Karst Hydrology -The Dye is Cast

Tom Aley, Ozark Underground Laboratory, Protem, Missouri 65733 USA

Introduction

I am delighted to have been asked to deliver the keynote address for a conference on the theme "The Water Below" with a sub-theme of "Whatever Goes Down, Comes Up". Protecting or improving the quality of karst groundwater requires public understanding of the highly vulnerable nature of karst groundwater and the fact that contaminants do not simply "go away", but instead, come back to cause problems for people and the environment.

I have three objectives that I will try and interweave into this presentation. First, I want to give you some perspective on the nature and extent of karst and karst water problems in the United States. Secondly, I want to illustrate some of the misconceptions that people commonly have about karst groundwater. Finally, I want to illustrate how we have used dye tracing in the United States to address the problems, educate the public, and improve the level of protection afforded karst water supplies.

The adage "whatever goes down, comes up" has been of great benefit in enhancing public concern about karst groundwater quality. In the 1980's the Virginia Cave Board published a lovely cave and karst poster that has been widely distributed to schools and other facilities. Emblazoned across the poster are the words "In Karstlands, What Goes Down Must Come Up". Among other features, the poster shows springs, caves, cave life, and sinkhole dumps. The poster and the adage have been effective in raising public awareness of the sensitivity of karst resources to groundwater contamination.

In 1973 I worked with a television station in Springfield, Missouri which produced a 30 minute-long television documentary entitled "Please Don't Drink the Water". In the film I guided the audience through features of Missouri's karstlands which included the boy's restroom at an elementary school (and a spring in a nearby city park to which the school septic system drained); a sinkhole dump and the associated spring (which supplied untreated water to an adjacent restaurant); and a sewage treatment plant that discharged wastewater to a sinking stream which contributed water to Big Spring, Missouri, 62 km away. A week before the program was to be shown the television station launched a search for "appropriate" music for the program. Their level of desperation was demonstrated when they asked if I would write and perform a karst water pollution ballad for them. We karst people must be multi-faceted! While the program was a national finalist in Emmy competition as the best independently produced documentary that year, it is perfectly clear that the music had nothing to do with the accolades. The words to "Cold, Clear, but Not Pure" follow (and were sung at the conference).

Oh won't you take care of our springs and our rivers,
Let them flow clear, let them flow free
The filth from our sewers, our feedlots, our dump grounds
Flows from the springs and slides to the sea

Chorus
Cold, clear, but not pure
There's one thing that's for sure
Ozark springs are beautiful things
Cold, clear, but not pure

Too much of our ground won't filter the water
Whatever goes down is going to come up
We must cleanse our waters up here on the surface
Whatever goes down is going to come up

We still have the time to make a beginning
We did it at Dora and Winona too
We can take care of our springs and our rivers
But it takes all our neighbors, me and you

Karst in the United States: the Setting

About 20% of the United States is underlain by karst aquifers. In many cases the surface expression of the karst is subtle due to a thick mantling of soil, residuum, or sometimes alluvium. The American karstlands are disproportionally located in southern and eastern portions of the nation. While many of the karst areas are rural with farms of 100 hectares or less, many cities and towns are located atop the karst. Major cities with karst and/or karst aquifers include San Antonio and Austin, Texas; St. Louis and Springfield, Missouri; Lexington and Louisville, Kentucky; and most of the cities of Florida including Miami and Orlando. Perhaps as much as a third of America's population lives atop karst aquifers.

Average annual precipitation in many of America's karst regions averages between 100 and 150 cm. Such areas typically yield 25 to 45 cm of annual runoff water, much of which passes into and through the karst groundwater system. As an illustration, much of the Ozarks, a large dissected plateau region which extends across much of Missouri, Arkansas, and smaller portions of Oklahoma, is underlain by limestones and dolomites. In these areas, about 75% of the annual water runoff passes into and through the karst groundwater system for at least some distance. Of the water which enters the groundwater systems of the Ozarks, about 75% of it enters through localized recharge zones (such as sinkholes and losing streams) which provide little natural cleansing for the introduced water. The impacts are rather predictable; whatever goes down, comes up. Add this to the large human populations (at least large compared with Australia) and the magnitude of the problem is clearly enormous, yet still poorly recognized by most of the public.

Whiskey's for Drinking, Water's for Fighting

Much of the western United States is arid. Karst and water are both limited in this region, and fights over water and water rights have often been of legendary proportion. In the West, one soon learns the axiom that "whiskey's for drinking and water's for fighting".

Americans have become infamous for their use of the courts and their concurrent love/hate relationship with attorneys. While not all of the fights over water have taken place in courtrooms, the courts have been arbitrators of many disputes involving water rights, water pollution, and various environmental problems associated with water resources. While some of the issues have been trivial, most have not. Furthermore, the American trend for both individuals and environmental organisations to use the court system to address water problems has often been extremely beneficial in protecting and restoring groundwater quality in karst areas, and in raising public awareness about karst groundwater issues.

Litigation requires proof. Proof requires experts. Experts require studies upon which to base their findings and ultimately their testimony. Some of the best of the karst water studies in the United States have resulted directly from litigation or the threat of litigation. Unlike the basic research studies commonly conducted by academic personnel, the litigation-associated studies must focus on crucial issues and avoid tangential topics regardless of how much such topics may intrigue the investigator. Conclusions must be credible and readily comprehensible to judges and jurors who often lack scientific backgrounds. Furthermore, conclusions and the investigations upon which they are based must be capable of withstanding both direct and flanking attacks by well prepared attorneys. Even though the litigation-associated studies are seldom reported upon in the scientific journals, they have benefited both public and professional knowledge and understanding of karst resources and the sensitivity of karst resources to human-derived damage.

The Dye Is Cast; Green Ducks and Better Methods

Groundwater tracing is often the most crucial tool in karst investigations, especially when litigation is involved. The tracer dyes can be used to answer the first two of the following questions and to provide information on the third:

1) Where does the water from a particular point go?
2) How long does it take for the water to reach particular points?
3) What happens to the water (and/or contaminants) along the way?

As we shall see, the older dye tracing methods do not provide the definitive data needed for work related to litigation.

Groundwater tracing with fluorescein dye began a hundred years ago. Large quantities of dye were commonly used, and dramatic visual effects sometimes resulted. One American ballad asks, "what do you do when the ducks turn green, do you say I'm sorry ma'am, or do you sneak off of her land?" Visual dye tracing work yielded some important discoveries, but yielded many failures amid the successes. While visual tracing work might have been acceptable in the past, it is clearly not the "state of the art" approach needed for karst investigation which may ultimately be presented in court.

Other fluorescent dyes became available subsequent to fluorescein and were used in limited cases. In the 1950's a few cavers and hydrologists began using packets containing activated carbon (alternately called charcoal). The activated carbon is able to adsorb and accumulate several of the tracer dyes, including fluorescein. Collected packets are treated with a solution of one or more strong bases dissolved in a water and alcohol matrix. This treatment, called elution, desorbs the tracer dye or dyes and, at great enough concentrations, make the dyes visible in the liquid mixture which had been poured over the activated carbon. The visual detection of the dyes is greatly enhanced by working in a dark room and shining a flashlight into the sample bottle. The "standard" sample bottles were often jars which had formerly contained baby food. Many very important groundwater traces were conducted in American karst areas using these approaches, although this work often identified only the major dye recovery locations and missed other points to which the tracer dyes flowed. Other materials adsorbed or trapped in the activated carbon packets often yielded color to samples, and inexperienced people sometimes reached incorrect conclusions about dye recovery locations.

While dye tracing using activated carbon samplers and visual detection methods was a great improvement over the "dump in lots of dye and look for color" approach, it had serious limitations for litigation work. One can visualise an attorney asking why the analysis used baby food jars and a flashlight in a dark room rather than a "scientific" method. I was once confronted in court by an attorney who had engaged a laboratory to make up about a dozen bottles with activated carbon in the bottom and solutions containing various dyes and mixtures of dyes lying atop the carbon. With theatrical flair he swirled the liquid in each of the bottles and then placed them in front of me and challenged me to tell him which bottle or bottles had fluorescein. I responded that his laboratory had unfortunately used unwashed carbon and that his swirling of the mixture had suspended the finer materials thus preventing me from properly assessing the samples. I further told him that I could not make any determination until all of the fine textured carbon particles had settled and that would be at least a few hours or maybe overnight. That ended the botched demonstration, yet if his laboratory had done a good job and he had not swirled the samples I might not have been able to correctly separate certain dye mixtures. Fortunately, my conclusions in this case were based upon visual observations which had been confirmed by instrumental analysis with a spectrofluorophotometer operated under a synchronous scan protocol. Still, the susceptibility of the typical dye tracing approach of the time to attack by clever attorneys was obvious. My purchase of the spectrofluorophotometer a few months earlier had clearly been an upgrade in techniques made in the nick of time.

Filter fluorometers are sometimes used to analyze for tracer dyes. However, variations in fluorescence background and interference with other fluorescence compounds limit the utility and credibility of these instruments for groundwater tracing investigations. A sample spiked with coffee or water in which broccoli has been cooked will produce a fluorescence peak when filter fluorometers are operated with the filters commonly used for detecting fluorescein dye. While these fluids are not commonly encountered in karst waters, they illustrate limitations of the filter fluorometers and the problems one could have using data from these instruments in court. I can readily visualize a sarcastic attorney asking if I relied upon an instrument which cannot discriminate between fluorescein dye, coffee, and broccoli water for determining where fluorescein dye was recovered and at what concentrations.

Appropriate spectrofluorophotometers can be programmed to set a number of parameters for analysis. Synchronous scan protocols are typically best suited to the analysis of tracer dyes in elutants from activated carbon packets or in water samples. The excitation and emission wavelengths are synchronously scanned at a bandwidth separation selected to maximize the fluorescence intensity of the dye or dyes which may be present; the emission fluorescence intensity is typically printed on an analysis graph. Each of the dyes has maximum fluorescence intensity within a relatively narrow wavelength range. The acceptable wavelength range of a particular dye is a function of the matrix of the sample (for example, whether it is a water sample or elutant from an activated carbon sampler).

The synchronous scan instruments are research-grade instruments and are very expensive. They are laboratory instruments; samples need to be brought to them. Furthermore, proper use requires experimentation and experience with the instrument; one cannot simply plug it in and push a few buttons. The great attribute of the instruments is that they can credibly detect many of the fluorescent tracer dyes in "real world" samples at concentrations as low as a few parts per trillion, and at lower concentrations in some samples. Such instruments have vastly expanded our abilities to do groundwater tracing work We are no longer limited to relatively simple groundwater traces. As an illustration, during the last ten years groundwater tracing has become a commonly used tool at hazardous waste sites located in American karst landscapes, and has proven particularly effective in groundwater tracing in the epikarstic zone.

Initially the cost of synchronous scan instruments and the associated facilities and experience needed to analyze dye samples drastically limited the number of dye traces which could utilize such equipment. In the late 1980's the Ozark Underground Laboratory began a program of assisting investigators in designing and conducting groundwater traces; part of this program was analysis of the resulting samples at our laboratory with our instrumentation and protocols. This has greatly facilitated good quality groundwater tracing work in the United States and Canada, and we occasionally receive samples from more exotic places such as Barbados and Nepal (but no samples as yet from Australia or New Zealand). A couple of other small laboratories in the United States have recently begun to provide some similar services.

Karst hydrologists involved with practical problem-solving work in the United States have conducted a large number of groundwater traces; the Ozark Underground Laboratory has directed or assisted with about 3,500 traces. Earlier I indicated how litigation issues essentially forced karst hydrologists involved with problem-solving work to improve groundwater tracing and dye analysis methodologies. In my opinion, litigation has been the most important force in moving American karst hydrologists into developing and implementing better scientific methods and techniques for groundwater tracing work. This is worth remembering as we consider the impact of attorneys upon society.

The Two Most Common Elements in Karst Areas

I am convinced that the two most common elements found in karst areas are calcium and stupidity. Calcium is more widely recognised than is stupidity. The direct connections between sinkholes and groundwater supplies should be fundamentally obvious, yet sinkholes seem to attract trash as readily as magnets attract iron filings.

With time we make some gains against karst stupidity, which we sometimes call ignorance to protect the guilty. For the last 30 years we have conducted educational field programs for high school and college students at the Ozark Underground Laboratory. Throughout this period I have routinely lectured at a sinkhole and asked if people believed that natural cleansing processes would effectively remove contaminants introduced into the sinkhole prior to the associated water discharging from springs 1.6 km away. For the first ten years or so the answers were invariably yes. People believed that the sinkhole and the connecting karst system should be expected to remove whatever contaminants might be present. For the next five years the answers were mixed and provided great discussions. For the last 15 years the answers invariably have been no. People now correctly recognize that the sinkhole and the connecting karst system should not be expected to remove contaminants. I think the change in answers we have noted is a direct result of groundwater pollution problems which have received enormous press attention. These problems have occurred in both karst and non-karst landscapes. I wonder what the answers to similar questions would be in Australia and New Zealand.

Harlan's Wells

Some people have a total lack of karst understanding. One of these people was Harlan. He called me on the phone a few years ago and asked me to help him with a pollution problem he was having on his property with his new well. I asked if he had an older well on the property, and he said of course he did; why else would he drill a new well. I guess we hydrologists can be rather dense. Anyway, I went to visit him and he took me to the new well, which he said had a total depth of 76 meters (of course the value was in feet since the United States still uses the English units). I asked how much casing the well had, and he replied 24 meters. Casing is the pipe that is placed around the outside of the well to prevent soil and near-surface waters from entering the well; it is of critical importance in protecting the quality of the groundwater. I asked Harlan why he only had 24 meters of casing, and he replied that this was the length that the state regulators said he had to have. That is the minimum value for karst areas in Missouri, yet I told Harlan that it was not an adequate amount in an area in which he already had contaminated groundwater. Harlan replied that he did not want anything more than the minimum; I suspected that he had used a similar strategy when IQ was being passed out.

I next asked Harlan where his old well was, fully expecting it to be on the opposite side of the house or at some other relatively distant point from the new well. Harlan pointed to a dilapidated box I had mistaken for a dog house and said "there". It was 3 meters from the new well. I was beginning to understand the problem better.

"How deep is your old well" I asked.

"46 meters" he answered.

"How much casing is there in the old well" I asked.

"Clear to solid bedrock, which is about 2.5 meters" he replied.

This "solid bedrock" concept has long bothered me. How can it be solid and still permit water to enter the well? There is a sign on the highway within a kilometer of Harlan's house advertising one of the local show caves; why is bedrock so routinely prefaced with the adjective "solid"? I've spent a career walking, crawling, and climbing through this "solid bedrock" I couldn't get through if it were solid.

Harlan's old well had served to introduce contaminants at least to the bottom of that well (46 meters below the surface). The new well, only 3 meters from the original well, served to move the contaminants down another 30 meters.

"Did you abandon the old well in accordance with state standards?" I asked . Such abandonment requires that the well be filled with a slurry of cement and water.

"Nope" he replied. "I thought I might want to use it to water the lawn sometime".

I responded by launching into my lecture number 17 on stupid things you should never do.

Harlan quickly interrupted. "I'm paying you by the hour, you know, and I know what the problem is. It's my filthy neighbor's septic tank. Their sewage just lays on the ground over there about 76 meters from my well."

"Well, that could be the problem" I said. "Now where is your septic system?"

"Works fine" he said.

"I still need to know where it is" I replied.

"Look, I told you it works fine. I never see anything on the surface" Harlan replied with increasing frustration at the mental slowness displayed by his "pay 'em by the hour" hydrologist.

I then explained that wells get their water out of the ground. If his septic system put water into the ground then there might be some groundwater connection between his toilet (and his neighbor's toilet) and his two wells. When he finally showed me the location of his septic system it was 38 meters from the two wells (half the distance of the neighbor's septic system).

Subsequent sampling of the new well found readily detectable optical brighteners, which are fluorescent dyes used in laundry soaps and detergents to make white clothes and linens appear "whiter". Optical brighteners are definitive proof of sewage contamination. Harlan thought we should figure out whose toilet he was drinking from, but I think I convinced him that neither toilet was a desirable water source.

I believe in doing thorough work, so I investigated a small duck pond located about 15 meters from the new well and a little closer than that to the old well. A spring discharged into the pond, but no water flowed out of the pond. It looked suspicious to me, and the water had a faint odor suggesting too many ducks and not enough dilution. The subsequent introduction of fluorescein dye into the pond turned Harlan's well a distinctive anti-freeze green color. He could stand at his kitchen window and look at a glass of water from his well (which he was still using for drinking water). The color in the glass of water would be the same as in the pond. I hoped Harlan would see a connection here.

On my first walk over to the duck pond and about 12 meters from the new well I stumbled over a rusty can sticking out of the ground in Harlan's front yard. Assuming that nobody would want a rusty can in the yard, I started kicking at it to get it loose.

"Don't dig there...you'll uncover the dump" complained Harlan.

My eventual report to Harlan was rather simple. A third well, with casing deeper than the bottom of either of the two existing wells, was not a viable strategy since very poor quality waters with elevated hydrogen sulfide concentrations are found at such depths in the area. The recommended strategy was to drill a new well 250 meters away and outside of the area with the existing wells, septic fields, dump, and duck pond. Water from the new well could then be piped to the house. Harlan's actions had in effect dedicated the vicinity of the existing wells to waste disposal.

A year later an attorney called me and said that he was considering representing Harlan in a lawsuit against a well driller for drilling a new polluted well. I was initially startled, since my recommended well location should have been far outside the contaminated area. Then I remembered Harlan, so I asked for the actual location of the new (third) well. Harlan had done it again; the new well was midway between the two septic field systems and, of course, had only the minimum requirement of 24 meters of casing.

Harlan never understood that when you mix dirty water with clean water you never get clean water. He never understood that in karst areas the surface and subsurface are intimately connected. He ultimately spent about $18,000 US on wells and water treatment attempts before filing for bankruptcy. His family members were repetitively sick for five years. On his 4 hectares of land Harlan had a microcosm of the water pollution problems typical of karst areas. He had wells placed in unsuitable locations and with inadequate amounts of casing for their locations. Based upon my experience in the Ozark karstlands about 60% of all private water wells are not always free of bacteria because the groundwater system is very directly connected with surface waters. He had a polluted well which had not be appropriate filled and abandoned (there are thousands of such wells, many of which cannot be found on the surface because of subsequent uses of the land). He had contamination from sewage (such problems are common with both municipal sewage systems and on-site systems). He had agricultural wastes from the duck pond; land application of manure from industrial-scale poultry and swine operations causes many problems in American karst areas. He had an old dump; dumps and landfills that leak into groundwater are also extremely common. Most unfortunate of all, there are many other "Harlans" living in karstlands.

The Cave House and Pointed Things that Hang Down

One day I received a phone call from an attorney representing a businessman who had a purchased a "cave house" in Arkansas. He had paid over three million dollars for the cave house, and the attorney said he now had a water problem which perhaps I could solve. I asked what, precisely, was a cave house? The attorney replied that it was a home built inside a natural cave, and that the ceiling had "lots of those pointed things that hang down" on it. The "water problem" was that the ceiling leaked even though the seller had "waterproofed" it by spraying it with an epoxy resin. The new owner reported that the ceiling had not leaked when he had negotiated the purchase in late September (the end of the summer dry season in Arkansas), but that the problem developed in December (the time at which winter rains start and plant evapotranspiration has essentially ended).

There were over 100 leaks in the ceiling of the cave house, and the owner had been forced to hire a young man to work as a caretaker. One of the young man's major duties was emptying 100 plus buckets. He complained that this job kept him very busy anytime it rained hard. We instituted studies of the drippage rates in anticipation of litigation, which displeased the caretaker since now he had to record data on when he emptied each bucket and how much water it contained. What fascinating data we gathered! Drippage rates and flow routes in the epikarstic zone vary dramatically among storms. Other studies disclosed innumerable other problems. One of these was particularly interesting and related to a fundamental problem with putting a wine cellar stocked with thousands of dollars of selected bottles of wine in a cave with relative humidity close to 100% and a large population of cave crickets. While not reported in the technical literature, it appears that cave crickets are fond of eating wine labels. The owner's stock of wine was reduced to two kinds: red and white.

Litigation on the cave house is still continuing. Since the property was sold as a "house", one of the issues involves implied warranties of inhabitability. Is a cave house with over 100 leaks in the ceiling inhabitable? Are the various problems "fixable"? The buyer has found that one cannot engineer around all karst problems.

The problems incurred by Harlan and the buyer of the cave house were largely due to their own actions. In contrast, many of the karst groundwater pollution problems in the United States result from industrial activities which discharge contaminants to points far beyond their property boundaries and impact innocent people.

The Pindall Landfill

Pindall, Arkansas is a rural community of about 50 people. An absentee landowner proposed construction of a 64 hectare landfill on an upland area about 1.6 km from town. The landfill proponent was not in the landfill business, and some people suggested that the landfill was likely to be used for disposal of pre-treatment wastes from large chicken processing facilities in the region since they had lots of waste and no disposal site. Several of the homes in town derive their drinking water from shallow wells in the floor of the losing stream valley in which the community is located. Other private wells were located in areas surrounding the proposed landfill site, and karst springs are abundant throughout the region. Mitch Hill Spring, a large spring tributary to the Buffalo River, is about 7 km southwest of the proposed landfill. The Buffalo River is a unit of the National Park System, and is the most significant national park in Arkansas.

The Ozark Underground Laboratory was hired by area residents to identify and assess the off-site hydrologic impacts which could result from a landfill at the site. Fluorescein dye was introduced into a small sinkhole 40 meters outside the boundary of the proposed landfill. The dye recoveries are summarized in Table 1. Note that the dye trace resulted in recoveries from eight different sampling stations (five of which were domestic wells) and that the dye recovery pattern demonstrated multi-directional flow from the sinkhole. I published a technical paper on this and other traces in the area (Aley, 1988).

Table 1. Successful Groundwater Traces from the Holder Sinkhole.

Dye Recovery Site	Distance and Direction from Dye Introduction Site (m)	Time of First Dye Arrival (Days After Introduction)
Holder Well	460m south	Within 5 days
Cannon Spring	3,475m southeast	3 to 5 days
Keith Spring	2,560m north	26 to 33 days
Young Well	1,600 m. north	38 to 41 days
Henson Well	1,585m north	38 to 41 days
Herron Well	1,615m north	38 to 41 days
Nichols Well	1,465m north	38 to 41 days
Mitch Hill Spring	7,165m southwest	About 61 days

The groundwater tracing program in the Pindall area demonstrated that there were major seasonal changes in groundwater flow directions in the regional karst aquifer. Under wet weather conditions much of the water from the proposed landfill site would flow northward to Pindall and beyond that community to Keith Spring. Under dry weather conditions the water discharged from Keith Spring would, for a few days, flow northward down a surface channel for about 335 meters and then sink into the ground. After a few days of this behavior, flow from Keith Spring would cease until major precipitation occurred. In a subsequent dye trace in the area we introduced a tracer dye at the sinking point downstream of Keith Spring and recovered it from Mitch Hill Spring 8.8 km southwest of the sinking point. Under these conditions, much of the water from the proposed landfill site would also flow southwestward to Mitch Hill Spring. The seasonal movement in the location of the general groundwater divide in the area is about 3.5 km. Of greater importance to the people of Pindall, the tracing work demonstrated that groundwater resources in a large area could be impacted by leachate from the proposed landfill. These groundwater resources included a deep public water supply serving portions of Pindall and two other communities. This well had been constructed with 150 meters of pressure grouted casing, yet three of our traces yielded dye in this well.

The Pindall issue attracted tremendous media attention. A press conference on the steps of the state capitol plus numerous interviews of the people of Pindall, their attorney, and me provided a unique forum for public education about karst hydrology. The issue twice appeared on the nightly national television news. Ultimately, the state of Arkansas (in a day-long meeting under the glare of local and network television cameras) reversed its initial approval of the landfill site and denied the landfill application. Cathy and I missed the celebration party; we were driving to another landfill case in a karst area of Alabama.

The Pindall issue illustrates several important points. First, groundwater movement in karst can be very complex and in some cases can change dramatically between wet and dry seasons. Secondly, the complexity of the groundwater system could not have been demonstrated without dye tracing studies, and the studies could not have withstood the bitter legal and political challenges had they not utilized synchronous scan analytical instruments. Thirdly, the American press and public are very much concerned with water quality issues, especially if the positions involve comprehensible studies such as groundwater tracing. One measure of American concern with water quality is the amount of shelf space devoted to bottled water in the grocery stores. I visited a large grocery in Mt Gambier; there is a lot of bottled water in Australian karst areas too.

Green Forest: the Fowl Deed

Green Forest, Arkansas, is a community of about 2,500 people located in a karst region. The town attracted a large chicken processing plant which, at the time of our involvement, discharged its wastewater with little or no pre-treatment to the town's sewage plant. One investigation indicated that about 95% of the waste load at the plant (as indicated by the biochemical oxygen demand) was derived from the chicken processing plant. In simple terms, the town's sewer plant was incapable of handling the amount and type of wastewater it was receiving from the industry. The result was the discharge of very poorly treated wastewater (which often was red and had long stringy things in it) to a losing stream. The losing stream routinely sank into the karst groundwater system about 1 km downstream of the sewage plant, introducing typical daily effluent volumes of 4.7 to 5.7 million liters per day of the wastewater into the regional karst aquifer. One could sometimes hear sucking and slurping sounds ten or twenty meters away from the point where much of the water sank in the channel of Dry Creek.

Several groundwater traces were conducted to delineate the areas impacted by the sewage discharge. Dyes were introduced into the discharge from the sewage treatment plant. Springs, surface streams, and private drinking water wells were monitored for the tracer dyes during these traces. The private wells ranged in depth from less than 30 to more than 300 meters. Under moderate to high flow conditions water is recharging the regional karst aquifer at many points and the sewage recharge in Dry Creek is only one of these. When these conditions exist, the dye tracing results indicate that the area where groundwater quality was degraded by the sewer plant discharges encompasses about 122 square km with about 52 square km of this region being heavily affected.. Under low flow conditions the sewage recharge in Dry Creek is one of only a few operating groundwater recharge zones and routinely recharges more water than any other recharge point in the area. When these conditions exist, the dye tracing results indicated that the area with degraded groundwater quality encompasses about 156 square km.

Dye from the dry weather dye introduction was recovered from 38 of the 87 sampling stations used during that study; 18 of the dye recovery stations were private wells. The dye monitoring stations were in two adjacent topographic basins. Typical travel times for the first arrival of tracer dyes at wells and springs was three to four weeks after dye introduction; dye was recovered from wells as far as 10.3 km from the point of dye introduction.

Water budget studies indicated that approximately 6% of the water in the affected 156 square km groundwater system had passed through the sewer plant. During my courtroom testimony an attorney for the chicken processing plant leaped upon the 6% value as demonstrating tremendous dilution provided by the karst groundwater system. I responded by telling him that I had conducted a study the previous weekend to put the 6% value in perspective. First, I had placed some tracer dye in the bowl of my toilet and then measured the concentration in the bowl. Next, I flushed the toilet and again measured the concentration of dye when the bowl had refilled. I found that, after one flush, my toilet bowl contained only 0.6% as much dye (and by inference, sewage) as it did before the flush. That meant that the groundwater system around Green Forest had 10 times more sewage in it than my toilet bowl did after one flush, and I didn't think you should drink out of a toilet bowl after one flush. Or for that matter even after two or three flushes, although I testified that I had a cat that did. For some reason the attorney did not pursue the dilution argument any further.

The Green Forest issue illustrated how rapidly and extensively pollutants can move through karst aquifers and destroy the utility of groundwater supplies. The dye tracing results were used to identify the area to be served with public water supply lines. Prior to the construction of these lines, dye tracing results were used to issue public health advisories and to identify areas to which clean water would be delivered on an emergency basis by the Arkansas National Guard. The dye tracing results were critical to the outcome of the litigation. During the litigation the dye tracing work withstood extremely well financed efforts to diminish or negate its integrity and credibility.

Public water supply lines were ultimately constructed to serve all residences in the polluted area. I estimated that the net economic loss from this one pollution source was on the order of $10 million US Much of the construction cost was paid with public funds, but landowners in the area lost previously useful wells and springs. A number of residents were convinced that they had suffered illnesses, and some were convinced that family members had died due to the pollution. A major court case resulted in which the chicken processing company was convinced of a number of counts of violation of the Federal Clean Water Act. There were subsequent out of court settlements between the chicken processing plant and area residents. Improvements were made in wastewater treatment by the chicken processing plant, yet clean water has not been restored to the affected karst aquifer. Subsequent to the successful litigation many chicken processing plants in Arkansas have significantly improved their wastewater treatment.

Green Forest also demonstrated the great hydrologic significance of losing streams. Let me provide some generalizations for the karstlands of the Ozarks. While losing stream and their adjacent valleys represent about 10% of the land area, they are responsible for about 40% of the total karst groundwater recharge. Furthermore, about 90% of the catastrophic sinkholes which form do so in losing stream valleys. Many of the losing stream valleys serve to transport water from one topographic basin to another through the associated karst conduits. Finally, waters entering the karst aquifers through losing stream valleys routinely receive poor quality natural cleansing. During our studies one of the polluted wells at Green Forest still supplied water to a home. The owner told me that there was chicken fat in his water, and took me into the house to show me a layer of fat floating on the water in the toilet tank. He reported that he needed to dip the fat off the toilet tank water about once a week to keep it from overflowing and making the toilet seat slippery. I can certainly relate to the undesirability of slippery toilet seats.

Summary

I have tried to present these case histories in a humorous fashion in hopes that this will make them more memorable. In reality, these stories are human and environmental tragedies. Through stupidity people have sickened their families and lost fortunes willed to them by parents who must have hoped for better outcomes. Tragedy for the environment and for local residents was narrowly averted at Pindall, yet enormous effort and expenses was necessary to protect the waters of the area from a poorly conceived proposal unsuited to that karst landscape. Green Forest was the greatest tragedy of all. A community desperate for jobs attracted an industry which paid poorly and polluted extensively. No nation is so wealthy that it can afford to destroy large segments of aquifers and injure citizens dependent upon groundwater supplies. We must do better for karst landscapes, whether in the United States or elsewhere in the world.

I do have some happier stories of cases where groundwater tracing studies identified problems early enough to prevent costly mistakes or tragedies. These have included:

1. Demonstrating that a proposed limestone quarry near Springfield, Missouri would adversely impact regional water supplies and the Ozark Cavefish. Rezoning needed for the quarry was not granted.
2. Showing that timber harvest on some temperate rain forest areas in southeast Alaska with deep epikarstic zones would yield appreciable sediment and suspended material loads via spring systems into salmon spawning streams. Both the trees and the salmon remain.
3. Demonstrating that groundwater flow rates through karst aquifers are commonly tens, hundreds, or thousands of times greater than suggested by conventional groundwater equations. The engineer for the landfill proponent at Pindall testified that it would take hundreds of years for leachate from the landfill to reach the aquifer. The dye demonstrated that it would take days, not hundreds of years. In another case very expensive groundwater recharge basins were being considered to recharge a karst aquifer; calculations indicated that the water would remain in the aquifer for many years. Dye tracing from some of the potential recharge basins demonstrated rapid travel to distant springs, with aquifer retention times on the order of days rather than years.
4. Numerous karst aquifer characterisations which have assisted in the development of effective strategies for cleanup of hazard waste sites.
5. Recharge area delineations for important caves, springs, and wells. One of the first steps in protecting the water resources of such features is identifying the source areas for the water.

Groundwater tracing using state of the art analysis is an extremely valuable tool for preventing groundwater problems and for enhancing public understanding of the environmental sensitivity of karst. In the United States professional grade groundwater tracing has been of tremendous value. I hope to see this valuable tool used by other people in other nations; I would be glad to help anyway I can.

Finally, I prepared the written form of this keynote address after seeing a number of show caves and returning to the United States. Several of the caves present wonderful opportunities for illustrating how karst waters behave and how caves and karst groundwater supplies are extremely important to people's daily lives. Few of these opportunity have been utilised. In karst areas the surface and the subsurface are intimately connected. What happens on the surface effects the subsurface; pollutants derived from the surface routinely degrade groundwater quality. What happens in the subsurface effects the surface; people at Green Forest up to 10 km from the sewer plant cannot make beneficial use of the waters beneath their lands. Many American show caves are making a transition from being solely in the entertainment industry into being at least partially in the "useful information" industry. I think it is a good strategy and could be effectively used in the southern hemisphere to enhance the relevance of a show cave visit.

Thank you again for the invitation to deliver this address to the 13th ACKMA Conference. May we all dye soon.

References

Aley, Thomas.  1988.  Complex radial flow in ground water in flat-lying residuum-mantled limestone in the Arkansas

Ozarks.  Proc. Second Conference on Environmental Problems in Karst Terranes and Their Solutions.  Assoc. of Ground Water Scientists and Engineers, pp. 159-170.