Ames Sink Tracer Test
Todd R. Kincaid, PhD
A rapid connection between Ames Sink and Indian, Sally Ward, and Wakulla Spring has been confirmed. The destination and travel-time of water entering Ames sink have long been contentious issues because the sink receives a large part of the surface water runoff from the city of Tallahassee. This tracing experiment has shown that the runoff water is flowing to the springs.
We began the Ames Sink test on August 11, just before Tropical Storm Bonnie unloaded its rain on the WKP. The tracer was injected into the stream flow at 7:55 PM. The stream stage was approximately .09 m (~0.3 ft). By 10:00 AM the next morning, the dye had flowed into the ground and the stage height had risen to greater than 1.83 m (~6.0 ft) due to the rain. Sampling had already begun at more than 10 points in the basin and by the end of the day we were continuously sampling 24 locations. Ten days later the tracer was detected at Indian spring. Within another 10 days it was also recorded at Sally Ward and Wakulla springs. Individual dye hits were also detected at B-tunnel and K-tunnel.
The goal of all our tracing experiments is to use as little dye as possible to achieve a detectable break-through curve at the sampling locations, where the amount of dye required is determined by the distance of the flow path, flow rate, rate of injection, and the background fluorescence. Minimizing the amount of dye reduces cost, potential impacts to local water wells, and residual levels of dye in the aquifer. Seven and one half kilograms (~16.5 lbs) of liquid uranine dye (C.I. Acid Yellow 73) were poured into the stream flow immediately above Ames Sink at 7:35 PM on August 11, 2004. The dye flowed rapidly into the sink and disappeared into the ground within 12 hours as the stream stage rose 2 meters ( ~6 ft) from rainfall associated with Tropical Storm Bonnie.
The Ames Sink test was the most ambitious trace we have attempted. The minimum distance between the injection point and the nearest probable receptor (Leon sinks Cave System at Big Dismal Sink) is 5.1 km (~3.2 miles), which is approximately twice as far as the pathways traced at Fisher Creek and Black Creek. The major springs, Wakulla, Indian, Sally Ward, McBride’s, and the St. Marks River Rise are 9.3, 8.1, 8.7, 9.2, and 15.13 km (5.8, 5.0, 5.4, 5.7 and 9.4 miles) away, respectively.
Automated water samplers were setup to collect samples at between 1 and 6 hour intervals at 12 locations: Abysmal, Susan, Force, Sullivan, Cheryl, Emerald, Upper River, Turner, and Ferrell-Turner sinks; K-Tunnel and the spring vent in Wakulla cave; and the St. Marks River rise. Grab samples were collected between 2 and 3 times per day, with the help of our project volunteers, at an additional 9 locations: Rectangle, Horse Lot, Ferrell-Road, and Meeting House sinks; and Indian, Sally Ward, McBride’s, Newport, and Rhodes springs. Sampling began on August 11 and continued through August 16 for the sinkholes closest to Ames and August 30 for the remaining locations setup with autosamplers. Grab sampling was continued at the springs through September 9.
The first indication of the tracer appeared at K-Tunnel, but only in one grab sample. Shortly thereafter, approximately 10 days after the injection, the strongest signal from the tracer was recorded at Indian Spring. A total of nineteen samples on a well defined break-through curve were recorded at Indian over the ensuing 13 days with a peak concentration of approximately 2.8 ppb. The beginnings of two additional break-through curves were recorded at Sally Ward spring beginning at approximately 15 days after the injection and at Wakulla spring beginning at approximately 20 days after the injection.
The tracer recoveries at each of the four locations is supported by analyses performed on a scanning spectrofluorophotometer wherein the synchronously-scanned emission spectrum for C.I.A.Y. 73 was observed and measured for each of the samples on the respective recovery curves. We are therefore confident in the designation of flow paths between Ames Sink and the three springs. The exact pathways and transit velocities, however, cannot be confidently calculated because complete recovery curves were not obtained and the tracer was not observed in any of the intermediate sinkholes between Ames Sink and the springs. One of the possible pathways lies between Ames Sink and some point in the downstream section of the Leon Sinks cave system that was shown previously to connect to K-Tunnel south of Turner sink.
The intermediate sinkhole data, or lack thereof, are largely unreliable because none of the sinkholes other than those intersecting the Leon Sinks cave system were properly instrumented with tubing leading to observably flowing water. Persistent high flow conditions in the basin kept visibility below the level deemed necessary to conduct safe diving operations by the Woodville Karst Plain Project divers during the months prior to the test. The high flow conditions through the basin resulting from the large rainfall events immediately after the injection likely contributed to significantly more tracer dilution than we had anticipated and potentially the activation of additional flow paths. These three factors could have led to both the lack of tracer recovery in the intermediate sinkholes as well as the low and variably detectable concentrations at Sally Ward and the Wakulla spring vent.
Another complicating factor only became apparent toward the end of the sampling period, when discussions with a property owner that had provided access for sampling led us to a sinkhole approximately 700 m (~2100 ft) further downstream of Ames sink. This sink was receiving overflow water from Ames and funneling it into an observable limestone outcrop into the aquifer. Flow measurements revealed that the newly identified feature (“Spider Sink”) recharges at least 10 times more flow into the aquifer than Ames Sink, at a rate of 1.8 cubic meters per second (~62.5 cubic feet per second). The existence of “Spider Sink” further increases the likelihood that the tracer was significantly diluted shortly after injection and would produce lower-concentration, broader break-through curves such as the ones the were partially recorded at Indian, Sally Ward and Wakulla springs.
We’ve conducted five successful quantitative tracer tests in the Woodville Karst Plain: Sullivan Sink – Cheryl Sink, Fisher Creek – Emerald Sink, Black Creek – Emerald Sink – Wakulla Spring, Emerald Sink – Wakulla Spring, and now Ames Sink – Indian, Sally Ward, and Wakulla springs. Each of these traces has demonstrated rapid connections between the disappearing streams in the basin and the major springs, thereby broadening our understanding of the extent of the springsheds that directly influence spring water quality. The most important conclusion from this work is that the karst features (caves) provide the controlling influence on flow paths and velocities in the Woodville Karst Plain and that these features provide similar controls on flow patterns in other karst basins in Florida. It is crucial, therefore, that we adequately delineate these features and address their influence on groundwater and surface water flow patterns if we are to effectively manage and protect our groundwater resources. In our next phase of work we will endeavor to reproduce the Ames Sink results and obtain complete break-through curves such that the key hydraulic parameters along the respective flow paths can be established; expand our studies to include the Spring Creek springs; and use the amassed data to develop a model of groundwater flow through the Woodville Karst Plain that both accurately embraces the karst flow patterns and provides a predictive tool for improved water resource management in the basin.
This project was funded by the Florida Geological Survey Hydrogeology Program and the Florida Springs Initiative. Dr. Rodney DeHan and Dr. David Loper of the Florida Geological Survey (FGS) and the Florida State University Geophysical Fluid Dynamics Institute (GFDI), respectively, have continued to work diligently as the project directors to support this research. The Northwest Florida Water Management District and the Florida Department of Environmental Protection provided the autosamplers that were crucial to the success of the project. Wakulla Springs State Park provided logistical support and lodging for our team members. The Global Underwater Explorers – Woodville Karst Plain Project provided professional highly-skilled cave divers that instrumented the springs and sinkholes with tubing, which was critical to effective sampling and without which our successes thus far would have been highly suspect. Gareth Davies of Cambrian Ground Water Co. is one of the world’s foremost experts in quantitative groundwater tracing and has skillfully directed all of our injection and sampling protocols. James McClean of the FGS, Brent Meyer of Hazlett-Kincaid, Inc, and Chris Werner of GFDI were the technical staff that ensured that the sampling equipment stayed up and running throughout the duration of the test. Cal Jamison, the Springs Ambassador for the Woodville Karst Plain, patiently led us to numerous sinkholes throughout the basin and helped negotiate access for sampling. Linda Jamison, Ed Chelette, and Paul Johnson volunteered to collect grab samples throughout the month-long test. All of these individuals and organizations as well as many more that I haven’t mentioned, for the sake of brevity, including the numerous property owners that graciously allowed us access to the sinkholes at all hours of the day and night, were invaluable to the success of this project. Our sincere thanks go out to everyone that has helped make this project a success.
Study made possible through the support of: