Ground water tracing report for Buttercup Creek property caves: Initial groundwater dye tracing study of the Buttercup Creek area, Cedar Park, Williamson County, Texas

Material Information

Ground water tracing report for Buttercup Creek property caves: Initial groundwater dye tracing study of the Buttercup Creek area, Cedar Park, Williamson County, Texas
Hauwert, Nico Mark
Warton, Mike
Mike Warton & Associates
Publication Date:


Subjects / Keywords:
Buttercup Blowhole Cave (Texas, United States) ( 30.496733, -97.837831 )
Buttercup Creek Cave (Texas, United States) ( 30.490486, -97.845408 )
Buttercup Drain Cave (Texas, United States) ( 30.491219, -97.846961 )
Convoluted Canyon Cave (Texas, United States) ( 30.4903, -97.844783 )
Hide-a-Way Cave (Texas, United States) ( 30.494183, -97.845156 )
Ilex Cave (Texas, United States) ( 30.489556, -97.847397 )
Nelson Ranch Cave (Texas, United States) ( 30.489158, -97.848444 )
Geology ( local )
Technical Report
serial ( sobekcm )
United States
30.496733 x -97.837831
30.490486 x -97.845408
30.491219 x -97.846961
30.4903 x -97.844783
30.494183 x -97.845156
30.489556 x -97.847397
30.489158 x -97.848444


General Note:
This report details findings of a groundwater quality sampling study of the Buttercup Creek area. The study shed light on the destination of flowing water in the area.
Open Access - Permission by Author(s)
General Note:
See Extended description for more information.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-01511 ( USFLDC DOI )
k26.1511 ( USFLDC Handle )
11522 ( karstportal - original NodeID )

USFLDC Membership

Karst Information Portal

Postcard Information



This item has the following downloads:

Full Text
This report details findings of a groundwater quality
sampling study of the Buttercup Creek area. The study shed
light on the destination of flowing water in the area.





CONTENTS: Preface ..,.Introduction Geomorphology and Karst Mechanics of Cave Development Endangered Invertebrate Species of the "Buttercup Creek" Area Karst Feature Inventory and Research Classification Status Three Cave Preserves -Management Report, Update Summary for 1997 Introduction and Exploration History of the Three Cave Sites Cave Preserve Management Plan Geology and Hydrogeology of the Three Cave Sites Endangered Species Count&Observational Surveys Synoptic Scale Weather Event -Account of Tornado Strike, MAY 27, 1997 Ground Water Quality Testing Program, MAY-JUNE 1997 Ground Water Quality Laboratory Test Results, MAY-JUNE 1997 Conclusions and Recommendations Acknowledgments References Exhibits Appendices Appendix A: Cave Descriptions And Maps Appendix B: Ground Water Dye-Tracing Project Report for the Buttercup Creek Cave System, including Ozark Underground Laboratory Results Synopsis 3'.,::.,.'(-.4 6 11 15... ", r'--16 16 24 26 29 31 32 41~. ~i.';"50 58 60 62 85




Ground Water Tracing Report for Buttercup Creek Property Caves: Initial Groundwater Dye Tracing Study of the Buttercup Creek Area Cedar Park, Williamson County, TexasSeptember 1997 Respectfully Submitted by: Mr. Nico M. Hauwert, Hydrogeologist&Mr. Mike Warton, Karst Specialist (AppendixBto: A STUDY OF CAVE HABITATS. POINT RECHARGE POTENTIAL. AND ASSOCIATED KARST LANDS OF THE BUTTERCUP CREEK DEVELOPMENT PROPERTIES. INCLUDING BUTTERCUP CREEK SECTION 4 AND PHASE V. CEDAR PARK. WILLIAMSON COUNTY. TEXAS.) 1.1 Introduction Field investigations in the subject site had progressed and discovery and mapping of all known karst features were completed. Initial suites of groundwater quality sampling in the Buttercup Creek area were also completed. Still, many questions and uncertainties remained regarding the nature of underground conduit connections and the hydrogeologic framework of this system. For more than a decade there' had been speculation surrounding the true nature of what many speleologists felt was a "cave system", with its inaccessible extents, its uniqueness of presence, and its diverse karst structures and cave developments, particularly when compared with adjoining upland areas of the Northern Edwards Plateau. The development of cave systems in this edge of the plateau is influenced by basic differences in the geologic strata and by faulting. Incision and erosion of the plateau margin over geologic time has also played a significant role in the development of the karstic terrain we observe in this area. As the excavation and exploration of more and more karst features led down through the Cedar Park Limestone towards small inaccessible fractures and solution drains developed along the top of the Bee Cave Marl, the frustrations of failed discoveries grew. Until this phase of the investigation, the destination of this flowing water was a major point of speculation. 1.2 Purpose The purpose of this investigation is to determine the flow route(s) for groundwater and recharge in the Buttercup Creek area and to measure the travel times. The conclusion of field investigations in the Buttercup Creek area has revealed the presence of several caves containing confined, perennial flowing streams. The discharge point or points for groundwater flowing under this area is unknown. Potential and suspect locations for discharge will likely only remain speculative in lieu of this study. This type of information is essential for efforts to protect the groundwater quality and preserve aquatic species that1


inhabit the conduits of this system. Water quality is of further importance to endangered invertebrate species habitat(s) present within the system, as the presence of groundwater in the system sustains the moisture regime needed for high quality habitat areas of the caves. Groundwater dye tracing has been used universally to directly measure groundwater flow routes and travel times. The nature of karst in the subsurface is most often complex and unpredictable, and cannot be assessed by studying surface topography alone. While surface streams may trend in a specific direction, subsurface cave streams often change directions abruptly. Groundwater dye tracing in karst terraines can help confirm suspected hydraulic connections between two points. 2.0 Geology and Karst Development of the Buttercup Creek Area The sediments forming the rock below the Buttercup Creek area were deposited about 100 million years ago in a relatively shallow, clear, marine environment (Rose, 1972). This environment was similar in many respects to the Bahamas and other Caribbean islands today, with its clear warm waters, reefs, shallow platforms, and a high diversity and abundance of aquatic organisms capable of extracting carbonate minerals from the water to incorporate into their protective casings and shells. During this period, known as the Cretaceous, the atmosphere was relatively warmer than today, consequently (due to melting of glaciers) the ocean levels were relatively higher and covered most of modern day Texas with shallow water. Most fossil remains found in this setting are those of shelled marine organisms: oysters such as Ceratostrean (previously Exogyra) texana, Trigonia sp., spiraled gastropods such as Turritella sp., tiny floating foramnifera such as Miliolid sp., and Dictyoconus walnutensis, as well as calcite-coated grains or oolites. About eight miles north of the Buttercup creek area, along the bank of the South San Gabriel River, the footprints of large meat-eating theropods and other dinosaurs, can be observed in the same layers (the contact interface between the Walnut and Glen Rose Formations) which is exposed in the Buttercup Creek project area. The rock units underlying the Buttercup Creek area include the Edwards Formation, the Comanche Peak Formation, the Walnut Formation, and the Glen Rose Formation (Figure 2.1, Rose, 1972; Moore, 1964, Rogers, 1969). There is no published description of karst features or cave development in the Buttercup Creek area prior to informal but documented cave surveys that began in 1986. Since then, specific karst studies, enhanced by the ability to physically access the subsurface to depths of 150 feet, have been instrumental in accurately delineating surface and subsurface strata. The stratigraphy observed within local caves, as discussed below, is based on interpretations by William Russell (1993) and Mike Warton (1997). The Edwards Formation, is a thick-bedded, chert-bearing limestone. Its appearance is often honeycombed due to the dissolution of fossils such as the long, tubular bivalve shell known as caprinids, and the spiraled shell Toucasia. Due to surface erosion, only a2



relatively small veneer of the Edwards Formation is present in the Buttercup Creek area. Figure 1 shows the outline of the Edwards Limestone Basal Unit outcropping in the Cedar Park Area. The Comanche Peak Formation is a hard, nodular limestone interbedded with marls. It has a thickness of about 60 feet along the eastern extent of Buttercup Creek near Highway 183 and grades into the overlying Edwards Limestone just southwest of the Williamsonrrravis county line (Rogers, 1969). Figure 2.1 Geologic Units of the Buttercup Creek area1111III111I11111111111111111111111111111111111111111111IIIIIIIIIIIII~III~I~IIIIIIIIIIIIIIIIII:liiilil:11111IlilEDWARDSICOMANCHE PEAK KEYS VALLEY MARL n ............. ",'' '",':11111111:1::lliIil~IIIIIIIIIIIIIIIIIIIIIIII.IIII.IIII .. 1111:111111.11.11111:11111:1::1::11::< 50 THICK-BEDDED, PERMEABLE LIMESTONE AND DOLOMITE 0-60iNODULAR LIMESTONE 30-40IMARL FREDERICKSBURG WALNUT WHITESTONE LENTIL CEDAR PARK LIMESTONE BEE CAVE MARL BULL CREEK LIMESTONE0-7035 4060 -70THICK-BEDDED OOLITIC LIMESTONE NODULAR LIMESTONE; PRINCIPAL CAVE-FORMING UNIT FOSSILIFEROUS MARL FOSSILIFEROUS NODULAR LIMESTONE TRINITY GLEN ROSE ALTERNATING DOLOMITIC MARLS AND RESISTANT BEDSThe Walnut Formation is 100 to 150 feet thick, composed of thick limestone and marl. It consists of the Keys Valley Marl Member, the geographically limited Whitestone Limestone Member, the Cedar Park Member, the Bee Cave Marl Member, and the Bull Creek Limestone Member. The variable thicknesses and composition of the Walnut in the Buttercup Creek area is due to the abrupt loss of the Whitestone Limestone and Keys Valley Marl to the south and facies changes in all members of the Walnut throughout the Buttercup Creek area. The Keys Valley Marl Member is 30 to 40-feet thick in much of the Buttercup Creek area, but abruptly disappears south and west of Buttercup Creek. The Keys Valley Marl contains a distinctive ammonite, Oxytropidoceras, about midway through the unit. Because of its clay-rich nature, the Keys Valley Marl is very poor for cave formation, if any cave development is present at all. The oolitic, thick-bedded Whitestone Limestone is limited to a small, five mile-wide area between Whitestone Quarry at FM 1431 and Lime Creek road, north of Buttercup creek, south of Cluck creek, and west of Highway 183 (Rogers, 1969). The Whitestone Lentil Member reaches a maximum thickness of about 70 feet within the Whitestone Quarry.4


Testudo Tube is believed to have developed within the Whitestone Limestone (Russell, 1993). The Cedar Park Member underlies the Whitestone Limestone and Keys Valley Marl, where those units are present. The Cedar Park is a nodular, fossiliferous limestone, often with abundant gastropod fossils. Russell (1993) identified the Cedar Park Limestone as the principal cave-forming member of the Buttercup Creek area. He divided it into an upper, massive-bedded unit; a middle, soluble unit that forms the upper level in many local caves; and a lower, massive-bedded unit that hosts many of the flowing stream passages in the Buttercup Creek area, perched upon the underlying Bee Cave Marl. The Bee Cave Member is a very clay-rich marl that forms a recessive slope in surface outcrops, but a smooth, white, granular exposure in cave walls. The cave stream in Marigold Cave appears to be developed near the base of the Bee Creek marl, just above the Bull Creek member. The Bull Creek Member is a nodular limestone underlying the Bee Cave Marl and overlying the Glen Rose Formation. The Bull Creek and Bee Cave Members maintain a combined thickness of 60 to 70 feet throughout the study area (Rogers, 1969). The Bull Creek and Bee Cave Members are largely identical to the Basal Nodular Member of the Edwards Group which can be traced to the San Antonio area (Small, Hanson, and Hauwert, 1996). Neither the Bee Cave nor Bull Creek Members are particularly cavernous, although some caves, such as Buttercup Creek Cave, have cut down into this horizon. Locally, the contact of the Walnut Formation and the underlying Glen Rose is marked by mudcracks, dinosaur footprints, and borings indicative of ancient hardgrounds (Kirkland, Banner, Moore, and others, 1996). The Glen Rose can be distinguished as the first dolomitic rock below the Walnut Formation (Moore, 1964). Because the upper Glen Rose consists of alternating soft marl layers and hard dolomitic beds, it is relatively resistant to dissolution and appears to act as a lower confining bed for local groundwater flow. One of the major local resurgences, "R-Bar-B" Ranch spring, appears to be positioned about 1 0 to 20 feet below the top of the Glen Rose Formation. The Walnut Formation has been generally considered a low-permeability, confining layer in the Austin area, yet it hosts a high density of caves in the Buttercup Creek area. At least three possibilities exist that might explain why the local abundance of caves within the Walnut is not observed in other areas: 1) Where the base level of local creeks cuts down to a base level below the Walnut Formation (low permeability, but easily eroded), vertical pits develop along fractures which are able to direct groundwater to lower, soluble layers within the Walnut Formation or upper Glen Rose Formation. As a result, the upper Glen Rose Formation is then the lower confining layer, which receives perched water flows from the Walnut through these drains. Where the permeable Edwards Formation5


is present and base levels of the creeks have not cut through the Walnut, the groundwater is more easily transmitted laterally and the vertical pit drains do not form. This explanation is favored since extensive caves in the San Antonio area are also developed along the Walnut Formation/Glen Rose Formation horizon at Natural Bridge caverns and Bracken Bat Cave (Abbott, 1973) where local creeks have eroded through the Walnut; or 2) The Walnut Formation experiences facies changes to the extent that it is more permeable in some areas and less permeable in others; or 3) Cave development occurs with the same density in the Walnut Formation as other areas (even where thick sections of the overlying Edwards is present) only the water-bearing conduits and pit drains are rarely encountered by widely spaced wells, and so are too deep and water-saturated to readily observe. Geologic structures in the Buttercup Creek area are faults and fractures associated with the Balcones Fault Zone. Varying interpretations of local faulting were presented by Rogers (1969) and ib unpublished maps by the Bureau of Economic Geology (1976). The fault interpretations presented in this report considered these maps, but are based on very extensive aboveand below-ground field observations by Mike Warton. The Buttercup Creek area is positioned above the western edge of the Northern Segment of the Edwards Aquifer. A regional study of groundwater flow in the Northern Segment by the Bureau of Economic Geology utilized widely-spaced water level elevations to postulate that regional flows move from southwest to northeast, toward major discharging springs along Interstate 35 (Senger, Collins, and Kreitler, 1990). The local hydrogeology of the Buttercup Creek area has not been studied to any detail, probably because, unlike other areas of the Edwards Aquifer, groundwater is not uniformly encountered throughout. Instead, groundwater flow within the Buttercup Creek area is limited to narrow conduits perched on various stratigraphic horizons above the marly upper Glen Rose Formation. 3.0 Study Methodology Two non-toxic dyes (tracers) were injected into two flowing cave streams (Marigold Cave and Whitewater Cave) under the Buttercup Creek area on Figure 4.1.0, a foldout map figure following page 10. All possible discharge points for local groundwater were monitored for the resurgence of the tracers. Access permission for "R-Bar-B" Springs was obtained from the owner, Roy Blizzard. Access permission for portions of Bull Creek were obtained from City of Austin Parks and Recreation. Prior to injecting the tracers, a copy of the proposed work plan was submitted to and approved by the U. S. Fish and Wildlife Service. (Appendix C).6


3.1 Tracer Injection In order to limit the quantity of tracer in the system at one time, and to limit interference between the traces, only two sites within the Buttercup Creek area were selected as injection locations for this study. Both sites were identified during earlier field studies and were chosen in part based on the presence of cave streams, knowledge of local geology and faulting, and their accessibility.a)Whitewater Cave. Whitewater Cave is located near the northern edge of the Buttercup Creek properties, and a few hundred feet south of Whitestone quarry. It contains an eastward flowing stream and a sump at a depth of 22 feet below the surface (see cave map in Attachment 1, of Warton Report, bound herewith). The water level in the sump appears to remain at a constant level, although the water appears to flow slowly. Whitewater cave contained very poor-quality air on the injection date. Three pounds of green sodium flourescein were injected into the flowing cave stream on August 9, 1997 at 2:45 p.m. b) Marigold Cave. Marigold Cave is an extensive cave and conduit system within the northeasterly section of known cave extents in the Buttercup Creek area (see cave map in Attachment 1, of Warton Report, bound herewith). The cave contains a perennial subsurface stream at a d~pth of 71 feet beneath the surface. The flow rate of this stream has not been measured, but has appeared to be relatively constant over a long period of observation. Its flow was visually estimated to be about 100 gallons per minute in July and August 1997. In this application, three pounds of liquid red Rhodamine WT tracer were injected directly into the cave stream at 4: 10 p.m. on August 9, 1997. The presence of poor quality air was detected in the cave on the injection date, however breathing conditions were tolerable. 3.2 Resurgence Monitoring As part of the study design plan, field reconnaissance was performed to identify major springs and locations on local spring-fed creeks in the periphery of the study area, where any dye tracers entering these tributaries could be intercepted. For this strategy, sufficient tracer had to be injected to insure some would reach the lower end of the watersheds, since the exact resurgent spring was not known at that time. Charcoal receptors were placed in the flow of creeks and springs to intercept and absorb any tracers reaching the monitored site(s). When analyzed at the laboratory, each charcoal receptor indicates the cumulative concentration of tracer flowing through the receptor. Receptors were placed in each creek for two weeks prior to the injection of the tracers, to document background, or baseline, levels of any constituents near the fluorescing wavelength of the tracers used. Fresh receptors were placed at all monitored sites prior to injection. After the tracers were injected, receptors were changed daily for the first week at sites likeliest to receive the7


tracers. Nearly all of the other monitored sites were changed on a weekly basis. The receptors were handled with dedicated latex gloves, carried in an iced cooler, and shipped to the contract laboratory for analysis. Ozark Underground Laboratory (OUL) in Protem, Missouri, performed the laboratory analysis on the samples in addition to providing their laboratory quality assurance and quality control. The monitored sites are described below (see Figure 4.1.0, a foldout map figure following page 10) : a) Cypress Creek. Cypress Creek was monitored at "R-Bar-B Ranch" spring and at a second location just upstream from Lake Travis. R-Bar-B Ranch spring is currently the headwater source for Cypress Creek, which appeared to flow at rates of about one to five cubic feet per second (500 to 2,000 gallmin) in July 1997. This spring lies 3 miles southwest of Whitewater Cave and about 3.5 miles southwest of Marigold Cave. Because of its location sub-parallel to faulting from the study area and high flow rate, R-Bar-B Ranch spring was identified as a likely discharge point for the groundwater below the Buttercup Creek project study area. According to the spring owner, R-Bar-B Ranch spring was merely a seep prior to his excavation of the spring head in 1973. Considerable flow also enters Cypress Creek between R-Bar-B Spring and Lake Travis. Therefore, Cypress Creek was also monitored just upstream of Lake Travis for tracers entering this creek at any point. b) Bull Creek Russell (1993) identified the Bull Creek watershed as a likely resurgence for the Buttercup Creek area, since the rock strata dip in this direction. Spicewood Springs is located in Bull Creek and has been estimated by the U.S. Geological Survey to flow about one to three million gallons per day. Other springs feed the upper tributaries of Bull Creek, such as "Fern Gully" Spring. Bull Creek was monitored at several sites, daily, at the junction of its two northernmost tributaries, and at the Spicewood Springs crossing just north of Oak Grove church. c) Long Hollow Creek During the field reconnaissance, Long Hollow Creek was identified as a possible resurgence point for groundwater below the Buttercup Creek area. Long Hollow Creek was monitored weekly at the bridge of FM 2769, near its mouth at Lake Travis. d) South Brushy Creek South Brushy Creek reaches elevations near the level of the cave streams about 3 miles east of the Buttercup Creek area. Regional groundwater flow within the northern segment of the Edwards Aquifer is believed to be eastward, toward springs in the major creeks near IH35 (Senger, Collins, and Kreitler, 1990). Therefore, South Brushy Creek was monitored weekly at the Palmer Lane bridge crossing. e) Lime Creek Whitewater Cave is located only a half mile east of portions of the Lime Creek watershed. However, Lime Creek is generally up gradient of the regional dip of the strata to the southeast, and perpendicular to the general northeast trending fractures and faults. Consequently, prior to injection, Lime Creek was not considered a likely site to8


receive the resurgence of tracers, but it was identified as a possible site. Lime Creek was monitored at its juncture with Fisher Hollow, near its mouth at Lake Travis.f)Hideaway Cave Hideaway Cave is located about 0.3 miles southeast of Whitewater Cave and about 0.8 miles west of Marigold Cave. It extends to a depth of about 50 feet below the ground surface. At a depth of about 38 feet below the surface, a water-filled passage known as Rhadine River is present. No receptors were monitored in this cave, although one trip was made on August 16 for a visual inspection and a tracer water sample.3.3 Quality Control and Assurance ProceduresThe quality control and assurance procedures utilized in this study will incorporate field duplicates, field control samples, laboratory blanks, laboratory spiked samples, and the testing of a portion of the sample containers for possible false-positive contaminants. To reduce the possibility for cross contamination, staff who handled the tracers were different from those collecting the charcoal receptors. Of the 64 samples submitted for analysis, 24 (37%) consisted of duplicate samples, and three consisted of control samples. The duplicate samples allowed two independent measurements for comparison. The control samples were either charcoal receptors which accompanied field personnel in the field, that were handled between sites in order to test for possible cross contamination, or material tests to insure that constituents similar to the tracers used were not present. Ozark Underground Laboratory (OUL) tested standard solutions of the tracers on a daily basis. OUL also tested one percent of the unused sample containers to assure that tracer contaminants are not present. Water grab samples were collected to verify the results of the receptors and measure the concentration of tracers at one point in time. The OUL Lab determined the level of tracer based on the fluorescent peaks measured using a Shimadzu RF-540 or RF-5000U Spectrofluorometer.3.4 Summary of Research on the Toxicology of the TracersIn designing this study, potential adverse effects to life supported by the groundwater and in caves were considered. Two tracers were selected that have been well-proven to be safe and relatively non-toxic, as discussed below. Although the groundwater system under study does not support many, if any, water users, the volume of tracers was kept low to avoid nuisance effects of visible levels of tracer. Furthermore, low levels of tracer were injected so as not to disturb or endanger aquatic life that mayor may not inhabit the system, such as the Jollyville salamander. The activities associated with this groundwater dye tracing study were believed to have no impact on any local cave invertebrates, such as the cave beetle Rhadine persephone, since the tracers were injected into the groundwater.9


The potential adverse properties of the tracers used were discussed in two reports (Smart, 1984 and Field, Wilhelm, Quinlan, and Aley, 1995) which examined existing toxicological research. Their studies indicated: a) Fluorescein and Rhodamine WT have been demonstrated to represent minimal carcinogenic and mutagenic hazards. 8ioaccumulation does not represent a hazard for aquatic ecosystems with either tracer. For the high, short-term concentrations associated with tracer injections there are no acute hazards. Smart recommended that persistent tracer concentrations not exceed 100 ug/l (about 100 parts per billion). b) For health considerations of humans and aquatic biota, an evaluation by the EPA's Office of Toxic Substances (Field and others, 1995) suggested that concentrations not exceed 1,000 to 2,000 ug/l (about 1,000 to 2,000 parts per billion) for periods of more than 24 hours. This assessment was based on a specialized chemical evaluation, utilizing structure activity relationships (SARs), developed by the U.S. Environmental Protection Agency.4.0 Results of the InvestigationThe five watersheds identified above were monitored for three weeks (21 days) following injection. Figure 4.1.0 shows the flow paths and travel times measured in this investigation. The laboratory results are presented in Appendix 8.a)Marigold CaveDye Injection Trace On Tuesday August 12, 1997, at 8:30 p.m., investigators observed a red dye discharging from R-8ar-8 Ranch spring that resembled the Rhodamine WT tracer injected at Marigold Cave on August 9. The charcoal receptors and water samples collected verified the arrival of the Rhodamine WT tracer (Figure 4.2). As verification, Rhodamine WT was also measured in downstream receptors near the mouth of Cypress Creek. A slight trace (0.644 ppb) of Rhodamine WT was measured in a water sample collected from Hideaway Cave on August 16, 1997. This slight trace could suggest a connection between Marigold Cave and Hideaway, although this conclusion cannot be drawn from this report, since the levels were low and were not verified by a second sample or visual observation.Measured Travel Time:About 3 days, between 50 hours and 74 hours.Trace length:2.9 miles.Estimated Duration of Visible levels at Resurgence:14 hoursb) Whitewater CaveDye Injection Trace. Water samples and charcoal receptors recovered from R-8ar-8 Ranch spring at 8:30 p.m. on August 15 indicated the presence of fluorescein tracer. The tracer was not observed at that time, probably due to the lack10


Figure 4.2 TRACER CONCENTRATIONS MEASURED AT R-BAR-B RANCH SPRINGS FOllOWING INJECTIONS AT MARIGOLD CAVE (August 1997)---J---.-, .. -----...I--.~---FE---+--I-=r---I-~----+--1--I----~--l--+-+----+---I---~"----n--.----t~~--~~~-j~ ..t-.j--. ... --.-+--+--_. J-: ,---__ \ __ J __1 I_ h __ . -l.]=I~]-----t-------f-.-.j--f---~--. ~-'n_' _----1-----1----21 31 4L __ I'__ __ __ ~_ --' 1---'ml~_!_ ---S~temb~_f-__ -l----I-I--l-_' _'-U~-jllt+....1 .]1 . .nl.E-. -. -. ~--------=t=rl=CCe..-Tl ...-fL.----, -+-eH=+_-~[-=:~~+-+-14---l--1312-----.--.~-+---+--1--1---+--1----1-,___ ._=CI + __ ~ ___ n .LL. _uu.-~-~l---1---1---. . ---1------1--1----1--------~~~~~!u----~_I 71 81 91 101 11____ ~--r--+---1----3.-0-~6jJ=r~Cl~=+--+-I--~~-,-i:JIL--l-f---+--I--+--J--+---1-[2-:000-I ~__ m ~t----1----1----1-1----. ----.c--+--......-------_r-l'1:00()]-'.--'--\------1------.--f----1--~--+---i-~--n_~I___I___~~-+---+~+----l----I---I----I--+--+ --1---+-------1----.---1--u __+ _______ \---__nin--1---1---\----------.---1---1-~---.--.--I--t--/----j---~--------~----.-----,.I----!-f--f----I-I --__m \-. -'n+--~=t=+tJl~~l-ffl_llJlH-+-H-~ I~~~arcealMConcentra~n M_~~sured ~~.!'rimary Receptor (ppb) ~~fncenfttralion ~!~aS~reld~n~~~Plor (ppb)____ ~ ~ __ =_--=~~ --~_---1-_--~~nded I September 1. 1997------~ ---\ -f ---f ----,----{--i--l--I-~i-H' -------. \I ---I ---l-1=1=~(.------j------wzCII.!!!s::'s ,2ME-~ ftI~ ---_ <.> --.--o ~s:: ---'0oClo'C~ ~ ~-------0--"s:: ~.Q>-----.--t-~ --l-~-t.. -.,g0:~-b ~ ~ ---~-l-j-~ .--.--l'fJ--~_ nl-.eo __ ._._. --.-----1_.-----1 . _. __.-1----I--l-+--~ --.--+---+----,_=,-~ __~~I415Time Following Injection (Days)~89 14/52d 101 14 151 16-.--4--1-r-r-r-r-j-----tt-__ _. __ ~ __ +------+--m. __._l-_nice/iabs.xls


of sunlight and possibly due to the green tracers resemblance to algae. At about 9:00 a.m. the next day, the green tracer was first observed discharging from R-Bar-B Ranch spring. On August 16, investigators entered Hideaway Cave, observed a distinct green color in the flowing stream, and collected a water sample. The charcoal receptors and water samples collected verified the arrival of the fluorescein tracer (Figure 4.3). Fluorescein was also measured in downstream receptors near the mouth of Cypress Creek. Measured Travel Time: Less than 7 days, between 125 hours and 148 hours Trace Length: 2.9 miles. Estimated Duration of Visible Levels at Resurgence: 54 hours Bull Creek, Lime Creek, Long Hollow, and South Brushy Creek were monitored for a period of 21 days following the tracer injections. No tracers were measured in these creeks. Groundwater flow from Marigold, Whitewater, and Hideaway Caves moves south along conduits developed parallel to the Cedar Park Fault. Groundwater in this system appears to generally perch either above or within the Walnut Formation and above the less permeable upper Glen Rose Formation. The R-Bar-B Ranch spring resurges about 10 to 20 feet below the top of the Glen Rose Formation. Groundwater flow is restricted to discrete subsurface channels and is not uniformly encountered across the study area. The deeply incised base level of Cypress Creek to the west seems to draw perched groundwater away from regional flow ot the Northern Segment of the Edwards Aquifer, and parallel to the direction of local fracturing. It is conceivable that flow conditions could change or bifurcate under different conditions than those of the period of study. For example, under higher aquifer flow conditions, it is possible that some groundwater could overflow in a different direction. Also, Mr. Blizzard reported that backhoe work was necessary to cause R-Bar-B Ranch springs to flow strongly. If this is true, it begs the question: Where did the other discharge flow previously? Levels of tracers at the resurgence point were maintained well below the recommended not-to-exceed levels of 1,000 to 2,000 parts per billion for any 24 hour period, based on analysis of water samples collected during the tracer resurgence. Rhodamine WT was measured at up to 336 parts per billion in grab water samples but declined to nonvisible levels within about 14 hours. Fluorescein remained at visible levels for a longer period (about 54 hours) but did not exceed 72 parts per billion in grab samples. Higher levels of tracer than were measured in grab samples may have reached the R-bar-B resurgence, although both pulses moved through the system relatively quickly. The source of groundwater feeding the cave streams of Marigold, Whitewater, Hideaway, and other caves in the Buttercup Creek area are not defined in this phase of the study. 13


Figure 4.3 TRACER CONCENTRATIONS MEASURED AT R-BAR-B RANCH SPRINGS FOLLOWING INJECTION AT WHITEWATER CAVE (August 1997)__ ~~I_n-,---+-~--+_-+~_+ ~~_._~_--1-------1---+-------I---~---1--+----+-----1------i----I--~------I--_~._m_ ~ __ u .. _-I-+---j----j----I.32=/Concentration in water (ppb)-1-----jR-Bar-B Ranch :springs Site-Whitewater_c~veTr,!c~+121 13114 15 _, __ ~~_~ ~ . _-1--1~j=~~--1-! ----t---~--. --. -~----+--1._I .-.I .=~L .~=1+ 1-i'i;_n-~~ -tL--==J ..1 __!:--E-~u __ ---t--t-1 -jJ= ------------------l-__ L-L __ ._ --------------1-E-J_J-f---Cumulative -_I_L __ ---_g:concentratlon In1Oc':% ;'-"Concentration Measured In Primary Rec~'p~~(PP~ __ -Parln~ad~;I~:~~rs :d--1con1centr[a!i]on IMeaS~!aed_in~~p~~Receptor (ppb)-1--.1-----1 :~IDuration : __ ~1_. f....ecePt",,:'ILeft In .'-=r-=t~---II_1~~;:,11_--~=_----------=__ ._. _. __ ._-...-~~---------,----[Monitoring Ended=J -_~~ --=--on SePt~mber1. 1997=-~. __I-~I--.----+--t-----I-.------f----!--t--~--~--1--+-+--'------~-I------~_--.---_-.--..... -----.----+~--I--.-__---<------.--------t----1-------j--t---j---+-----.---~-~I------t_____.--~---L.J---c--r---.---.--+-----+-f----+-=-J-=1i=~--l~~eEff-=~:~~_.----3.0~O}--L--I--+-----I-------I---- . --I---.~-.-~ z(QIII--.~ ----~ o '";; Ne!a;-_. _. ~ -5liOOQL~--8 ---I--E-~c:Q>:=-.~ ~Gl C-_._-g--~-..-1-LL ~(U --.-..-__ z 11 21 31_4SUTime Following Injection (Days)w-r-r~-171181~1~~~-~-~ip~w.s12712812913013113~~-~3~1 35iFLbs.xls


However, three insurgent sources are hypothesized to recharge the local cave systems: (1) infiltration and storage in to the porous Cedar Park member along the Buttercup Creek drainage; (2) collective recharge from the upland karst features; and (3) perennial flows into the groundwater system from spring sources in the Cluck Creek drainage near Highway 183. 15


5.0 References Cited Abbott, Patrick L., 1973, The Edwards Limestone in the Balcones Fault Zone. South Central Texas: Unpublished Univ. of Texas Ph. D. Dissertation. 122 pp. Kirkland, Brenda L, Jay L. Banner, Clyde H. Moore, Cory Hoffman, Ben Pursell, and Rosario Vasquez, 1996, Cretaceous Cyclic Platform Carbonates of Central Texas: South Central Section Meeting of the Geological Society of America 1996 Field Trip no. 3. 36 pp. Moore, Clyde H., 1964, Stratigraphy of the Fredricksburg Division, South-Central Texas: Univ. of Texas Bureau of Economic Geology Report of Investigations No. 52. 48 pp. Rogers, Margret A, 1969, Stratigraphy and Structure of the Fredricksburg Division (Lower Cretaceous). Northeast Quarter Lake Travis Quadrangle. Travis and Williamson Counties. Texas. Unpublished Univ. of Texas M.A Thesis. 49 pp. Rose, Peter, 1972, Edwards Group, Surface and Subsurface, Central Texas: Univ. of Texas Bureau of Economic Geology Report of Investigations NO.7 4. 198 pp. Russell, William H., 1993, The Buttercup Creek Karst: Report prepared for the University Speleological Society. 76 pp. Senger, Rainer K., Edward W. Collins, and Charles W. Kreitler, 1990, Hydrogeology of the Northern Segment of the Edwards Aquifer, Austin Region: Univ. of Texas Bureau of Economic Geology Report of Investigations No. 192. 58 pp. Small, Ted A, John A Hanson, and Nico M. Hauwert, 1996 .. Geologic Framework and Hydrogeologic Characteristics of the Edwards Aquifer Outcrop (Barton Springs Segment), NE Hays and SW Travis Counties, Texas: U.S. Geological Survey Water Resource Investigation 96-4306.16