University of Kentucky UKnowledge

Kentucky Geological Survey Information Circular Kentucky Geological Survey

2013 Kentucky Geological Survey Procedures for Groundwater Tracing Using Fluorescent Dyes James C. Currens University of Kentucky, [email protected] Right click to open a feedback form in a new tab to let us know how this document benefits oy u.

Follow this and additional works at: https://uknowledge.uky.edu/kgs_ic Part of the Hydrology Commons

Repository Citation Currens, James C., "Kentucky Geological Survey Procedures for Groundwater Tracing Using Fluorescent Dyes" (2013). Kentucky Geological Survey Information Circular. 17. https://uknowledge.uky.edu/kgs_ic/17

This Report is brought to you for free and open access by the Kentucky Geological Survey at UKnowledge. It has been accepted for inclusion in Kentucky Geological Survey Information Circular by an authorized administrator of UKnowledge. For more information, please contact [email protected]. Kentucky Geological Survey James C. Cobb, State Geologist and Director University of Kentucky, Lexington

Kentucky Geological Survey Procedures for Groundwater Tracing Using Fluorescent Dyes

James C. Currens

Dr. Quinlan’s First Principle for Groundwater Tracing in Karst: “Dye-tests should be designed so that there is always a positive result—somewhere.” Alexander and Quinlan (1996)

Information Circular 26 https://doi.org/10.13023/kgs.ic26.12 Series XII, 2013 Our Mission Our mission is to increase knowledge and understanding of the mineral, energy, and water resources, geologic hazards, and geology of Kentucky for the benefit of the Commonwealth and Nation.

© 2006 University of Kentucky Earth Resources—OurFor further information Common contact: Wealth Technology Transfer Officer Kentucky Geological Survey 228 Mining and Mineral Resources Building University of Kentucky Lexington, KY 40506-0107 www.uky.edu/kgs

ISSN 0075-5591

Technical Level

Technical Level General Intermediate Technical

General Intermediate Technical

ISSN 0075-5583 Contents Introduction...... 1 A Philosophy to Dye By...... 1 A Cautionary Note...... 2 Qualitative Tracing Protocol...... 2 Inform Regulatory Agencies and Other Researchers...... 2 Expendable Equipment and Supplies...... 2 Construction of Gumdrops...... 3 Construction of Receptors (Bugs)...... 3 Quality Control and Quality Assurance...... 4 Initial Field Work...... 4 Scouting...... 4 Deploy Gumdrops and Receptors...... 5 Logistical Considerations...... 5 Conducting a Trace...... 6 Calculating the Mass of Tracer to Use...... 6 Handling Tracer in the Field...... 7 Servicing Receptors...... 8 Determination of Results...... 9 Prepare Receptors...... 9 Analyze Receptors...... 9 Operation of Fluorometers...... 9 Criteria for Determination of Tracer Recovery in Passive Activated-Carbon Receptors...10 Recording and Documenting Results...... 10 Quantitative Tracing Protocol...... 10 Inform Regulatory Agencies and Other Researchers...... 11 Discharge Measurement Is Essential...... 11 Selecting Tracer Recovery Equipment...... 11 Selecting a Fluorometer...... 11 Deploying an Autosampler...... 11 More Programming Considerations...... 12 Prepare Quantitative Tracer Aliquots...... 13 Conduct the Trace...... 13 Analyze the Concentration and Calculate the Relevant Parameters...... 13 Calculate the Percent Recovery of the Tracer, the Time of Travel, and the Mean Velocity...... 15 Additional Thoughts on Qualitative Groundwater Dye Tracing...... 16 Summary ...... 18 Selected Bibliography...... 19 Appendix A: Characteristics of Commonly Used Tracers...... 20 Appendix B: Tracer Mass Calculation Spreadsheet...... 22 Appendix C: Matrix for Estimating Mass of Groundwater Tracer for Connectivity in Karst (10 ppb)...... 23 Appendix D: Matrix for Estimating Mass of Groundwater Tracer for Connectivity in Karst (100 ppb)...... 24 Appendix E: Matrix for Estimating Mass of Groundwater Tracer for Time of Travel in Karst (10 ppb)...... 25 Appendix F: Matrix for Estimating Mass of Groundwater Tracer for Time of Travel in Karst (100 ppb)...... 26 Contents (Continued) Appendix G: University of Kentucky, Kentucky Geological Survey, Groundwater Trace Monitoring and Analysis Data Report...... 27 Appendix H: Groundwater Trace Injection Report Form...... 28 Appendix I: Definitions...... 29 Figures 1. Photograph of a gumdrop dye receptor anchor...... 3 2. Example of a simple activated-charcoal bug...... 4 3. Illustration of packaging dye for transport in the same vehicle as dye receptors...... 5 4. Photograph of Sulforhodamine B being poured into a sinking stream...... 7 5. Textbook example of Sodium Fluorescein injection into a sinking stream...... 7 6. Photograph of Optical brightener (Tinopal CBS-X) being poured into a sinking stream cave entrance...... 8 7. Photograph of introduction of a quantitative trace into an inundated swallow hole...... 14 8. Photograph of automatic water-sampling machines set up to collect two 24-bottle series of samples...... 14 9. Photograph of analog filter fluorometer...... 14 10. Photograph of synchronous scan fluorescence spectrophotometer...... 15 11. Photograph of submersible fluorometer with 30 m of cable...... 16 Tables 1. Organic dyes used as tracers by the Kentucky Geological Survey...... 6 2. Serial dilution for Rhodamine WT to produce a 100 μg/L working solution...... 15 1 Kentucky Geological Survey Procedures for Groundwater Tracing Using Fluorescent Dyes James C. Currens Introduction scientific journals describing qualitative ground- Karst terrain often develops from an ances- water techniques, the repeatability of the results, tral landscape of surface-flowing streams, which or the analytical and quality-control techniques. leaves behind a relict pattern of the surface wa- Many case histories and methodologies are avail- tershed divides. If caves only developed in ances- able in speleological journals and agency reports, tral watersheds, then groundwater tracing, for the however. These protocols are the nuts and bolts of purpose of groundwater basin mapping, would conducting successful and defensible qualitative be unnecessary. But lithologic, structural, and hy- traces. The guidelines that follow promote an intui- drologic factors conspire to ensure that some caves tive understanding of how to apply the techniques extend headward faster than their neighbors and and a sense of what the techniques can and cannot encroach upon adjacent groundwater basins to pi- do. rate drainage under the original surface divides. In Organic fluorescent dyes were developed for many areas, groundwater basin boundaries have commercial applications (only Rhodamine WT been significantly reorganized, to the point that was designed for water tracing) and were subse- there is little relationship to the ancestral surface quently applied to groundwater problems by vari- watershed boundaries. ous researchers beginning in the early 1900’s. The Groundwater dye tracing is a powerful hy- ideal groundwater tracer is soluble in cold water, drogeologic tool for resolving these ambiguities is nontoxic to plants, animals, and people, does not when used in well-planned experiments. Like any occur in the natural environment, is not present as powerful tool, it also has its pitfalls, however, and a pollutant, does not degrade over the duration of when misapplied can cause considerable problems. the trace, is easily detected at low concentrations, These protocols standardize the methods used by is imperceptible downstream of the resurgence, the Kentucky Geological Survey for groundwater does not adsorb onto the aquifer substrate, does tracing with fluorescent organic dyes. Although not have to be continuously monitored, and is in- the methods prescribed in this document are rigor- expensive. Such a tracer does not exist, but several ous, they are nevertheless necessary to assure the fluorescent dyes come close. Researchers are con- accuracy of the groundwater tracing data gathered tinually trying other dyes, with varying degrees of by KGS. success. Appendix A summarizes the characteris- tics of six dyes commonly used for several types of A Philosophy to Dye By tracing problems. If you use another tracer, above Historically, groundwater tracing has been all else, be certain the tracer is safe (nontoxic) for conducted with a variety of substances varying in your intended purpose. sophistication from wheat chaff to radioactive iso- Qualitative groundwater tracing methods are topes. The most commonly used tracers are fluo- widely accepted by the karst hydrology commu- rescent dyes (Alexander and Quinlan, 1996), and nity. Some researchers have criticized qualitative this report is concerned only with them. A full de- groundwater tracing (Field, 2003), whereas other scription of many types of tracers may be found in practitioners defend it vigorously (Worthington Davis and others (1985). There are few articles in and Smart, 2003). In my experience, groundwater dye tracing is highly efficient and reliable. Qualita- 2 A Cautionary Note tive tracing is the appropriate tool for determining sults at surrounding springs are also needed. Al- point-to-point connectivity. Quantitative tracing is though knowing that a trace from a swallow hole appropriate when time of travel must be accurately was not detected at a specific spring is valuable in- determined, when hydrograph separation analysis formation, a positive result at another spring will must be conducted, or when conduit hydraulics strongly support your conclusion. must be determined for a flow path that is already known. Quantitative tracing requires considerable A Cautionary Note logistics, is labor intensive, and has its own vulner- The following guidelines are intended for use abilities. The process requires both collecting sam- in Kentucky by the staff of the Kentucky Geologi- ples and measuring flow past the discharge point. cal Survey. They have been modified from other Submersible fluorometers or automatic water sam- sources for the hydrogeology and the logistics of plers are typically deployed at the spring closest to working in Kentucky and are not intended to be the injection site, along with a stage recorder and comprehensive or universally applicable. Use these equipment to measure discharge. Although the re- guidelines in your project area outside of Kentucky sulting breakthrough curve from analysis of tracer only after thorough review and testing has shown concentration in a series of water samples conclu- they are appropriate for your area. sively shows that the tracer detected is the tracer injected, it is impractical for most reconnaissance Qualitative Tracing Protocol studies covering large areas. Before tracer is injected, significant prepara- Qualitative tracing is the right tool when: tion is required. A tracing project will more likely • Connectivity between a swallow hole, or be successful and defensible if you adhere to some other inflow point, and a spring, or other basic rules. discharge point, must be determined. • The watershed for a spring must be Inform Regulatory Agencies and mapped. Other Researchers • A suspected source of contamination must During the project planning stage, inform lo- be linked to a spring. cal environmental and emergency officials in writ- • Suspected leakage from an impoundment ing about the planned work. During field work, must be confirmed. each tracer injection must also be reported. When • An approximate groundwater flow veloc- tracing is restarted following a significant hiatus, a ity is sufficient. reminder is highly recommended. Current regula- Quantitative tracing is the right tool when: tions require submitting your planned tracer injec- • The spring or other potential recovery tion to the Kentucky Division of Water via an on- point is already known. line form at dep.gateway.ky.gov/dyetrace/login. • The velocity of groundwater flow in con- aspx [accessed 01/18/2012]. duits or other hydraulic parameters must Expendable Equipment and Supplies be precisely measured. (1) Use laboratory-grade activated carbon (Fisher • Various hydraulic coefficients and param- Scientific catalog number 05 685A, 6- to 14- eters must be determined. mesh) for charcoal receptors. Negative results at a monitored spring can al- (2) Fabric receptors should be either untreated ways be questioned in the absence of corroborating cotton swab or bleached cotton broadcloth data. There are many possible reasons for a nega- test fabric (Testfabrics Inc. catalog number tive result; for example, the dye may be discharged 419). to an unknown spring. In addition to being less de- (3) Use only laboratory-grade alcohol, basic com- fensible, a missed trace is undesirable because the pounds, and other chemicals, excluding the same tracers cannot be used in the same ground- tracers. water basin again for a long time or until a major (4) Use Smart Solution, which is a mixture by precipitation event occurs. Keep in mind that to volume of 5:2:3 of 1-propanol, concentrated bracket a basin boundary with traces, positive re- Qualitative Tracing Protocol 3

NH4OH, and distilled water as the elutant, for (3) Field-expedient anchors may be fashioned all tracers. from material found onsite when hiding the (5) If the initial result for a trace that used Fluo- anchor is necessary. They may also be used as rescein or Eosine is a weak positive, then a temporary anchors when a spring is discov- 5 percent solution of KOH in 70 percent iso- ered during reconnaissance and no gumdrop propyl (by mass) may be used on residual is available. charcoal to elute more dye. The use of the re- (4) Field-expedient anchors must be clean and served remainder of the charcoal for this pur- free of apparent sources of contamination. pose should be carefully considered. They should be replaced with a gumdrop as (6) Purchase groundwater tracers from estab- soon as possible if there is a chance they could lished chemical suppliers. Specify the Color be buried by sediment. Index and common name. Also ask the sup- (5) Gumdrops must have a lanyard of dull-color plier to report the fraction of the product that nylon cord that will be tied to an immov- is active ingredient percentage by mass. able object above the estimated greatest flood (7) All other materials are common hardware stage. and general-purpose items. (6) Passive tracer receptors should be attached to the gumdrop with a section of electrical wire Construction of Gumdrops or fishing trotline clip, as available. (1) Gumdrop anchors are preferred, and must (7) Gumdrops that are visible to casual observers be used when the receptor could be buried must have an identification tag attached that by sediment. Gumdrops must be made of a asks that the gumdrop be replaced as it was dense anchoring material, such as a brick or and provides contact information. concrete cast for the purpose. Use the design of Quinlan (1986) (Fig. 1). An eight- to 10-gage, Construction of Receptors (Bugs) galvanized steel wire should extend vertically (1) The envelope for charcoal receptors must be from the weight, and have loops bent in it for constructed of synthetic window screen. The attaching receptors. charcoal compartment should be a flat rectan- (2) Keep a supply of gumdrops in the vehicle gular packet approximately 3 x 2 in., with a when scouting or deploying receptors. Keep tapered loop extending from one edge for af- them away from tracer materials. fixing the completed receptor to the gumdrop (Fig. 2). (2) The receptor envelope can be assembled by sewing or stapling. Stainless steel or brass sta- ples are preferred. (3) The window-screen envelope must be filled with a minimum of 1 teaspoon and a maxi- mum of 3 teaspoons of charcoal (approximate mass of 15 g). (4) If broadcloth is used, a “bow tie” is fashioned from a 1.5-in.-wide by 12- to 18-in.-long strip of fabric. A loop is formed in the middle by tying an overhand knot on a bight. (5) Place receptors in zippered plastic bags for storage, protection, and transport to and from the field. Each bag may contain two receptors, one of cotton and one of charcoal. Figure 1. A gumdrop dye receptor anchor complete with a lan- (6) Put the individual receptor bags in a large zip- yard, an identification tag printed with a “replace as you found pered plastic bag and the larger bag inside a it” message, and bugs, both cotton and charcoal. Two steel light-resistant container, such as an ice chest. wires are used to provide attachment points for multiple bugs. 4 Initial Field Work

(2) Trip blanks are standard receptors (charcoal or cotton) placed without an individual pack- age between the outer bag and the bags with the individually packaged receptors. They are treated and analyzed in the same way as the sample receptors. One trip blank must be used for each field trip in which receptors are recovered. (3) Background blanks are standard receptors deployed in the field prior to any tracer- in jection. They are treated and analyzed in the same way as the sample receptors. One nega- tive background blank should be recovered from each spring prior to initiating a ground- water trace. (4) Duplicates are second or third receptors de- ployed at springs expected to have high rates of human or animal visitation and at all other springs when tampering actually occurs. The duplicate receptor should be well hidden and placed in a different location than the primary receptor. (5) Replicates are second or third receptors de- ployed at springs to provide charcoal for addi- tional analysis. The replicate receptor should be hung with the sample receptor at exactly Figure 2. Example of a simple activated-charcoal bug. The ta- the same location. As an alternative, a slightly per at the top facilitates attaching the bug to the gumdrop with larger envelope may be used for the replicate, vinyl-clad multistrand copper wire. Staples should be brass or provided water circulates freely through the stainless steel. package. (6) Duplicates and replicates of traces thought (7) Charcoal receptors may be used in place of to be positive, and which are not needed to cotton fabric receptors, if a scanning spectro- replace damaged or lost receptors, may be photofluorometer is available and if the tracer submitted to a second laboratory for replicate has a well-defined narrow range of -wave analysis. length for fluorescence emission. (7) The presence of fluorescence in the eluent Quality Control and Quality Assurance from a charcoal receptor deployed for back- (1) The following types of blank receptors should ground determination will exclude the inten- represent 5 to 10 percent of the receptors de- tional use of the corresponding tracer in that ployed over the course of the tracing project. groundwater basin as long as the background A range is allowed because a trace that is re- presence persists. covered quickly requires fewer receptors, whereas the minimum number of blanks and Initial Field Work background receptors is comparatively con- Scouting stant. (1) Conduct a thorough field reconnaissance. (a) Trip blanks The initial search should be by automobile. (b) Background Springs are easily hidden by vegetation and (c) Duplicate may not be seen from a vehicle. Search all sus- (d) Replicates pect areas on foot. Initial Field Work 5

(2) Scout reaches of streams that likely function and at all other springs if tampering actually as local base level in the project area by boat, occurs. or when possible, on foot. (3) Use topographic and geologic maps and aerial Logistical Considerations photography to search for springs and swal- (1) Before going to the field, repackage tracers low holes. Large springs can be located with into incremental aliquots in small, sturdy, un- topographic maps, but these maps are not breakable containers. Use half-liter Nalgene very useful for finding small springs. Aerial bottles for powder and gallon jugs for tracers photographs should be used when available, that are to be mixed in the field. The exterior and are the most helpful if taken when veg- of all reusable containers must be thoroughly etation was dormant. washed after being refilled. (4) Record the locations of field sites (springs, (2) Use aliquots of tracer packaged in quantities swallow holes, sinking streams, and other that approximate the mass of tracer desired hydrologically important features) on a form and that can be composited from the pack- and on either a topographic map or in a Geo- aged units. Increments of Fluorescein should graphic Information System database. be 25, 50, 75, 100, and 200 g, for example. This (5) Ask residents of the study area about the lo- allows for adjustments in tracer mass and cations of springs; this can save a significant minimizes handling of the tracers. amount of time. Knock on doors, be polite, (3) The tracer packaging should be in four layers and fully inform people of what you are doing (Fig. 3). Put bottles inside a zippered plastic and why. The source of the water discharg- bag, and the bagged bottles inside another ing from a karst spring on their property is a zippered plastic bag while wearing dispos- great mystery to many people, and they will able gloves. Put the individual packages in be more willing to help if you provide them a garbage bag and the entire package inside with the results. As a bonus, you will be able a plastic tub or dedicated ice chest. Several to educate them about the methods you are small bottles can be placed in one large zip- using so they will not be concerned if they see pered plastic bag. Put one pair of disposable colored water discharging from their springs. gloves inside the garbage bags, and extra pairs in a zippered plastic bag inside the plastic tub. Deploy Gumdrops and Receptors (4) Do not transport tracer to the field in the bulk (1) All springs in the study area that are likely to shipping container. receive tracer, and some unlikely resurgences, should be located and bugged. (2) Carry receptors and anchors when on recon- naissance trips and set background bugs as new springs are found. (3) Deploy the gumdrop in the channel of the spring run so that the dye receptor is totally immersed in the outflow, for maximum expo- sure to the resurgent tracer. (4) Take precautions so that the receptor is not buried by sediment, stranded by low flow, or dislodged by wildlife, livestock, or vandals. If possible, place the receptor in the shade. (5) Always place “failsafe” bugs at intervals along base-level streams that are apparent boundar- Figure 3. Packaging dye for transport in the same vehicle as ies to unconfined flow. dye receptors. The hand symbols represent packages of dis- (6) Deploy two receptors at springs expected to posable gloves and the green bottle shapes are the dye inside have high rates of human or animal visitation zippered plastic bags. The large bag should be placed in a plastic storage tub. See text for further details. 6 Conducting a Trace

(5) Do not open any of the tracer packages while rithms were derived from the correlation of inside the vehicle. mass injected (as calculated) versus mass re- (6) Tracer and receptors can be transported in covered for 209 groundwater traces in karst. the same vehicle without compromising the Mass to inject is based on Sodium Fluorescein. results, but the practice is discouraged. When Qualitative traces (point to point): transported together to and from the field, Eq. 1: M=19 (LQC)0.95 both dye and bugs must be carefully pack- Quantitative traces (time of travel): aged, and the vehicle interior must be thor- Eq. 2: M=0.73 (TQC)0.97 oughly cleaned. Where: M=mass of tracer to use, in grams (7) Vehicles used to transport tracers must be ex- L=distance in meters amined with an ultraviolet light before use to Q=discharge in m3/sec verify absence of tracer from door handles, T=expected travel time in seconds gear shifters, steering wheels, etc. Any sur- C=desired concentration at the face showing fluorescence must be cleaned spring in g/m3 or ppb. with water containing bleach. Keep a record (2) An Excel workbook, “Tracer Mass Calcula- of when the vehicle was cleaned and if any tion.xls,” has been prepared that uses these tracer contamination was found with the ve- equations to produce a matrix for estimating hicle mileage log. the mass of Fluorescein to use. The goal is to produce a specified peak concentration at the Conducting a Trace spring in the parts per billion (μg/L or mg/ Calculating the Mass of Tracer to Use m3) range. Worksheets have been developed Choosing the type and correct mass of dye for peak concentration at the spring or 10 ppb is perhaps the most difficult decision made in (0.01 mg/m3) and 100 ppb (0.1 mg/m3). Cop- groundwater tracing. Table 1 lists the dyes used ies of these spreadsheets are in Appendix B. by the Kentucky Geological Survey. The objective (3) Other worksheets in the workbook are set is to produce a concentration at the spring that is up to use a second equation for quantitative invisible to the human eye, yet detectable by in- traces, which estimates an approximate flow strumentation. A trace that fails because too little velocity from prior qualitative traces. The tracer was used is nearly as bad as an easily visible maximum time of travel is estimated from trace because of the wasted time, expense, and the these data and is used with the discharge to recurrence of risk from a second trace. Use enough calculate the quantity of tracer to inject. dye, but try to anticipate the consequences of using (4) If a proposed trace has the possibility of flow- too much. ing to either a nearby spring or a very distant (1) This protocol uses the equations of Worthing- spring, use the median distance. The intent is ton and Smart (2003, p. 287–295). Their algo- to provide enough tracer mass for a positive

Table 1. Organic dyes used as tracers by the Kentucky Geological Survey. Common Name Color Index Color in Solution Excitation (nm) Emission (nm) Fluorescein Acid Yellow 73 brilliant green 4851 5131 Rhodamine WT Acid Red 388 vivid orange-pink 5581 5651 Sulforhodamine B Acid Red 52 vivid orange-pink 5481 5781 Tinopal CBS-X Fabric Brighter 351 blue-white 3602 4502 Eosine Acid Red 87 greenish orange to red 5121 5381 Solophenyl* Direct Yellow 96 greenish yellow 397 490 Pyranine D&C Green 8 green (pH adjusted) 4552 5011 1Aley (1999) 2Ford and Williams (1989) *Although formally renamed Solophenyl, the older name, Diphenyl Brilliant Flavine 7GFF, is still widely used. Conducting a Trace 7

result at the more distant spring while reduc- (3) Although tracers delivered from the manufac- ing the coloration at the nearer spring, should turer as powder may be injected dry, they are the trace in fact follow the shorter route. much easier to handle if premixed with water. (5) When conducting numerous traces in a study (4) Mixing with water onsite can be done by plac- area, use alternate dyes for traces that may ing the dye in a suitable jug and adding water flow to the same spring (Figs. 4–6). This reduc- from the inflow to the sink or a potable source. es the chance that residual dye in the system Any heavy plastic container can be used, but could result in a false positive. In dry weather, a carboy is preferable for large quantities of this practice shortens the time needed to com- tracer. If bleach bottles are used, they must be plete the tests. thoroughly rinsed before use. Do not use milk jugs, because they are too fragile. Handling Tracer in the Field (5) Never inject Solophenyl (Direct Yellow 96) (1) Before injecting any tracer, notify the Ken- dry. It is difficult to dissolve and will lay in tucky Division of Water–Groundwater Branch clumps in the stream bed. via fax (502-564-0111), telephone, or e-mail. (6) Use boots and gloves that can be easily washed If by e-mail, send a copy to the project man- or disposed of. ager at KGS (www.uky.edu/kgs/water). If (7) Always keep the wind to your back and the by telephone, write down who you spoke to, tracer container downwind when pouring the time, and which tracers were discussed. into the water. If someone reports strangely colored water (8) Potential injection points that can receive nat- discharging from a spring, Division of Water ural run-in include sinking streams, swallow officials will then call you to determine if it is holes, dolines, karst windows, caves, and dry just a groundwater dye trace before raising (stormwater disposal) wells. an alarm. Agency staff may be aware of other (9) All of the locations mentioned in (8), and also persons conducting dye traces in the water- auger holes and trenches, can be traced with shed, which obviously could result in very hauled, potable water. Wet any soil or organic confusing results. If you cause a problem for the material in the injection feature with an abso- water protection agency, they will cause a problem lute minimum of 300 gal of clean, potable wa- for you! ter before injection. Flush the tracer with an (2) Do not open any of the packaging of the tracers additional 600 to 2,000 gal. while the packages are inside a vehicle.

Figure 4. Sulforhodamine B being poured into a sinking stream Figure 5. Textbook example of Sodium Fluorescein injection in Hardin County. Notice the stance of the researcher above into a sinking stream. The dark object at the distant terminus the water, with feet away from the tracer. The wind direction of the dye is the swallow hole. Fort Knox Military Reservation. should be toward the camera. Photo by David Lutz. Photo by David Lutz. 8 Conducting a Trace

(13) If you have injected tracer, do not touch re- ceptors or the containers they are packaged in until you bathe and change clothes. Servicing Receptors (1) The default schedule for changing receptors is once a week. This is the optimum frequency for logistics, dye adsorption, and the security of the receptors. (2) If a receptor is to be changed more frequently than once every 24 hours, a duplicate recep- tor should be deployed before the tracer ar- rives. The duplicate must be left in place until the tracer has passed or changed each week, whichever is sooner. (3) Do not leave receptors in the field for longer than 2 weeks. Long deployments make failure of the trace more likely because of degrada- tion, loss, or theft of the receptor. (4) Always replace one set of receptors with an- other until you are certain the tracer has come through. Continue to exchange receptors until the trace is detected and the concentration has returned to pre-trace levels or the program is discontinued. (5) Many traces take longer than expected and may be lost if you end the experiment prema- turely. If the conceptual model does not pre- dict slow travel times, after 6 weeks of nega- Figure 6. Optical brightener (Tinopal CBS-X) being poured tive results the trace may be considered lost. into a sinking stream cave entrance, Bourbon County. Photo (6) Handle the receptors with clean hands washed by J.C. Currens. immediately before handling the bug, or wear disposable gloves. To protect the receptors (10) Prior to injecting tracer into trenches or auger from contamination, always place them in a holes (epikarstic dye injection points), estab- sample bag immediately after removal from lish that the excavation will accept inflow. the gumdrop. Never lay a receptor on the The minimum acceptable rate of infiltration is ground or on top of the gumdrop. They are 1 L/min/m2. easily lost when not in a bag. (11) Do not inject tracer into a water well without a (7) Carry a clean, empty zippered plastic bag in which written memorandum of understanding signed by to place the old receptor. Keep a supply of fresh the well owner stating that tracer concentrations bags if there is any chance of litigation. Other- may remain high for an extended period. Also se- wise, save the clean bag that the new receptor cure the permission of the Kentucky Division was taken from and use it for the recovered of Water before injecting in any well. The only receptor at the next site. Keep the empty bag exception is stormwater disposal dry wells closed and do not put anything else in it. (EPA Class V). (8) Label the exterior of the bag with a water- (12) Wash your hands thoroughly after handling proof marker. Provide the site name, the date, tracers to avoid cross contamination from and the time. Do not insert any other material door handles and steering wheels, for exam- into the bag with the receptor. ple. Determination of Results 9

(9) Use a black permanent marker to label the centration is desired. A temperature of 80°C is bags and do not use a colored pen. adequate. (10) Transport receptors inside the individual sample bags placed inside a larger sample Analyze Receptors bag. Place the large bag inside a lightproof (1) Open the receptor envelope with a staple container. puller or cut off a corner with scissors. Pour (11) Do not ice receptors unless there is going to about half of the charcoal (exactly 5 gm if be a 24-hour or longer delay in preparation semiquantitative comparison is desired) into for analysis. If using water ice to chill recep- a disposable plastic condiment cup or a glass tors, be certain each individual zippered plas- jar that is designated for the purpose (washed tic bag is sealed and each large bag is tightly with bleach and thoroughly rinsed). sealed so that meltwater is excluded. (2) Pour enough eluent into the container to cover (12) If traces are to be conducted on the same trip the charcoal. Use exactly 10 ml if it is impor- that receptors are changed, do not inject any tant to have a semiquantitative comparison of tracer until all receptors have been serviced. one receptor to another. The recovered receptors should be completely (3) Allow elution to proceed for no less than 20 repackaged and a trip blank included before minutes and no more than 1 hour. If an ap- the tracer is handled. Always have one person proximate breakthrough curve from receptors handle the tracer and another handle the re- that have been changed every 30 minutes is ceptors. needed, the timing of elution must be consis- tent, and 20 minutes is recommended. Determination of Results (4) Gently pour the elutant from the container into a clean, disposable cuvette and analyze. Prepare Receptors (5) All trace receptors collected should be eluted (1) If receptors must be stored longer than 24 and analyzed. Leaving some of the receptors hours before analysis, keep them on ice or until a later date is not recommended because refrigerated. The larger zippered plastic bag there may be more than one positive. must be waterproof if iced. The temperature (6) Confirm all visually positive traces by deter- should be near 5°C. mining fluorescence of the elutant with a filter (2) To prepare the receptors for analysis, wash fluorometer or scanning spectrophotometer. the window-screen packet or cotton test fabric (7) Examine cotton receptors under a handheld strip thoroughly with a high-pressure stream ultraviolet light. Use UV-protective eyeglass- of clean, nonchlorinated water. Do not use es. chlorinated tap water. (8) It is highly recommended that cotton recep- (3) Do not put the receptors down on a counter- tors that are positive by visual inspection be top or lay them in a sink. Use latex gloves or confirmed with a scanning spectrophotom- wash your hands thoroughly before and after eter with a dry-sample adapter. Analyzing an handling receptors. You can place the clean activated carbon receptor that was deployed receptors back in the individual zippered in place of another deployment of the cotton plastic bags in which they were delivered. is acceptable. (4) Put a tag on the receptors or affix them to a (9) Record all visually positive traces and those clean sample bag with the identification infor- detected by fluorometer on the Groundwater mation clearly recorded. Hang the receptors Trace Analysis Report. in a dark room or storage cabinet to air dry. Never allow more than one receptor to be si- Operation of Fluorometers multaneously involved in relabeling because (1) Operate the Varian Cary Eclipse scanning they may become mixed up. spectrophotometer according to the manu- (5) Optionally, the receptors can be oven dried to facturer’s hardware and software operating uniform moisture content if long-term storage manuals dated May 2000 and February 2000, or semiquantitative comparison of tracer con- respectively. 10 Quantitative Tracing Protocol

(2) Operate the Turner Designs SCUFA (Self- (4) A positive trace must have the minimum flu- Contained Underwater Fluorescence Appara- orescence equivalent of 1 ppb concentration tus) filter fluorometer according to the manu- and at least 10 times the ambient background facturer’s user’s manual dated April 15, 2000. fluorescence intensity. (3) Operate the Turner Designs model 10-070A (5) The tracer should be present in more than one water-resistant, field-portable filter fluorom- receptor for the trace to be considered posi- eter according to the manufacturer’s Operat- tive. Exception may be made for receptors ing and Service Manual dated October 1981. that are deployed in poorly accessible loca- (4) Calibrate the instruments used in a project ac- tions, such as underground, for longer than cording to the manufacturer’s instructions at the standard 1 week. An exception may also the beginning of its usage period. be made for trace distances less than 1,000 m. (5) Battery-powered equipment should have (6) After a trace to a spring has been conducted, fresh batteries installed before calibration wait until a negative receptor has been recov- when possible, and if depleted by the calibra- ered from that spring before using the same tion process, replace batteries immediately tracer again in the basin. before deployment. (6) Print the analytical report for archiving when Recording and Documenting Results possible. Keep a digital copy of the report files (1) Keep the receptors in a secure room, prefer- permanently on a suitable storage medium, ably in a locked cabinet, when they are not in and on the hard drive of a dedicated comput- the possession of someone who will sign or er. has signed the chain of custody bug sheet. (7) Complete the Groundwater Trace Analysis (2) The attached receptor recovery sheet func- Report, which is the official record of the trace. tions as a chain-of-custody form. Spaces are Retain the printouts from the instruments and provided for the signatures of the persons the archived remainder of the charcoal and relinquishing and receiving the receptors. If cloth receptors for 3 years or until after the the same person is performing more than one end of the project, when a final report is draft- task, that person will sign at all of the lines. ed. (3) When the receptors are received, the person making the delivery signs the relinquishing Criteria for Determination of Tracer custody line and the receiving person signs Recovery in Passive Activated- the appropriate line. Ask the person making Carbon Receptors the delivery if he or she wants a photocopy as (1) A spectrophotometer emission scan must a receipt. show a peak at the appropriate wavelength of (4) If the receptors are to be analyzed by another the injected dye in the sample matrix. The pH staff person, the person who received the re- of the matrix affects the wavelength within a ceptors should have the receiving staff person known and predictable range. Deviation from sign the appropriate line. Give a photocopy of the ideal wavelength of ±2 nm is acceptable. the form to the person making the delivery. (2) The spectrophotometer scan must reveal a (5) When the analysis of the listed receptors is peak with the same curve shape as standard complete, indicate the results on the Ground- solutions of the tracer, or a composite of sev- water Trace Analysis Report and sign the line eral known tracers. attesting to the findings. Make a photocopy (3) The tracer should be absent from all back- for the GIS database manager, and deliver the ground receptors collected at likely resurgent original to the person conducting the trace. springs before the injection. If repeated use of a tracer is unavoidable, the receptors must Quantitative Tracing Protocol have known and predictable background flu- Most of the precautions discussed under orescence intensity. Qualitative Groundwater Tracing can also be ap- plied to quantitative work. It is assumed that the Quantitative Tracing Protocol 11 researcher knows from previous qualitative work be summed and the total discharge during the where the trace is going to go. Quality control and 30-minute interval would be multiplied by the quality assurance are just as important for quanti- prevailing concentration. If making discharge tative tracings as for qualitative, perhaps more so. measurements manually, collect a sample at the beginning of each discharge measurement Inform Regulatory Agencies and as well as a final sample. Other Researchers Just as with qualitative work, inform local Selecting Tracer Recovery Equipment environmental and emergency officials in writing (1) In addition to the discharge, the prevailing about the planned project. Current regulations re- tracer concentration must be determined. This quire submitting your planned tracer injection to can be accomplished by collecting samples or the Kentucky Division of Water via an online form can be observed in the field directly with por- at dep.gateway.ky.gov/dyetrace/login.aspx [ac- table fluorometers. cessed 06/08/2012]. (2) Containers for tracers should be glass bottles or culture tubes, regardless of sampling meth- Discharge Measurement Is Essential od. (1) Discharge measurement is essential to accu- (3) Keep water samples containing tracer out rately calculate the center of mass used to de- of direct sunlight, preferably on ice in an ice termine time of travel. Although a semiquan- chest. Close the caps securely when using ice, titative trace can be useful, the resulting time and lay the bottles on their side. of travel determined from first arrival or peak concentration can be significantly different Selecting a Fluorometer from the center-of-mass time of travel. (1) A fluorometer capable of determining the (2) Discharge can be obtained in several ways. concentration of the tracer is essential for this (a) Use an existing gaging station for which work. a rating curve has already been estab- (a) Most fluorometers require tracer con- lished. centration standards to be mixed, with (b) Make a series of discharge measurements which the machine is calibrated. See the during the tracer recovery. discussion on mixing standards. (c) Install a staff gage and make discharge (b) Most field-deployable fluorometers are measurements at a range of stage depths. filter fluorometers, and they can be con- Use these data to develop a rating curve figured for only one dye at a time. so that wading is not necessary to man- (2) Decide if you will use a field fluorometer or ually obtain the discharge during the an autosampler. trace. (3) Portable fluorometers are available that use a (d) Record the stage continuously by install- flow-through cell inserted into the machine. ing a stage recorder that is correlated The other type is a submersible fluorometer. with the staff gage. (4) If multiple dyes are being used simultaneous- (e) Install a discharge control structure ly or sequentially in rapid succession, a scan- (flume or weir) with appropriate stage ning fluorometer is essential, which dictates recorder. If personnel are not able to that water samples be collected. monitor the tracer recovery, (d) or (e) is compulsory. Deploying an Autosampler (1) There are two common and several less com- (3) Correlate the concentration of the recovered mon makes of autosamplers, and it is possible tracer with the total discharge between each to build your own. time period. For example, the discharge is de- (2) When you have obtained an autosampler, it termined every 10 minutes, but tracer concen- should first be set up in the laboratory and tration is recorded every 30 minutes. The three tested. Any worn or broken parts should be interim discharge measurements following replaced, desiccant pack refreshed, and bat- determination of the dye concentration would 12 Quantitative Tracing Protocol

teries tested for amperage and for the ability (h) After calibration and testing are com- to sustain a charge if set to be triggered by a plete, turn the machine off and replace rise in stage. the battery with a completely charged (3) Practice programming the machine. The most one. Use a lead-acid battery if the ma- common source of failure to collect samples chine will be deployed for a week or more is a programming error, followed by jamming between inspections. Nickel-cadmium of the distributor and holes in the intake hose. batteries have a tendency to discharge (4) Thoroughly clean the hoses and bottles in spontaneously, but when fresh have a soapy water. Rinse with a dilute solution of higher amp/hour rating than lead-acid household bleach followed by complete rins- batteries. ing with distilled water. (i) After restoring power, press “exit pro- (5) When deploying, consider the following pre- gram” and “start sampling.” Enter a brief cautions: interval to start time. When the start time (a) Choose a location for the machine that arrives, if an actuator is being used, some is above base-flow elevation as high as indication should be displayed that the the machine will pump, typically 20 feet machine is disabled. Use the latch func- above the intake. tion on the water sensor. (b) Choose a location that is at least partly (j) Cover the pump controller, fasten any obscured by underbrush or other cover security cables, and run a chain or wire and that has a tree or other anchor point rope through the handles, around the to which the machine can be chained. anchor point, and meet the ends at the (c) Test the machine by collecting a manual padlock, along with the security cables. sample and advancing the distributor to the last bottle. Be certain that the distrib- More Programming Considerations utor will clear the mouths of the sample (1) Larger volumes of sample require either bottles all the way around. larger sample bottles (plastic 1 liter for ISCO (d) If using a water-level trigger of any type, brand) or multiplexing. Multiplexing is the make sure the machine turns on when use of more than one bottle for a sample, or the trigger is touched by water. alternately, more than one sample in a single (e) Be certain the clocks on the autosam- bottle. pler, field fluorometer, stage recorder, (2) The more bottles multiplexed per sample, the and wristwatches are all synchronized. fewer samples that can be collected for a given Havoc will ensue if the dye injection was period. Avoid multiplexing for quantitative recorded in daylight-saving time while tracing. the sampler and stage recorder are on (3) The first time a quantitative trace is attempt- standard time. ed, estimate the travel time using results from (f) Calibrate the sample volume; do not de- any qualitative tracing results available. Sam- pend on the autovolume functions, but pling should be timed to begin at about half of use them if the existing calibration is the expected travel time. By spacing samples a grossly incorrect. If insufficient sample is half hour to one hour apart, the tracer is more collected, check the intake hose for cuts likely to be sampled, which provides data to and pinholes. If too much sample is col- plan more precise experiments later. lected, you may have entered the wrong (4) If more than one quantitative trace is planned, hose dimensions or bottle type. subsequent sampling intervals should be (g) Repeat calibration until the desired vol- shorter. Shorter sampling intervals signifi- ume is obtained. When calibrating, keep cantly increase the detail and the accuracy of the autosampler pump at the same eleva- dye-recovery calculations. tion as when the machine is deployed. (5) Construct a chart of travel time/stage or trav- el time/discharge rating curve for the quanti- Quantitative Tracing Protocol 13

tative traces. Use the chart for programming of the tracer, yet have sufficient concentration the start time and the sample intervals on the so that the falling limb of the breakthrough autosampler. When enough data to construct curve is distinct. See Appendices D and E. such a rating curve have been measured, the (4) Always know exactly the mass of the tracer task is nearly finished, in any case. actually injected. If conducting a constant-rate (6) The ideal sampling sequence would begin injection, collect a sample of the tracer as it is about three sampling intervals before the trac- introduced into the groundwater. er arrives, and the sampling interval would match the period on the stage recorder. Conduct the Trace (7) If only one attempt is possible, consider us- (1) After the recovery location is identified and ing multiple autosamplers. For example, one all necessary equipment is in place, verify that sampler could be set to sample hourly, so as the clocks are synchronized and the sampler to have a high probability of bracketing the or fluorometer is properly programmed. tracer breakthrough curve. A second sampler (2) Set the delayed start time of an autosampler could be programmed to sample at a shorter to allow enough time for personnel to travel interval, which is the best estimate of the trav- to the injection site plus half of the expected el time. Alternatively, two or more samplers travel time of the tracer. Program the sample could be programmed to sample in sequence collection interval to be in multiples of the at shorter intervals for however long neces- stage recording interval. sary to assure the breakthrough of tracer is (3) Have personnel and equipment on hand to covered and that the available equipment will measure discharge, or apply the previously permit. If one of the short-interval samplers developed rating equation to the stage data fails, the data from a widely spaced interval being recorded. sampler could provide some backup. Widely (4) Travel to the injection site and introduce the spaced data may be all that is needed to de- tracer (Fig. 7). Be careful to keep powdered fine the slope of the breakthrough curve on tracers downwind. Hold the mouth of the the trailing side of the peak concentration. container near the water surface. Rinse the bottle with water and pour any residue into Prepare Quantitative Tracer Aliquots the sinking flow as quickly as possible. Keep (1) Determine which tracers are to be used. Be in mind that the percentage of dye recovered aware that many tracers have bulking agents is valuable information in addition to the cen- (commonly cornstarch) mixed with the active ter of mass. ingredient, and the percent mass of the com- (5) Put the empty bottle and used gloves in a bag, ponents is not readily available. and place the bag in the larger dye container. (2) Determine the percent active ingredient or Save the bottle for reuse. Return to the tracer plan to treat samples as if the mass injected is recovery site at the earliest opportunity. 100 percent active ingredient. Rhodamine WT (6) Upon arrival at the recovery site, verify that is 20 percent by volume, at a specific gravity the sampler or fluorometer and stage recorder of 1.19 g/cm3 (varies from 1.1 to 1.2). Use Flu- are running (Fig. 8). If time permits, wait until orescein (Uranine) as if it is 100 percent active colored water arrives to be certain the sampler ingredient. Treat Eosine and Sulforhodamine has come on in time. B the same way. If both the standard and the samples are treated as 100 percent active in- Analyze the Concentration and Calculate gredient, the dye recovery percentage and the the Relevant Parameters arrival time of the center of mass will be the (1) Standards are required for all fluorometers same. that are used to measure the concentration of (3) Calculate the mass to be injected based on fluorescent tracers (Figs. 9–10). Mixing stan- the expected discharge during the recovery dards is not difficult, but is time-consuming, period. The goal is to minimize the visibility and blunders are possible. 14 Quantitative Tracing Protocol

Figure 7. Introduction of a quantitative trace into an inundated swallow hole, Hardin County. The water was over 5 m deep and the boat was positioned using landmarks. A garden hose was inserted into the swallow hole, and a funnel was used to pour the Rhodamine WT into the hose. The total head in the swallow hole was less than at the surface, and the dye was quickly drawn down the hose. Photo by David Lutz.

(2) Determine if the tracer you are using is the factory container thoroughly each time 100 percent active ingredient or something the initial quantity of tracer is obtained, but if else. If it is less than pure, calculate the grams using Rhodamine WT, let the container be still per kilogram of actual dye and prepare your long enough for foam to dissipate. dilution series using that mass of active ingre- (4) The most difficult step is obtaining the pi- dient. Alternatively, prepare the standards as pet of concentrated Rhodamine WT from the if the tracer is 100 percent active ingredient. original container. Use a to-detain pipet with (3) Keep one factory-packaged container for white graduations. A pipet with multiple quantitative traces only. Mix the contents of graduations is easier to use than a single-vol- ume pipet at this point in the process. Be as ac- curate as possible. Blow out the residual dye in the pipet, but do not rinse it into the mixing flask.

Figure 9. Analog filter fluorometer. The instrument shown can Figure 8. Automatic water-sampling machines set up to collect be configured for Fluorescein and Rhodamine WT. Photo by two 24-bottle series of samples. Photo by J.C. Currens. J.C. Currens. Quantitative Tracing Protocol 15

(8) The most efficient way to prepare dilution se- ries repeatedly is to set up a cookbook table listing the volume of diluted tracer in so many milliliters of distilled water to create the de- sired concentration (Table 2). Also see Kilpat- rick and Cobb (1985), Kilpatrick and Wilson (1989), and Wilson and others (1986). (9) The standards should bracket the expected range of concentrations to be measured. Stan- dards above 100 μg/L may have to be special- ly made, as the published dilution series do not have aliquots above 100 μg/L. Figure 10. Synchronous scan fluorescence spectrophotom- (10) Choice of which series to use is dependent on eter. The system is entirely digital. Photo by J.C. Currens. the glassware and storage available for the re- sidual solution from each step in the series. If (5) Use amber glass bottles for all standards. The a large number of sets of standards is expect- bottles should be 2 L or larger for the first, sec- ed to be mixed, then the first or second line ond, and third dilution and decreasing in vol- could be used. At KGS, the last series in Table ume to 250 ml for the lesser concentrations, 2 will produce more than enough batches of except that the 1 μg/L concentration will re- the final concentration. quire a 1 L bottle. (11) Follow the procedures recommended by the (6) A dilution of 100 μg/L is the working solu- fluorometer manufacturer to calibrate the ma- tion. It may require a 4 L bottle. This aliquot chine. can last a long time and be the source of all subsequent series for that tracer. Calculate the Percent Recovery of the (7) Organic fluorescent dyes decrease in concen- Tracer, the Time of Travel, and the tration quickly when put into a new bottle. Mean Velocity In my experience, subsequent aliquots of the (1) Measure the concentration of dye in the sam- same concentration are more stable when put ples, or download the concentration data into a used and unwashed bottle that former- from your submersible fluorometer (Fig. 11) ly held the same concentration. I always keep or other field fluorometer. my 10 μg/L standard in the same 10 μg/L (2) Multiply the concentration by the discharge bottle and the 5 μg/L standard in the 5 μg/L during the interval between samples to de- bottle, and so on. This simple convention has termine mass conveyed during the sampling extended the shelf life of standards and the interval. accuracy of the calibrations made with them. (3) Calculate the percentage of tracer recovered. Keep your standards in a dark cabinet. Sum the mass recovered for each sample time

Table 2. Serial dilution for Rhodamine WT to produce a 100 μg/L working solution.* Initial Final First Dilution Second Dilution Third Dilution Concentration Concentration Volume Dye Volume Volume Dye Volume Volume Dye Volume Working (ml) Water (ml) (ml) Water (ml) (ml) Water (ml) Solution Rhodamine WT 20% by 50 3,792 20 3,500 20 3,500 100 μg/L volume. SG 25 2,585 20 3,000 20 3,000 100 μg/L 3 1.19 g/cm 20 2,068 20 3,000 20 3,000 100 μg/L 20 1,158 10 2,000 10 2,000 100 μg/L *Modified from Wilson and others (1986). 16 Additional Thoughts on Qualitative Groundwater Dye Tracing

for recording their location once found. Aer- ial photographs and online satellite imagery should be used when available, but are not very helpful unless the picture was taken when the vegetation was dormant. All likely dye recovery points and some unlikely resur- gences should be located and bugged. Always place “failsafe” bugs at intervals along the re- gional base-level streams. The most common cause of a lost trace is an unmonitored spring. Figure 11. Submersible fluorometer with 30 m of cable. The (3) The existence and locations of springs can be device requires a 12-volt auxiliary power supply and is pro- quickly obtained from area residents, and a grammed and downloaded using software installed on a laptop few minutes of casual conversation can save computer. Photo by Collie Rulo, Kentucky Geological Survey. significant time. Knock on doors, be polite, and fully inform people what you are doing increment. Divide the mass recovered by the and why. Most people, even the very skepti- mass injected. cal, will eventually be won over. The source of (4) Determine the overall length of the break- the water discharging from their karst spring through curve in units of time. Sum up the is a great mystery to many people, and they time units needed to represent half of the will be delighted to help if you provide them mass of tracer injected. The elapsed time at with the results. A bonus benefit to talking to which half of the total tracer injected has been residents is the opportunity to educate them accounted for is the center of mass. about the methods you are using and the ap- (5) Calculate the velocity by dividing the distance pearance and safety of the dye. The well water from the injection point to the recovery point of one Logan County resident was colored by by the time of travel. Adjustments for sinuos- a trace I conducted, but because I had taken ity of the flow route may be needed. the time several months earlier to let the com- (6) Calculate the cross-section area of the conduit munity know what to expect, he was neither by dividing the instantaneous discharge by alarmed nor angry. the mean velocity. If the cross section is as- (4) Inform local environmental and emergency sumed to be circular, a diameter may be cal- officials about your work without fail! In Ken- culated. tucky, you are required to give the Division of Water in Frankfort advance notice. If you Additional Thoughts on are doing more than a couple of traces, obtain a user ID and password for the online report- Qualitative Groundwater ing form at dep.gateway.ky.gov/dyetrace/ Dye Tracing login.aspx [accessed 06/08/2012]. If some- (1) Do not embark on a groundwater-tracing one reports colored water discharging from a project just because it might be fun or it is the spring, emergency response officials will then only test that has not been tried. To be cost- call you to determine if it is just a groundwa- and technically effective, formulate a hypoth- ter dye trace before raising an alarm. Agency esis to test and contemplate if knowing the an- staff may be aware of other persons conduct- swer will justify the effort needed to conduct ing dye traces in the watershed, which obvi- the trace. ously could result in very confusing results. (2) Conduct a thorough field reconnaissance on After a significant hiatus in your program, foot. Springs are easily concealed by vegeta- you should remind the agency people. If tion and may not be visible from a car or boat. you cause a problem for the water protection Topographic maps are less useful than pho- agency, you will be found and and they will tographs for finding springs, but are handy cause a problem for you. Additional Thoughts on Qualitative Groundwater Dye Tracing 17

(5) Every dye trace should be designed to have a bug that is negative for the dye you wish to a positive result—somewhere. Although use before tracing to the same spring with it knowing that a swallow hole does not flow again. to your spring is valuable information, you (9) Always replace one set of bugs with another will always wonder if the dye was diluted or until you are absolutely certain the dye has adsorbed, rather than discharged to an un- come through. Many traces take longer than known spring. Furthermore, a positive result expected and may be lost by prematurely at another spring will strongly support your ending the experiment. Of course, after 4 to conclusion that the injection point is indeed 6 weeks of negative results, you will have to outside of your groundwater basin. Keep in decide if the trace is lost. mind that to constrain the basin boundary, it (10) Be very careful about how the bugs are placed. is necessary to bracket it with traces flowing At the spring, deploy the bug in a shaded area in opposite directions. You need positive re- and where it is not likely to be covered by silt. sults to surrounding springs. If tampering with the bug is likely at a site, (6) Deploy a set of background bugs before be- or the bug could be lost to high flow, place a ginning tracing, especially when optical second or third hidden bug. The frequency of brighteners are being used as tracers, or if bug changes is dependent upon the precision there is a chance any fluorescent dye will be a needed for groundwater travel time. Chang- contaminant in the aquifer. I commonly carry ing bugs frequently, however, may result in bugs and anchors on reconnaissance trips and insufficient time for dye adsorption onto the set background bugs as I find new springs. bug if the peak dye concentration for the trace High background fluorescence can result is highly diluted at the spring. Do not change from a variety of sources such as antifreeze, bugs more frequently than every 24 hours un- domestic sewage, and crop seed markers. The less you have access to a scanning fluorom- ubiquity of several tracers in the groundwater eter. Leaving bugs in the field for long periods from commercial applications is beginning to (more than 2 weeks) may result in their deg- interfere with their usefulness. radation, loss, or theft. For qualitative traces, (7) Conduct your first tests over short distances, changing bugs on a weekly schedule is the op- and then work your way headward in the timum for logistics, dye adsorption, and the basin. The initial groundwater traces can security of the bugs. be ongoing with the continuing reconnais- (11) Always place bugs before handling the dye. sance, provided the first traces are over short If the bugs must be transported in the same distances and well within the interior of the vehicle as the dye, be certain both containers hypothesized basin. This builds your knowl- are redundantly sealed and spillproof. edge of the flow system so that when you be- (12) Use enough dye, but try to anticipate the con- gin tracing along the basin divides, you can sequences of using too much. Choosing the make an educated guess as to which springs correct volume of dye is perhaps the most dif- the dye will flow. If carefully planned, this ficult decision made in groundwater dye trac- procedure can reduce the number of springs ing. Traces that are unexpectedly short are you will need to bug, or enable you to conduct the worst culprits, because the amount of dye multiple groundwater traces simultaneously. needed to reach the hypothesized resurgence By running many traces simultaneously, you is commonly more than enough to very viv- can save much time and labor. idly color the nearby actual resurgence. See (8) When conducting repeated traces, try to alter- (3) above. Algorithms to calculate the amount nate dyes that may flow to the same spring. needed are found in the appendices, but may This reduces the chance that some residual still result in amounts that are too small for dye in the system may result in a false posi- very long traces or high flow. A trace that is tive, and in dry weather, shortens the amount lost because too little dye was used is just as of time needed to complete the tests. Collect 18 Summary

annoying as one that flowed to an unmoni- some tracers, and therefore the detection of tracer tored spring. at very low fluorescence intensity may be suspect. For the vast majority of tracing projects, the time Summary lost adopting stringent minimal concentrations, The goal of the KGS groundwater tracing pro- for example, is thought to be greater than the in- tocol is to produce reliable results without impos- frequent occasions when a trace must be repeated. ing procedures that are excessively time-consum- The KGS protocol does not preclude, however, ap- ing. The methods are appropriate for studies not plying more stringent standards (for example, a involving litigation or where ambiguous traces can standardized mass of charcoal to elute) when the be redone. For point-to-point qualitative traces, the hypothesis being tested requires it. The protocol important result is connectivity, whereas precise presented in this document assures that any fluo- determination of tracer concentration is valuable rescence detected in the receptor at the wavelength but less time-efficient at this point in a study. KGS of the tracer is from the tracer. Use enough dye, but also recognizes there are compounds in ground- not too much. water that fluoresce at wavelengths near those of Selected Bibliography 19

Selected Bibliography monitoring: National Water Well Association, Alexander, E.C., and Quinlan, J.F., 1996, Practi- 26 p. cal tracing of groundwater with emphasis on Quinlan, J.F., 1989, Ground-water monitoring in karst terranes, in Eckenfelder Inc., Guidelines karst terranes: Recommended protocols and for wellhead and springhead protection area implicit assumptions: U.S. Environmental delineation in carbonate rocks: U.S. Environ- Protection Agency, Environmental Monitor- mental Protection Agency, Groundwater Pro- ing Systems Laboratory, 600/X-89/050, 77 p. tection Branch, Region 4, Atlanta, Georgia, Smart, C., and Worthington, S.R.H., 2004, Water 904-B-97-003, appendix B, 38 p. tracing, in Gunn, J., ed., Encyclopedia of caves Aley, T., and Fletcher, M.W., 1976, The water trac- and karst science: New York, Fitzroy Dear- er’s cookbook: Missouri Speleology, v. 16, no. born, Taylor and Francis Group, p. 769–771. 6, 36 p. Smart, P.L., 1984, A review of the toxicity of twelve Davis, S.N., Campbell, D.J., Bentley, H.W., and fluorescent dyes used for water tracing: The ­Flynn, T.J., 1985, Groundwater tracers: U.S. National Speleological Society Bulletin, Jour- Environmental Protection Agency, Office of nal of Caves and Karst Studies, v. 46, no. 2, Research and Development, and National p. 21–33. Water Well Association, 200 p. Spangler, L.E., Byrd, P.E., and Thrailkill, J., 1984, Field, M.S., 2003, Tracer-test planning using the ef- Use of Optical Brightener and Direct Yellow ficient hydrologic tracer-test design (EHTD) dyes for water tracing in the Inner Bluegrass program: U.S. Environmental Protection karst region, Kentucky: The National Speleo- Agency, National Center for Environmental logical Society Bulletin, Journal of Caves and Assessment–Office of Research and Develop- Karst Studies, v. 46, no. 2, p. 10–16. ment, EPA/600/R-03/034, 175 p. Thrailkill, J., Byrd, P.E., Sullivan, S.B., Spangler, Ford, D.C., and Williams, P.W., 1989, Karst geo- L.E., Taylor, C.J., Nelson, G.K., and Pogue, morphology and hydrology: London, Unwin K.R., 1983, Studies in dye-tracing techniques Hyman Ltd., 601 p. and karst hydrogeology: University of Ken- Jones, W.K., 1984, Dye tracer tests in karst areas: tucky Water Resources Research Institute, Re- The National Speleological Society Bulletin, search Report 140, 89 p. Journal of Caves and Karst Studies, v. 46, Thrailkill, J., Spangler, L.E., Hopper, W.M., Jr., no. 2, p. 3–9. ­McCann, M.R., Troester, J.W., and Gouzie, Kilpatrick, F.A., and Cobb, E.D., 1985, Measure- D.R., 1982, Groundwater in the Inner Blue- ment of discharge using tracers: U.S. Geologi- grass karst region, Kentucky: University of cal Survey, Techniques of Water-Resources Kentucky Water Resources Research Institute, Investigations, TWI 03-A16, 52 p. Research Report 136, 108 p. Kilpatrick, F.A., and Wilson, J.F., Jr., 1989, Mea- Wilson, J.F., Jr., Cobb, E.D., and Kilpatrick, F.A., surement of time of travel in streams by dye 1986, Fluorometric procedures for dye trac- tracing: U.S. Geological Survey, Techniques ing: Techniques of water-resources investiga- of Water-Resources Investigations, TWI 03- tions of the United States Geological Survey: A9, 27 p. Applications of Hydraulics, book 3, chapter Mull, D.S., Smoot, J.L., and Liebermann, T.D., A12, 34 p. 1988, Dye tracing techniques used to deter- Worthington, S.R.H., and Smart, C.C., 2003, Em- mine groundwater flow in a carbonate aquifer pirical determination of tracer mass for sink system near Elizabethtown, Kentucky: U.S. to spring tests in karst, in Beck, B.F., ed., Pro- Geological Survey Water-Resources Investi- ceedings of the Ninth Multidisciplinary Con- gations Report 87-4174, 95 p. ference, Sinkholes and the Engineering and Quinlan, J.F., ed., 1986, Qualitative water-tracing Environmental Impacts of Karst: American with dyes in karst terranes: Practical karst hy- Society of Civil Engineers, Geotechnical Spe- drogeology, with emphasis on groundwater cial Publication 122, p. 287–295. 20 Appendix A: Characteristics of Commonly Used Tracers OH 4 1 1 1 Sulforhodamine B (Acid Red 52) 3520-42-1 purple-red solution vivid orange-pink high very low unknown charcoal & autosampler 1-propanol & NH 0.08 ppb hydraulic fluid, anhydrous ammonia good organic detritus quenched in low pH very low 548 nm 572 nm 563 nm 578 nm $22.50/lb OH 4 1 1 Rhodamine WT (Acid Red 388) 37299-86-8 purple-red solution vivid orange-pink high very low minimal charcoal & autosampler 1-propanol & NH 0.01 ppb organics, crop seed > 25% by weight clays lower at higher temperature very low 558 nm 572 nm 550 nm 565 nm $33.00/lb Revised December 15, 2005 1 1 Fluorescein (Acid Yellow 73) 518-47-8 rust-brown powder brilliant green high low none charcoal & autosampler ethyl alcohol & 5% KOH 0.1 ppb chlorophyll, antifreeze good clays quenched at pH < 6 very high 485 nm 508 nm 500 nm 513 nm $30.00/lb

Appendix A: Characteristics of Commonly Used Tracers Appendix Aley (1999) Ford and Williams (1989) Name C.I. CAS Number Color Fluorescence Color Visibility in Water (naked eye) Toxicity Carcinogenic & Mutagenic Hazard Detector Types Recommended Elutant Minimum Detection Limit Background Interference Solubility Adsorption Fluorescence Stability Photochemical Decay Excitation Wavelength Emission Wavelength in Water KGS Emission Wavelength in Elutant (Varian Eclipse) Emission Wavelength in Elutant (from literature) Cost (2005) 1 2 Appendix A: Characteristics of Commonly Used Tracers 21 -

Solophenyl (Direct Yellow 96) Name has been changed not found yellow powder greenish yellow low low unknown cotton UV lamp; 365 nm “long set ting” 1.0 ppb none requires prolonged mixing none known quenched in low pH low 397 nm 490 nm nm nm $25.50/lb OH 4 1 1 1 Eosine (Acid Red 37) 17372-87-1 brown-green powder greenish orange to red high very low minimal charcoal & autosampler 1-propanol & NH 0.04 ppb pentachlorophenol good minor adsorption quenched in low pH very high 512 nm 533 nm 523 nm 538 nm $22.00/lb - Revised December 15, 2005 2 2 Tinopal CBS-X (Fabric Brightener 351) 38755-42-1 yellow-white powder blue-white very low low none cotton & autosampler UV lamp; 365 nm “long set ting” 0.1 ppb detergent 25 g/L organic detritus quenched in low pH high 360 nm 435 nm 435 nm 450 nm $22.50/lb Appendix A: Characteristics of Commonly Used Tracers Appendix Aley (1999) Ford and Williams (1989) Name C.I. CAS Number Color Fluorescence Color Visibility in Water (naked eye) Toxicity Carcinogenic & Mutagenic Hazard Detector Types Recommended Elutant Minimum Detection Limit Background Interference Solubility Adsorption Fluorescence Stability Photochemical Decay Excitation Wavelength Emission Wavelength in Water KGS Emission Wavelength in Elutant (Varian Eclipse) Emission Wavelength in Elutant (from literature) Cost (2005) 1 2 22 Appendix B: Tracer Mass Calculation Spreadsheet

Appendix B: Tracer Mass Calculation Spreadsheet

Access the Excel spreadsheet at kgs.uky.edu/kgsweb/olops/pub/kgs/IC26_12_Appendix_B.xls.

The effectiveness multiplier and equation to convert mass of tracer to the common concentrate solution of Rhodamine WT are to adjust the mass calculations for less effective or diluted tracers. The mass read from the matrix should be multiplied by the appropriate effectiveness. For example, the liquid equivalent of 119 g of tracer for Rhodamine is 500 ml. The mass of Tinopal CBS-X is 3 times, or 357 g for the same matrix value. Appendix C: Matrix for Estimating Mass of Groundwater Tracer for Connectivity in Karst (10 ppb) 23 X 1 X 1 X 1 X 3 X 4 76 147 216 284 352 418 484 550 615 679 744 808 872 935 998 100 1,062 1,125 1,187 1,312 1,929 2,535 3,134 3,726 2.832 69 133 196 257 318 378 438 497 556 615 673 731 789 846 903 960 90 1,017 1,074 1,187 1,745 2,294 2,835 3,371 2.549 62 119 175 230 284 338 392 445 497 550 602 653 705 756 808 859 910 960 80 1,062 1,560 2,051 2,535 3,015 Effectiveness Multiplier Fluorescein Eosine SR-B Brightener DY 96 2.265 54 105 154 203 251 298 345 392 438 484 530 576 621 666 711 756 801 846 935 70 1,375 1,806 2,233 2,655 1.982 47 91 133 175 216 257 298 338 378 418 458 497 536 576 615 653 692 731 808 60 1,187 1,560 1,929 2,294 1.699 39 76 112 147 182 216 251 284 318 352 385 418 451 484 517 550 582 615 679 998 50 1,312 1,622 1,929 1.416 To convert grams to ml ((Mass/1.19 s.g.)/0.2) ml=(g/1.19 g/ml)/0.2 of 20% RWt; grams X 4.2 32 62 91 119 147 175 203 230 257 284 311 338 365 392 418 445 471 497 550 808 40 1,062 1,312 1,560 1.133 24 47 69 91 112 133 154 175 196 216 237 257 278 298 318 338 358 378 418 615 808 998 30 1,187 0.850 17 32 47 62 76 91 =10 μg/L 3 105 119 133 147 161 175 189 203 216 230 244 257 284 418 550 679 808 =100 μg/L 20 3 =1 mg/L 0.566 3 Spreadsheet: James C. Currens, KGS, 2004 1 g/m 1 mg/L=1,000 μg/L 0.1 g/m 0.01 g/m 13 24 36 47 58 69 80 91 101 112 123 133 144 154 165 175 185 196 216 318 418 517 615 15 0.425 9 17 24 32 39 47 54 62 69 76 83 91 98 105 112 119 126 133 147 216 284 352 418 10 0.283 7 13 20 26 32 38 44 50 56 62 67 73 79 85 91 96 8 102 108 119 175 230 284 338 0.227 4 7 10 13 17 20 23 26 29 32 35 38 41 44 47 50 53 56 62 91 4 119 147 175 0.113 , T=sec 0.01 Estimated Discharge at the Expected Resurgence, CFS (Top Row), CMS (Bottom Row) 3

3 2 4 5 7 9 2 10 12 13 15 17 18 20 21 23 24 26 27 29 32 47 62 76 91 0.057 75g 1 2 3 4 4 5 6 7 8 9 9 10 11 12 13 13 14 15 17 24 32 39 47 1 /sec, C=g/m 3 0.028 50g This estimate is dependent on the accuracy of your prediction distance to resurgence and its discharge. 152 305 457 610 762 914 (Worthington and Smart, 2003) 1,067 1,219 1,372 1,524 1,676 1,829 1,981 2,134 2,286 2,438 2,591 2,743 3,048 4,572 6,096 7,620 9,144 0.95 CFS > Meters Est. Dist. 25g 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500 6,000 6,500 7,000 7,500 8,000 8,500 9,000 Feet 10,000 15,000 20,000 25,000 30,000 Est. Dist. Appendix C: Matrix for Estimating Mass of Groundwater Tracer for Connectivity in Karst (10 ppb) Appendix C: Matrix for Estimating Mass of Groundwater Tracer Desired concentration is 10 ppb or 0.01 g/m Equation: M=19 (LQC) Units: M=grams, L=meters, Q=m 1 hour=3,600 sec, day=86,400 m=meters Minimum mass: 24 Appendix D: Matrix for Estimating Mass of Groundwater Tracer for Connectivity in Karst (100 ppb) X 1 X 1 X 1 X 3 X 4 679 100 1,312 1,929 2,535 3,134 3,726 4,314 4,897 5,477 6,054 6,628 7,199 7,768 8,334 8,899 9,461 2.832 10,022 10,581 11,695 17,191 22,594 27,929 33,211 615 90 1,187 1,745 2,294 2,835 3,371 3,903 4,431 4,956 5,477 5,996 6,513 7,028 7,540 8,051 8,560 9,068 9,574 2.549 10,581 15,554 20,442 25,269 30,048 550 80 1,062 1,560 2,051 2,535 3,015 3,490 3,962 4,431 4,897 5,362 5,824 6,284 6,742 7,199 7,654 8,108 8,560 9,461 2.265 13,907 18,278 22,594 26,867 Effectiveness Multiplier Fluorescein Eosine SR-B Brightener DY 96 484 935 70 1,375 1,806 2,233 2,655 3,074 3,490 3,903 4,314 4,723 5,130 5,535 5,939 6,341 6,742 7,142 7,540 8,334 1.982 12,250 16,100 19,902 23,666 418 808 60 1,187 1,560 1,929 2,294 2,655 3,015 3,371 3,726 4,079 4,431 4,781 5,130 5,477 5,824 6,169 6,513 7,199 1.699 10,581 13,907 17,191 20,442 352 679 998 50 1,312 1,622 1,929 2,233 2,535 2,835 3,134 3,431 3,726 4,021 4,314 4,606 4,897 5,188 5,477 6,054 8,899 1.416 11,695 14,457 17,191 284 550 808 To convert grams to ml ((Mass/1.19 s.g.)/0.2 ml=(g/1.19 g/ml)/0.2 of 20% RWt; grams X 4.2 40 1,062 1,312 1,560 1,806 2,051 2,294 2,535 2,775 3,015 3,253 3,490 3,726 3,962 4,197 4,431 4,897 7,199 9,461 1.133 11,695 13,907 216 418 615 808 998 30 1,187 1,375 1,560 1,745 1,929 2,112 2,294 2,475 2,655 2,835 3,015 3,193 3,371 3,726 5,477 7,199 8,899 0.850 10,581 147 284 418 550 679 808 935 20 =10 μg/L 1,062 1,187 1,312 1,437 1,560 1,684 1,806 1,929 2,051 2,172 2,294 2,535 3,726 4,897 6,054 7,199 3 =100 μg/L 0.566 3 =1 mg/L 3 112 216 318 418 517 615 711 808 903 998 Spreadsheet: James C. Currens, KGS, 2004 1 g/m 1 mg/L=1,000 μg/L 0.1 g/m 0.01 g/m 15 1,093 1,187 1,281 1,375 1,468 1,560 1,653 1,745 1,929 2,835 3,726 4,606 5,477 0.425 76 147 216 284 352 418 484 550 615 679 744 808 872 935 998 10 1,062 1,125 1,187 1,312 1,929 2,535 3,134 3,726 0.283 62 8 119 175 230 284 338 392 445 497 550 602 653 705 756 808 859 910 960 0.227 1,062 1,560 2,051 2,535 3,015 32 62 91 4 119 147 175 203 230 257 284 311 338 365 392 418 445 471 497 550 808 0.113 1,062 1,312 1,560 Estimated Discharge at the Expected Resurgence, CFS (Top Row), CMS (Bottom Row) , T=sec 0.1 3

3 2 17 32 47 62 76 91 105 119 133 147 161 175 189 203 216 230 244 257 284 418 550 679 808 0.057 75g 9 17 24 32 39 47 54 62 69 76 83 91 98 1 105 112 119 126 133 147 216 284 352 418 /sec, C=g/m 3 0.028 50g This estimate is dependent on the accuracy of your prediction distance to resurgence and its discharge. 152 305 457 610 762 914 (Worthington and Smart, 2003) 1,067 1,219 1,372 1,524 1,676 1,829 1,981 2,134 2,286 2,438 2,591 2,743 3,048 4,572 6,096 7,620 9,144 0.95 CFS > Meters Est. Dist. 25g 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500 6,000 6,500 7,000 7,500 8,000 8,500 9,000 Feet 10,000 15,000 20,000 25,000 30,000 Est. Dist. Appendix D: Matrix for Estimating Mass of Groundwater Tracer for Connectivity in Karst (100 ppb) Appendix D: Matrix for Estimating Mass of Groundwater Tracer Desired concentration is 100 ppb or 0.1 g/m Equation: M=19 (LQC) Units: M=grams, L=meters, Q=m 1 hour=3,600 sec, day=86,400 m=meters Minimum mass: Appendix E: Matrix for Estimating Mass of Groundwater Tracer for Time of Travel in Karst (10 ppb) 25 X 1 X 1 X 1 X 3 X 4 28 55 82 108 134 160 186 211 237 263 288 313 339 364 389 414 464 514 762 100 1,008 1,251 2.832 25 50 74 97 121 144 168 191 214 237 260 283 306 329 351 374 419 464 688 910 90 1,130 2.549 23 44 66 87 108 129 150 170 191 211 232 252 273 293 313 334 374 414 614 812 80 1,008 2.265 Effectiveness Multiplier Fluorescein Eosine SR-B Brightener DY 96 20 39 58 76 95 113 131 150 168 186 204 222 240 258 275 293 329 364 539 713 885 70 1.982 17 34 50 66 82 97 113 129 144 160 175 191 206 222 237 252 283 313 464 614 762 60 1.699 14 28 42 55 68 82 95 108 121 134 147 160 173 186 199 211 237 263 389 514 639 50 1.416 12 23 34 44 55 66 76 87 97 108 118 129 139 150 160 170 191 211 313 414 514 40 1.133 To convert grams to ml ((Mass/1.19 s.g.)/0.2) ml=(g/1.19 g/ml)/0.2 of 20% RWt; grams X 4.2 9 17 25 34 42 50 58 66 74 82 90 97 105 113 121 129 144 160 237 313 389 30 0.850 6 12 17 23 28 34 39 44 50 55 60 66 71 76 82 87 97 108 160 211 263 20 0.566 =10 μg/L 3 =100 μg/L 3 =1 mg/L 4 9 3 13 17 21 25 30 34 38 42 46 50 54 58 62 66 74 82 121 160 199 15 0.425 Spreadsheet: James C. Currens, KGS, 2004 1 g/m 1 mg/L=1,000 μg/L 0.1 g/m 0.01 g/m 3 6 9 12 14 17 20 23 25 28 31 34 36 39 42 44 50 55 82 108 134 10 0.283 2 5 7 9 12 14 16 18 20 23 25 27 29 31 34 36 40 44 66 87 8 108 0.227 1 2 4 5 6 7 8 9 10 12 13 14 15 16 17 18 20 23 34 44 55 4 0.113 0.01 1 1 2 2 3 4 4 5 5 6 6 7 8 8 9 9

3 2 10 12 17 23 28 , T=sec Estimated Discharge at the Expected Resurgence, CFS (Top Row), CMS (Bottom Row) 0.057 3 0 1 1 1 2 2 2 2 3 3 3 4 4 4 4 5 5 6 9 12 14 1 0.028 75g /sec, C=g/m 3 1,524 3,048 4,572 6,096 7,620 9,144 10,668 12,192 13,716 15,240 16,764 18,288 19,812 21,336 22,860 24,384 27,432 30,480 45,721 60,961 76,201 (sec) 50g CFS > (Worthington and Smart, 2003) to Spring Est. Total This estimate is dependent on the accuracy of your prediction travel time to resurgence and its discharge. 0.97 152 305 457 610 762 914 25g 1,067 1,219 1,372 1,524 1,676 1,829 1,981 2,134 2,286 2,438 2,743 3,048 4,572 6,096 7,620 Meters is assumed to be 0.1 m/sec estimate TOT from distance 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500 6,000 6,500 7,000 7,500 8,000 9,000 Feet 10,000 15,000 20,000 25,000 Estimated Distance Appendix E: Matrix for Estimating Mass of Groundwater Tracer for Time of Travel in Karst (10 ppb) of Travel for Time Appendix E: Matrix for Estimating Mass of Groundwater Tracer Desired peak concentration is 10 ppb or 0.01 g/m Velocity Equation: M=0.73 (TQC) Units: M=grams, L=meters, Q=m 1 hour=3,600 sec, day=86,400 m=meters Minimum mass: 26 Appendix F: Matrix for Estimating Mass of Groundwater Tracer for Time of Travel in Karst (100 ppb) X 1 X 1 X 1 X 3 X 4 263 514 762 100 1,008 1,251 1,493 1,734 1,974 2,213 2,451 2,925 3,397 3,866 4,334 4,801 7,114 9,404 2.832 237 464 688 910 90 1,130 1,348 1,566 1,782 1,998 2,213 2,641 3,067 3,491 3,913 4,334 6,423 8,490 2.549 211 414 614 812 80 1,008 1,203 1,397 1,590 1,782 1,974 2,356 2,736 3,114 3,491 3,866 5,729 7,574 2.265 Effectiveness Multiplier Fluorescein Eosine SR-B Brightener DY 96 186 364 539 713 885 70 1,056 1,227 1,397 1,566 1,734 2,069 2,403 2,736 3,067 3,397 5,033 6,654 1.982 160 313 464 614 762 910 60 1,056 1,203 1,348 1,493 1,782 2,069 2,356 2,641 2,925 4,334 5,729 1.699 134 263 389 514 639 762 885 50 1,008 1,130 1,251 1,493 1,734 1,974 2,213 2,451 3,632 4,801 1.416 108 211 313 414 514 614 713 812 910 40 1,008 1,203 1,397 1,590 1,782 1,974 2,925 3,866 1.133 To convert grams to ml (Mass/1.19 s.g.)/0.2) ml=(g/1.19 g/ml)/0.2 of 20% RWt; grams X 4.2 82 160 237 313 389 464 539 614 688 762 910 30 1,056 1,203 1,348 1,493 2,213 2,925 0.850 55 108 160 211 263 313 364 414 464 514 614 713 812 910 20 1,008 1,493 1,974 0.566 =10 μg/L 3 =100 μg/L 3 =1 mg/L 3 42 82 121 160 199 237 275 313 351 389 464 539 614 688 762 15 1,130 1,493 0.425 Spreadsheet: James C. Currens, KGS, 2004 1 g/m 1 mg/L=1,000 μg/L 0.1 g/m 0.01 g/m 28 55 82 108 134 160 186 211 237 263 313 364 414 464 514 762 10 1,008 0.283 23 44 66 87 8 108 129 150 170 191 211 252 293 334 374 414 614 812 0.227 12 23 34 44 55 66 76 87 97 4 108 129 150 170 191 211 313 414 0.113 0.1 6

3 2 12 17 23 28 34 39 44 50 55 66 76 87 97 108 160 211 , T=sec Estimated Discharge at the Expected Resurgence, CFS (Top Row), CMS (Bottom Row) 0.057 3 3 6 9 12 14 17 20 23 25 28 34 39 44 50 55 82 1 108 0.028 75g /sec, C=g/m 3 1,524 3,048 4,572 6,096 7,620 9,144 10,668 12,192 13,716 15,240 18,288 21,336 24,384 27,432 30,480 45,721 60,961 (sec) 50g CFS > (Worthington and Smart, 2003) to Spring Est. Total This estimate is dependent on the accuracy of your prediction travel time to resurgence and its discharge. 0.97 152 305 457 610 762 914 25g 1,067 1,219 1,372 1,524 1,829 2,134 2,438 2,743 3,048 4,572 6,096 Meters is assumed to be 0.1 m/sec estimate TOT from distance 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 6,000 7,000 8,000 9,000 Feet 10,000 15,000 20,000 Estimated Distance Appendix F: Matrix for Estimating Mass of Groundwater Tracer for Time of Travel in Karst (100 ppb) of Travel for Time Appendix F: Matrix for Estimating Mass of Groundwater Tracer Desired peak concentration is 100 ppb or 0.1 g/m Velocity Equation: M=0.73 (TQC) Units: M=grams, L=meters, Q=m 1 hour=3,600 sec, day=86,400 m=meters Minimum mass: Appendix G: Kentucky Geological Survey, Groundwater Trace Monitoring and Analysis Data Report 27 Comment Date: Date: Not recovered. Not recovered. Not determined. NR Conc. ND Water Time OB Ambiguous. Not applicable. Cotton X DY Analysis Results NA Date of Field Work: Eosine SRB Charcoal R WT. Positive background. B+ Fluor. Person Receiving Trace Receptors: Person Conducting Analysis: - Pull drop Gum Explanation Bug destroyed / bug not changed. Collect Cotton BD Date: Date: Char. Negative background. Field Personnel: B– Field Tasks Cotton Bug missing. BM Hang Char. Groundwater Trace Monitoring and Analysis Data Report Monitoring and Groundwater Trace - Rig drop Gum Appendix G: University of Kentucky, Kentucky Geological Survey Appendix G: University of Kentucky, No. Visit Record Location: Name or ID Trip Blank (always)

24-hr Clock Time: Project or Area: √ Bug changed or task performed. – Negative (< 1 ppb). + Positive (> Signatures Are Required: Person Delivering Receptors: Person Preparing Receptor: 28 Appendix H: Groundwater Trace Injection Report Form

Appendix H: Groundwater Trace Injection Report Form Circle, √, X, or underline wherever possible Injection Site Location

Reporters: Report date: Field ID: Trace serial no.: Site name: Type of karst feature: County: Quadrangle: Corridor or project: ״ . ׳ ° :W longitude ״ . ׳ ° :Geodetic location: N latitude UTM location: Zone: , easting , northing Elevation: , Accuracy est. GPS Topo Other Additional forms completed for this location: Spring Inspection Cover-Collapse Report Bug Sheet Karst Feature Inventory

Tracer Injection Weather or field conditions:

Date of injection: Time (military) of injection: Amount of tracer: Batch or lot no. (optional):

Tracer Used Eosine-Y (Acid Red 87) Sulforhodamine B (Acid Red 52) Rhodamine WT (Acid Red 388) Tinopal CBS-X (Fabric Brightener 351) Sodium Fluorescein (Acid Yellow 73) Other tracer (specify) Solophenyl (Direct Yellow 96)

Tracer Recovery Name of positive spring(s): AKGWA no.: Detector placement: Date: Time (military): Detector recovery: Date: Time (military): Other positive sites? For a complete list of all monitored sites, see the Monitoring Report dated:

Comment or Interpretation Appendix I: Definitions 29

Appendix I: Definitions Aliquot: A portion of a larger whole, especially a or till) that formed the roof of a soil void or conduit sample taken for chemical analysis. at the soil-bedrock interface, or spanned a grike Base flow: Normal rate of outflow of groundwa- (fissure) or other karst void in the bedrock. ter from a spring that is not influenced by rapid Cuvette: A test tube or chamber that holds a sam- recharge from a storm or snowmelt. Base flow is a ple and fits into a receiving chamber of an analyti- steady or slowly decreasing discharge rate and has cal machine. uniform physical water-quality characteristics (i.e., low turbidity). Discharge measurements should be Dissolution sinkhole (synonym: doline): Sinkhole made between the complete recession of any peaks resulting exclusively from gradual dissolution of from the runoff of a storm up to the next runoff the bedrock and removal of the dissolved rock and event. Summer base flow in Kentucky is from mid- insoluble residuum via the sinkhole throat and June through mid-October, whereas winter base karst aquifer conduits. Dissolution sinkholes may flow is from mid-December through mid-March. be totally buried and filled, or the bedrock can be totally exposed. Most dissolution sinkholes have Bedrock-collapse sinkhole: Formed by the col- the classic bowl-shaped contour, with a variable lapse of a bedrock roof into an underlying cave. thickness of soil or other unconsolidated residual Bedrock collapse in karst is rare, but is the origin of covering the bedrock. Also see epikarst. some sinkholes, principally karst windows. Eluent: The solution applied to activated carbon Cave: A natural opening created by dissolution or charcoal to extract fluorescent tracing dyes ad- erosion of bedrock and large enough for an adult sorbed on the charcoal. Common mixtures are person to enter. The flow of water in a cave may 5 percent by volume potassium hydroxide and be year-round, seasonal, high-flow (flood) only, 70 percent isopropyl alcohol, the balance being or permanently dry. A diameter of 50 cm and a distilled water, and for Smart Solution, 50 percent length (depth) of 2 m are approximate minimal di- of 1-propanol to 20 percent ammonia hydroxide mensions. In Kentucky, caves can exceed 30 m in (NH4OH) to 30 percent water, by volume. There width and hundreds of kilometers in aggregated are other solutions. The choice depends on the dye length. Orientation of the cave passage in space is being extracted. not definitive; therefore, an open air pit with mini- mal overhanging ledges is a cave if the minimum Elutant: The liquid with tracer decanted from the dimensions are met. The orientation relevant to the charcoal after elution. outcrop is significant in that an opening 2 m wide Epikarst (synonym: subcutaneous zone): The in- and 0.6 m deep in a cliff is an overhang or rock terval below the organic soil and above the mass shelter, not a cave. of largely unweathered soluble bedrock, consist- Conduit: A tubular opening created by dissolution ing of highly corroded bedrock, residuum, subsoil, of the bedrock, which carries, can carry, or has car- float, and unconsolidated material of other origins. ried water. Conduits have a minimum diameter of Thickness of the epikarst varies from absent to a re- 1 cm up to a maximum diameter of 0.5 m. Flow in ported 30 m. The epikarst is important for the stor- a conduit may be year-round, seasonal, high-flow age and transport of soil water and groundwater only, or permanently dry. The minimum diameter in the karst system and is relevant to foundation is the approximated critical diameter of 1 cm, at stability. which fundamental changes in the mechanisms of Epikarstic dye injection point: A shallow excava- carbonate dissolution and groundwater flow -oc tion, typically a trench dug with a backhoe, but in- cur. cluding auger holes into soil and very shallow ex- Cover-collapse sinkhole: Formed by the collapse cavations into bedrock. Tracer is mixed with water of the unconsolidated cover (soil, residuum, loess, in the trench or a tank, then poured into the hole. 30 Appendix I: Definitions

Additional water is added until the movement of broadcloth. Tracer receptors are fastened to a gum- the tracer into the epikarst is assured. drop or field-expedient anchor. Fluorescent dye (synonym: tracer): One of sev- Ponor: See sinkhole throat. eral organic dyes that fluoresce under short-wave- Qualitative trace: Tracer experiment to establish length light, particularly when dissolved in water the point-to-point connectivity, or flow vector, or other solvent. Detection of the dye is confirmed from the input point of the tracer to the resurgence. by analysis of elutant or water samples with a fluo- It identifies only the presence of tracer in the wa- rometer. ter (Smart and Worthington, 2004). The method of Gumdrop: Anchor fashioned from concrete and sampling and determination of the tracer is imma- heavy-gage wire that suspends a passive dye re- terial to the concept, but is generally assumed to be ceptor above the bottom of the channel. Usually a passive tracer receptors or bugs. lanyard is tied from the gumdrop to a higher eleva- Quantitative trace: Combines concentration mea- tion so that the dye receptor can be changed during surements of tracer in water at the resurgence with high water. flow-rate measurements, permitting the compila- Karst aquifer: A body of soluble rock that conducts tion of mass-flux curves. The area under the mass- water principally via a connected network of tribu- flux curve represents the total mass of tracer mea- tary conduits, formed by the dissolution of the rock, sured at a site (Field, 2002; Smart and Worthington, which drain a groundwater basin and discharge to 2004). The incremental mass flux of tracer, the total at least one perennial spring. The conduits may be mass of tracer recovered, and the center of mass of partly or completely water-filled. The karst aquifer the tracer breakthrough curve are among the quan- may also have primary (intergranular) and second- titative measures calculated from the concentration ary porosity (fracture), which is saturated with wa- and discharge data. ter when below the potentiometric surface. Semiquantitative trace: The concentration of the Karst terrain (synonym: karst terrane): A land- tracer in water samples collected over time is de- scape generally underlain by limestone or dolo- termined as in a quantitative trace. Discharge is mite, in which the topography is chiefly formed by not included in the calculation. The resulting time dissolving the rock and which may be character- versus concentration curve provides additional ized by sinkholes, sinking streams, closed depres- data for recognition of the tracer, and in some cases sions, subterranean drainage, and caves (Monroe, can aid in determining characteristics of the traced 1970). The term “terrain” implies that only the sur- route (Smart and Worthington, 2004). face is considered, whereas “terrane” includes the Sinkhole: Any closed depression in soil or bed- subsurface (caves or aquifer) as a single system. rock formed by the erosion and transport of earth Karst also forms on gypsum and salt bedrock, al- material from below the land surface, which is cir- though not in Kentucky. cumscribed by a closed topographic contour and Karst valley: A mid-sized to valley-scale closed de- drains to the subsurface. Morphologies of sink- pression otherwise meeting the definition of a sink- holes formed in soluble rock include dissolution hole but also enclosing and including more than sinkhole (gently sloping depression wider than one smaller sinkholes and/or a sinking stream. deep), karst window (sinkhole exposing an under- ground stream), vertical shaft (depression in bed- Passive tracer receptor (synonyms: bugs, dye de- rock much deeper than wide and roughly circular tector, dye receptor, passive charcoal detector): in plan), and grike (depression in soil and bedrock Consists of two general types. The activated car- much deeper than wide and roughly lenticular in bon receptor is constructed of a few grams of co- plan). Also see cover-collapse sinkhole and bed- conut shell charcoal enclosed in a mesh bag typi- rock-collapse sinkhole. cally made of nylon screen. The cotton receptor is a section of untreated surgical cotton or bleached Sinkhole cluster area: A group of two or more sinkholes clustered so that the average spacing Appendix I: Definitions 31 among them is closer than the average spacing of perennial stream varying in flow rate from rivulet sinkholes in the immediate area as a whole. A sink- to rivers. hole cluster is likely to have a common groundwa- Sinkhole watershed: An area bounded by a pro- ter basin. jected line demarcating a change in slope from to- Sinkhole throat (informal usage): Outlet or outlets ward the center of the sinkhole to away from the for a sinkhole allowing runoff from the sinkhole sinkhole, which represents a local topographic watershed to flow into the ground. Not all sinkhole drainage divide. Precipitation falling on the sur- throats have a discernable opening or an opening face sloping toward the sinkhole is likely to run large enough to enter, but some have large dimen- into the sinkhole throat or infiltrate the soil and sions and all are a sink point for an intermittent or move through subsoil conduits to the throat.