst, in 1995. They begin the article by stating that many remediation efforts in karst areas have been unsuccessful, particularly those that rely on groundwater extraction for cleanup. They have failed because the source, storage and transport characteristics are different for karst than for other aquifers. Instead, understanding of the complex nature of karst storage and transport is necessary for toxic cleanup. They even admit that with some contaminants, in some settings – cleanup may not be possible. Remediation in karst aquifers includes proper site characterization, and an understanding of contaminant transport mechanisms.

Site characterization means comprehensive surveys and dye tracings that monitor all springs and seeps in the area as potential areas of resurgence -- not ones that monitor "downstream" only. In addition, they note the importance of monitoring water chemistry at different times of the year to determine if recharge comes from bedrock or subsurface waters.

The characteristics of contaminant transport are important as well. Different types of contaminants have different flow characteristics. L-NAPL (light non-aqueous phase liquid) organic compounds (like petroleum) are hydrophobic and tend not to be flushed out. Rather, because they float on top, they are absorbed and remain in the "interstitial spaces of the aquifer matrix" (Pg. 209.) where they remain as contaminant sources for long periods of time. Dense NAPL compounds (D-NAPL) can flow by gravity through the aquifer. These chemicals (PCE, TCE and Cresote are examples) oftensink to depths and remain as long-term contaminant sources. Plumes from these concentrated sources of toxin continue to contaminate the aquifer. Knowing the transport and "resting" characteristics of contaminants can aid in locating them in the aquifer so that they might be removed. But Barner and Uhlman conclude by saying that many times the source of contamination cannot be removed, for physical or economic reasons. When this is the case, they suggest implementing groundwater filtration technologies at the spring resurgence instead.

Leaks, spills or deliberate dumping of toxic or explosive chemicals are a particularly serious hazard in karst areas. Nicholas Crawford of Kentucky University wrote Toxic And Explosive Fumes Rising From Carbonate Aquifers: A Hazard For Residents Of Sinkhole Plains in the first multidisciplinary conference on sinkholes. According to Crawford, contamination problems are aggravated in karst areas by the practice of disposing of solid and liquid wastes in sinkholes where they are prone to being washed directly into the aquifer. Not only are these materials a threat to water supplies but, when they vaporize they can become highly concentrated in the cave atmosphere rising through fractures in the limestone to enter homes and businesses on the land surface. Chemicals that leak from underground tanks may be carried into the caves below by water that percolates into the soil following storm events. Most caves become completely water-filled at some point down stream. This can result in trapped floating chemicals, especially LNAPLs such as gasoline. Fumes may then fill the cave and rise into buildings on the surface. Occasionally homes in urban areas must be evacuated because fumes reach explosive levels in basements (Crawford, 1984a).

He describes that case of residents of Bowling Green KY , who since 1969, have had problems with explosive fumes from caves rising into their homes. Leaking underground storage tanks are undoubtedly the culprits. Gasoline floating on subsurface streams can travel several miles from the site filling a cave passage with explosive fumes. Non-explosive but, toxic fumes are also a problem: Benzene, Methylene Chloride, toulene, xylene, and alphatic hydrocarbons were detected in cave air using a portable organic vapor analyzer. Spills, leaks and deliberate dumping of toxic and explosive chemicals are thought to be the cause. The author proposes that maps of groundwater flow be developed along with better storage and transport procedures and that emergency response techniques be improved. (Crawford 1984a)

The extent to which a shallow karst aquifer is contaminated depends on whether it receives primarily diffuse or concentrated recharge, on proximity and types of sources of contamination, and the thickness of overlying soils. For example, karst springs and water wells supplied entirely by distributed recharge through thick regolith may be free of contaminants and serve as good sources of potable water. Unfortunately, most karst springs and water wells receive concentrated recharge from a nearby area where sources of contamination are present (Crawford and Whallon, 1985).

Another example of research in this area include the case of highway runoff in Knoxville TN and Frederick, MD, where stormwater runoff seems to flow directly into sinkholes and to the karst conduits below. In an effort to trace contaminant pathways, Stephenson, Zhou, Beck and Green used dye tracing at the Tennessee site to demonstrate that the water does, in fact, pass through the I-40 sinkhole and resurgence at Holston Spring. They found that the lag time for peak flows was, on average, one hour! A stormwater treatment mechanism was recommended that uses peat, sand and rock to remove contaminants by sedimentation, filtration and absorption (Stephenson, et al. 1997).



In the article, "Source Identification investigations of petroleum contaminated groundwater in the Missouri Ozarks," James B. Fels detailed a Missouri Dept of Natural Resources program to investigate leaking underground storage tank sites in the state, where petroleum contamination that had been found in groundwater was not associated with a known site. Many methods that were employed in this study: initial site visits, field reconnaissance, intense background information research leading to the construction of an initial model of site hydrogeology. In addition a door to door survey of local residents was conducted to locate well or groundwater odors. A potentiometric map to determine direction of regional flow and location of any groundwater divides was also generated. Finally, a field search for any potentially responsible parties within the boundaries was conducted. Once located site specific investigations of potentially responsible parties methods included: groundwater tracing at all potentially responsible party sites including photoionization detector lamp to detect the volatile compounds.

In addition measuring surface resistivity to determine depth to water, rock , bedrock and possibly areas with petroleum contamination was performed. Finally, petroleum fingerprinting in areas with multiple potentially responsible parties helped to identify individual culprits. This thorough methodology has resulted in quite successful levels of identification, punishment and cleanup.


Agriculture in general is a source of much contamination. In "Agricultural Chemicals at the Outlet of a Shallow Carbonate Aquifer" G.K. Felton studies the effects agricultural chemicals that are applied to the soil in the course of normal agricultural practice. Runoff from precipitation or irrigation carries a large portion of these chemicals to stream sinks, swallow holes or into ground fractures where they become concentrated as pollutants.

A study site of 4,700 acres in the Inner Bluegrass region of Kentucky was chosen. It contains over 40 water wells, 400 sinkholes, 2 karst windows and a sinking stream. Using aerial photos and dye tracing techniques, boundaries of the cacthments were determined. Land use was also determined from the aerial photos with some field checking. The outlet of the drainage basin is at Garrett’s String and is fed by two branches. Statistical analysis showed levels of nitrate and phosphate that were variable. Peaks were not coincident with fertilizer application, rather they occurred during periods of high soil-water content. In addition, variations in triazines implied that there was also an annual cycle of concentration. Fecal contamination much higher than acceptable for drinking water standards, with high levels of coliform and streptococci. According to the authors, what remains to be done are more focused, time-series analysis to better determine the peak variations in contamination and the relationships in transport. In addition, monitoring should be better coordinated with field tests and timing of chemical application. Finally, observation of animal land use should be conducted to determine the source of the high coliform levels, whether it is domestic or ferile.

Management of chemical application is one way to protect aquifers in agricultural regions. But management requires ascertaining present-day contamination levels so that improvements can be detected. In Logan County Kentucky, James Currens planned to test the efficacy of using a "Best Management Practices" program to improve the groundwater in this farming region. Logan County was chosen because groundwater contamination sources are limited to agriculture. In preparation for the implementation and monitoring of the program, the UKY geologist mapped the watershed and characterized the pre-BMP water quality through copious testing. Mapping was conducted using dye tracing techniques. Once watershed boundaries were determined, land use in the watershed was cataloged , according to aerial photo interpretation. 70% of the land was found to be used for some type of agriculture. Next, water quality and discharge were monitored at 10-minute intervals. Tests determined that non-point pollution by chemicals used in agriculture was ubiquitous and that the diffuse-flow areas of the basin served as sinks for chemicals and were a source of long-term contamination. With this information well-documented, Currens planned to implement the BMP program and monitor the water for improvement.

Almonas Gutkauskas writes in the Swedish environmental magazine AMBIO about the case of Lithuania which in 1991 underwent an economic collapse. Until then agriculture growing steadily with the benefit of commercial pesticides and herbicides. Coincidental with that was a deterioration in ground water. In Lithuania, 25% of the 3.71 million residents get their water from wells in karst aquifers. In addition, thirty three large scale hog complexes (range of 12-54000 head) have been constructed and their pollution potential is comparable to adding 5 million more residents. As the economy improved, farmers began to acquire chemical inputs again. To combat potential contamination , the Karst Regio Fund Tatula was established in 1993 to develop a model program to train farmers in sustainable agriculture and provide low-interest government loans. The incredible drop in non-point source pollution was enough to get governmental buy in for this program.


Sinkhole Formation and Collapse

Sinkhole Flooding

Ponding of storm water runoff in sinkholes is a part of natural hydrologic systems in karst regions. Ponding occurs at large storm events when: 1) the rate of storm water runoff exceeds sinkhole drainage capacities, 2) the capacity of the cave system to transmit storm water is exceeded and water must be stored temporarily in sinkholes, and 3) there is a backwater effect on groundwater flow due to sinkholes with bottoms lower than the level of surface streams at flood stage. Unfortunately, structures are often built within sinkhole flood plains in urban areas where flooding problems may be greatly aggravated by increased rates of runoff caused by land use changes, especially from impervious roofed or paved areas, decreased storage due to sinkhole grading and filling, and clogging of sinkhole drains by debris and silt. (Crawford, Lewis and Tucker, 1998)

Sinkhole drains are sometimes unclogged in an attempt to reduce sinkhole flooding. Also, stormwater runoff may be routed into drainage wells drilled into the aquifer. One of the most effective solutions to the problem of sinkhole flooding is the establishment of sinkhole flood plain easements

Sinkhole Collapse

While some sinkholes form slowly by solution of the underlying carbonate rock, other sinkholes develop as a result of the collapse of surface or near-surface material. There are two basic types of sinkhole-forming collapses: 1) bedrock collapses, and 2) regolith collapses. Bedrock collapses are rare and generally occur due to enlargement of cave passages in limestone. The enlargement causes the roofs above the passages to weaken and eventually collapse to create sinkholes. Regolith collapses are much more common than bedrock collapses and generally result from regolith falling into openings in the underlying limestone. In areas where the water table is usually above the regolith-bedrock contact, collapses often occur when the water table drops below the regolith-bedrock contact, either during droughts or during high-volume pumping. (Newton, 1984)

Physically, a collapse in this case is caused by loss of buoyant support for the regolith arches that span openings in the limestone. Collapses are also caused by removal of saturated regolith down the opening, enlarging the arch, and eventually causing collapse at the land surface. When the water table fluctuates above and below the regolith-bedrock contact, collapse may result from repeated wetting and drying. Regolith collapses also may occur in situations where the water table is usually below the regolith-bedrock contact. In "Review of Induced Sinkhole Development," J. G. Newton outlines these causes of sinkholes resulting from a decline of water level due to groundwater use. He also discusses the occurrence of sinkhole collapse due to surface construction.

According to Newton, construction includes the erection of a structure, modification of the land surface, or to the diversion of water. Construction, and its grading of the surface, thins the unconsolidated deposits above existing sinkholes. Putting weight on these thinned unconsolidated roofs can result in roof failure. (Newton, 1984.) In addition, vibration or shock from blasting associated with construction can hasten collapse. Diverting water may increase discharge and subsurface erosion and the creation of cavities. The same thing may happen due to broken underground pipes.

Joseph A. Fischer and Joseph J. Fischer were consulted to examine a 242-acre housing development in northwest NJ called Wyndham Farms. The site was farmland and comprised of primarily carbonate rock and their were both existing and presumed sinkholes. After examining aerial photographs and conducting on-site inspections they made some recommendations. The recommendations that they made to avoid problems of flooding and subsidence included: laying out housing sites to preserve existing grading and pattern of storm water flow. In addition, runoff will travel through a series of grassy swales. Sinkholes that presently are repositories for overland flow will be cleaned and surrounded with gabion baskets to maintain the integrity of the sinkhole and purify the flow. These will direct surface water flow to the subsurface. Steel should be added to all foundations and runoff from roof drains will be directed away from the buildings. (Fisher and Fisher, 1997)

White, Aron and White looked at the influence of urbanization on sinkhole development in central Pennsylvania. In this case they look at suffusional sinkholes and the land-uses that aggravate their development. Once a cavity has been created between bedrock and overlying soil, the roof of soil can begin to wash away, sometimes only held in place by matted roots of surface plants. According to White, et al. the most important factor influencing sinkhole formation is stormwater runoff and its modification by urbanization. Roofs, streets, driveways, and other impervious surfaces increase flashy overland flow and tend to concentrate runoff. This in conjunction with soil removal hastens sinkhole collapse. They suggest that land-use planners take great care in identifying sinkhole areas before building begins. By using geologic, topographic and soil maps sinkhole areas and potential sinkhole areas can be detected. They suggest that sinkhole prone areas be allowed to grow thick vegetative cover to support the soil. In addition, adequate storm drains should be constructed that direct water in to the bedrock, rather than into the soil surface.

In a study of urbanization in the St. Louis area, and the development of sinkholes, Ripp and Baker had similar recommendations. Recent development has brought under construction sinkhole prone areas of St. Louis that were passed up during previous construction phases. As urban sprawl continues, the potential for damage to new homes grows. For remediation and management, Ripp and Baker suggest that first, all agencies: city, county and district, cooperate in sharing information on sinkhole locations. In areas prone to sinkhole development visual observation, boring, test pits and geophysical methods should be used for on-site investigation. Once identified, action may not be necessary. However, if collapse seems potentially imminent, they recommend the use of structural plugs (a concrete plug in the bedrock) and backfilling of the sinkhole. Alternately, a graded drain may be used when the drainage to the sinkhole must be maintained. If the sinkhole is used as a storm sewer (!) then the authors recommend the use of a drop inlet. A drop inlet is a concrete standpipe inserted into a "cleaned" sinkhole. the sinkhole is then backfilled with sand, crushed concrete and crushed rock. The authors do note that this method may, in fact, accelerate sinkhole collapse but is preferred by the Department of public works because it reduces the load on municipal storm drains.


A common theme in all of the above articles seems to be the problem of delimiting and mapping catchment areas and potential sinkhole areas. Some states have undertaken mapping projects. Johnson, Kenneth S. & James Quinlan described in "Regional Mapping of Karst Terrains in Order to Avoid Potential Environmental Problems." that the Oklahoma Geological Survey took on the mapping of karst and associated environmental problems for engineers, construction and remediation uses. They produced the following map







detailing the geologic areas of Oaklahoma that are potentially susceptible to karst processes.


According to the literature about contaminant transport knowledge about karst and the transport of water in karst aquifers in critical in order to prevent pollution. The first step is mapping regional geology and hydrology, before construction or change in land use. Dye tracing studies and spring basin boundaries should be done as well. Models of flow transport not only can help find potential sources of contamination but also may be helpful in removing contaminants once a spill has occurred or an LUST has been discovered. In addition, public and private education about and enforcement of appropriate disposal measures is essential to preventing pollution. Septic system treatments, dumping restrictions and the removal of underground storage tanks should all be done proactively to prevent pollution. In construction, maintaining natural grades and sinkhole structures seems to be recommended for slowing the development of urban- based sinkholes.

However, what would be most effective, in this authors opinion, would be to stay AWAY from karst in the first place. Growth restrictions and infill development may be one answer. However, as the pressure to build from the megacities of the Eastern Seaboard moves populations onto carbonate aquifers (or consolidates huge farming operations in karst terrain) where there had previously been only low-impact agricultural land-use, it seems that pollution and sinkhole formation are an inevitable consequence.



Bibliography and References

Barner, Wendell and Kristine Uhlman. 1995. "Contaminant Transport Mechanisms in

Karst Terrains and Implications on Remediation." In Beck, Barry F. Ed. 1995. Karst Geohazards; Engineering and Environmental Problems in Karst Terrane. Rotterdam, Netherlands: Balkema. Pp. 207-212.

Beck, Barry F., Editor. 1984. Sinkholes: Their Geology, Engineering, and Environmental

Impact: Proceedings of the First Multidisciplinary Conference on Sinkholes. Orlando, Florida.

Covers the geologic framework and mechanisms of development, site studies, susceptibilities, environmental impacts, case studies and engineering recommendations for development in sinkhole areas. 429 pages.

Beck, Barry F. Ed. 1995. Karst Geohazards; Engineering and Environmental Problems in

Karst Terrane. Rotterdam, Netherlands: Balkema.

Beck and Stephenson, Editors. 1997. The Engineering Geology and Hydrogeology of

Karst Terranes. Rotterdam, Netherlands: Balkema.

Proceedings of the 6th multidisciplinary conference on sinkholes and the engineering and environmental impacts of karst, held in Springfield Missouri in 1997. International in perspective and comprehensive in scope: sinkholes, hydrogeology, groundwater contamination, protection, flooding, etc, etc…

Crawford, Nicholas. 1984a. "Toxic and explosive fumes rising from carbonate aquifers: a

hazard for residents of sinkhole plains." In Beck, Barry F., Editor. 1984. Sinkholes: Their Geology, Engineering, and Environmental Impact: Proceedings of the First Multidisciplinary Conference on Sinkholes. Orlando, Florida. A.A. Balkema, Rotterdam, Netherlands.

Ground goes BOOM! Seriously, this study of toxic and explosive fumes in wells, sinkholes and from homes in Bowling Green, Kentucky reveals the extent and range of contaminant transport.


Crawford, N.C. 1984b. "Sinkhole Flooding Associated with Urban Development upon

Karst Terrain: Bowling Green, Kentucky". In Beck, Barry F., Editor. 1984. Sinkholes: Their Geology, Engineering, and Environmental Impact: Proceedings of the First Multidisciplinary Conference on Sinkholes. Orlando, Florida. A.A. Balkema, Rotterdam, Netherlands, 283-292.

Crawford, N.C. and A.J. Whallon 1985. Hydrologic Hazards in Karst Terrain. U.S.

Geological Survey Water Fact Sheet, 2.

Currens, James. 1995. "Mass Flux Of Agricultural Nonpoint-Source Pollutants In A

Conduit Flow Dominated Karst Aquifer, Logan County, Kentucky." In Karst Geohazards; Engineering and Environmental Problems in Karst Terrane. Rotterdam, Netherlands: Balkema. Pp.179-187.


Daoxian, Yuan, Zhu Dehao et al. 1991. Karst of China. Beijing, China. Geological Publishing House.

History of karst, major karst types, caves and environmental consideration of karst in China.

Day, M. J. 1993. "Human impacts on Caribbean and Central American Karst." In

Williams, P.W., Editor. 1993. CATENA Supplement: 25, Karst Terrains, Environmental Changes , Human Impact. Germany: Catena Verlag. Pp. 109 – 125.

Non North-American view of human impact.

Elhatip, H. 1997. "The influence of karst features on environmental studies in Turkey."

In Environmental Geology Vol 31(1/2). Pp. 27 – 33.

Turkey is 1/3 karst. Awareness of the environmental issues in karst areas is relatively new. Description of some common dumping practices that need to be curtailed.

Fels, James B. 1997. "Source Identification investigations of petroleum contaminated

groundwater in the Missouri Ozarks." In Beck and Stephenson, Editors. 1997. The Engineering Geology and Hydrogeology of Karst Terranes. Rotterdam, Netherlands: Balkema.

The problem of leaking underground storage tanks and the contamination of karst aquifers.

Felton, G. K. 19 Agricultural Chemicals at the outlet of a shallow carbonate aquifer. in

Transactions of the American society of Agricultural Engineers. Vol (39)3:873-882.

Introduction to the transport mechanisms and chemical constituents in Agricultural pollution with particular attention paid to karst.

Fischer, Joseph A. and Joseph J. Fischer. 1997. "Wyndham Farms – A Karst Case

History." In Beck and Stephenson, Editors. 1997. The Engineering Geology and Hydrogeology ofKarst Terranes. Rotterdam, Netherlands: Balkema.

Gentile, R.J. "Geology and utilization of underground space in metropolitan Kansas City

Area, USA." In Environmental Geology. Vol. 29(1/2) Pp. 11 – 16.

OH my goodness! Almost 4000 people commute to work everyday UNDER Kansas City, which has developed old mines into office and warehouse spaces. It’s vibration free, low-cost, insulated and quiet but, sheesh! how scary.

German, E.R. 1996. Analysis of Nonpoint-source Ground-Water Contamination in

Relation to Land Use: Assessment of Nonpoint-Source Contamination in Central Florida. USGS Water Supply Paper 2381-F. prepared in cooperation with the Florida Department of Environmental Protection.

A report of the findings of a study to determine the effects of three different land-uses on Florida’s ground-water quality. Compares citrus, mining and urban land-use for contamination potential and water quality.

Gutkauskas, Almonas. 1997. "The reorganization of Lithuanian Agriculture Towards

Sustainable Farming" in Ambio. Royal Swedish Academy. Pp. 442 – 444.

The economic collapse of the early 1990’s caused a decline in agricultural chemical use and agricultural production. Coincidentally, ground water quality improved markedly at this same time. Now that the economy is improving, the government is training farmers in "sustainable" , non-chemical methods of farming.

Hardwick, P & J. Gunn. 1993. "The Impact of Agriculture on Limestone Caves."

In Williams, P.W., Editor. 1993. CATENA Supplement: 25, Karst Terrains, Environmental Changes , Human Impact. Germany: Catena Verlag. Pp. 235 – 249.

Heath, Ralph C. 1998. Basic Groundwater Hydrology. USGS Water Supply Paper 2220.

Prepared in cooperation with the North Carolina Department of Natural Resources and Community Development.

Hydrologic cycle and groundwater basics. Nice two color graphics. Includes lots of definitions for neophytes and technical information for experts.


Hoke, John A. & Carol M. Wicks. 1997. "Contaminant transport in karst aquifers." In

Beck and Stephenson, Editors. 1997. The Engineering Geology and Hydrogeology of Karst Terranes. Rotterdam, Netherlands: Balkema.

Process of transport: prediction flow is more possible on a regional rather than conduit-level scale. Description of linear systems model.

Johnson, Kenneth S. & James Quinlan. 1995. "Regional Mapping of Karst Terrains in

Order to Avoid Potential Environmental Problems." In Cave and Karst Science, Transactions of the British Cave Research Association. Vol. 21(2).

The Oklahoma Geological Survey takes on the mapping of karst and associated environmental problems for engineers, construction and remediation uses. Focus on data collection, mapping and questions of scale.

Johes, Russell L. and Arlin Bostian. 1989. Developing Management Practices for

Preventing Residues of Agricultural Chemicals in Drinking Water Wells. In Ground Water Management Resources. Fall 1989. Pp. 75 – 78.

Management in non-karst aquifer. Very conservative and pro-chemical. Yucky.

Newton, J.G. 1984. "Review of induced sinkhole development." In Beck, Barry F.,

Editor. 1984. Sinkholes: Their Geology, Engineering, and Environmental

Impact: Proceedings of the First Multidisciplinary Conference on Sinkholes. Orlando, Florida.

Description of induced sinkholes, their formation and collapse. Focus on problems caused by construction.

Ripp, Bryan J. & John Baker. 1997. "Urbanization in karst sinkhole terrain – A St. Louis

perspective." In Beck and Stephenson, Editors. 1997. The Engineering Geology and Hydrogeology of Karst Terranes. Rotterdam, Netherlands: Balkema.

The interface between urban sprawl and sinkhole development. Techniques for identification, mapping and remediation.

Smith, D.I. 1993. "The Nature of Karst Aquifers and their Susceptibility to Pollution" in

Williams, P.W., Editor. 1993. CATENA Supplement: 25 Karst Terrains, Environmental Changes , Human Impact. Germany: Catena Verlag. Pp. 41-58.

Focus on Hydrologic processes unique to karst.




Stephenson, J. Brad, W.F. Zhou, Barry Beck & Tom Green, 1997. "Highway Stormwater

Runoff in Karst Areas – Preliminry Results of Baseline Monitoring and Design of a Treatment System for a Sinkhole in Knoxville, Tennessee." in Beck and Stephenson, Editors. 1997. The Engineering Geology and Hydrogeology of

Karst Terranes. Rotterdam, Netherlands: Balkema.

White, William B. 1988. Geomorphology and Hydrology of Karst Terrains. London:

Oxford Press.

THE book on the basics of karst landforms and processes. Good diagrams references. And it’s authored by "Mr. Karst".

White, Elizabeth, Gert Aron & William White. 1984. "The influence of urbanization on

sinkhole development in central Pennsylvania." In Beck, Barry F. 1984. Sinkholes: Their Geology, Engineering, and Environmental Impact: Proceedings of the First Multidisciplinary Conference on Sinkholes. Orlando, Florida.

The anatomy of suffusion sinkholes and the effects of urbanization on their formation. Some remediation techniques as well.

Williams, P.W. 1993. Environmental Change and Human Impact on Karst Terrains: An

Introduction. In CATENA Supplement, 25 Karst Terrains, Environmental Changes , Human Impact. Germany: Catena Verlag. Pp. 1 – 19.

Good intro to anthropogenic karst change… see below.

Williams, P.W. 1993. Editor. CATENA Supplement, 25 Karst Terrains, Environmental

Changes , Human Impact. Germany: Catena Verlag.

Supplement to the soil science journal Catena. Good introduction to nature of karst and its sensitivity. Articles range from articles on karst in different regions with a focus on impact of agriculture, tourism, quarrying, acid rain, agriculture on karst terrain.

Williams, P.W. (1983). "The Role of the Subcutaneous Zone in Karst Hydrology."

Journal of Hydrology, 61, 45-67.