<%BANNER%>

Chemical and Isotopic Evidence for Exchange of Water between Conduit and Matrix in a Karst Aquifer: An Example from the ...


PAGE 1

CHEMICAL AND ISOTOPIC EVIDENCE FOR EXCHANGE OF WATER BETWEEN CONDUIT AND MATRIX IN A KARST AQUIFER: AN EXAMPLE FROM THE SANTA FE RIVER SINK/RISE SYSTEM By BROOKE ELIZABETH SPROUSE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Brooke E. Sprouse

PAGE 3

iii ACKNOWLEDGMENTS I would like to thank my adviso r Jon Martin for all of his help and insight with this project. I would especially like to thank La uren Smith and Jennifer Martin for all of their support and willingness to battle the ticks, mos quitoes, and gators to help me collect the water samples. Thanks for helping me keep my sanity. Finally I would like to thank my parents for all of thei r love and support thr oughout all my endeavors both here at UF and throughout my life.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT......................................................................................................................v ii CHAPTER 1 INTRODUCTION........................................................................................................1 Background Studies......................................................................................................2 Study Area/Geologic Background................................................................................7 Study Area.............................................................................................................7 Temperature and Climate......................................................................................7 Physiography.........................................................................................................9 Stratigraphy/Hydrostratigraphy.............................................................................9 Previous Studies of the Santa Fe River.......................................................................13 2 METHODS.................................................................................................................16 Water Sampling..........................................................................................................16 Surface Water......................................................................................................16 Ground Water......................................................................................................18 Analyses......................................................................................................................1 9 Field Measurements.............................................................................................19 Chemical and Isotopic Measurements.................................................................20 Computer Modeling....................................................................................................21 3 RESULTS...................................................................................................................22 Stage, Precipitation and Discharge.............................................................................22 Drought Stage.............................................................................................................24 Field measurements......................................................................................24 Chemical and isotopic composition.............................................................24 Intermediate Stages.....................................................................................................25 Surface Water......................................................................................................25 Field measurements......................................................................................25

PAGE 5

v Chemical and isotopic compositions............................................................26 Ground Water......................................................................................................28 Field measurements......................................................................................28 Chemical and isotopic compositions............................................................28 Flood Stage.................................................................................................................29 Surface Water......................................................................................................29 Field measurements......................................................................................29 Chemical and isotopic compositions............................................................30 Ground Water......................................................................................................30 Field measurements......................................................................................30 Chemical and isotopic compositions............................................................31 4 DISCUSSION.............................................................................................................42 Variation in Chemical and Isotopic Composition......................................................42 Chemical Variations in Karst Windows and Wells.............................................43 Isotopic Variations in the Karst Windows and Wells.........................................48 Sr Mixing Model.........................................................................................................50 Saturation and Mixing Calculations...........................................................................54 Nutrient Loading.........................................................................................................56 5 CONCLUSIONS........................................................................................................59 APPENDIX A DAILY PRECIPITATION AND STAGE RECORDS..............................................61 B WATER CHEMISTRY DATA..................................................................................66 BIOGRAPHICAL SKETCH.............................................................................................75

PAGE 6

vi LIST OF TABLES Table page 1-1 Stratigraphic and hydrostratigraphic units of the Santa Fe River Basin................13 2-1 Sample collection dates and the ri ver stage at the time of collection....................17 2-2 Total depth and location of the wells.....................................................................18 2-3 River stage levels at the time of ground water sample collections........................19 3-1 Water levels and discharge m easurements at the Sink and Rise...........................24 4-1 The HCO3 -/Ca2+ ratios in the karst windows and the wells...................................45 4-2 Percentage of surface water in the karst windows.................................................54 4-3 Percentage of surf ace water in the wells................................................................55

PAGE 7

vii LIST OF FIGURES Figure page 1-1 Map of the Santa Fe River Basin...........................................................................10 1-2 Map of the study area, including all of the karst windows and wells....................11 3-1 Water elevation of the Santa Fe River elevation collected at O’Leno State Park.23 3-2 Precipitation amounts collected at O’ Leno State Park during the study period....23 3-3 a&b. Ca2+ concentrations.......................................................................................32 3-4 a&b. Sr2+ concentrations .......................................................................................33 3-5 a&b.Mg2+ concentrations ......................................................................................34 3.6 a&b. Na+ concentrations.......................................................................................35 3-7 a&b. Clconcentrations..........................................................................................36 3.8 a&b. SO4 2concentrations......................................................................................37 3-9 a&b. Alkalinity concentrations..............................................................................38 3-10 a&b.Saturation Index Ratios..................................................................................39 3-11 a&b. 18O values ...................................................................................................40 3-12 a&b. 87Sr/86Sr ratios ..............................................................................................41 4-1 Graph of the Na/Cl ratios of th e karst windows for each sample period...............46 4-2. 87Sr/86Sr vs. Sr and 87Sr/86Sr vs.1/Sr models for all karst windows and wells.......53

PAGE 8

vii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CHEMICAL AND ISOTOPIC EVIDENCE FOR EXCHANGE OF WATER BETWEEN CONDUIT AND MATRIX IN A KARST AQUIFER: AN EXAMPLE FROM THE SANTA FE RIVER SINK/RISE SYSTEM By Brooke Sprouse August 2004 Chair: Jonathan B. Martin Major Department: Geological Sciences Karst aquifers are characterized by condu it porosity, fractures, and intergranular matrix porosity. Although classical karst aquifers have low matrix porosity, where permeability is high, ground water may exchange between conduits and matrix porosity. Because conduits can be fille d with surface water entering th rough a sink or swallet this exchange may lead to contamination of th e aquifer from the surface. An important process of karst hydrogeology is the exte nt that surface and ground water exchange, particularly by mixing of matrix and conduit water, and the pro cesses that can control this mixing. In North-central Florida the Upper Santa Fe River enters a 36m deep sinkhole at the River Sink. Less than 8 km south of the si nk the headwaters of the Lower Santa Fe River occurs at the River Rise. Between th e Sink and the Rise, numerous karst windows provide an opportunity to examine the move ment of water between the conduit and the

PAGE 9

viii matrix. In this study, the Sink, Rise, 12 ka rst windows, and 6 monitoring wells were sampled four times between May 2002 a nd May 2003 at drought, baseflow, and flood conditions. At drought conditions, high solute co ncentrations in water collected from the karst windows reflect evaporat ive loss of water and suggest that conduits contain water that originates from the matrix. Solute c oncentrations indicate 45-55% of the water originates from the matrix on the basis of two end-member mixing models. At flood conditions, concentrations reflect dilution and movement of surface water through the conduit and mixing models indicate the ka rst windows contain almost 100% surface water. At baseflow conditions the fracti on of surface water depends on the antecedent conditions, specifically if baseflow is preceded by drought or flood conditions. Following drought conditions, low solute con centrations suggest th at increasing water flow in the conduit and low hydraulic head of the matrix relative to the conduit allows loss of water to the matrix. At these condi tions mixing models suggest that surface water makes up approximately 90-95% of the wate r in the karst windows. Following flood conditions, increased concentrations in the karst windows suggest wa ter flows from the matrix to the conduit. Mixing models suggest the karst windows contain 75-80% surface water. Solute concentrations decrease by 50% in one of the wells in the study area during flooding and may be the result of flow along high permeability zones, which appears to be affected by loss of water from the conduit. All other wells have solute concentrations that change little during the study period and a ppear to be unaffected by surface water. Based on these data, the water chemistry in the karst windows appears to be affected by river levels and weather conditi ons to a greater extent than the water in the matrix.

PAGE 10

1 CHAPTER 1 INTRODUCTION Karst aquifers provide important water re sources worldwide by providing water to more than 25% of the world’s population. In the United States, karst terrains make up 20% of the land surface and karst aquifers supply 40% of the ground water used for drinking water (Karst Waters In stitute, 2001). These aquifers are particularly vulnerable to contamination because they have a wide range of porosity. Unlike darcian-type aquifers, heterogeneous porosity develops in karst aquifers through dissolution of matrix rocks, resulting in large so lution channels that typically have turbulent flow. Karst aquifers are often considered a trip le porosity system composed of fractures, conduits, and intergranu lar matrix porosity (Worthington, 1999). These three types of porosity may develop to different degrees de pending on karst history and age. Conduits consist of dissolution features having large aper tures, typically 10mm to tens of meters. Fracture porosity and the interg ranular porosity typically ha ve apertures of less than 10mm (White, 1999). Intergranular and fracture porosity can be difficult to distinguish, therefore, matrix will refer here to both fr actures and intergranular porosity. In regions with low matrix permeability, conduits can transm it most of the potable water. In regions with elevated matrix porosity and permeabilit y, large volumes of water can flow through the intergranular porosity of the matrix rock s (Smart and Ford, 1986) and in these areas most water may be stored in the matrix porosity. Water can flow rapidly from the surface into sinkholes and then into conduits, thereby providing allogenic recharge (White, 1999). Mixi ng of water in conduits and

PAGE 11

2 matrix can lead to rapid contamination of th e water in the intergranular porosity, which may be a major source of potable water. U nderstanding the extent of exchange between the two different flow systems within karst aq uifers is thus critical to determine the sources of spring waters and the potential for their contamination. While there have been numerous studie s focusing on karst aquifers, questions remain about the rates and extent of inte ractions between surface water and ground water and the mixing of water between the conduits and the matrix. The purpose of this study is to examine this interac tion and to develop a better understanding of the hydrologic processes occurring in karst aq uifers characterized by high ma trix porosity. In addition, this study will develop techniques that incor porate natural chemical compositions, such as the use of 87Sr/86Sr, as natural tracers in karst aquifers. This study attempts to answer several general questions: 1. How does variation in surface water levels influence mixing of water in conduits and matrix? What is the extent of mixing at various river water levels? 2. Can mixing models be developed for the extent of conduit/matrix exchange at various conditions? 3. How is the intergranular porosity wate r chemistry altered through water-rock interactions as water moves th rough the conduit and the matrix? 4. How does the distance from the conduit affect matrix water chemistry? 5. What role do antecedent conditions pl ay in both conduit and matrix water chemistry? Background Studies Early studies using a variety of methods including potentiomet ric maps, dye-tracer studies, hydrographs, chemographs, and wate r chemistry (e.g., Shuster and White, 1971; Atkinson 1977; Thrailkill, 1985; Vervier, 1990; Thrailkill et al. 1991; Padilla, 1994; and Halihan et al 1998) found that springs in karst aquifers can be divided into those that are

PAGE 12

3 sourced from conduits in which water flows through large underground passages and those that are sourced by diffuse flow from intergranular poros ity (Shuster and White, 1971 and Atkinson, 1977). Later work (Dea n, 1999; Martin and Dean, 2001) showed that a gradation exists between the two types of springs. In aquifers, potentiometric maps provide information about the ground water flow pattern. The presence of conduits in karst aq uifers may alter the expected flow pattern determined by these maps (Thrailkill, 1985) a nd, as a result, dye-t racer tests are often used to determine ground water flow pattern s in karst aquifers (e.g. Thrailkill, 1985; Thrailkill et al. 1991). Fluorescein and RhodamineWT ar e often used as the tracer and are typically injected directly into a sink or swallet. These dye tracer experiments can define a ground water basin, conduit geomet ry, resurgence points from sinks and swallets, and response to storm even ts (Thrailkill, 1985 and Thrailkill et al. 1991; Hisert, 1994). Resurgence of dyes provides informa tion about the source of water at a given spring as well as linear travel time through the system. These studies cannot determine absolute flow paths, however, but instead pr ovide a straight-line connection between the injection point and resurgence point. Storm hydrographs and the chemistry of spring discharge provide additional information about flow in karst aquifers. Storm hydrographs represent the response of the spring to the influx of storm water and subsequent recession as springs return to baseflow conditions. Hydrographs can be compared with changes in spring water chemistry (chemographs) to distinguish betw een conduit springs and diffuse flow springs (Shuster and White, 1973). Chemical studies of spring discharge have also been used to study surface water/ground water exchange (Redwine and Howell, 2001). Precipitation

PAGE 13

4 and ground water typically have different chemical concentra tions as a result of waterrock reactions of the ground water. Precipitation typically has low concentrations of solutes and salts; however, ground water may ha ve high solute concentrations due to water-rock interactions in the matrix. In some karst aquifers, however, the geometry and geology of the region make chemical separation of spring types difficult. For example, in the Inner Bluegrass Karst Region in Kentucky both conduit and diffuse springs are recharge d through limestone, but the conduit springs are associated with ka rst features, deep integrated conduit systems and large catchment areas while the diffuse springs are associated with shallow flow paths and small catchment areas (Scanlon and Thrailkill, 1987). The chemical compositions of both spring types are simila r during low flow due to percolation and chemical reactions near the recharge zone. During high flow events the compositions are similar due to surface runoff through sinkholes recharging the conduit springs and short flow distances to the diffuse springs. S canlon and Thrailkill (1987) concluded that the chemical composition of spring water is a ffected not only by c onduit size but also by recharge type and flow path lengths in regions where the physiographical properties associated with springs do not differ between conduit-fed and diffuse-fed. In recent years, isotopic compositions ha ve been used to distinguish between surface water and ground water in karst aquifers. The primary isotope systems used are 18O and D (e.g., Frederickson and Criss, 1999; Greene, 1997; Lakey and Krothe, 1996). The isotopes of oxygen and hydrogen have been used to trace lateral movements across a ground water basin (Greene, 1997) as well as spring discharge (Lakey and Krothe, 1996). When 18O and D values are compared to the established meteoric water

PAGE 14

5 line, sources of waters to springs and th e influx of water following a storm can be determined (Lakey and Krothe, 1996). Several studies have used 87Sr/86Sr isotopic ratios to st udy ground water in karst aquifers (Katz and Bullen, 1996; Katz et al. 1997; Cao et al. 1999; and Woods et al. 2000). Strontium isotope ratios have been us ed in studies of surface/ground water mixing in carbonate systems (Katz and Bullen, 1996; Katz et al ., 1998; Cao et al. 1999; Woods et al. 2000; Dogramaci and Herczeg, 2002) and mi xing with other ground water sources in non-carbonate systems (McNutt et al ., 1990; Johnson et al. 2000). Strontium can be used for these studies because the 87Sr/86Sr ratio of dissolved Sr depends on the amount of dissolution of the solid aquifer material and its isotopic ratio. The only radiogenic Sr isotope is 87Sr, which is the produc t of the decay of 87Rb. Depending on their age, rocks containing abundant Rb-bearing minerals will have more 87Sr and a higher 87Sr/86Sr ratio than rocks with few Rb-bearing minerals, such as carbonates (Faure, 1986). The 87Sr/86Sr ratios can be coupled with Sr2+ concentrations to provide a two endmember model. Woods et al. (2000) used such a model to provide evidence of mixing between water in the Upper Castle Hayne limestone aquifer with water in the younger Surficial Aquifer in the vicinity of the North Carolina coast. In Florida, the ratios have been used to discriminate ground water sources based on 87Sr/86Sr ratio-age plotting (Cao et al. 1999) and to examine surface water/ground water interactions (Katz and Bullen, 1996; Katz et al. 1997&1998). In the Leon Sinks Geological Area near Tallahassee, Florida groundwater samples from we lls and the deepest sinkholes have 87Sr/86Sr ratios that plot along the Sr seawater age curve, suggesting the waters originate from the Oligocene-age limestones of the Upper Floridan Aquifer (Cao et al. 1999). Samples

PAGE 15

6 from shallow sinkholes have ratios above the modern seawater ratio of 0.70907, suggesting the waters originate from the younge r clastics of the Water Table aquifer (Cao et al. 1999). Katz and Bullen (1996) and Katz et al. (1997) examined lake water and ground water interactions in the mantled karst region of Nort h Florida using the 87Sr/86Sr isotope ratios, and found ground wa ter was associated with less radiogenic Sr than the water in the lakes. Ground water with high 87Sr/86Sr ratios was interpreted to either have a source from lake water or from ra pid movement through the aquifer limiting equilibration with the limestone (Katz et al. 1997). While many methods have been employed in the study of karst aquifer systems, a number of these have focused on the disc harge of a spring or springs to identify characteristics of the aquifer. In many of these studies, flow paths between surface water sources (i.e. a swallet) and the spring are unknown. In additio n, there have been studies of surface water/ground water exchange but fe w studies focusing on the interaction of water in a porous matrix with water in c onduits. A lack of abunda nt features along the flow path may be the reason for few detailed studies of conduit/matrix exchange. This study utilizes the major ion concentrations (Ca2+, Mg2+, Sr2+, Na+, Cl-, and SO4 2-) and 87Sr/86Sr isotopic ratios of wate r at known recharge and discharge points and from numerous karst windows and wells along the flow path connecting the recharge and discharge points. Consequently, the focus of this project is to examine the interaction of water in the matrix and water in the c onduit where numerous karst windows provide access to the conduit water.

PAGE 16

7 Study Area/Geologic Background Study Area The study area is along the Santa Fe River, in north-central Florida the second longest river flowing across nor th-central Florida and a majo r tributary to the Suwannee River (Figure 1-1). The ri ver basin covers ~3500 km2 and occurs in three physiographic regions, the Northern Highlands to the east and the Central Highlands and Gulf Coast Lowlands to the west (Hunn and Slack, 1983, Me yer, 1962). The river flows west from Santa Fe Lake for approximately 50 km then it disappears into a 36 m deep sinkhole known as the River Sink. Some of the water that enters the Sink eventually reemerges approximately 8 km downstream at the River Rise (Hisert, 1994); however, at times some fraction of this water appears to be lost to the matrix as the wa ter travels through the conduit while at other times, water is gained from the matrix (Hisert, 1994; Dean, 1999; Martin, 2003; Smith et al. 2002; Smith, 2004). The water discha rging at the River Rise is probably a mixture of water flow ing into the Sink and water lost from the matrix to the conduits (Hisert, 1994). Between the Sink a nd the Rise there are a number of karst windows present (Figure 1-2), and their wate r compositions reflect the physical and chemical processes that occur between the conduit and the matrix (e.g. Dean, 1999; Martin and Dean, 1999; Martin, 2003; Smit h, 2002; Smith, 2004). This study extends the previous work through sampling and analysis of previously unsampled karst windows, as well as six wells recently drilled and completed at depths of the conduits. Temperature and Climate North-central Florida is cl assified as a humid sub-tr opical climate (Meyer, 1962), with an average annual daytime temperature of 21 C. During January and December, the coldest winter months, the average daytime temperature is 14 C, and during August, the

PAGE 17

8 warmest month, the daytime temperature averages 27 C. The average annual rainfall in the Santa Fe River basin is 140 cm (Hunn and Slack, 1983). Most rainfall occurs between June and September, and the least am ount of rainfall occurs during the winter months (Meyer, 1962). In the summer, rain fall commonly comes from short afternoon thunderstorms that are the result of warm ai r rising over the land and drawing cool moist air inland from the Gulf of Mexico and the Atlantic Ocean. In the winter rainfall commonly comes from extra-tropical frontal syst ems. These storms t ypically last longer than the summer storms but do not occur as frequently. The study area recently experienced a threeyear drought that ended in the late fall and winter of 2002/2003. During the drought, the average yearly rainfall from January 1, 1999 to December 31, 2001 was 93 cm/year at O’ leno State Park, which is approximately 33% less than the annual average (Suwa nnee River Water Management District archives). In contrast, in the years preceding the drought (1996-1998) the average rainfall at O’Leno was 147cm/yr. The river le vel at O’leno State Park for 1999 to the end of 2001 ranged between 9.6 and 10.3 meters above sea level (masl). Prior to that, the average river level at O’Leno State Pa rk from 1994 through 1998 was 10.86 masl (SRWMD). For most of 2002, the river level was approximately 9.6 masl until midAugust when the river rose to 10 masl reachi ng a maximum at the end of December at 11 masl. In the first half of 2002 only 35 cm of precipitation occurred in the study area; however, between July and December 2002 an ad ditional 70 cm of rainfall fell, resulting in 105 cm of rain for the year. During th e fall and winter of 2002-2003 a moderate El Nino Southern Oscillation (ENSO) period caused greater than normal precipitation levels

PAGE 18

9 throughout Florida and the Southeast with 60 cm of rain between November 1, 2002 and the end of March 2003. Physiography The Santa Fe River Basin is located along the Western Valley and the Northern Highlands physiographic provinc es (White, 1970). The Western Valley is one of several large lowland areas within the Central Highla nds. Several gaps within the valley allow drainage to the Gulf Coastal Lowlands, with the Santa Fe River flowing through the High Springs Gap to a confluence with the Suwannee River (White, 1970). The Northern Highlands represent the eastern portion of the river basin and have elevations greater than 30 masl. The eastern boundary of the basin is Trail Ridge, a north-south trending ridge that extends through central Florida. In the center of the basin lies an escarpment called the Cody Scarp that marks the boundary betw een the Western Valley and the Northern Highlands. This scarp is th e erosional edge of the Miocene Hawthorn Group and marks the retreating edge of a formerly high plai n that sloped northward. Similar to most streams crossing the Cody Scarp, the Santa Fe River flows into a sinkhole and reemerges at a first magnitude spring approxima tely 5km to the south (Fig. 1-2.). Stratigraphy/Hydrostratigraphy The lithology of Florida is composed of carbonate rocks that are pre-Miocene in age and mixed siliciclastic and carbonate rocks that are Miocen e and younger (Table 11), with few outcrops. The oldest exposed units in the study area are Eocene carbonates of the Ocala Limestone. The lower Ocala is composed of grainst ones to packstones and in some regions may be partially to completely dolomitized (Scott, 1992). The upper portion of the unit is muddy, granular limestone s (packstones to wackestones) and can be soft and friable (Scott, 1992). Above the Eo cene rocks of the Floridan Aquifer are the

PAGE 19

10 O’Leno State Park Figure 1-1. Map of the Santa Fe River Basin with the arrow showing O’Leno State Park and the study area, which is shown in detail in Figure 1-2. Modified from Hunn and Slack (1983).

PAGE 20

11 Figure 1-2. Map of the study area, including all of the karst windows and wells. Modified from Ginn (2002). 1 7 2 3 4 6 W e ll s Hawg Sink Vinzants Landing Hornsby Spring Treehouse Spring

PAGE 21

12 Miocene-age rocks of the Hawthorn Group. The Hawthorn Group contains primarily interbedded sands, clayey sands, sandy clays, clays, and carbonates (Grozos et al 1992) and is composed of the Hawthorn Formation and the Alachua Formation (Hunn and Slack, 1983). The upper most units in the st udy area are Pliocene-Ple istocene sediments that range from sands, sandy clays and carbona tes and in North Florida include the Nashua Fm., Cypresshead Fm., Miccosuk ee Fm., and Undifferentiated PleistoceneHolocene sediments. The undifferentiated se diments include marine sediments, eolian sand dunes, fluvial deposits, fresh water carbona tes, and sediment mixtures that cover most of Florida (Scott, 1992). Ground water is found in three aquifers that include, from upper to lowermost, the Surficial, Intermediate, and Floridan aquife rs. The Surficial Aquifer is a water table aquifer within Pleistocene-Holocene sands (Hunn and Slack, 1983). Throughout most of the Santa Fe River basin, the water tabl e is approximately 3 m below land surface; however, in the eastern portion of the basin the water table may be up to 9 m below land surface. Within the study area the Surficial Aquifer is approximately 5 m in thickness. Below the Surficial aquifer the Hawthorn Group carbonates contain the Intermediate Aquifer. Where present, the Hawthorn Fm acts as a confining unit to the underlying Floridan Aquifer, but where missing, the Flor idan Aquifer is unconfined and is recharged directly from the surface. The Floridan A quifer is the primary source of drinking water for most of northern Florida. The aquifer consists of porous limestone that can be divided primarily into five units that are, from oldest to youngest Lake City Limestone, Avon Park Limestone, and Ocala Limestone (all Eocene age), Suwannee Limestone

PAGE 22

13 (Oligocene), and in some places lower Mio cene limestones. The Eocene Ocala limestone is the major unit present in the study area. Table 1-1. Stratigraphic and hydrostratigraphic units of the Santa Fe River Basin. Age Stratigraphic Unit Hydrostratigraphic Unit Lithology Thickness (m) Holocene Pleistocene Pliocene Undifferentiated sediments Surficial Aquifer Fine sands and gravel 0-25 Pliocene to Miocene Alachua Formation Miocene Hawthorn Formation Intermediate Aquifer/Confining Bed Interbedded sands and clays Carbonates 0-45 Oligocene Suwannee Limestone Eocene Ocala, Avon Park and Oldsmar Limestones Floridan Aquifer Porous limestone and dolomite 325-425 Paleocene Cedar Keys Formation Sub-Floridan confining bed Limestone with some clays and evaporites ? Adapted from Meyer (1962), Hunn and Slack (1983), and Dean (1999). Previous Studies of the Santa Fe River In one of the earliest studies of the Sa nta Fe River basin in O’leno State Park Skirvin (1962) attempted to determine the underground flow path of the river after it sinks into the subsurface at the River Sink. Based upon the tannic color of water emerging at the Rise, he sugge sted that most water enteri ng the Sink reemerges at the Rise. During this study, the river was te mporarily dammed near the Sink, but water continued to flow from the Rise, suggesting groundwater enters the system between the Sink and Rise.

PAGE 23

14 Hunn and Slack (1983) published a report on the hydraulic properties and water quality of the Santa Fe River basin, including the area within O’leno State Park. In the karstic region of the basin, the Floridan Aqui fer is the primary groundwater source and is recharged directly through rainfall. They found that where the Floridan Aquifer is unconfined, it supplies groundwater to surface dr ainage features within the basin through large springs. In confined regions of the basin, however, discharge to streams is small and the Santa Fe River receives discharge only downstream from Worthington Springs. These sources of water are reflected in wa ter quality in the co nfined and unconfined regions. Streams in the unconfined wester n region have greater concentrations of calcium, magnesium and bicarbonate, and are le ss tannic than streams in the confined region to the east. The higher chemical conc entrations in the west ern region appear to reflect a water source from the Floridan Aquifer. More comprehensive studies of the Santa Fe River between the Sink and the Rise have been conducted within the past ten ye ars to better understand the underground flow of the river, the interacti on of surface water and groundwater and the exchange of water between conduits and matrix porosity (Hisert, 1994; Dean, 1999; Martin and Dean, 1999; Smith et al 2002; Martin 2003; Smith, 2004). Thes e studies have used various natural tracers including 222Rn, 18O values, major element chemistry (i.e. Ca2+, Mg2+ Cl-), and temperature and the injected tracer SF6 (Hisert, 1994). Hisert (1994) was able to determine a connection between the River Sink and Sweetwater Lake, using an injection of SF6, but required a second injection to connect Sweetwater Lake to the River Rise (Fig. 1-2). A connecti on with a single injection between the Sink and the Rise has never been obtained. Cave divers have recently

PAGE 24

15 verified portions of this c onduit system within the park (O ld Bellamy Cave Exploration, 2001). In addition to SF6 Hisert (1994) analyzed the chemical composition (Ca2+, Mg2+, Na+, K+, Al3+, 18O, and 222Rn) at 23 locations between the Sink and Rise. Through the various experiments, Hisert (1994) determin ed that some mixing between surface water from the conduit and ground water in the ma trix occurs. The fraction of ground water relative to surface water in the karst windows was not determined. Dean (1999) measured the chemical composition of the water at the Sink, Sweetwater Lake, the Rise at various river levels and chemical compositions of two wells, one near the Sink and th e other near the Rise. These data suggest that surface water and ground water mixing varied at diffe rent flow conditions with more surface water flowing into the matrix porosity during high flow periods. Dean (1999) suggested that high spatial and temporal resolution of surface and ground water sampling are needed to better understand the extent of mixing and the locations where mixing occurs in the system.

PAGE 25

16 CHAPTER 2 METHODS Measurements of physical hydrologic parameters and water chemistry included daily river stage, daily precipi tation, and the chemical composition of 19 water samples. The staff at O’leno State Park collected the river stage data upstream from the Sink and precipitation data at the park entrance. These data were obtained through the Suwannee River Water Management District. Analysis of the water samples included field and laboratory measurements. Temperature, pH, specific conductance, dissolved oxygen (DO), and turbidity were each measured in th e field. Major ions, dissolved species and isotopes were measured in the laboratory, including Na+, K+, Ca2+, Mg2+, Cl-, SO4 2-, NO2 2-, NO3 2-, NH3, alkalinity, silica, Sr2+ concentration, and 87Sr/86Sr ratios and 18O values. Water Sampling Surface Water Water samples from the Sink, Rise, nine karst windows, a swallet upstream from the Sink, and two springs south of the Rise were collected four times between May 2002 and May 2003 at river stage levels vary ing between drought, high and intermediate discharge conditions (Table 2-1). The samp les were collected over time periods ranging from several days to a few weeks with samples collected sequentially from upstream to downstream. The length of time for each samp le period was, in some cases, dependent on weather conditions but an atte mpt was made to collect the same water parcel entering

PAGE 26

17 the Sink and traveling through the conduit base d on previously determined travel times between the Sink and Rise (Dean, 1999). Samples were collected from a depth of approximately 0.5 meter and near the resurgence point of the spring, when possibl e. A Geotech Geopump 2 peristaltic pump was attached to PVC tubing to collect the water samples. Prior to sample collection, the tubing was purged with at least 2L of water from the karst window that was to be sampled. A free-flow cell was used to collect the purged water as well as to take field measurements. During each sample period tw o field duplicate and 1-2 field instrument blanks were collected. At each location, five se parate containers were used to collect the samples. For the isotopic analyses, samples we re collected in 30ml glass Qorpak bottles. All other samples were collect ed in polyethylene bottles ra nging in size from 125 ml to 1L. Samples to be analyzed for alkalinity, Cl-, orthophosphate, NO2 -, and SO4 2had no preservatives added, but samples for silica and soluble reactive phosphorous were later filtered in the laboratory. The NOx, NH3, total phosphorous and to tal Kjeldahl nitrogen samples were acidified with sulfuric acid to a pH level of <2 and the sample for the total metals was acidified with nitric acid to a pH level of <2. All samples were kept in a cooler on wet ice while in the field and were then refrigerated at 4 C at the laboratory. Table 2-1. Sample collection dates and th e river stage at the time of collection. Sample Period Date River Stage (masl) 1 May 8, 2002June 5, 2002 9.8-9.6 2 January 15 & 16, 2003 10.7410.71 3 Feb. 24 – March 5, 2003 11.63~12 4 April 28May 1, 2003 10.49

PAGE 27

18 Ground Water In January 2003, six 2-inch monitoring we lls were drilled throughout O’leno State Park. Three of the wells were installed along the regional ground water flow path and three were installed near the location of the conduit between Sweet water Lake and the Rise, which has been mapped through cave divi ng exploration (Figure 1-2). The wells were drilled to approximately 30 m total dept hs and screened over a 6 m depth (Table 22). Table 2-2. Total depth and location of the wells used in this projec t. Distance is unknown for wells 2 and 7. Well Total depth of well (m) Screened interval (m) Direction to conduit Distance from Conduit (m) 1 23 23-17 NE 475 2 30 30-24 SW 3 28 28-22 NE 30 4 30 30-23 W 115 6 31 31-25 W 85 7 30 30-24 SW Between January and May 2003, three sets of water samples were collected from the monitoring wells. The January set include d only wells 1, 2 and 7 while the other two sets included all six wells. During this time period, the river stage varied between intermediate river level and high river level (Table 2-3). The first and third sample periods took place during intermediate stag e levels and the second sampling period took place during the March high river level (Table 2-1). The wells were sampled in conjunction with the surface water sample se ts 2,3, and 4 (Tables 2-1&2-3) and were generally collected several days following the surface water collections.

PAGE 28

19 The ground water samples were collected using a Redi-Flo w2 variable speed submersible pump. Before sample collection the depth to the water level was measured to determine the depth at which to place the pump, which was set approximately 1m below the water table. Sampling followed purging of at least three well volumes. The field measurements of temperature, pH, specific conductance, DO, and turbidity were made while the wells were purged. Purging was considered complete when all of the parameters stabilized such that three consecutive measurements of temperature were within 0.2 C, pH were within 0.2 standard units, specific conductance were within 5.0% of the reading, DO was no greater than 20% of saturation at the field measured temperature, and turbidity was no gr eater than 20 NTUs. When stabilization occurred, five sets of sa mples were collected for each well in the same manner as previously described for surface water. Table 2-3. River stage levels at the time of ground water sample collections. Sample Period Date River Stage 2 February 5, 2003 10.43 masl 3 March 19, 2003 13 masl 4 April 28, 2003 10.49 masl Analyses Field Measurements Measurements of the field parameters for both surface water and ground water were made with the same equipment. Measuremen t of pH was made with an Orion portable pH meter Model #250A calibrated at the st art of each sampling period using 7.0 and 9.0 pH buffers. Specific conductance and temper ature were measured with an ATI Orion portable conductivity meter M odel #130. Dissolved oxygen and turbidity were measured

PAGE 29

20 with YSI Model 55 handheld dissolved oxyge n and temperature system and LaMotte 2020 turbidimeter, respectively. Chemical and Isotopic Measurements For the first sample peri od, the analyses of the Ca2+, Mg2+, Na+, Cl-, and SO4 2were carried out using an ion chromat ograph at the University of Flor ida, Gainesville, Florida. For samples periods 2-4 analyses for the con centration of major ions nutrients, alkalinity, and silica were carried out by PPB Environmenta l Labs, Inc. in Gainesville, Florida. Analyses were done according to Environmen tal Protection Agency regulations for each particular ion. Ca2+, Mg2+, and Na+ were measured using an inductively coupled plasma (ICP) mass spectrometer and K+ was measured by atomic adsorption. The analysis of the 18O values of the samples was done using a Prism II Isotope Ratio Mass Spectrometer (IRMS) at the Depart ment of Geological Sciences, University of Florida attached to an automated multiprep preparation system. Two hundred L of each sample were pipetted into a glass vial and the headspace was filled with CO2 gas within a glove bag. The water was equilibrated with CO2 for 12 hours in the multiprep system and the samples were analyzed automatically by the mass spectrometer. Results are reported in standard delta notation relative to SMOW. Strontium isotope measurements were made to determine the 87Sr/86Sr isotopic ratios of the waters following separation of Sr using ion exchange columns. Concentration of Sr2+ was measured using isotope dilution. To determine both the Sr2+ concentration and the 87Sr/86Sr ratio, samples were analy zed on a thermal ionization mass spectrometer (TIMS). The standard used was NBS 987, which had errors of 0.7-1.1%.

PAGE 30

21 Computer Modeling The Saturation Index values of the water samples were determined using the program PHREEQC (Parkhurst and Appelo, 1999). The program uses the chemical composition and temperature of each water samp le to determine the saturation indices of various minerals including calcite for each sample. PHREEQC can also be used to determine the amount of calcite needed to dissolve in one liter of water to cause the water to reach saturation. Additionally, mixing tw o water samples in given proportions to calculate a mixture similar to measured samp les can be measured to determine mixing fractions of surface water and ground water at various river levels and conditions.

PAGE 31

22 CHAPTER 3 RESULTS Stage, Precipitation and Discharge Before the initial sampling period, the study area experienced a three-year drought, resulting in river stages ranging between 9.89.6 meters above sea level (masl) (Figures 3-1 and 3-2). During the El Nio event of autumn 2002 through early spring 2003, 114cm of precipitation fell, raising river stag e levels to a maximum stage of 14.43 masl on March 13, 2003. Throughout the study period the average river level was 10.35 masl. Based on this, river levels lower than 10 ma sl are considered low flow conditions and river levels above 11masl are considered high flow conditions. River levels of approximately 10.5 are considered baseflow c onditions and are intermediate between low flow and high flow conditions. Continuous water level measurements coll ected by Martin (2003) were used to determine discharge measurements for the Sink and Rise. The Sink discharge was based on a rating curve obtained from the Suwannee River Water Management District (Rating No 3. for Station No. 02321898, Santa Fe River at O’Leno State Park) and the discharge at the Rise was based on a rating curve determined by Screaton et al (2004). During the drought conditions of May 2002, the calculations yield negative discharge rates at the Sink and the Rise (Table 3-1), indicating that the rating curves are not suitable for low water levels because the river is not actively flowing. During this time no water enters the Sink because the entire river was captu red by the sinkhole at Vinzant’s Landing. The

PAGE 32

23 Precipitation Amounts 0 2 4 6 8 101/1/02 2/1/02 3/1/02 4/1/02 5/1/02 6/1/02 7/1/02 8/1/02 9/1/02 10/1/02 11/1/02 12/1/02 1/1/03 2/1/03 3/1/03 4/1/03 5/1/03DatePrecipitation (cm) 1 2 3 4 highest discharge measurements for the Sink and Rise occurred dur ing the March flood event and were 45.41 and 40.2 m3/s, respectively. Figure 3-1. Graph of the Santa Fe River elev ation collected at O’ Leno State Park during the course of the study period. The a rrows indicate the times of sampling. Figure 3-2. Graph of precipita tion collected at O’Leno Stat e Park during the study period. The arrows indicate the times of sampling. River Stage8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.001/1/02 2/1/02 3/1/02 4/1/02 5/1/02 6/1/02 7/1/02 8/1/02 9/1/02 10/1/02 11/1/02 12/1/02 1/1/03 2/1/03 3/1/03 4/1/03 5/1/03DateStage (masl) 1 2 3 4

PAGE 33

24 Table 3-1. Water levels and discharge measur ements at the Sink and Rise at the time of sampling. Location Date Water Level (m) Discharge, Q (m3/s) 5/14/02 9.76 -11.42 1/15/03 10.31 11.61 3/3/03 11.95 45.41 Sink 4/28/03 10.49 3.7 5/14/02 9.31 -1.39 1/17/03 10.11 11.2 3/5/03 11.11 40.2 Rise 5/1/03 9.94 7.71 Drought Stage Field measurements Vinzant’s Landing, the River Sink, Rive r Rise, eight karst windows and two springs were sampled during the drought c onditions. Conductivity values ranged from a low of 319 S at the Sink to a high of 645 S at Ogden Lake. Water temperatures were 23 C 28 C with the Rise having the lowest temperature and Treehouse Spring the highest. The lowest pH value was 7.03 at both Hawg and Twohole and highest pH value was 7.72 at Vinzants Landing. The dissolv ed oxygen (DO) content ranged from 0.1 mg/L at Ogden Lake, which at the time of sampling was nearly covered with duckweed, to 6.45 mg/L at Hornsby Spring. Chemical and isotopic composition The concentrations of cations were highe st during the low river stage of the drought. The chemical compositions of th e karst windows between the Sink and Rise were similar along the flow path with th e exception of Ogden Lake, which had higher concentrations of all solutes compared with the other karst windows. In most cases, the lowest cation concentrations occurred at the River Sink or Vinzant’s Landing. The

PAGE 34

25 concentrations of most karst windows and spri ngs fell between the va lues of the Sink or Vinzant’s Landing and Ogden Lake values (Figs. 3-3a, 3-4a, 3-5a, and 3-6a). In addition, the concentrations are higher at Sweetwater Lake than at the Rise and the downstream springs. The anion concentrations follow a pa ttern similar to the cation concentrations with Vinzant’s Landing and the Sink having the lowest concen trations of the solutes and Ogden Lake having the highest concentration (Figs. 3-7a a nd 3-8a). The alkalinity concentrations are more varied along the flow path than most of the solutes, and the lowest concentration occurs at Twohole sink and the highest at Hornsby Spring (Fig. 39a). Both the SO4 2and alkalinity concentrations are higher at Sweetwate r Lake than the River Rise. Based on the cation and anion data all of the samp les are at or near saturation with respect to calcite (Fig. 3-10a). The isotopic data do not vary much along the flow path and have small ranges in values compared to the major elemental data. The lightest 18O concentration is -3.06‰ at Hornsby Spring and the Sink has the heav iest concentration of –1.71‰ (Fig. 3-11a), although most of the samples are between –2.6 to –2.1‰. The 87Sr/86Sr ratios are lowest during this sample period with Paraners Bran ch Sink having the least radiogenic ratio and the Sink the most radiog enic (Fig. 3-12a). Intermediate Stages Surface Water Field measurements During the second sample period in Janua ry 2003, conductivity m easurements were typically 130-160 S/cm with the exception of Hawg Sink, Twohole Sink and Hornsby Spring, which had higher values of 368, 235, and 252 S/cm, respectively. In late

PAGE 35

26 April/early May 2003, the conductivity in the karst windows is higher than in January, ranging from 225-420 S/cm with the highest values at Sweetwater Lake, the Rise, Treehouse Spring and Hornsby Spring. The pH values in January were approximately 6 while the pH in April was approximately 7. During the colder, winter event in January, the water temperature of the karst windows was 10-11 C, but Hawg Sink is slightly warmer at 15.5 C. In April the temperature of the surface waters increased to 21-23 C. The DO content in January was 5.5-6, with Hawg having a very low level of 1.63, and was 1.1 to 3.8 in April. Chemical and isotopic compositions The chemical concentrations in January and April were intermediate compared to the concentrations during the drought and fl ood sample times. The concentrations in January were generally lower than the concen trations in April and were similar to the flood conditions while the concentrations in Ap ril were closer to the drought conditions. In January the Ca2+, Mg2+, and Sr2+ concentrations in the karst windows increase 63% between the River Sink and Ravine Sink a nd remain similar along the flow path. However, Hawg and Twohole Sinks and Hornsby Spring typically had higher concentrations (Figs. 3-3a, 34a, 3-5a, and 3-6a). The Na+ concentration at Hawg Sink, however, was lower compared with the other ka rst windows. In April, the concentrations of the cations were higher than the concentr ations in January with the exception of Na+, which was lower than January. The concentrat ions of most of the cations increased along the flow path from Paraners Branch Sink toward the River Rise, although the increase was small for Mg2+ and Sr2+ and there was no increase in Na+ (Figs. 3-3a, 3-4a, 3-5a, and 3-6a).

PAGE 36

27 Both the SO4 2and alkalinity concentrations were lower in January than in April (Figs. 3-8a and 3-9a). The concentrations of these two solutes increase ~63% between River Sink and Ravine Sink and were relativel y equal along the flow path. As with the cation concentrations, the SO4 2and alkalinity concentrations were higher at Hawg and Twohole Sinks as compared to the other kars t windows along the flow path. In April the concentrations of the SO4 2and alkalinity increased along the flow path from Paraners Branch Sink toward the River Rise. In contrast to the SO4 2and alkalinity, the Clconcentrations in January did not decrease between the first and second sampling periods, but remained elevated and were equal al ong the flow path (Fig. 3-7a). Hawg Sink, however, had a lower concentration than other locations. In April, the Clconcentrations were lower than January. The concentrations were slightly higher in the karst windows than at Vinzant’s Landing and the Sink but th ere is no increase in concentration along the flow path. However, there was a small sp ike in the concentration at Ravine Sink. During January, most of th e karst windows were undersat urated with respect to calcite (Fig. 3-10a), with Hawg Sink, Twohole Sink and Horn sby Spring being closest to saturation. In contrast, the samples in April we re much closer to sa turation with calcite and approached the saturati on index values from May. In contrast to the major element chemistry, the 18O values in January and April was similar along the flow path. The values in January were slightly depleted in 18O than in April from Vinzant’s Landing to Pa raners Branch Sink (Fig. 3-11a), but were enriched from Hawg Sink to Hornsby Spring. The ratios of 87Sr/86Sr, however, were similar to the major element chemistry with the samples from January plotting closer to

PAGE 37

28 the March data and the samples from April plot ting closer to the May data (Fig. 3-12a). In January the 87Sr/86Sr ratios were more radi ogenic than in April. Ground Water Field measurements The field measurements of the ground water in the wells were similar for the two intermediate sample periods. The conductiv ity of the ground water was typically 415530 S/cm, although Well 2 had the highest conductivity value in January (1009 S/cm) and Well 1 was highest in April (907 S/cm). The average pH of the water during both events was 6.7 and the average temperature was 21-22 C. As with the conductivity, the temperature of Well 2 was highest in January (26 C) and Well 1 was highest in April (26 C). The DO content was low for mo st of the wells at 0.15-0.22mg/L. Chemical and isotopic compositions The chemical concentrations of the wells during January and April were similar for most of the wells; however, Well 2 had highe r concentrations of all major elements during both sample events (Figs. 3-3b, 3-4b, 3-5b, and 3-6b). Wells 1 and 7 had higher concentrations than Wells, 3, 4 and 6 for Ca2+ and Well 7 had higher concentrations of Sr2+ and Mg2+ in April. The highest Sr2+ concentration in January was in Well 1. The Na+ concentrations were similar between th e wells in April; however, Well 7 had a slightly higher concentration. As with most of the cations the anion concentrations of the ground water during sampling of the interm ediate stages were similar between the two sampling periods (Figs. 3-7b, 3-8b, and 39b). Well 2 had the highest concentrations of Cland Na+. The Clconcentrations were higher in Well 7 than Wells 1,3,4, and 6, but the SO4 2concentrations were similar in these 5 wells. The alkalinity concentrations

PAGE 38

29 were lower in Well 2 than the other wells. Th e highest concentrations of alkalinity for both sample periods were in Well 7. During both January and April the ground water from each well was near saturation with respect to calcite, however, Well 7 was supersaturated with calcite in April (Fig. 3-10b). The isotopic compositions for ground wa ter during both intermediate stage periods were also similar. The 18O values were light in all of the wells in January and April and the values remained relatively uncha nged (Fig. 3-11b). Af ter several attempts, the ratios of 87Sr/86Sr could not be calculated for Well 2 in January but were calculated fro wells 1 and 7 (Fig. 3-12b). The ratio in Well 1 was less radiogenic in January than in April. In addition, Well 1 had the most radiogenic 87Sr/86Sr ratio in April as compared to the other wells. Flood Stage During the high water levels of the Santa Fe River at O’Leno Stat e Park, all of the karst windows were sampled with the excep tion of Twohole sink and Treehouse Spring. These sites were inaccessible because of fl ooded roads and high waters and swift currents in the river. All of the wells were sampled at this time. Surface Water Field measurements Field measurements of the karst windows during the flood stage were less variable than during the drought conditions. Conductivity was approximately 80 S/cm at most locations, and the temperature of the water was 17 C with the exception of Hornsby Spring, which was 18.8 C. The surface water pH levels were approximately 5 and DO levels were 6 mg/L, although Hawg was 2.88 mg/L.

PAGE 39

30 Chemical and isotopic compositions The solute concentrations in March are th e lowest of the four sample periods. Unlike the low flow period, the concentrati ons at Ogden Lake were not anomalously higher than the other karst windows and, for the most part, the solute concentrations were similar at each location along the flow path (F igs. 3-3a, 3-4a, 3-5a, and 3-6a). Hawg Sink, however, had higher concentrations of Ca2+, Mg2+, and Na+ than the other karst windows. As with the cations, the anion concentrations fell within a narrow range and were similar along the flow path (Figs. 37a, 3-8a, and 3-9a). Again, Hawg Sink had higher concentrations than the other karst windows. Based on the PHREEQC calculations all of the samples were undersaturated with resp ect to calcite (Fig. 3-10a). The 18O values were the lightest during this sample period (Fig. 3-11a), with the exception of the Rise, Treehouse Spring, and Hornsby Spring. The 87Sr/86Sr ratios were the most radiogenic and the Sink and Ogden La ke had the highest ra tios (Fig. 3-12a). The low ratio at the River Rise may be due to error during Sr separation and analysis. Ground Water Field measurements Field measurements of the physical para meters of the ground water in the wells were higher than those measur ed in the surface water. Conductivity of the wells was 408-488 S/cm and the temperature was 21 C except for Well 2 (25.8 C). The pH levels during flooding were close to 7 and the DO content of most of the wells was 0.13-0.33 mg/L. Well 3 and Well 4 had higher DO concentrations of 1.64 and 4.06 mg/L, respectively.

PAGE 40

31 Chemical and isotopic compositions All of the wells had similar chemical compositions except for Well 2, which had higher concentrations of some of the catio ns (Figs. 3-3b, 3-4b, 3-5b, and 3-6b), although the concentrations of these solutes in Well 2 were lower than during the intermediate sampling periods. The Ca2+ and Sr2+ concentrations were lowest in Well 2 during this sample period, however; the Mg2+ and Na+ concentrations were hi gher in Well 2 than in the other wells. The Cland SO4 2concentrations were highest in Well 2 (Figs. 3-7b and 3-8b). These concentrations were lower th an those from the intermediate sampling periods. The alkalinity con centrations were similar in Wells 3,4,6 and 7 with Well 1 having the highest concentra tion and Well 2 the lowest (F ig. 3-9b). All of the ground water samples were at or near saturati on with respect to calcite (Fig. 3-10b). The isotopic data of the ground water were similar to the intermediate sample periods. The 18O values were lightest in Wells 3,4,6, and 7 (Fig. 3-11b). The ratios of 87Sr/86Sr were less radiogenic than the surface water (Fig. 3-12b). Wells 2,3, and 4 had the highest ratios and Wells 1 and 7 had the lowest.

PAGE 41

32 0 10 20 30 40 50 60 70 80 02000400060008000Ca Concentrations (Sinks) May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)Ca (mg/L)Distance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsby 80 100 120 140 160 Well 1Well 2Well 3Well 4Well 6Well 7Ca Concentration (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)Ca (mg/L)Location Figure 3-3 a. Ca2+ concentrations of karst windows vs. distance from Vinzants Landing for each sample set and b.Ca2+ concentrations of the wells for sample sets 2-4. a. b.

PAGE 42

33 0 0.5 1 1.5 2 02000400060008000Sr Concentration (Sinks) May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)Sr (ppm)Distance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsbya. 0 0.5 1 1.5 2 2.5 Well 1Well 2Well 3Well 4Well 7Sr Concentration (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)Sr (ppm)Location Figure 3-4a. The Sr2+ concentrations of the karst wi ndows vs. distance from Vinzants Landing and b. the concentrations of Sr2+ in the wells for each sample period. b. a.

PAGE 43

34 0 5 10 15 20 25 02000400060008000Mg Concentrations (Sinks) May(9.8-9.6masl) Jan.(10.75-10.45masl) March(11.65-13masl) April(10.49masl)Mg (mg/L)Distance (m) Vinzants Sink Ogden Ravine Big Paraners Jim JugHawgTwohole Sweetwater Rise Treehouse Hornsby 0 5 10 15 20 25 30 Well 1Well 2Well 3Well 4Well 6Well 7Mg Concentration (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)Mg (mg/L)Location Figure 3-5a. The Mg2+ concentrations in the karst windows vs. distance from Vinzants Landing and b. the concentrations of Mg2+ of each well for each sample period. a. b.

PAGE 44

35 0 5 10 15 20 25 30 Well #1Well #2Well #3Well #4Well #6Well #7Na Concentration (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)Na (mg/L)Location Figure 3.6a. The Na+ concentrations in the karst wi ndows vs. distance from Vinzants Landing and b. Na+ concentrations of the wells for each sample period. 5 10 15 20 25 30 02000400060008000Na Concentrations (Sinks) May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)Na (mg/L)Distance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsby a. b.

PAGE 45

36 Figure 3-7a. The Clconcentrations in the karst wi ndows vs. distance from Vinzants Landing and b. Clconcentrations in each well for each sample period. a. b. 10 20 30 40 50 60 02000400060008000Cl Concentrations (Sinks) May (9.8-9.6masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)Cl (mg/L)Distance (m)VinzantsSink OgdenRavine Big Jim Paraners Jug Twohole Hawg SweetwaterRiseTreehouse Hornsby 0 10 20 30 40 50 Well 1Well 2Well 3Well 4Well 6Well 7Cl Concentration (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)Cl (mg/L)Location a.

PAGE 46

37 0 20 40 60 80 100 120 140 02000400060008000SO4 Concentrations (Sinks) May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.64-13masl) April(10.49masl)SO4 (mg/L)Distance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Horsnby Figure 3.8a.The SO4 2concentrations in the karst wi ndows vs. distance from Vinzants Landing and b. SO4 2concentrations of each we ll for each sample period. a. 0 50 100 150 200 250 300 350 Well #1Well #2Well #3Well #4Well #6Well #7SO4 Concentration (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)SO4 (mg/L)Location b.

PAGE 47

38 0 50 100 150 200 02000400060008000Alkalinity (Sinks) May(9.8-9.8masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)Alkalinity (mg/L)Distance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsby 100 150 200 250 300 Well #1Well #2Well #3Well #4Well #6Well #7 Alkalinity (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)Alkalinity (mg/L)Location Figure 3-9a. The alkalinity concentrations of the karst windows vs. distance from Vinzants Landing and b. alkalinity concentrations of the wells for each sample period. a. b.

PAGE 48

39 -5 -4 -3 -2 -1 0 1 02000400060008000Saturation Index (Sinks) May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.64-13masl) April(10.49masl)Saturation IndexDistance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsbya -1 -0.5 0 0.5 1 1.5 2 2.5 Well #1Well #2Well #3Well #4Well #6Well #7Saturation Index (Wells) Jan.(10.45masl) March(13masl) April(10.49masl) Saturation IndexLocationb. Figure 3-10a. The Saturation Index ratios of the karst windows vs. distance from Vinzants Landing and b. the Saturation Index ratios of the wells for each sample period.

PAGE 49

40 -3.5 -3 -2.5 -2 -1.5 0200040006000800018O (Sinks) May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)18O Distance (m)Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsby -5 -4 -3 -2 -1 0 1 Well 1Well 2Well 3Well 4Well 6Well 718O (Wells) Jan.(10.45masl) March(13masl) April(10.49masl)18O (ppm)Location Figure 3-11a. 18O values of karst windows vs. dist ance from Vinzants Landing and b. The 18O values in each well for each sample period. a. b.

PAGE 50

41 0.7075 0.708 0.7085 0.709 0.7095 02000400060008000Sr/86Sr Ratios May(9.8-9.6masl) Jan.(10.74-10.45masl) March(11.65-13masl) April(10.49masl)87Sr/86SrDistance (m) Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Twohole Sweetwater Rise Treehouse Hornsby 0.7076 0.7078 0.708 0.7082 0.7084 0.7086 Well 1Well 2Well 3Well 4Well 787Sr/86Sr Ratios Jan.(10.45masl) March(13masl) April(10.49masl)87Sr/86SrLocation Figure 3-12a. The 87Sr/86Sr ratios of the karst windows vs. distance from Vinzants Landing for each sample period and b. 87Sr/86Sr ratios of each well for each sample period. a. b.

PAGE 51

42 CHAPTER 4 DISCUSSION Variation in Chemical and Isotopic Composition The water in the matrix porosity can disso lve the limestone of the Floridan Aquifer increasing concentrations of Ca2+, Mg2+, and Sr2+ in the ground water. In contrast, conduit water may have low Ca2+, Mg2+, and Sr2+ concentrations if it comes directly from surface water. Consequently, the flow of water from the matrix to the conduit may increase Ca2+, Mg2+, and Sr2+ concentrations of the kars t windows. Conversely water from the conduit flowing to the matrix will de crease these solute concentrations in the matrix. Therefore high solute concentrations in the karst windows may reflect flow of ground water from the matrix to the conduit while low solute concentrations may reflect flow of surface water through th e conduit to the karst windows. Chloride and sodium are chemically conservative during reactions with the carbonate rocks of karst aquifers and thus changes in their concentrations should be decoupled from changes in Ca2+, Mg2+, and Sr2+ concentrations. The main sources of Cland Na+ in the natural waters of Florida (M addox, 1992) include 1) introduction from seawater along the coastal transition zone and 2) from marine aerosol s in precipitation. Because the study area is located inland, introduction of seawater is negligible as a source of Cland Na+, so their primary source in the stud y area is from marine aerosols in precipitation. The concentrations of Cland Na+ can increase through evaporation or decrease by dilution from precipitation. Precipitation a nd evaporation should in crease or decrease

PAGE 52

43 concentrations of both solutes proportionately so that the Na/Cl ratio would not change. Sodium concentration may change through i on exchange with clays from the Hawthorn Formation. Chloride concentrations will not be effected by these reactions and thus changes in Na+ by this process could be observed as variations in the Na/Cl ratio. The effects of cation exchange are likely to be much smaller than those caused by evaporative effects or dilution by precipitation. The concentration of SO4 2can also be used as a tracer; however, like the Ca2+, Mg2+, and Sr2+ it is a non-conservative tracer involv ed in chemical reactions with the matrix rocks. The sources of SO4 2in Florida include 1) dissolution of gypsum and anhydrite 2) oxidation of rocks with sulfide-bear ing minerals such as pyrite and 3) marine aerosols and acidic precipitation from ai rborne sulfur oxides (Maddox, 1992). Chemical Variations in Karst Windows and Wells The high concentrations of Cl-, Na+, Ca2+, Mg2+, Sr2+, and SO4 2in the karst windows in May 2002 may be due to introductio n of water from the matrix that has higher solute concentrations, evaporation from the karst windows because of low precipitation, or a combination of th e two. The high concentrations of Ca2+, Sr2+, Mg2+, and SO4 in the karst windows are expected for wa ter that has reacted with aquifer rocks and flowed from the matrix to the condu it. However, the concentrations of Cland Na+ are higher than the concentrations in the gr ound water in the wells at baseflow indicating evaporation must also be occu rring to increase the concen trations of these non-reactive solutes. Another possible explanation fo r high concentrations of Ca2+ and SO4 2during this time may be from upwelling of water that has dissolved gypsum and anhydrite in the lower portions of the Floridan Aquifer. In Katz et al (1999) it was suggested that calcite

PAGE 53

44 dissolution yields HCO3 and Ca2+ concentrations in a 2:1 ra tio in the Upper Floridan Aquifer water according to the equation CaCO3 + H2CO3 = Ca2+ + 2HCO3 (1) Katz et al (1999) suggested that HCO3 -/Ca2+ ratios less than 2 were attributed to an alternative source of Ca2+ and SO4 2-, such as gypsum. The HCO3 -/Ca2+ ratios in May are greater than two, however, ranging from 2.08 to 2.75 in the karst windows (Table 4-1) and 3.38 and 3.33 at Vinzants Landing and the River Sink, respectively. During this time all of the karst windows were at saturation or slightly supersaturat ed with respect to calcite (Fig. 3-10a). Extended periods of evaporation with l ittle to no precipitation during the drought will concentrate the so lutes in the water. Over time, these solutes can reach or exceed their saturation points, resulting in mineral precipitation. Calcite precipitation (e.g. reverse of reaction 1) would increase the HCO3 -/Ca2+ ratio and may cause the elevated HCO3 -/Ca2+ ratios at Vinzant’s Landi ng and the Sink. Based on the concentrations of the reactive and non-reactive solutes, it ap pears that duri ng the drought there was loss of water from the matrix to th e conduit as well as evaporation in the karst windows resulting in the increased solute concentrations. In January the increased concentrations in the karst windows from the Sink to the Rise suggest an input of water with higher solute concentrations downgradient from the River Sink. Dean (1999) found that at river stages of 10.74, 10.5 and 10.44 masl, SO4 2concentrations increased by 53.2%, 50.2% a nd 49.3% between the Sink and Sweetwater Lake, which he attributed to an input of water from the matrix to the conduit. The HCO3 -/Ca2+ ratios in January were, for the most pa rt, less than two (Tab le 4-1) and could reflect upwelling of water that has dissolved gypsum. The SO4 2concentrations are low,

PAGE 54

45 however, suggesting the ratios probably refl ect mixing with dilute surface water (e.g. Katz et al. 1999). Table 4-1. The HCO3 -/Ca2+ ratios in the karst windows and the wells. Location May January March April Vinzants 3.38 1.07 1.44 2.28 Sink 3.33 1.08 1.05 2.25 Ogden 2.08 1.01 2.11 2.26 Ravine 0.99 1.26 1.91 Paraners 2.29 1.07 1.23 2.26 Jim 2.62 1.00 0.96 2.25 Jug 2.75 1.00 1.42 2.27 Hawg 2.46 2.27 1.74 2.05 Twohole 2.2 1.81 2.04 Sweetwater 2.14 1.25 1.05 2.16 Rise 2.26 1.29 0.99 2.89 Hornsby 2.16 1.72 1.40 2.07 Treehouse 2.49 1.26 1.91 Well 1 2.36 2.14 2.40 Well 2 1.36 1.29 1.27 Well 3 2.32 2.36 Well 4 2.26 2.20 Well 6 2.12 2.31 Well 7 2.59 2.25 2.31 In contrast to the other solute data, the Clconcentrations are not diluted in the karst windows in January and it is unclear at this time why the Clconcentrations remain elevated following the increased precip itation in the fall of 2002. Unlike the Clconcentrations, the Na+ concentrations decrease by 16-30% between May 2002 and January 2003. The change in the Na+ concentration compared to no change in Clsuggests that during drought conditions th ere may have been another source of Na+ to the water other than the marine aerosols in preci pitation. One possible source could be from cation exchange with the clay minerals of the Hawthorn Formation, which, in central Alachua County, have cation exchange cap acities ranging between 6 and 46 meq/100 g (Rose, 1989). A possible explanation for the disproportionate decrease in the Cland Na+

PAGE 55

46 concentrations in January is that during drought conditions Na+ in the smectite may exchange with the high Ca2+ concentrations in the Santa Fe River, which would result in a high Na/Cl ratio (Fig. 4-1). W ith increased precipitation, the Ca2+ concentration in the river would be dilu ted, releasing the Ca2+ previously held in the exchange sites and uptaking Na+ into the clays. This would result in a decrease in the Na/Cl ratio. The lower Na/Cl ratios in January further s upport cation exchange with the Hawthorn Group as a possible source of Na+ during low flow. 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2Vinzants Sink Ogden Ravine Big Paraners Jim Jug Hawg Two Hole Sweetwater Rise Treehouse HornsbyNa/Cl Ratios (Sinks) May January March AprilNa/Cl (mole ratio)Location Figure 4-1. Graph of the Na/Cl ratios of the karst windows for each sample period. Although the river stages were approximat ely the same in April and January the concentrations of Ca2+, Mg2+, Sr2+, and SO4 2are higher in April th an in January but the Na+ and Clconcentrations are lower. Flooding in March may cause these differences in solute concentrations in April. During the flooding, water is lost from the conduit to the matrix and may be held in the matrix por osity, allowing the water to react with the carbonate rocks of the aquifer. As the floodwaters recede, the hydraulic head in the conduit becomes lower than the matrix (Marti n, 2003) allowing the reac ted water to flow a.

PAGE 56

47 into the conduit. The lo wer concentrations of Cland Na+ in April may be a reflection of dilute floodwaters that were held in the matrix, with elevated Ca2+, Mg2+, and Sr2+ concentrations from dissolution reactions. Solute concentrations do not change as mu ch from one sample time to the next in water collected from all wells, as compared to the karst windows except Well 2 (Figs. 31b – 3-9b). Some of the wells, however, do ha ve some small changes in concentrations between low and high flow with some solute s becoming more diluted and other solutes more concentrated. The changes in concentrat ions of the solutes are different for each well, suggesting that lo ss of water from the conduit is heterogeneous and varies at the local scale. The HCO3 -/Ca2+ ratio of Well 2 ranges between 1.27 and 1.36 for the study period (Table 4-1), suggesting upwelling of deeper gr oundwater according to the model of Katz et al (1999). However, the lower concentrations of all of the solutes in March suggest the low HCO3 -/Ca2+ ratio reflects mixing with dilute surface water that has an HCO3 -/Ca2+ ratio of 1.05. The low solute concentrations in Well 2 in March suggest that at high discharge rates at the River Sink, surface wa ter may leave the conduit and flow to the well, possibly along a zone of high permeability. In January and April the HCO3 -/Ca2+ ratios in Well 2 are less than two and correspond with high SO4 2concentrations of 305 and 243 mg/L, respectively and indicate an alternative source of Ca2+ and SO4 2-, such as gypsum, from water deep in the Floridan Aquifer. During intermediate, ba seflow conditions, the wa ter in Well 2 appears to represent ground water from th e Floridan Aquifer. The solu te concentrations in Well 2 are higher than the concentrations in the other wells and may reflect a deeper water

PAGE 57

48 source. The differences in the ground water chemistry throughout the study area demonstrate the complex nature of karst aqui fers in that the composition of ground water in karst terrains can vary on the local scale. 87Sr/86Sr ratios as tracers in carbonate aquifer studies Sr isotopes are particularly useful in st udies of aquifers in marine carbonates because their 87Sr/86Sr reflects the87Sr/86Sr ratio of the seawater at the time of deposition (DePaolo and Ingram, 1985; Hess et al. 1986). The 87Sr/86Sr ratio of seawater at the beginning of the Cretaceous was 0.7072. This value increased to 0.70775 at the Cretaceous-Tertiary boundary (Burke et al. ; DePaolo and Ingram, 1985) and then decreased into the late Eocene to a valu e of 0.7076. From the late Eocene into the Quaternary the 87Sr/86Sr ratio in seawater increased to the modern ratio of 0.70907 (Burke et al. 1982; DePaolo and Ingram, 1985; Hess et al. 1986). During the Oligocene and Eocene Periods, when many of the carbonate rocks of the Floridan Aquifer were deposited, the ratio remained relatively constant at 0.7077-0.7078. Thus, dissolution of this material should cause groundwater ra tios to approach th ese low values. Isotopic Variations in the Karst Windows and Wells Water from the karst windows and wells appe ar to reflect mixing between Sr from the carbonates, Sr from rainfall and Sr from the Hawthorn Group rocks. The 87Sr/86Sr ratios in the wells are 0.7078 and 0.7079 prior to the flooding, which suggest that the 87Sr in the water in the wells is enriched from di ssolution of the calcite of the matrix rocks and the ratios are representative of ground water in the matrix. During low flow, the values of the 87Sr/86Sr ratios in the karst window s are 0.7078, similar to those of Eocene/Oligocene seawater and presumably th e carbonate rocks of th e Floridan Aquifer

PAGE 58

49 (Fig. 3-12a). These values s uggest that water in the karst windows originates from the matrix as has been suggested based on the solute concentrations. During high flow conditions, the Sr isotope ratios in the karst windows reach the highest values of all the sa mples of 0.7091 to 0.7094. At the Sink and Vinzant’s Landing the ratio is higher than the modern s eawater value (0.70907) and may reflect the radiogenic ratios associated with the clays of the Hawthorn, which are present upstream from the River Sink. The 87Sr/86Sr ratio in Well 1 remains within the Eocene/Oligocene seawater curve during high flow (Fig. 3-12b); however the 87Sr/86Sr ratios of Wells 3,4, and 7 have slightly more radiogenic suggesti ng some mixing with wate r that has been in contact with radiogenic 87Sr/86Sr ratios such as the clays of the Hawthorn Group. During the two times of sampling during in termediate conditions, samples from the karst windows collected in January have more radiogenic 87Sr/86Sr ratios than in April 2003. The differences in values suggest that the January samples are more influenced by the clays of the Hawthorn Formation than th e April samples, which have ratios that reflect influence by the Florid an Aquifer carbonates. The karst windows in January 2003 are undersaturated with respect to calcite (Fig. 3-10a). This undersaturation implies there has been little dissolution of the carbonates of the Floridan Aquife r, thereby limiting the decrease in the 87Sr/86Sr ratios and increase in Sr2+ concentrations. In April 2003, the 87Sr/86Sr ratios in the karst windows are closer to the values of Eocene/Oligocene seawater than in January, consistent with re actions with the carbona tes of the Floridan Aquifer. These reactions may have occurr ed during flooding in March when water from the conduit would react with the high Sr2+ concentrations and low 87Sr/86Sr ratios of the aquifer rocks. As the floodwater recedes, th e reacted water in the matrix would flow to

PAGE 59

50 the conduit, causing the higher Sr2+ concentrations and less radiogenic87Sr/86Sr ratios observed in April than in January. The slightly more radiogenic 87Sr/86Sr ratio of Well 1 in April may be a reflection of the floodwater held in the matrix that did not appear until sampling in April. Sr Mixing Model The Sr mixing model assumes two end-member mixing using the Sr2+ concentration and the 87Sr/86Sr ratios of selected end-members. One end member used for this model is the value of the River Sink sample dur ing the flooding in March 2003. This value is assumed to represent the most pristine surface water sample and thus would be the closest to average precipitation in the region. The other end member used in the model is water from Well 1 in January 2003, due to its high solute concentrations and low radiogenic 87Sr/86Sr ratio, suggesting this water re presents the ground water in the region. The strontium mixing model used is based on Faure (1986) and Woods et al (2000), assumes mixing of the end members of surface water and ground water following an equation in the form: b Sr a Sr Sr/M M 86 87 (2) where: (87Sr/86Sr)M and [Sr]M are the 87Sr/86Sr ratio and concentration of the mixture, respectively, a is the slope of the line from the equation: B A A 86 87 86 87 B ASr Sr Sr Sr/ Sr Sr/ Sr Sr a (3) and b is the y-intercept from the equation: B A B 86 87 B A 86 87 ASr Sr Sr Sr/ Sr Sr Sr/ Sr b (4)

PAGE 60

51 where Sr A and Sr B and (87Sr/86Sr)A and (87Sr/86Sr)B are the concentrations and ratios of end members A (the River Sink in March) a nd B (Well 1 in January) respectively. The ratio of 87Sr/86Sr is plotted versus Sr and versus 1/Sr (Fig. 4-2). These figures suggest there is two end member mixing, but the extreme isotope values occur during extreme flood conditions. From the Sr isotope mixing model a quantit ative percentage of surface water in a given sample can be determined by calculating the f parameter for mixing assuming a composition for the end members. In this model, f defines a mixture of two solutions A and B. The equation for f, based on Faure (1986), assumes twocomponent mixing using the formula: f= B B A MX X X X (5) where XM, XA, and XB are the concentrations of Sr2+ for the sample and end members A (River Sink) and B (Well 1), respectively. These calculations indicate that water w ith the lowest fraction sourced from the surface water occurred in the ka rst windows in May 2002 (Table 42) and averaged 32%. The highest proportion of wate r originating from the River Sink in the karst windows occurred in January, March, and April 2003 (Table 4-2). During the March flooding, the model suggests nearly 100% of the water in the karst windows is from the Sink. At river stages of 11.75 or greater, the travel time between the Sink and Rise is less than one day (Dean, 1999; Martin, 2003) meaning there is rapi d flushing of water through the conduit. This rapid transport of water would prevent water from leaving the matrix and entering the conduit. Thus the low concentrations in the karst windows pr obably reflect dilution by precipitation.

PAGE 61

52 The f parameter calculations indicate approxi mately 90% of water in the karst windows in January 2003 originated from the Sink surface water end member but in April 2003 52-71% originated from the Sink. Although the river stages were approximately equal during these sample tim es the karst windows in January contain 19%-38% more water from the Sink than in April. As the river level began to rise with increased precipitation in Fall 2002, water entered the Sink and flowed through the conduit. The hydraulic head in the conduit wa s higher than the matrix, allowing water to flow from the conduit to the matrix and result in water that is more dilute than in April. In contrast, the water in the karst windows in April reflects flow from the matrix to the conduit as the flooding receded. These result s suggest that the antecedent conditions along with river stage are importa nt for water chemistry. From the f parameter calculation, the estimat ed proportion of water originating from surface water in Well 7 is similar be tween January, March and April 2003 (Table, 4-3) with only 25-30% of the water originating from ground water suggesting that surface water may reach this well. The well, however is saturated with re spect to calcite for each sample event and suggests that undersatur ated surface water dissolves calcite as it moves through the matrix toward the well. Wells 3 and 4 also appear to have high percentages of water from th e Sink end member for both March and April, which would lead to the more radiogenic 87Sr/86Sr ratios and low Sr2+ concentrations in the wells at

PAGE 62

53 Sink Well 1 Figure 4-2. 87Sr/86Sr vs. Sr and 87Sr/86Sr vs.1/Sr models for all ka rst windows and wells. The arro ws indicate the end members. Ogden Lake River Rise 0102030405060 1/Sr (ppm) 0.7075 0.708 0.7085 0.709 0.7095 0.71 00.511.522.5 Mixing Model May (Sinks) Jan.(Sinks) March(Sinks) April (Sinks) Jan.(Wells) March(Wells) April(Wells)87Sr/86SrSr (ppm)

PAGE 63

54 Table 4-2. The calculations of the percentage of surface water in the karst windows. Location May (%) January (%) March (%) April (%) Vinzants 59 Sink 76 72 100 85 Ogden -6 97 100 Ravine 90 99 52 Big 27 91 Paraners 32 Jim 37 90 99 71 Jug 37 Hawg 40 91 68 Twohole 41 80 63 Sweetwater 5 90 99 54 Rise 30 90 100 Treehouse 14 Hornsby 8 those times. The mixing calculation suggest s that Well 2 is composed of 95% surface water in March. This is supported by low solu te concentrations at that time and suggests that at times of high flow in the river, water can be flushed toward the well. In April, the Sr2+ concentration in Well 2 is greater than the Well 1 end member, resulting in the negative f value at that time. Well 2 has higher concentrations of all of the solutes than Well 1 and suggests that water from deeper po rtions of the aquifer may be upwelling into Well 2. This possible source water for Well 2 is also evidenced by the low HCO3 -/Ca2+ ratios of less than two. Saturation and Mixing Calculations To better understand the com position of the water in Well 2, which appears to be more affected by flooding than the other wells calculations of cal cite dissolution and mixing

PAGE 64

55 Table 4-3. The calculations of the percentage of surface water of the wells Location January (%) March (%) April (%) Well 1 0 46 95 Well 2 95 -33 Well 3 92 Well 4 92 94 Well 7 71 75 70 were done using PHREEQC. The program was us ed to simulate the reaction of one liter of a known solution of water with a given amou nt of calcite in order to determine how much calcite needs to dissolve to reach equi librium. The saturation index of Well 2 in March is -0.40, which is close to saturation. For the simulation, water from the Sink in March with a saturation index of –4.2 was r eacted with calcite until equilibrium was reached, which required dissolution of 1.4 x 10-3 moles of Ca2+. In March, the water in Well 2 is greater by1.7 x 10-3 moles of Ca2+ than the water in the Sink, similar to the amount calculated. This similarity suggests that Well 2 is a mixture of water from the Sink that has dissolved calcite and ground water from the matrix The large decreases in solute concentrations in Well 2 in March were not seen at the other wells. Although the high solute concentrations in the well appear to reflect ground water from deep within the aquifer at baseflow conditions, the large changes during flooding suggest loss of water from the conduit to the matrix. For this reason, PHREEQC was used to check the fractions of surface water and ground water in Well 2 in March determined using the Sr2+ concentrations. The 87Sr/86Sr model suggests that the composition of Well 2 in March is approximate ly 95% surface water. Based on this, the same end-members used in the Sr model were mixed at varying proportions. The results show that a solution of 25-30% Sink water and 70-75% Well 1 water was closest to the

PAGE 65

56 measured data, which is less than the 95% surf ace water estimated in the Sr model. This discrepancy may be due to the elements us ed in the calculation. The PHREEQC model takes into account the concentr ations of all of the major el ements in the solutions while the Sr model is based only on the concentrations of Sr2+ in the end-members and the mixture. The ground water in Well 2 differs from Well 1 and, therefore, the composition of Well 1 may not be an accurate representation of the waters that are mixing at Well 2. Because the Sr2+ concentrations of Well 1 and Well 2 are similar at baseflow conditions, the percentage of surface wate r in Well 2 calculated from the Sr model may be more accurate than the fractions determined using PHREEQC. Nutrient Loading The discharge at the Sink and Rise were calculated using water level data collected by Martin (2003) (Table 3-1) and, along with the NO3 and PO4 concentrations, were used to determine the amount of nutrient load ing to the system (Tables 4-4 & 4-5). The discharge at the Sink in March (45.41 m3/s) is greater than the Rise (40.2 m3/s) indicating that water is lost to the matrix. The NO3 and PO4 loading at the Sink are 0.75 x 103 mg/s and 11.54 x 103 mg/s, respectively and are 0.48 x 103 mg/s and 9.4 x 103 mg/s at the Rise. Less NO3 and PO4 discharges at the Rise than enters the Sink during the flooding, which is consistent with loss of water to the matrix. In January the discharge at the Rise is slightly less than the Sink (11.2 m3/s vs. 11.6 m3/s) so there is some loss of water to the matrix. This loss of water is reflected in the lower nutrient loading rates at the Rise (Tables 4-4 & 4-5). In April, however the discharge at the Rise (7.71 m3/s) is greater than the Sink (3.70 m3/s) indicating water flows from the matrix to the conduit. If this additional water has higher NO3 and PO4 concentrations it would be expected that nutrient loading to the Rise is greater than at the Sink. In April, the nutrient loading at

PAGE 66

57 the Rise (2.78 x 103 mg/s NO3 and 1.78 x 103 mg/s PO4 3-) is nearly two times the loading at the Sink (1.27 x 103 mg/s NO3 & 0.67 x 103mg/s PO4 3-) even though the concentrations of the nutrients are similar be tween the two locations (Tables 4-4 and 4-5). There is a decrease in the PO4 concentration at both the Sink and the Rise in April compared to the concentrations in January and March. The decrease in PO4 3concentrations coupled with an increase in loading of both nut rients suggest that as water flows from the conduit with incr easing discharge it is held in the intergranular porosity of the matrix. As the system returns to basefl ow this water flows back into the conduit, increasing the nutrient loading to the Rise. Table 4-4. The discharge and NO3 2loading calculations at the Sink and Rise. Location Date Water Level Q NO3 NO3 Loading (m) (m3/s) (mg/L) (mg/s x 1000) 5/14/02 9.76 -11.42 0.009 -0.103 1/15/03 10.81 11.61 0.031 0.36 3/3/03 11.95 45.41 0.016 0.73 Sink 4/28/03 10.49 3.70 0.344 1.27 5/14/02 9.31 -1.39 0.077 -0.107 1/17/03 10.11 11.2 0.024 0.27 3/5/03 11.11 40.2 0.012 0.48 Rise 5/1/03 9.94 7.71 0.360 2.78

PAGE 67

58 Table 4-5. The discharge and PO4 loading calculations of the Sink and Rise. Location Date Water Level Q PO4PO4-Loading (m) (m3/s) (mg/L) (mg/s x 1000) 5/14/02 9.76 -11.42 0.132 -1.5 1/15/03 10.81 11.61 0.105 1.22 3/3/03 11.95 45.41 0.254 11.54 Sink 4/28/03 10.49 3.7 0.181 0.67 5/14/02 9.31 -1.39 0.116 -0.16 1/17/03 10.11 11.2 0.103 1.15 3/5/03 11.11 40.2 0.234 9.4 Rise 5/1/03 9.94 7.71 0.137 1.06

PAGE 68

59 CHAPTER 5 CONCLUSIONS The heterogeneous characteristics of karst aquifer systems can leave ground water vulnerable to contamination from surface water. This heterogeneity allows water to move between the large solution channels of the c onduit porosity and the smaller pore spaces of the intergranular porosity of the ma trix. As water enters a c onduit through a sink or swallet some of it may be lost to the matrix if the hydraulic head in the condu it is greater than the matrix. If, however, the hydraulic head is high er in the matrix than the conduit, water can flow from the matrix to the conduit. Because surface water may contaminate ground water supplies it is important to understand the processes and circumstances that can allow surface water to enter the ground water matrix. Chemical analysis of surface wa ter, in the form of sinks and springs, and ground water from wells can be used to eval uate the exchange of water from the two systems. In the Santa Fe River Sink/Rise syst em it appears that both the water level of the river as well as the antecedent conditions play a role in the movement of water between the conduit and the matrix. Under low river levels and drought conditions such as May 2002, no water enters the River Sink and flows through the conduit or out of th e River Rise. As a result, the water chemistry in the karst windows reflects ground water from the matrix. As the river level rises with increased ra infall, more water enters the Sink and flows through the conduit. The hydraulic head in the conduit becomes higher than the matrix and surface water flows to the matrix. The increased amount of surface water in the subsurface is reflected in dilute solute

PAGE 69

60 concentrations of the karst windows. Duri ng flooding in March 2003, water in the karst windows reflects a composition of nearly 100% su rface water. As the flooding subsides, the system returns to baseflow conditions. Following flooding, water from the matrix enters the conduit and the karst windows. Because of the different circumstances preceding baseflow, the chemistry of the water in the karst window s is different. After prolonged periods of drought conditions an increased pr ecipitation allows the conduit to quickly fill with water, increasing its hydraulic head, po ssibly over that in the matrix However, following periods of high flow and flood conditions, the conduit quickly loses some water as precipitation decreases and discharge at the River Sink decreases. The response of the conduit and matrix to changes in precipitation a nd flow rates are important in understanding the hydrologic characteristics of karst aquifer systems. The matrix appears less responsive to change s in the flow conditions and precipitation amounts than the conduit. The chemistry of grou nd water collected in the wells, for the most part, did not change much from one sample time to the next. However, the significant difference in the chemistry of Well 2 from baseflow to flood conditions indicates the complex, heterogeneous nature of karst aquife rs. In addition, the Sr model indicates high percentages of surface water in the wells du ring the study period. Dissolution of matrix material can increase the solute concentrations in the ground wa ter and making it appear that no surface water is present. Although there has been more extensive chemical analyses of the karst windows in the Sink/Rise system, it is now important to better understand the effects of flow conditions on the matrix with more sample collection at higher temporal resolutions.

PAGE 70

APPENDIX A DAILY PRECIPITATION AND STAGE RECORDS

PAGE 71

62 Daily Precipitation and Stage for 2002 Month ppt(cm) masl 1-Jan 0 9.83 19-Feb 0 10.22 9-Apr 0 10.25 2-Jan 0.508 9.82 20-Feb 0 10.22 10-Apr 0 10.24 3-Jan 0.305 9.82 21-Feb 0 10.22 11-Apr 1.22 10.22 4-Jan 0 9.81 22-Feb 0.508 10.22 12-Apr 0.36 10.22 5-Jan 0 9.81 23-Feb 1.473 10.21 13-Apr 0.10 10.22 6-Jan 0.508 9.81 24-Feb 0 10.20 14-Apr 0.41 10.22 7-Jan 0 9.80 25-Feb 0 10.22 15-Apr 0 10.22 8-Jan 0 9.80 26-Feb 0 10.24 16-Apr 0 10.22 9-Jan 0 9.80 27-Feb 0 10.23 17-Apr 0 10.23 10-Jan 0 9.80 28-Feb 0 10.22 18-Apr 0 10.23 11-Jan 0 9.79 1-Mar 0 10.22 19-Apr 0 10.23 12-Jan 0.813 9.79 2-Mar 1.19 10.23 20-Apr 0 10.22 13-Jan 0 9.79 3-Mar 5.03 10.24 21-Apr 0 10.21 14-Jan 5.461 9.79 4-Mar 5.00 10.32 22-Apr 0 10.20 15-Jan 0 9.88 5-Mar 0 10.43 23-Apr 0 10.17 16-Jan 0 9.87 6-Mar 0 10.49 24-Apr 0 10.15 17-Jan 0 10.28 7-Mar 0 10.52 25-Apr 10.13 18-Jan 0 10.29 8-Mar 0 10.53 26-Apr 0 10.11 19-Jan 0.432 10.30 9-Mar 0 10.53 27-Apr 0 10.05 20-Jan 0.076 10.30 10-Mar 0 10.53 28-Apr 10.01 21-Jan 0.000 10.30 11-Mar 0 10.51 29-Apr 0 9.95 22-Jan 1.219 10.30 12-Mar 0 10.50 30-Apr 9.92 23-Jan 0.076 10.32 13-Mar 0.610 10.47 1-May 9.89 24-Jan 0 10.33 14-Mar 0.051 10.45 2-May 9.87 25-Jan 0 10.34 15-Mar 0 10.43 3-May 0 9.84 26-Jan 0 10.33 16-Mar 0 10.42 4-May 9.81 27-Jan 0 10.33 17-Mar 0 10.40 5-May 9.75 28-Jan 0 10.34 18-Mar 0 10.40 6-May 9.74 29-Jan 0.737 10.34 19-Mar 0 10.39 7-May 0 9.73 30-Jan 0 10.40 20-Mar 0 10.39 8-May 9.72 31-Jan 0.305 10.43 21-Mar 0 10.39 9-May 9.71 1-Feb 0 10.40 22-Mar 0.356 10.36 10-May 9.70 2-Feb 0 10.39 23-Mar 0 10.34 11-May 9.69 3-Feb 0 10.39 24-Mar 0 10.34 12-May 9.68 4-Feb 0 10.38 25-Mar 0 10.33 13-May 9.67 5-Feb 0 10.37 26-Mar 0.279 10.33 14-May 1.07 9.69 6-Feb 0 10.37 27-Mar 0 10.32 15-May 9.67 7-Feb 1.219 10.30 28-Mar 0 10.31 16-May 9.66 8-Feb 0.076 10.28 29-Mar 0 10.30 17-May 9.65 9-Feb 0 10.28 30-Mar 0 10.29 18-May 9.64 10-Feb 0 10.28 31-Mar 0 10.28 19-May 0.152 9.65 11-Feb 0 10.28 1-Apr 0 10.23 20-May 0.051 9.63 12-Feb 0 10.26 2-Apr 0.356 10.26 21-May 0 9.60 13-Feb 0 10.26 3-Apr 0.483 10.26 22-May 0 9.62 14-Feb 0.127 10.26 4-Apr 0 10.26 23-May 0 9.62 15-Feb 0 10.25 5-Apr 0 10.26 24-May 0 9.62 16-Feb 0 10.25 6-Apr 0 10.26 25-May 0 9.59 17-Feb 0 10.25 7-Apr 0 10.26 26-May 0 9.59 18-Feb 0 10.22 8-Apr 0 10.25 27-May 0.051 9.57

PAGE 72

63 28-Ma y 0 9.57 18-Jul0.1279.45 7-Se p 0 10.37 29-May 0 9.57 19-Jul 0 9.45 8-Sep 0 10.41 30-May 0.152 9.57 20-Jul 2.01 9.48 9-Sep 0 10.37 31-May 1.27 9.55 21-Jul 0.254 9.50 10-Sep 0 10.35 1-Jun 0.152 9.56 22-Jul 0 9.51 11-Sep 0 10.32 2-Jun 9.55 23-Jul 0.051 9.48 12-Sep 0 10.29 3-Jun 9.54 24-Jul 0.076 9.48 13-Sep 0 10.09 4-Jun 1.016 9.54 25-Jul 0.076 9.45 14-Sep 2.26 10.26 5-Jun 9.54 26-Jul 0.025 9.45 15-Sep 0.508 10.28 6-Jun 0.305 9.54 27-Jul 0.254 9.48 16-Sep 0 10.29 7-Jun 0.178 9.54 28-Jul 0 9.45 17-Sep 0 10.32 8-Jun 0.254 9.52 29-Jul 0 9.48 18-Sep 0 10.33 9-Jun 0 30-Jul 0 9.46 19-Sep 0 10.36 10-Jun 0 9.53 31-Jul 0 9.42 20-Sep 0 10.43 11-Jun 0 9.51 1-Aug 0 9.42 21-Sep 0 10.47 12-Jun 0 9.51 2-Aug 0 9.45 22-Sep 0 10.53 13-Jun 0.686 9.50 3-Aug 0 9.84 23-Sep 0 10.57 14-Jun 0.025 9.50 4-Aug 0 10.16 24-Sep 0 10.54 15-Jun 0.025 9.50 5-Aug 0.0254 10.14 25-Sep 0.91 10.51 16-Jun 0 9.48 6-Aug 0 10.16 26-Sep 0.30 10.48 17-Jun 0 9.47 7-Aug 0.1524 10.14 27-Sep 0.53 10.45 18-Jun 0.305 9.47 8-Aug 0 10.23 28-Sep 0 10.43 19-Jun 0.025 9.47 9-Aug 0 10.23 29-Sep 0 10.42 20-Jun 1.499 9.45 10-Aug 0 10.24 30-Sep 0 10.42 21-Jun 0.203 9.48 11-Aug 0 10.24 1-Oct 0 10.41 22-Jun 0.864 9.48 12-Aug 1.93 10.23 2-Oct 0 10.42 23-Jun 0 9.48 13-Aug 0.254 10.23 3-Oct 0 10.40 24-Jun 0 9.45 14-Aug 0 10.20 4-Oct 0 10.40 25-Jun 2.515 9.47 15-Aug 0 10.20 5-Oct 0 26-Jun 0.025 9.48 16-Aug 0 10.17 6-Oct 0 27-Jun 0 9.50 17-Aug 0 10.17 7-Oct 0 10.36 28-Jun 0.533 9.51 18-Aug 0 10.19 8-Oct 0 10.33 29-Jun 2.210 9.51 19-Aug 0 10.19 9-Oct 0.23 10.33 30-Jun 0.254 9.52 20-Aug 0 10.19 10-Oct 0 10.31 1-Jul 0 9.51 21-Aug 0 10.17 11-Oct 0 10.31 2-Jul 0 9.50 22-Aug 0.025 10.14 12-Oct 0 10.33 3-Jul 0.660 9.50 23-Aug 0 10.14 13-Oct 0.41 10.33 4-Jul 1.727 9.50 24-Aug 0 10.13 14-Oct 0 10.30 5-Jul 0.102 9.50 25-Aug 0 10.13 15-Oct 0 10.30 6-Jul 0 9.51 26-Aug 0 10.10 16-Oct 0.23 10.30 7-Jul 0 9.50 27-Aug 0.43 10.07 17-Oct 0 10.29 8-Jul 0 9.50 28-Aug 5.21 10.14 18-Oct 0 10.10 9-Jul 0 9.49 29-Aug 0 10.05 19-Oct 0 10.29 10-Jul 0 9.51 30-Aug 2.39 10.05 20-Oct 0 10.29 11-Jul 0 9.48 31-Aug 0.025 10.16 21-Oct 0 10.28 12-Jul 0.051 9.46 1-Sep 0 10.16 22-Oct 0 10.28 13-Jul 2.69 9.46 2-Sep 1.65 10.20 23-Oct 0 10.26 14-Jul 1.65 9.51 3-Sep 0 10.20 24-Oct 0 10.26 15-Jul 0.305 9.51 4-Sep 0 10.23 25-Oct 0 10.25 16-Jul 0 9.51 5-Sep 0.56 10.33 26-Oct 0 10.23 17-Jul 0 9.45 6-Sep 1.6002 10.36 27-Oct 0 10.23

PAGE 73

64 31-Oct 0 10.20 21-Dec 0 10.62 1-Nov 0 10.20 22-Dec 0 10.61 2-Nov 0 10.19 23-Dec 0 10.58 3-Nov 0 10.19 24-Dec 4.37 10.58 4-Nov 0 10.18 25-Dec 0.025 10.60 5-Nov 0 10.17 26-Dec 0 10.63 6-Nov 0.279 10.16 27-Dec 0 10.72 7-Nov 0 10.14 28-Dec 0 10.80 8-Nov 0 10.13 29-Dec 0 10.90 9-Nov 0 10.14 30-Dec 0 10.96 10-Nov 2.210 10.14 31-Dec 1.80 10.99 11-Nov 0 10.14 12-Nov 0.051 10.14 13-Nov 1.02 10.17 14-Nov 0.025 10.19 15-Nov 0.000 10.23 16-Nov 8.407 10.25 17-Nov 0.737 10.30 18-Nov 0 10.35 19-Nov 0 10.43 20-Nov 0 10.49 21-Nov 0.279 10.54 22-Nov 10.57 23-Nov 10.57 24-Nov 10.57 25-Nov 10.57 26-Nov 10.55 27-Nov 10.53 28-Nov 10.50 29-Nov 10.48 30-Nov 10.47 1-Dec 10.45 2-Dec 10.43 3-Dec 10.42 4-Dec 10.41 5-Dec 1.68 10.40 6-Dec 0.432 10.39 7-Dec 0 10.38 8-Dec 0 10.38 9-Dec 1.24 10.38 10-Dec 3.12 10.39 11-Dec 0.025 10.40 12-Dec 0.000 10.42 13-Dec 2.032 10.48 14-Dec 10.50 15-Dec 10.53 16-Dec 10.57 17-Dec 10.60 18-Dec 10.62 19-Dec 10.63 20-Dec 0.635 10.63

PAGE 74

65 Daily precipitation and stage records for 2003 Date ppt (cm) masl 1-Jan 2.4384 11.02 19-Feb 0 11.42 9-Apr 10.85 2-Jan 0 11.04 20-Feb 0 11.72 10-Apr 10.86 3-Jan 0.0254 11.05 21-Feb 0 11.83 11-Apr 10.86 4-Jan 0 11.09 22-Feb 1.6002 11.84 12-Apr 10.85 5-Jan 0 11.16 23-Feb 0 11.78 13-Apr 10.80 6-Jan 0 11.18 24-Feb 0 11.63 14-Apr 10.75 7-Jan 0 11.17 25-Feb 0 11.54 15-Apr 10.71 8-Jan 0 11.13 26-Feb 0 11.48 16-Apr 10.68 9-Jan 0 11.07 27-Feb 0.1016 11.43 17-Apr 10.65 10-Jan 0.1524 28-Feb 0.1778 11.40 18-Apr 10.62 11-Jan 0 10.90 1-Mar 1.4478 11.45 19-Apr 10.59 12-Jan 0 10.84 2-Mar 0.4826 11.65 20-Apr 10.56 13-Jan 0.1016 10.81 3-Mar 0.1524 11.87 21-Apr 10.55 14-Jan 0.1524 10.77 4-Mar 1.0414 22-Apr 10.54 15-Jan 0 10.74 5-Mar 0.0508 23-Apr 10.51 16-Jan 0 10.71 6-Mar 0 24-Apr 10.50 17-Jan 0 10.68 7-Mar 3.2258 13.11 25-Apr 10.50 18-Jan 0 10.65 8-Mar 0 26-Apr 10.48 19-Jan 0 9-Mar 3.4544 27-Apr 10.48 20-Jan 0 10.60 10-Mar 0.0254 13.41 28-Apr 10.49 21-Jan 0 10.59 11-Mar 14.01 29-Apr 10.49 22-Jan 0.5334 10.57 12-Mar 14.22 30-Apr 10.49 23-Jan 0.0254 10.56 13-Mar 14.34 1-May 0.4826 10.48 24-Jan 0 10.54 14-Mar 14.16 2-May 10.45 25-Jan 0 10.53 15-Mar 13.86 3-May 26-Jan 0 10.53 16-Mar 13.58 4-May 10.43 27-Jan 0 10.49 17-Mar 1.27 13.25 5-May 10.43 28-Jan 0 10.50 18-Mar 0.0508 6-May 10.41 29-Jan 0 10.48 19-Mar 7-May 10.40 30-Jan 0 10.48 20-Mar 0.0762 8-May 10.39 31-Jan 0 10.46 21-Mar 0.0254 12.19 9-May 10.39 1-Feb 0 10.46 22-Mar 11.99 10-May 2-Feb 0 10.46 23-Mar 0.1016 11.87 11-May 3-Feb 0 10.45 24-Mar 11.69 12-May 0.2794 4-Feb 1.27 10.45 25-Mar 11.55 13-May 5-Feb 0 10.43 26-Mar 11.42 14-May 6-Feb 0 10.43 27-Mar 2.54 11.32 15-May 7-Feb 4.826 10.46 28-Mar 11.27 16-May 8-Feb 0.0762 10.53 29-Mar 11.29 17-May 9-Feb 1.1684 10.60 30-Mar 11.32 18-May 1.143 10-Feb 0.3556 10.78 31-Mar 1.27 11.29 19-May 3.6322 11-Feb 0 10.78 1-Apr 11.28 20-May 0.254 12-Feb 0 10.83 2-Apr 11.25 13-Feb 0 10.86 3-Apr 11.25 14-Feb 0 10.87 4-Apr 11.19 15-Feb 0 10.88 5-Apr 11.12 16-Feb 7.3914 10.86 6-Apr 11.05 17-Feb 0.1778 11.04 7-Apr 10.97 18-Feb 0 11.16 8-Apr 10.90

PAGE 75

APPENDIX B WATER CHEMISTRY DATA

PAGE 76

67 Data collected between May 8,2002 to June 5, 2002 Location Cl SO4 Ca Na Mg Alk. SiO2 18O Sr 87Sr/86Sr NO3 PO4 Temp. pH Cond.DO (mg/L) (mg/L) (mg/L) (mg/L)mg/L)(mg/L )(mg/L)(‰) (mg/L) (mg/L)(mg/L)(C) Vinzants 12.7 22.7 46.6 7.5 12.4 157.514.0 -2.3 0.7 0.707942 0.018 0.120 24.6 7.72 358 4.98 Sink 23.8 16.7 38.0 11.5 10.7 126.55.3 -1.7 0.4 0.708141 0.009 0.132 26.6 7.37 319 1.72 Ogden 58.9 139.5 72.6 28.8 20.8 151.38.4 -1.9 1.7 0.707892 0.009 0.132 26.6 7.21 645 0.1 Big 22.1 65.3 60.7 12.6 15.0 163.413.2 -2.6 1.2 0.707882 0 0.122 25.2 7.42 463 2.78 Paraners 22.8 64.2 59.4 12.7 14.7 136.012.7 -2.6 1.1 0.707858 0.02 0.125 26 7.61 456 1.33 Jim 23.1 63.2 59.0 12.8 14.7 154.613.1 -2.6 1.0 0.707890 0 0.127 24.7 7.13 454 1.7 Jug 25.3 58.9 56.1 13.5 14.1 154.112.3 -2.4 1.0 0.707877 0 0.122 24.4 7.14 440 2.64 Hawg 25.6 57.9 55.1 13.5 13.9 135.712.4 -2.3 1.0 0.707907 0.009 0.127 24.3 7.03 437 0.51 Twohole 27.2 55.4 53.9 14.1 13.7 118.711.7 -2.1 1.0 0.707888 0.105 0.131 24.6 7.03 429 0.23 Sweetwater 22.9 95.8 69.1 13.1 16.4 148.314.7 -2.9 1.5 0.707859 0.119 0.103 24.1 7.65 512 0.72 Rise 30.3 73.2 57.8 15.1 14.1 130.511.4 -2.3 1.1 0.707899 0.077 0.116 22.8 7.33 458 2.04 Treehouse 64.8 12.8 14.1 140.3 -1.9 1.4 0.707812 27.5 7.5 475 5.44 Hornsby 12.8 79.2 74.4 9.4 13.5 184.912.4 -3.1 1.5 0.707849 0.067 0.052 23 7.54 498 6.46

PAGE 77

68 Data collected between January 15,2003 and February 5,2003 Location Cl SO4 Ca Na Mg K Alk. SiO2 18O Sr 87Sr/86SrNO3 PO4 Temp.pH Cond. DO (mg/L) (mg/L) (mg/L) (mg/L)mg/L)(mg/L)(mg/L)(mg/L)(‰) (mg/L) (mg/L)(mg/L)(C) Vinzants 23.0 6.8 11.5 10.703.45 1.25 12 7.8 -2.8 0.0290.14 10.4 5.2 131.1 8.7 Sink 22.8 6.5 11.1 10.503.37 1.23 12 7.8 -2.8850.468 0.70792 0.0310.14 10 6.1 131.1 8.8 Ogden 23.1 7.8 11.9 10.703.56 1.25 12 7.6 -2.8250.070 0.70883 0.0310.15 10.2 6.36 136 8.6 Big 23.6 17.7 14.9 10.604.04 1.19 16 8.2 -2.9 0.0310.15 10 6.03 160.7 8.2 Ravine 23.8 17.6 16.1 11.204.33 1.23 16 8.6 -2.825 0.03 0.15 11 5.23 164.9 8.3 Paraners 23.3 17.4 15.0 10.704.09 1.20 16 7.8 -2.77 0.0270.14 10 6.2 160.7 8.3 Jim 23.6 16.4 16.0 11.004.29 1.23 16 8.7 -2.73 0.188 0.70821 0.0310.15 10.6 6.29 164.8 8.1 Jug 23.1 18.8 16.0 11.104.30 1.24 16 8.8 -2.725 0.0280.15 10.9 6.02 163.9 5.6 Hawg 16.7 24.0 60.0 9.17 6.26 0.94 136 13.3 -2.4850.174 0.70819 -0.0010.16 15.5 6.66 368 1.6 Two Hole 26.3 28.7 28.7 12.105.57 1.43 52 10.5 -2.28 0.338 0.70805 0.0040.16 12 6.66 235 3.1 Sweetwater 23.5 18.1 16.0 10.704.22 1.21 20 8.5 -2.69 0.188 0.70811 0.0280.14 10 6.2 165.4 4.6 Rise 22.8 15.3 15.5 10.404.08 1.21 20 8.6 -2.83 0.180 0.70819 0.0240.15 11 6.19 161.9 4.3 Treehouse 26.2 20.1 19.1 11.004.62 1.26 24 8.8 -2.8 0.0410.15 11.1 6.48 182.4 7.3 Hornsby 19.7 35.7 32.6 9.93 5.97 1.24 56 10 -2.81 0.03 0.13 13.2 6.54 252 2.5 Well #1 8.4 2.0 107.0 3.79 3.25 0.24 252 10.5 -3.38 1.626 0.70784 0 0.09 21.5 6.7 485 0.2 Well #2 46.0 305.0 152.0 26.9027.401.71 206 15 -3 0.0650.59 26.3 6.76 1009 2.3 Well #7 13.4 14.6 101.0 7.06 5.41 0.62 262 9.1 -3.58 0.488 0.70793 0.0020.14 20.4 6.51 530 0.2

PAGE 78

69 Data collected March 3, 2003 to March 19,2003 Location Cl SO4 Ca Na Mg K Alk. SiO2 18O Sr 87Sr/86Sr NO3 PO4 Temp. pH Cond. DO (mg/L) (mg/L) (mg/L) (mg/L)mg/L)(mg/L)(mg/L)(mg/L)(‰) (mg/L) (mg/L)(mg/L) (C) Vinzants 14.3 2 6.96 5.5 2.15 1.34 10 3.3 -3.32 0.0080.217 17.2 5.98 84.5 6.08 Sink 12.7 2 7.64 6.29 2.41 1.72 8 3.4 -2.460.025 0.709481 0.0160.286 17.4 5.35 79.2 5.77 Ogden 12.8 2 7.59 5.76 2.32 1.55 16 3 -2.530.02730.709434 0.0150.294 17.2 4.98 79 5.7 Ravine 12.8 2 7.92 5.82 2.36 1.53 10 3.1 -2.670.03790.709099 0.02 0.294 17.3 4.75 81.9 5.52 Paraners 12.9 2 8.14 6.29 2.46 1.68 10 3 -2.49 0.0170.294 17.1 5.04 82 5.54 Jim 13.1 2 8.31 6.16 2.48 1.69 8 2.6 -2.360.03360.709091 0.0120.300 17.4 5.05 82.6 5.43 Jug 13.2 2 8.48 6.32 2.48 1.68 12 2.7 -2.37 0.0130.292 17.6 4.9 84.3 5.5 Hawg 15.7 7.2 16.1 7.5 2.93 1.57 28 3 -2.54 0.0220.253 18.4 5.58 123.2 2.88 Sweetwater 11.2 2 7.6 5.34 2.22 1.44 8 2.9 -2.450.03910.709101 0.0080.305 16.5 4.78 71.5 5.56 Rise 11.3 2 8.12 5.71 2.32 1.6 8 2.8 -2.350.03010.707725 0.0120.338 17.1 4.67 72.5 5.54 Hornsby 11.6 2 11.4 5.74 2.45 1.7 16 3.1 -2.42 0.0210.292 18.8 5.11 86.8 4.35 Well #1 8.3 2 113 4.33 2.21 0.09 242 7.8 -3.340.88990.7078735 0.0010.075 21.9 6.8 484 0.33 Well #2 21.3 114 79.1 14.1 13.8 1.43 102 1.4 -2.760.11030.7080629 0 0.142 25.8 7.08 488 0.14 Well #3 5.3 3 91.4 3.68 1.82 0.09 212 10 -4.110.148 0.708141 0.0570.063 21.8 6.93 409 1.64 Well #4 8.2 4.4 91 5.08 2.18 0.24 206 8.2 -4.160.148 0.708127 0.0340.063 21.4 6.97 408 4.06 Well #6 7.2 2.5 100 3.87 1.89 0.33 212 6.1 -4.33 0 0.078 21.1 6.8 435 0.22 Well #7 13.1 17.7 87.2 7.53 5.78 0.57 196 6.9 -3.71 0.0040.151 20.7 6.98 422 0.16

PAGE 79

70Data collected April 28, 2003 to May 1, 2003 Location Cl SO4 Ca Na Mg K Alk. SiO2 Sr 87Sr/86Sr 18O NO3 PO4 Temp.pH Cond.DO (mg/L) (mg/L) (mg/L) (mg/L)(mg/L)(mg/L)(m g/L)(mg/L)(mg/L) (‰) (mg/L)(mg/L)(C) Vinzants 15.1 13.1 33.7 7.27 6.79 0.94 77 11.8 -2.59 0.301 0.255 20.8 7.06 230 3.24 Sink 15.5 12.8 35 7.38 6.8 0.94 78.9 12.2 0.2670.708033 -2.33 0.344 0.244 21.4 7.07 235 3.8 Ogden 16.3 16.5 36 8.24 7.36 1 81.5 12 -2.21 0.383 0.233 22.1 6.99 242 3.74 Ravine 19.2 50 52 9.96 9.68 0.96 99.3 13.1 0.7960.707894 -1.96 0.329 0.211 22.7 6.99 225 3.74 Paraners 17.5 27.9 43.1 8.57 8.23 0.96 97.2 12.5 -2.38 0.336 0.238 22 7.04 300 2.57 Jim 17.1 23.8 42.3 8.45 8.07 0.97 95.1 12.6 0.4930.707955 -2.58 0.322 0.241 22.1 7.02 293 2.45 Jug 17.3 27.7 40 8.31 7.64 0.96 90.7 12.4 -2.68 0.395 0.227 22.1 7.02 287 2.1 Hawg 18.7 37.5 43.8 8.85 8.23 0.96 90 11.9 0.5350.707937 -2.73 0.307 0.227 21.8 6.67 295 2.1 TwoHole 19.7 40.9 47.1 9.29 8.76 1.01 96 12.2 0.6150.707927 -2.72 0.366 0.238 22.1 6.85 315 2.04 Sweetwater 17.8 49.3 51.9 8.85 9.27 0.96 112 12.9 0.7640.707887 -2.71 0.367 0.189 22.1 6.76 338 1.69 Rise 18.2 62.9 57.1 9.36 9.82 0.96 108 12.6 -2.96 0.36 0.172 21.8 6.84 377 1.14 Treehouse 14.4 64 64.8 8.59 10.1 1.03 124 12.8 -3.05 0.423 0.180 22.4 7.06 400 2 Hornsby 12.2 64.6 69.5 7.72 10.1 1.05 144 13.2 -3.11 0.349 0.153 23.1 6.99 420 0.31 Well #1 7.3 20 97.6 3.26 1.39 0.15 234 9.4 0.1080.708194 -3.34 0.014 0.095 26 6.92 907 0.22 Well #2 40.7 242 148 25.2 27.8 1.87 188 13.5 2.16 0.707843 -2.83 0.022 0.111 22 6.87 448 0.13 Well #3 5.6 10 88 3.31 1.67 0.41 208 10.5 -4 0.07 0.078 21.8 6.9 416 1.27 Well #4 8.2 10 94.5 4.41 1.51 0.31 208 8.1 0.1290.708112 -4.25 0.054 0.078 21.7 6.92 423 2.08 Well #6 6 20 90.9 2.95 1.22 0.37 210 5.8 -4.39 0.001 0.045 21.2 6.94 416 0.17 Well #7 14.9 11.6 111 4.41 4.61 0.62 256 10.2 0.5080.7079129-3.68 0.035 0.149 20.8 6.79 538 0.14

PAGE 80

71 LIST OF REFERENCES Atkinson, T.C., 1977, Diffuse flow and conduit flow in limestone terrain in the Mendip Hills, Somerset (Great Britain), J. Hydrol., 35: 93-110. Burke, W.J., Denison, R.E., Heatherington, E. A., Koepnik, R.B., Nelson, H.F., Otto, J.B. Variation of seawater 87Sr/86Sr throughout Phanerozoi c time, Geology, 10: 516519. Cao, H., Cowart, J.B., Osmond, J.K., 1999, Ur anium and strontium isotopic geochemistry of karst waters, Leon Sinks Geological Area, Leon County, Florida, Cave and Karst Science, 26: 101-106. Dean, R.W., 1999, Surface and groundwater mixi ng in a karst aquifer: An example from the Floridan Aquifer: Gaines ville, Florida, University of Florida, MS Thesis 74p. DePaolo, D.J. and Ingram, B.L., 1985, High-resolution stratigraphy with strontium isotopes, Science, 227: 938-941. Desmarais, K. and Rojstaczer, S., 2002, Infe rring source waters from measurements of carbonate spring response to storms, J. Hydrol., 260: 118-134. Dogramaci, S.S. and Herczeg, A.L., 2002, St rontium and carbon isotope constraints on carbonate-solution interacti ons and inter-aquifer mixi ng in groundwaters of the Semi-arid Murray Basin, Austra lia, J. Hydrol., 262: 50-67. Faure, G., 1986, Principles of Isotope Geology, John Wiley & Sons, New York, 464p. Frederickson, G.C. and Criss, R.E., 1999, Isotope hydrology and residence times of the unimpounded Meramec River Basin, Missouri, Chem. Geol., 157: 303-317. Ginn, B., 2002, Using temperature and water elevation measurements to model conduit properties in karst aquifers : An example for the Santa Fe River Sink/Rise System, Florida, University of Florida, 23p. Greene, E.A., 1997, Tracing recharge from sinking streams over spatial dimensions of kilometers in a karst aquifer, Ground Water, 35: 898-904. Grozos, M., Ceryak, R., Allison, D., Cooper, R., Weinberg, M., Macesich, M., Enright, M.M, and Rupert, F., 1992, Carbonate Units of the intermediate aquifer system in the Suwannee River water Management Di strict, Florida. Florida Geological Survey Open File Report No. 54.

PAGE 81

72 Grubbs, J.W., 1998, Recharge rates to the Uppe r Floridan Aquifer in the Suwannee River Water Management District: Florid a, U.S. Geological Survey, 30 p. Halihan, T., Wicks, C.M., and Engelin, J.F., 1998, Physical response of a karst drainage Basin to flood pulses: exam ple of the Devil’s Icebox cave system (MO, USA), J. Hydrol., 204: 24-36. Hess, J., Bender, M.L., Schilling, J.G., 1986, Evolution of the ratios of strontium-87 to Strontium-86 in seawater from Cretace ous to Present, Science, 231: 979-984. Hisert, R.A., 1994, A multiple tracer appro ach to determine the ground and surface water relationships in the wester n Santa Fe River, Columbia County, Florida: Ph.D. Dissertation, University of Florida, 212p. Hunn, J.D., and Slack, L.J., 1983, Water resource s of the Santa Fe River Basin, Florida, U.S. Geological Survey, Water-Resources Investigations Report 83-4075, 105 p. Johnson, T.M., Roback, R.C., McLing, T., Bullen, T.D., DePaolo, D.J., Doughty, C., Hunt, R.J., Smith, R.W., Cecil, L.D ., Murrell, M.T., 2000, Ground water “fast paths” in the Snake River Plain aquifer: Radiogenic isotope ratios as natural groundwater tracers, Geology, 28: 871-874. Karst Waters Institute, Karst Water Institute Home Page, 2001, 24 June 2002
PAGE 82

73 Martin, J.B., and Dean, R.W., 1999, Temperatur e as a natural tracer of short residence times for ground water in karst aquifers: In: A.N. Palmer, M.V. Palmer and I.D. Sasowsky (Eds), Karst Modeling, Karst Waters Institute Special Publication, Charlestown, West Virginia, No. 5: p. 236-242. Martin J.B., and Dean, R.W., 2001, Exchange of water between conduits and matrix in the Floridan Aquifer, Chem. Geol., 179: 145-155. Martin, J.B. and Screaton, E.J., 2000, Exch ange of matrix and conduit water with examples from the Floridan Aquifer: U.S. Geological Survey Water-Resources Investigations Report 01-4011, p. 38-44. Martin, J.M., 2003, Quantification of the ma trix hydraulic conductivity in the Santa Fe River Sink/Rise system with implicati on on the exchange of water between the matrix and conduits: Gainesville, Fl, Univ ersity of Florida, MS Thesis, 80p. McNutt, R.H., Frape, S.K., Fritz, P., Jones, M.G., and MacDonald, I.M., 1990, The 87Sr/86Sr values of Canadian Shield brines and fracture minerals with applications to groundwater mixing, fracture history, and geochronology, Geochim. et Cosmo. Acta., 54: 205-215. Meyer, F.W., 1963, Reconnaissance of the geology and ground water resources of C\Columbia, County, Florida, Florida Geol ogical Survey, Report of Investigations No. 30, Padilla, A., Pulido-Bosch, A., and Mangin, A., 1994, Relative importance of baseflow and quickflow from hydrographs of karst spring, Ground Water, 32: 267-277. Parkhurst, D.L. and Appelo, C.A., 1999, Us er’s guide to PHREEQC (version 2); a computer program for speciation, batch -reaction, one-dimensional transport and inverse geochemical calculations, U.S. Geological Survey, Water Resources Investigations Report 99-4259, 312 p. Redwine, J.C. and Howell, J.R., 2002, Geoc hemical methods for distinguishing surface water from groundwater in the K nox Aquifer System, Env. Geo., 42: 485-491. Rose, S., 1989, The heavy-metal adsorption ch aracteristics of Hawthorn Formation (Fl, USA) sediments, Chem Geol. 74: 365-370. Scanlon, R.R. and Thrailkill, J., 1987, Chem ical similarities am ong physically distinct spring types in a karst terrain, J. Hydrol., 89: 259-279. Scott, T.M., 1992, A geological overview of Florida. Florida Geological Survey, Open File Report No. 50. Screaton, E.J., Martin, J.B., Ginn, B., a nd Smith, L.A., 2004, Conduit properties and karstification in the Santa Fe River SinkRise system of the Floridan Aquifer, Ground Water, 42: 338-346.

PAGE 83

74 Shuster, E.T., and White, W.B., 1971, Seasonal fl uctuations in the chemistry of limestone Springs: A possible means for characte rizing carbonate aquifers, J. Hydrol., 14: 93128. Skirvin, R.T., 1962, The underground course of the Santa Fe River near High Springs, Florida: Gainesville, Fl, Universi ty of Florida, MS Thesis, 55p. Smart, C.C. and Ford, D.C., 1986, Structure a nd function of a conduit aquifer, Can. J. of Earth Science, 23: 919-929. Smith, L.A., Martin, J.B., and Screaton, E.J ., 2001, Surface water control of gradients in the Floridan Aquifer: Observations from th e Santa Fe River Sink-Rise system,: In: J.B. Martin, C.M. Wicks, and I.D. Sasowsky (Eds), Hydrogeology and Biology of Post-Paleozoic Carbonate Aquifers, Karst Waters Institute Special Publication, Charlestown, West Virginia, No. 7: p. 44-48. Smith, L.A., 2004, Using 222Rn as a tracer of mixing be tween surface and ground water in the Santa Fe River Sink/Rise system: Gainesville, Fl, University of Florida, MS Thesis, 63p. Thrailkill, J., 1985, Flow in a limestone a quifer as determined from water tracing and water level in wells, J. Hydrol., 78: 123-136. Thrailkill, J., Sullivan, S.B., and Gouzie, D.R., 1991, Flow parameters in shallow conduit-flow carbonate aquifer, Inner Bluegrass Karst Re gion, Kentucky, USA, J. Hydrol., 129: 87-108. Vervier, P., 1990, Hydrochemical characteri zation of the water dynamics of a karstic system, J. Hydrol., 121: 103-117. White, W.A., 1970, The geomorphology of the Fl orida Peninsula, Geological Bulletin No. 51, Florida Geological Survey. White, W.B., 1999, Conceptual models for karstic aquifers: In: A.N. Palmer, M.V. Palmer and I.D. Sasowsky (Eds), Karst Modeling, Karst Waters Institute Special Publication, Charlestown, We st Virginia, No 5: 11-16. Woods, T.L., Fullagar, P.D., Spruill, R.K., Sutton, L.C., 2000, Strontium isotopes and Major elements as tracers of ground wate r evolution: Example from the Upper Castle Hayne Aquifer of North Ca rolina, Ground Water, 38: 762-771. Worthington, S.R.H., 1999, A comprehensiv e strategy for understanding flow in carbonate aquifers: In: A.N. Palmer, M.V. Palmer and I.D. Sasowsky (Eds), Karst Modeling, Karst Waters Institute Special Publication, Charlestown, West Virginia, No 5: 11-16.

PAGE 84

75 BIOGRAPHICAL SKETCH Brooke Sprouse was born in Huntington, We st Virginia, in October 1977. After graduating from Huntington East H.S. in 1995 she attended Furman University in Greenville, South Carolina. She graduated with a B.S. in earth and environmental sciences in 1999. Following a two-year hiatus from school, she began graduate studies in geology at the University of Florida in th e fall of 2001. She is currently seeking employment in the environmental/geological scie nces fields in Charlotte, North Carolina.


Permanent Link: http://ufdc.ufl.edu/UFE0005343/00001

Material Information

Title: Chemical and Isotopic Evidence for Exchange of Water between Conduit and Matrix in a Karst Aquifer: An Example from the Santa Fe River Sink/Rise System
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0005343:00001

Permanent Link: http://ufdc.ufl.edu/UFE0005343/00001

Material Information

Title: Chemical and Isotopic Evidence for Exchange of Water between Conduit and Matrix in a Karst Aquifer: An Example from the Santa Fe River Sink/Rise System
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0005343:00001


This item has the following downloads:


Full Text












CHEMICAL AND ISOTOPIC EVIDENCE FOR EXCHANGE OF WATER
BETWEEN CONDUIT AND MATRIX IN A KARST AQUIFER: AN EXAMPLE
FROM THE SANTA FE RIVER SINK/RISE SYSTEM
















By

BROOKE ELIZABETH SPROUSE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004


































Copyright 2004

by

Brooke E. Sprouse















ACKNOWLEDGMENTS

I would like to thank my advisor Jon Martin for all of his help and insight with this

project. I would especially like to thank Lauren Smith and Jennifer Martin for all of their

support and willingness to battle the ticks, mosquitoes, and gators to help me collect the

water samples. Thanks for helping me keep my sanity. Finally I would like to thank my

parents for all of their love and support throughout all my endeavors both here at UF and

throughout my life.















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES ....................................................... ............ .. ............ vi

LIST OF FIGURE S ......... ....................... ............. ........... vii

ABSTRACT ........ ........................... .. ...... .......... .......... vii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Background Studies ............................................... ............. .... .......... 2
Study A rea/G eologic B background ......................................... .......................................7
S tu dy A rea .................. .................................................... ............... 7
Tem perature and Clim ate .................................. ...........................................7
P hysiography ................................................... ..................... 9
Stratigraphy/H ydrostratigraphy................................... .............................. ...... 9
Previous Studies of the Santa Fe River .................................................... ............... 13

2 M E T H O D S .......................................................................................................1 6

W after Sam pling ................................................................16
Surface W after .................. ................................. ....... .. ............ 16
G round W after ..................................... .............. ....................... 18
A n aly se s...................................................... 19
Field M easurem ents ................ .............................. .... .... .................. 19
Chemical and Isotopic M easurements...................... ...... ....................20
Com puter M modeling .......... ........ ........... ................. .... ... .....21

3 R E S U L T S .............................................................................2 2

Stage, Precipitation and Discharge ....... .................... ........................22
D brought Stage ....................................................... 24
Field m easurem ents ......... .. ......... .. .... ................. ......... .. 24
Chem ical and isotopic com position ................................... ............... 24
Intermediate Stages........... ....... ................................ .. 25
Surface W after ............. .................. ......... .. .. ............ 25
Field measurements ....... .................. ............ ..... ..... ................ 25










Chemical and isotopic compositions............... ...........................................26
G round W after ................................................................. ... ......... 28
Field m easurem ents ......................................................... .. ................28
Chemical and isotopic compositions............... ...........................................28
Flood Stage ......................... .............................29
Surface W after ........... ................ .. ... ........... ...... .... ........29
Field m measurements .............. ................. ........................... 29
Chemical and isotopic compositions ...... ..... ............ .................... ...30
Ground Water ............. ...................... .......................30
Field m easurem ents .............. ................. ........................... 30
Chemical and isotopic compositions ...... ..... ............ .................... ...31

4 DISCU SSION ........... ................ .. ......... ............ ... ..... 42

Variation in Chemical and Isotopic Composition ...............................................42
Chemical Variations in Karst W windows and W ells..........................................43
Isotopic Variations in the Karst Windows and Wells ......................................48
Sr M ixing M odel ............................................................. ...............50
Saturation and M ixing Calculations ........................................ ....................... 54
N utrient L loading ........................................... ............. ...... ...............56

5 CON CLU SION S .................................. .. .......... .. .............59

APPENDIX

A DAILY PRECIPITATION AND STAGE RECORDS ..........................................61

B W ATER CHEM ISTRY DATA ........................................................ ............... 66

B IO G R A PH IC A L SK E TCH ..................................................................... ..................75






















v
















LIST OF TABLES


Table page

1-1 Stratigraphic and hydrostratigraphic units of the Santa Fe River Basin...............13

2-1 Sample collection dates and the river stage at the time of collection.................. 17

2-2 Total depth and location of the wells.................... ............ .. ............. 18

2-3 River stage levels at the time of ground water sample collections.....................19

3-1 Water levels and discharge measurements at the Sink and Rise .........................24

4-1 The HCO3/Ca2+ ratios in the karst windows and the wells................................45

4-2 Percentage of surface water in the karst windows..........................................54

4-3 Percentage of surface water in the wells................................... ............... 55
















LIST OF FIGURES


Figure page

1-1 M ap of the Santa Fe River B asin. ................. ....................................... .......... 10

1-2 Map of the study area, including all of the karst windows and wells .................. 11

3-1 Water elevation of the Santa Fe River elevation collected at O'Leno State Park .23

3-2 Precipitation amounts collected at O'Leno State Park during the study period ....23

3-3 a& b C a2 concentrations .......................................................................... ...... 32

3-4 a& b. Sr2 concentrations .............................................. ............................. 33

3-5 a& b .M g2+ concentrations ........................................................... .....................34

3.6 a& b N a+ concentrations ............................................................ .....................35

3-7 a& b C con centration s............................................................... .....................36

3.8 a& b. S0 42 concentrations............................... ........................................... 37

3-9 a& b. A lkalinity concentrations ........................................ ......................... 38

3-10 a& b.Saturation Index Ratios....... ... ............ .................... ........... 39

3-11 a& b 6180 values ....................... .... ........................... .. ...... .... ...... ...... 40

3-12 a& b 7Sr/ 6Sr ratios ...................... .. ........ ................ ............ .... ............4 1

4-1 Graph of the Na/Cl ratios of the karst windows for each sample period ..............46

4-2. 87Sr/86Sr vs. Sr and 87Sr/86Sr vs. 1/Sr models for all karst windows and wells.......53
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

CHEMICAL AND ISOTOPIC EVIDENCE FOR EXCHANGE OF WATER BETWEEN
CONDUIT AND MATRIX IN A KARST AQUIFER: AN EXAMPLE FROM THE
SANTA FE RIVER SINK/RISE SYSTEM

By

Brooke Sprouse

August 2004

Chair: Jonathan B. Martin
Major Department: Geological Sciences

Karst aquifers are characterized by conduit porosity, fractures, and intergranular

matrix porosity. Although classical karst aquifers have low matrix porosity, where

permeability is high, ground water may exchange between conduits and matrix porosity.

Because conduits can be filled with surface water entering through a sink or swallet this

exchange may lead to contamination of the aquifer from the surface. An important

process of karst hydrogeology is the extent that surface and ground water exchange,

particularly by mixing of matrix and conduit water, and the processes that can control this

mixing.

In North-central Florida the Upper Santa Fe River enters a 36m deep sinkhole at

the River Sink. Less than 8 km south of the sink the headwaters of the Lower Santa Fe

River occurs at the River Rise. Between the Sink and the Rise, numerous karst windows

provide an opportunity to examine the movement of water between the conduit and the









matrix. In this study, the Sink, Rise, 12 karst windows, and 6 monitoring wells were

sampled four times between May 2002 and May 2003 at drought, baseflow, and flood

conditions. At drought conditions, high solute concentrations in water collected from the

karst windows reflect evaporative loss of water and suggest that conduits contain water

that originates from the matrix. Solute concentrations indicate 45-55% of the water

originates from the matrix on the basis of two end-member mixing models. At flood

conditions, concentrations reflect dilution and movement of surface water through the

conduit and mixing models indicate the karst windows contain almost 100% surface

water. At baseflow conditions the fraction of surface water depends on the antecedent

conditions, specifically if baseflow is preceded by drought or flood conditions.

Following drought conditions, low solute concentrations suggest that increasing water

flow in the conduit and low hydraulic head of the matrix relative to the conduit allows

loss of water to the matrix. At these conditions mixing models suggest that surface water

makes up approximately 90-95% of the water in the karst windows. Following flood

conditions, increased concentrations in the karst windows suggest water flows from the

matrix to the conduit. Mixing models suggest the karst windows contain 75-80% surface

water. Solute concentrations decrease by 50% in one of the wells in the study area during

flooding and may be the result of flow along high permeability zones, which appears to

be affected by loss of water from the conduit. All other wells have solute concentrations

that change little during the study period and appear to be unaffected by surface water.

Based on these data, the water chemistry in the karst windows appears to be affected by

river levels and weather conditions to a greater extent than the water in the matrix.














CHAPTER 1
INTRODUCTION

Karst aquifers provide important water resources worldwide by providing water to

more than 25% of the world's population. In the United States, karst terrains make up

20% of the land surface and karst aquifers supply 40% of the ground water used for

drinking water (Karst Waters Institute, 2001). These aquifers are particularly vulnerable

to contamination because they have a wide range of porosity. Unlike darcian-type

aquifers, heterogeneous porosity develops in karst aquifers through dissolution of matrix

rocks, resulting in large solution channels that typically have turbulent flow.

Karst aquifers are often considered a triple porosity system composed of fractures,

conduits, and intergranular matrix porosity (Worthington, 1999). These three types of

porosity may develop to different degrees depending on karst history and age. Conduits

consist of dissolution features having large apertures, typically 10mm to tens of meters.

Fracture porosity and the intergranular porosity typically have apertures of less than

10mm (White, 1999). Intergranular and fracture porosity can be difficult to distinguish,

therefore, matrix will refer here to both fractures and intergranular porosity. In regions

with low matrix permeability, conduits can transmit most of the potable water. In regions

with elevated matrix porosity and permeability, large volumes of water can flow through

the intergranular porosity of the matrix rocks (Smart and Ford, 1986) and in these areas

most water may be stored in the matrix porosity.

Water can flow rapidly from the surface into sinkholes and then into conduits,

thereby providing allogenic recharge (White, 1999). Mixing of water in conduits and









matrix can lead to rapid contamination of the water in the intergranular porosity, which

may be a major source of potable water. Understanding the extent of exchange between

the two different flow systems within karst aquifers is thus critical to determine the

sources of spring waters and the potential for their contamination.

While there have been numerous studies focusing on karst aquifers, questions

remain about the rates and extent of interactions between surface water and ground water

and the mixing of water between the conduits and the matrix. The purpose of this study

is to examine this interaction and to develop a better understanding of the hydrologic

processes occurring in karst aquifers characterized by high matrix porosity. In addition,

this study will develop techniques that incorporate natural chemical compositions, such

as the use of 87Sr/86Sr, as natural tracers in karst aquifers. This study attempts to answer

several general questions:

1. How does variation in surface water levels influence mixing of water in conduits
and matrix? What is the extent of mixing at various river water levels?

2. Can mixing models be developed for the extent of conduit/matrix exchange at
various conditions?

3. How is the intergranular porosity water chemistry altered through water-rock
interactions as water moves through the conduit and the matrix?

4. How does the distance from the conduit affect matrix water chemistry?

5. What role do antecedent conditions play in both conduit and matrix water
chemistry?

Background Studies

Early studies using a variety of methods including potentiometric maps, dye-tracer

studies, hydrographs, chemographs, and water chemistry (e.g., Shuster and White, 1971;

Atkinson 1977; Thrailkill, 1985; Vervier, 1990; Thrailkill et al., 1991; Padilla, 1994; and

Halihan et al, 1998) found that springs in karst aquifers can be divided into those that are









sourced from conduits in which water flows through large underground passages and

those that are sourced by diffuse flow from intergranular porosity (Shuster and White,

1971 and Atkinson, 1977). Later work (Dean, 1999; Martin and Dean, 2001) showed

that a gradation exists between the two types of springs.

In aquifers, potentiometric maps provide information about the ground water flow

pattern. The presence of conduits in karst aquifers may alter the expected flow pattern

determined by these maps (Thrailkill, 1985) and, as a result, dye-tracer tests are often

used to determine ground water flow patterns in karst aquifers (e.g. Thrailkill, 1985;

Thrailkill et al., 1991). Fluorescein and RhodamineWT are often used as the tracer and

are typically injected directly into a sink or swallet. These dye tracer experiments can

define a ground water basin, conduit geometry, resurgence points from sinks and

swallets, and response to storm events (Thrailkill, 1985 and Thrailkill et al., 1991; Hisert,

1994). Resurgence of dyes provides information about the source of water at a given

spring as well as linear travel time through the system. These studies cannot determine

absolute flow paths, however, but instead provide a straight-line connection between the

injection point and resurgence point.

Storm hydrographs and the chemistry of spring discharge provide additional

information about flow in karst aquifers. Storm hydrographs represent the response of

the spring to the influx of storm water and subsequent recession as springs return to

baseflow conditions. Hydrographs can be compared with changes in spring water

chemistry (chemographs) to distinguish between conduit springs and diffuse flow springs

(Shuster and White, 1973). Chemical studies of spring discharge have also been used to

study surface water/ground water exchange (Redwine and Howell, 2001). Precipitation









and ground water typically have different chemical concentrations as a result of water-

rock reactions of the ground water. Precipitation typically has low concentrations of

solutes and salts; however, ground water may have high solute concentrations due to

water-rock interactions in the matrix.

In some karst aquifers, however, the geometry and geology of the region make

chemical separation of spring types difficult. For example, in the Inner Bluegrass Karst

Region in Kentucky both conduit and diffuse springs are recharged through limestone,

but the conduit springs are associated with karst features, deep integrated conduit systems

and large catchment areas while the diffuse springs are associated with shallow flow

paths and small catchment areas (Scanlon and Thrailkill, 1987). The chemical

compositions of both spring types are similar during low flow due to percolation and

chemical reactions near the recharge zone. During high flow events the compositions are

similar due to surface runoff through sinkholes recharging the conduit springs and short

flow distances to the diffuse springs. Scanlon and Thrailkill (1987) concluded that the

chemical composition of spring water is affected not only by conduit size but also by

recharge type and flow path lengths in regions where the physiographical properties

associated with springs do not differ between conduit-fed and diffuse-fed.

In recent years, isotopic compositions have been used to distinguish between

surface water and ground water in karst aquifers. The primary isotope systems used are

6180 and 6D (e.g., Frederickson and Criss, 1999; Greene, 1997; Lakey and Krothe,

1996). The isotopes of oxygen and hydrogen have been used to trace lateral movements

across a ground water basin (Greene, 1997) as well as spring discharge (Lakey and

Krothe, 1996). When 6180 and 6D values are compared to the established meteoric water









line, sources of waters to springs and the influx of water following a storm can be

determined (Lakey and Krothe, 1996).

Several studies have used 87Sr/86Sr isotopic ratios to study ground water in karst

aquifers (Katz and Bullen, 1996; Katz et al., 1997; Cao et al., 1999; and Woods et al.,

2000). Strontium isotope ratios have been used in studies of surface/ground water mixing

in carbonate systems (Katz and Bullen, 1996; Katz et al., 1998; Cao et al., 1999; Woods

et al., 2000; Dogramaci and Herczeg, 2002) and mixing with other ground water sources

in non-carbonate systems (McNutt et al., 1990; Johnson et al., 2000). Strontium can be

used for these studies because the 87Sr/86Sr ratio of dissolved Sr depends on the amount

of dissolution of the solid aquifer material and its isotopic ratio. The only radiogenic Sr

isotope is 8Sr, which is the product of the decay of 87Rb. Depending on their age, rocks

containing abundant Rb-bearing minerals will have more 87Sr and a higher 87Sr/86Sr ratio

than rocks with few Rb-bearing minerals, such as carbonates (Faure, 1986).

The 87Sr/86Sr ratios can be coupled with Sr2+ concentrations to provide a two end-

member model. Woods et al. (2000) used such a model to provide evidence of mixing

between water in the Upper Castle Hayne limestone aquifer with water in the younger

Surficial Aquifer in the vicinity of the North Carolina coast. In Florida, the ratios have

been used to discriminate ground water sources based on 87Sr/86Sr ratio-age plotting (Cao

et al., 1999) and to examine surface water/ground water interactions (Katz and Bullen,

1996; Katz et al., 1997&1998). In the Leon Sinks Geological Area near Tallahassee,

Florida groundwater samples from wells and the deepest sinkholes have 87Sr/86Sr ratios

that plot along the Sr seawater age curve, suggesting the waters originate from the

Oligocene-age limestones of the Upper Floridan Aquifer (Cao et al., 1999). Samples









from shallow sinkholes have ratios above the modern seawater ratio of 0.70907,

suggesting the waters originate from the younger plastics of the Water Table aquifer (Cao

et al., 1999). Katz and Bullen (1996) and Katz et al. (1997) examined lake water and

ground water interactions in the mantled karst region of North Florida using the 87Sr/86Sr

isotope ratios, and found ground water was associated with less radiogenic Sr than the

water in the lakes. Ground water with high 87Sr/86Sr ratios was interpreted to either have

a source from lake water or from rapid movement through the aquifer limiting

equilibration with the limestone (Katz et al., 1997).

While many methods have been employed in the study of karst aquifer systems, a

number of these have focused on the discharge of a spring or springs to identify

characteristics of the aquifer. In many of these studies, flow paths between surface water

sources (i.e. a swallet) and the spring are unknown. In addition, there have been studies

of surface water/ground water exchange but few studies focusing on the interaction of

water in a porous matrix with water in conduits. A lack of abundant features along the

flow path may be the reason for few detailed studies of conduit/matrix exchange. This

study utilizes the major ion concentrations (Ca2+, Mg2+, Sr2+, Na+, C1-, and SO42-) and

87Sr/86Sr isotopic ratios of water at known recharge and discharge points and from

numerous karst windows and wells along the flow path connecting the recharge and

discharge points. Consequently, the focus of this project is to examine the interaction of

water in the matrix and water in the conduit where numerous karst windows provide

access to the conduit water.









Study Area/Geologic Background

Study Area

The study area is along the Santa Fe River, in north-central Florida the second

longest river flowing across north-central Florida and a major tributary to the Suwannee

River (Figure 1-1). The river basin covers -3500 km2 and occurs in three physiographic

regions, the Northern Highlands to the east and the Central Highlands and Gulf Coast

Lowlands to the west (Hunn and Slack, 1983, Meyer, 1962). The river flows west from

Santa Fe Lake for approximately 50 km then it disappears into a 36 m deep sinkhole

known as the River Sink. Some of the water that enters the Sink eventually reemerges

approximately 8 km downstream at the River Rise (Hisert, 1994); however, at times some

fraction of this water appears to be lost to the matrix as the water travels through the

conduit while at other times, water is gained from the matrix (Hisert, 1994; Dean, 1999;

Martin, 2003; Smith et al., 2002; Smith, 2004). The water discharging at the River Rise is

probably a mixture of water flowing into the Sink and water lost from the matrix to the

conduits (Hisert, 1994). Between the Sink and the Rise there are a number of karst

windows present (Figure 1-2), and their water compositions reflect the physical and

chemical processes that occur between the conduit and the matrix (e.g. Dean, 1999;

Martin and Dean, 1999; Martin, 2003; Smith, 2002; Smith, 2004). This study extends the

previous work through sampling and analysis of previously unsampled karst windows, as

well as six wells recently drilled and completed at depths of the conduits.

Temperature and Climate

North-central Florida is classified as a humid sub-tropical climate (Meyer, 1962),

with an average annual daytime temperature of 210C. During January and December, the

coldest winter months, the average daytime temperature is 140C, and during August, the









warmest month, the daytime temperature averages 27C. The average annual rainfall in

the Santa Fe River basin is 140 cm (Hunn and Slack, 1983). Most rainfall occurs

between June and September, and the least amount of rainfall occurs during the winter

months (Meyer, 1962). In the summer, rainfall commonly comes from short afternoon

thunderstorms that are the result of warm air rising over the land and drawing cool moist

air inland from the Gulf of Mexico and the Atlantic Ocean. In the winter rainfall

commonly comes from extra-tropical frontal systems. These storms typically last longer

than the summer storms but do not occur as frequently.

The study area recently experienced a three-year drought that ended in the late fall

and winter of 2002/2003. During the drought, the average yearly rainfall from January 1,

1999 to December 31, 2001 was 93 cm/year at O'leno State Park, which is approximately

33% less than the annual average (Suwannee River Water Management District

archives). In contrast, in the years preceding the drought (1996-1998) the average

rainfall at O'Leno was 147cm/yr. The river level at O'leno State Park for 1999 to the end

of 2001 ranged between 9.6 and 10.3 meters above sea level (masl). Prior to that, the

average river level at O'Leno State Park from 1994 through 1998 was 10.86 masl

(SRWMD). For most of 2002, the river level was approximately 9.6 masl until mid-

August when the river rose to 10 masl reaching a maximum at the end of December at 11

masl. In the first half of 2002 only 35 cm of precipitation occurred in the study area;

however, between July and December 2002 an additional 70 cm of rainfall fell, resulting

in 105 cm of rain for the year. During the fall and winter of 2002-2003 a moderate El

Nino Southern Oscillation (ENSO) period caused greater than normal precipitation levels









throughout Florida and the Southeast with 60 cm of rain between November 1, 2002 and

the end of March 2003.

Physiography

The Santa Fe River Basin is located along the Western Valley and the Northern

Highlands physiographic provinces (White, 1970). The Western Valley is one of several

large lowland areas within the Central Highlands. Several gaps within the valley allow

drainage to the Gulf Coastal Lowlands, with the Santa Fe River flowing through the High

Springs Gap to a confluence with the Suwannee River (White, 1970). The Northern

Highlands represent the eastern portion of the river basin and have elevations greater than

30 masl. The eastern boundary of the basin is Trail Ridge, a north-south trending ridge

that extends through central Florida. In the center of the basin lies an escarpment called

the Cody Scarp that marks the boundary between the Western Valley and the Northern

Highlands. This scarp is the erosional edge of the Miocene Hawthorn Group and marks

the retreating edge of a formerly high plain that sloped northward. Similar to most

streams crossing the Cody Scarp, the Santa Fe River flows into a sinkhole and reemerges

at a first magnitude spring approximately 5km to the south (Fig. 1-2.).

Stratigraphy/Hydrostratigraphy

The lithology of Florida is composed of carbonate rocks that are pre-Miocene in

age and mixed siliciclastic and carbonate rocks that are Miocene and younger (Table 1-

1), with few outcrops. The oldest exposed units in the study area are Eocene carbonates

of the Ocala Limestone. The lower Ocala is composed of grainstones to packstones and

in some regions may be partially to completely dolomitized (Scott, 1992). The upper

portion of the unit is muddy, granular limestones (packstones to wackestones) and can be

soft and friable (Scott, 1992). Above the Eocene rocks of the Floridan Aquifer are the







10







r"~n- -p-~-- -._ ._ _
840 30* BsI




HAMILTONI

0 MADISON W
BAKER
SUWANNEE COLUMBIA

o o P a r k
300- LAFAYETTE



SUWANNEE
RIVER WATER DIXIE
MANAGEMENT GILCHRI Gainesville
DISTRICT ALACHUA

\ olv VY \ ALEVY 1
LEVY RIVER
So BASIN





0 10 20 30 MILES
O 20 40 KILOMETERS





Figure 1-1. Map of the Santa Fe River Basin with the arrow showing O'Leno State Park
and the study area, which is shown in detail in Figure 1-2. Modified from
Hunn and Slack (1983).


















Big
Sink,
b


Parener's
SSink


Jim's
Sink


Two Hole
Sink


U.S.
441


Regional Ground
water Flow
Direction


0 1
Kilometers
N) Wetlands
SSurface water
(River or sinkhole)

Mapped cave system


-7
U


Road
Wells


Figure 1-2. Map of the study area, including all of the karst windows and wells.
Modified from Ginn (2002).











Miocene-age rocks of the Hawthorn Group. The Hawthorn Group contains primarily

interbedded sands, clayey sands, sandy clays, clays, and carbonates (Grozos et al, 1992)

and is composed of the Hawthorn Formation and the Alachua Formation (Hunn and

Slack, 1983). The upper most units in the study area are Pliocene-Pleistocene sediments

that range from sands, sandy clays and carbonates and in North Florida include the

Nashua Fm., Cypresshead Fm., Miccosukee Fm., and Undifferentiated Pleistocene-

Holocene sediments. The undifferentiated sediments include marine sediments, eolian

sand dunes, fluvial deposits, fresh water carbonates, and sediment mixtures that cover

most of Florida (Scott, 1992).

Ground water is found in three aquifers that include, from upper to lowermost, the

Surficial, Intermediate, and Floridan aquifers. The Surficial Aquifer is a water table

aquifer within Pleistocene-Holocene sands (Hunn and Slack, 1983). Throughout most of

the Santa Fe River basin, the water table is approximately 3 m below land surface;

however, in the eastern portion of the basin the water table may be up to 9 m below land

surface. Within the study area the Surficial Aquifer is approximately 5 m in thickness.

Below the Surficial aquifer the Hawthorn Group carbonates contain the Intermediate

Aquifer. Where present, the Hawthorn Fm. acts as a confining unit to the underlying

Floridan Aquifer, but where missing, the Floridan Aquifer is unconfined and is recharged

directly from the surface. The Floridan Aquifer is the primary source of drinking water

for most of northern Florida. The aquifer consists of porous limestone that can be

divided primarily into five units that are, from oldest to youngest, Lake City Limestone,

Avon Park Limestone, and Ocala Limestone (all Eocene age), Suwannee Limestone









(Oligocene), and in some places lower Miocene limestones. The Eocene Ocala limestone

is the major unit present in the study area.

Table 1-1. Stratigraphic and hydrostratigraphic units of the Santa Fe River Basin.
Age Stratigraphic Hydrostratigraphic Lithology Thickness
Unit Unit (m)

Holocene Undifferentiated Surficial Aquifer Fine sands and 0-25
Pleistocene sediments gravel
Pliocene
Pliocene to Alachua
Miocene Formation Intermediate Interbedded 0-45
Aquifer/Confining sands and clays
Miocene Hawthorn Bed Carbonates
Formation

Oligocene Suwannee
Limestone 325-425
Eocene Ocala, Avon Porous
Park and Floridan Aquifer limestone and
dolomite
Oldsmar
Limestones
Paleocene Cedar Keys Sub-Floridan Limestone with
Formation confining bed some clays and ?
evaporites
Adapted from Meyer (1962), Hunn and Slack (1983), and Dean (1999).

Previous Studies of the Santa Fe River

In one of the earliest studies of the Santa Fe River basin in O'leno State Park

Skirvin (1962) attempted to determine the underground flow path of the river after it

sinks into the subsurface at the River Sink. Based upon the tannic color of water

emerging at the Rise, he suggested that most water entering the Sink reemerges at the

Rise. During this study, the river was temporarily dammed near the Sink, but water

continued to flow from the Rise, suggesting groundwater enters the system between the

Sink and Rise.


between the

Sink and Rise.









Hunn and Slack (1983) published a report on the hydraulic properties and water

quality of the Santa Fe River basin, including the area within O'leno State Park. In the

karstic region of the basin, the Floridan Aquifer is the primary groundwater source and is

recharged directly through rainfall. They found that where the Floridan Aquifer is

unconfined, it supplies groundwater to surface drainage features within the basin through

large springs. In confined regions of the basin, however, discharge to streams is small

and the Santa Fe River receives discharge only downstream from Worthington Springs.

These sources of water are reflected in water quality in the confined and unconfined

regions. Streams in the unconfined western region have greater concentrations of

calcium, magnesium and bicarbonate, and are less tannic than streams in the confined

region to the east. The higher chemical concentrations in the western region appear to

reflect a water source from the Floridan Aquifer.

More comprehensive studies of the Santa Fe River between the Sink and the Rise

have been conducted within the past ten years to better understand the underground flow

of the river, the interaction of surface water and groundwater, and the exchange of water

between conduits and matrix porosity (Hisert, 1994; Dean, 1999; Martin and Dean,

1999; Smith et al, 2002; Martin 2003; Smith, 2004). These studies have used various

natural tracers including 222Rn 6180 values, major element chemistry (i.e. Ca2+, Mg2+ C-

), and temperature and the injected tracer SF6 (Hisert, 1994).

Hisert (1994) was able to determine a connection between the River Sink and

Sweetwater Lake, using an injection of SF6, but required a second injection to connect

Sweetwater Lake to the River Rise (Fig. 1-2). A connection with a single injection

between the Sink and the Rise has never been obtained. Cave divers have recently









verified portions of this conduit system within the park (Old Bellamy Cave Exploration,

2001). In addition to SF6 Hisert (1994) analyzed the chemical composition (Ca2+, Mg2+

Na+, K+, Al3+, 180, and 222Rn) at 23 locations between the Sink and Rise. Through the

various experiments, Hisert (1994) determined that some mixing between surface water

from the conduit and ground water in the matrix occurs. The fraction of ground water

relative to surface water in the karst windows was not determined.

Dean (1999) measured the chemical composition of the water at the Sink,

Sweetwater Lake, the Rise at various river levels and chemical compositions of two

wells, one near the Sink and the other near the Rise. These data suggest that surface

water and ground water mixing varied at different flow conditions with more surface

water flowing into the matrix porosity during high flow periods. Dean (1999) suggested

that high spatial and temporal resolution of surface and ground water sampling are

needed to better understand the extent of mixing and the locations where mixing occurs

in the system.














CHAPTER 2
METHODS

Measurements of physical hydrologic parameters and water chemistry included

daily river stage, daily precipitation, and the chemical composition of 19 water samples.

The staff at O'leno State Park collected the river stage data upstream from the Sink and

precipitation data at the park entrance. These data were obtained through the Suwannee

River Water Management District. Analysis of the water samples included field and

laboratory measurements. Temperature, pH, specific conductance, dissolved oxygen

(DO), and turbidity were each measured in the field. Major ions, dissolved species and

isotopes were measured in the laboratory, including Na K Ca2+, Mg2+, C-, S042-,

NO22-, NO32-, NH3, alkalinity, silica, Sr2 concentration, and 7Sr/86Sr ratios and 6180

values.

Water Sampling

Surface Water

Water samples from the Sink, Rise, nine karst windows, a swallet upstream from

the Sink, and two springs south of the Rise were collected four times between May 2002

and May 2003 at river stage levels varying between drought, high and intermediate

discharge conditions (Table 2-1). The samples were collected over time periods ranging

from several days to a few weeks with samples collected sequentially from upstream to

downstream. The length of time for each sample period was, in some cases, dependent

on weather conditions but an attempt was made to collect the same water parcel entering









the Sink and traveling through the conduit based on previously determined travel times

between the Sink and Rise (Dean, 1999).

Samples were collected from a depth of approximately 0.5 meter and near the

resurgence point of the spring, when possible. A Geotech Geopump 2 peristaltic pump

was attached to PVC tubing to collect the water samples. Prior to sample collection, the

tubing was purged with at least 2L of water from the karst window that was to be

sampled. A free-flow cell was used to collect the purged water as well as to take field

measurements. During each sample period two field duplicate and 1-2 field instrument

blanks were collected. At each location, five separate containers were used to collect the

samples. For the isotopic analyses, samples were collected in 30ml glass Qorpak bottles.

All other samples were collected in polyethylene bottles ranging in size from 125 ml to

1L. Samples to be analyzed for alkalinity, C1-, orthophosphate, NO2-, and SO42- had no

preservatives added, but samples for silica and soluble reactive phosphorous were later

filtered in the laboratory. The NOx, NH3, total phosphorous and total Kjeldahl nitrogen

samples were acidified with sulfuric acid to a pH level of <2 and the sample for the total

metals was acidified with nitric acid to a pH level of <2. All samples were kept in a

cooler on wet ice while in the field and were then refrigerated at 40 C at the laboratory.

Table 2-1. Sample collection dates and the river stage at the time of collection.
River Stage
Sample Period Date (masl)

1 May 8, 2002- June 5,
2002

2 January 15 & 16, 2003 10.74- 10.71

3 Feb. 24 -March 5, 2003 11.63- 12

4 April 28- May 1, 2003 10.49









Ground Water

In January 2003, six 2-inch monitoring wells were drilled throughout O'leno State

Park. Three of the wells were installed along the regional ground water flow path and

three were installed near the location of the conduit between Sweetwater Lake and the

Rise, which has been mapped through cave diving exploration (Figure 1-2). The wells

were drilled to approximately 30 m total depths and screened over a 6 m depth (Table 2-

2).

Table 2-2. Total depth and location of the wells used in this project. Distance is unknown
for wells 2 and 7.
Well Total depth of Screened interval Direction to conduit Distance from
well (m) Conduit
(m) (m)
1 23 23-17 NE 475

2 30 30-24 SW
3 28 28-22 NE 30
4 30 30-23 W 115
6 31 31-25 W 85
7 30 30-24 SW

Between January and May 2003, three sets of water samples were collected from

the monitoring wells. The January set included only wells 1, 2 and 7 while the other two

sets included all six wells. During this time period, the river stage varied between

intermediate river level and high river level (Table 2-3). The first and third sample

periods took place during intermediate stage levels and the second sampling period took

place during the March high river level (Table 2-1). The wells were sampled in

conjunction with the surface water sample sets 2,3, and 4 (Tables 2-1&2-3) and were

generally collected several days following the surface water collections.









The ground water samples were collected using a Redi-Flow2 variable speed

submersible pump. Before sample collection the depth to the water level was measured

to determine the depth at which to place the pump, which was set approximately lm

below the water table. Sampling followed purging of at least three well volumes.

The field measurements of temperature, pH, specific conductance, DO, and

turbidity were made while the wells were purged. Purging was considered complete

when all of the parameters stabilized such that three consecutive measurements of

temperature were within 0.20C, pH were within 0.2 standard units, specific

conductance were within 5.0% of the reading, DO was no greater than 20% of

saturation at the field measured temperature, and turbidity was no greater than 20 NTUs.

When stabilization occurred, five sets of samples were collected for each well in the same

manner as previously described for surface water.

Table 2-3. River stage levels at the time of ground water sample collections.
Sample Period Date River Stage
2 February 5, 2003 10.43 masl
3 March 19, 2003 13 masl
4 April 28, 2003 10.49 masl

Analyses

Field Measurements

Measurements of the field parameters for both surface water and ground water were

made with the same equipment. Measurement of pH was made with an Orion portable

pH meter Model #250A calibrated at the start of each sampling period using 7.0 and 9.0

pH buffers. Specific conductance and temperature were measured with an ATI Orion

portable conductivity meter Model #130. Dissolved oxygen and turbidity were measured









with YSI Model 55 handheld dissolved oxygen and temperature system and LaMotte

2020 turbidimeter, respectively.

Chemical and Isotopic Measurements

For the first sample period, the analyses of the Ca2+, Mg2+, Na C1, and S042- were

carried out using an ion chromatograph at the University of Florida, Gainesville, Florida.

For samples periods 2-4 analyses for the concentration of major ions, nutrients, alkalinity,

and silica were carried out by PPB Environmental Labs, Inc. in Gainesville, Florida.

Analyses were done according to Environmental Protection Agency regulations for each

particular ion. Ca2+, Mg2+, and Na+ were measured using an inductively coupled plasma

(ICP) mass spectrometer and K+ was measured by atomic adsorption.

The analysis of the 6180 values of the samples was done using a Prism II Isotope

Ratio Mass Spectrometer (IRMS) at the Department of Geological Sciences, University

of Florida attached to an automated multiprep preparation system. Two hundred ptL of

each sample were pipetted into a glass vial and the headspace was filled with CO2 gas

within a glove bag. The water was equilibrated with CO2 for 12 hours in the multiprep

system and the samples were analyzed automatically by the mass spectrometer. Results

are reported in standard delta notation relative to SMOW.

Strontium isotope measurements were made to determine the 87Sr/86Sr isotopic

ratios of the waters following separation of Sr using ion exchange columns.

Concentration of Sr2+ was measured using isotope dilution. To determine both the Sr2+

concentration and the 87Sr/86Sr ratio, samples were analyzed on a thermal ionization mass

spectrometer (TIMS). The standard used was NBS 987, which had errors of 0.7-1.1%.









Computer Modeling

The Saturation Index values of the water samples were determined using the

program PHREEQC (Parkhurst and Appelo, 1999). The program uses the chemical

composition and temperature of each water sample to determine the saturation indices of

various minerals including calcite for each sample. PHREEQC can also be used to

determine the amount of calcite needed to dissolve in one liter of water to cause the water

to reach saturation. Additionally, mixing two water samples in given proportions to

calculate a mixture similar to measured samples can be measured to determine mixing

fractions of surface water and ground water at various river levels and conditions.














CHAPTER 3
RESULTS

Stage, Precipitation and Discharge

Before the initial sampling period, the study area experienced a three-year drought,

resulting in river stages ranging between 9.8-9.6 meters above sea level (masl) (Figures

3-1 and 3-2). During the El Nifio event of autumn 2002 through early spring 2003,

114cm of precipitation fell, raising river stage levels to a maximum stage of 14.43 masl

on March 13, 2003. Throughout the study period, the average river level was 10.35 masl.

Based on this, river levels lower than 10 masl are considered low flow conditions and

river levels above 1 Imasl are considered high flow conditions. River levels of

approximately 10.5 are considered baseflow conditions and are intermediate between low

flow and high flow conditions.

Continuous water level measurements collected by Martin (2003) were used to

determine discharge measurements for the Sink and Rise. The Sink discharge was based

on a rating curve obtained from the Suwannee River Water Management District (Rating

No 3. for Station No. 02321898, Santa Fe River at O'Leno State Park) and the discharge

at the Rise was based on a rating curve determined by Screaton et al (2004). During the

drought conditions of May 2002, the calculations yield negative discharge rates at the

Sink and the Rise (Table 3-1), indicating that the rating curves are not suitable for low

water levels because the river is not actively flowing. During this time no water enters

the Sink because the entire river was captured by the sinkhole at Vinzant's Landing. The










highest discharge measurements for the Sink and Rise occurred during the March flood

event and were 45.41 and 40.2 m3/s, respectively.

River Stage


15.00
S14.00 3
S13.00 2 4
E 12.00 -
., 11.00 1
S 10.00
9.00
8.00
N NNC N N (N (N (N (N (N (N (N CO CO CO CO CO
0 00 0 0 0 0 0 0 0 0 0 00 0 0 0 0
Si(N M LO CO r- CO O O 0 N (N M q LO

Date


Figure 3-1. Graph of the Santa Fe River elevation collected at O'Leno State Park during
the course of the study period. The arrows indicate the times of sampling.


Precipitation Amounts


-.10
E
C
8 6
5 4
"' 2
o. 0


C) C) C) C) C) C) C) C) C) C) C) C) CO CO CO CO CO
0 00 0 0 0 0 0 0 0 0 0 0 00 0 0
'- C1 C M LO (0 N CO O( 0 -- WN MO C LO

Date


Figure 3-2. Graph of precipitation collected at O'Leno State Park during the study period.
The arrows indicate the times of sampling.


S3
2 4
1h 2l.. ~~ 1^^ l^~~









Table 3-1. Water levels and discharge measurements at the Sink and Rise at the time of
sampling.
Water Level Discharge, Q
Location Date () (M 3/s)
(m) (m /s)
5/14/02 9.76 -11.42
Sink 1/15/03 10.31 11.61
Sink
3/3/03 11.95 45.41
4/28/03 10.49 3.7
5/14/02 9.31 -1.39
1/17/03 10.11 11.2
Rise
3/5/03 11.11 40.2
5/1/03 9.94 7.71

Drought Stage

Field measurements

Vinzant's Landing, the River Sink, River Rise, eight karst windows and two

springs were sampled during the drought conditions. Conductivity values ranged from a

low of 319 [tS at the Sink to a high of 645 [tS at Ogden Lake. Water temperatures were

230C 280C with the Rise having the lowest temperature and Treehouse Spring the

highest. The lowest pH value was 7.03 at both Hawg and Twohole and highest pH value

was 7.72 at Vinzants Landing. The dissolved oxygen (DO) content ranged from 0.1

mg/L at Ogden Lake, which at the time of sampling was nearly covered with duckweed,

to 6.45 mg/L at Hornsby Spring.

Chemical and isotopic composition

The concentrations of cations were highest during the low river stage of the

drought. The chemical compositions of the karst windows between the Sink and Rise

were similar along the flow path with the exception of Ogden Lake, which had higher

concentrations of all solutes compared with the other karst windows. In most cases, the

lowest cation concentrations occurred at the River Sink or Vinzant's Landing. The


River Sink or Vinzant's Landing. The









concentrations of most karst windows and springs fell between the values of the Sink or

Vinzant's Landing and Ogden Lake values (Figs. 3-3a, 3-4a, 3-5a, and 3-6a). In addition,

the concentrations are higher at Sweetwater Lake than at the Rise and the downstream

springs. The anion concentrations follow a pattern similar to the cation concentrations

with Vinzant's Landing and the Sink having the lowest concentrations of the solutes and

Ogden Lake having the highest concentration (Figs. 3-7a and 3-8a). The alkalinity

concentrations are more varied along the flow path than most of the solutes, and the

lowest concentration occurs at Twohole sink and the highest at Hornsby Spring (Fig. 3-

9a). Both the S042- and alkalinity concentrations are higher at Sweetwater Lake than the

River Rise. Based on the cation and anion data all of the samples are at or near saturation

with respect to calcite (Fig. 3-10a).

The isotopic data do not vary much along the flow path and have small ranges in

values compared to the major elemental data. The lightest 6180 concentration is -3.06%o

at Hornsby Spring and the Sink has the heaviest concentration of -1.71%o (Fig. 3-1 la),

although most of the samples are between -2.6 to -2.1%o. The 87Sr/86Sr ratios are lowest

during this sample period with Paraners Branch Sink having the least radiogenic ratio and

the Sink the most radiogenic (Fig. 3-12a).

Intermediate Stages

Surface Water

Field measurements

During the second sample period in January 2003, conductivity measurements were

typically 130-160 [[S/cm with the exception of Hawg Sink, Twohole Sink and Hornsby

Spring, which had higher values of 368, 235, and 252 [[S/cm, respectively. In late










April/early May 2003, the conductivity in the karst windows is higher than in January,

ranging from 225-420[tS/cm with the highest values at Sweetwater Lake, the Rise,

Treehouse Spring and Hornsby Spring. The pH values in January were approximately 6

while the pH in April was approximately 7. During the colder, winter event in January,

the water temperature of the karst windows was 10-1 10C, but Hawg Sink is slightly

warmer at 15.50C. In April the temperature of the surface waters increased to 21-230C.

The DO content in January was 5.5-6, with Hawg having a very low level of 1.63, and

was 1.1 to 3.8 in April.

Chemical and isotopic compositions

The chemical concentrations in January and April were intermediate compared to

the concentrations during the drought and flood sample times. The concentrations in

January were generally lower than the concentrations in April and were similar to the

flood conditions while the concentrations in April were closer to the drought conditions.

In January the Ca2+, Mg2+, and Sr2+ concentrations in the karst windows increase 63%

between the River Sink and Ravine Sink and remain similar along the flow path.

However, Hawg and Twohole Sinks and Hornsby Spring typically had higher

concentrations (Figs. 3-3a, 3-4a, 3-5a, and 3-6a). The Na+ concentration at Hawg Sink,

however, was lower compared with the other karst windows. In April, the concentrations

of the cations were higher than the concentrations in January with the exception of Na+,

which was lower than January. The concentrations of most of the cations increased along

the flow path from Paraners Branch Sink toward the River Rise, although the increase

was small for Mg2+ and Sr2+ and there was no increase in Na+ (Figs. 3-3a, 3-4a, 3-5a, and

3-6a).









Both the S042- and alkalinity concentrations were lower in January than in April

(Figs. 3-8a and 3-9a). The concentrations of these two solutes increase -63% between

River Sink and Ravine Sink and were relatively equal along the flow path. As with the

cation concentrations, the S042- and alkalinity concentrations were higher at Hawg and

Twohole Sinks as compared to the other karst windows along the flow path. In April the

concentrations of the S042- and alkalinity increased along the flow path from Paraners

Branch Sink toward the River Rise. In contrast to the S042- and alkalinity, the C1-

concentrations in January did not decrease between the first and second sampling periods,

but remained elevated and were equal along the flow path (Fig. 3-7a). Hawg Sink,

however, had a lower concentration than other locations. In April, the C1- concentrations

were lower than January. The concentrations were slightly higher in the karst windows

than at Vinzant's Landing and the Sink but there is no increase in concentration along the

flow path. However, there was a small spike in the concentration at Ravine Sink.

During January, most of the karst windows were undersaturated with respect to

calcite (Fig. 3-10a), with Hawg Sink, Twohole Sink and Hornsby Spring being closest to

saturation. In contrast, the samples in April were much closer to saturation with calcite

and approached the saturation index values from May.

In contrast to the major element chemistry, the 6180 values in January and April

was similar along the flow path. The values in January were slightly depleted in 6180

than in April from Vinzant's Landing to Paraners Branch Sink (Fig. 3-11 a), but were

enriched from Hawg Sink to Hornsby Spring. The ratios of 87Sr/86Sr, however, were

similar to the major element chemistry with the samples from January plotting closer to










the March data and the samples from April plotting closer to the May data (Fig. 3-12a).

In January the 87Sr/86Sr ratios were more radiogenic than in April.

Ground Water

Field measurements

The field measurements of the ground water in the wells were similar for the two

intermediate sample periods. The conductivity of the ground water was typically 415-

530 [tS/cm, although Well 2 had the highest conductivity value in January (1009 [[S/cm)

and Well 1 was highest in April (907[LS/cm). The average pH of the water during both

events was 6.7 and the average temperature was 21-220C. As with the conductivity, the

temperature of Well 2 was highest in January (260C) and Well 1 was highest in April

(260C). The DO content was low for most of the wells at 0.15-0.22mg/L.

Chemical and isotopic compositions

The chemical concentrations of the wells during January and April were similar for

most of the wells; however, Well 2 had higher concentrations of all major elements

during both sample events (Figs. 3-3b, 3-4b, 3-5b, and 3-6b). Wells 1 and 7 had higher

concentrations than Wells, 3, 4 and 6 for Ca2+ and Well 7 had higher concentrations of

Sr2+ and Mg2+ in April. The highest Sr2+ concentration in January was in Well 1. The

Na+ concentrations were similar between the wells in April; however, Well 7 had a

slightly higher concentration. As with most of the cations, the anion concentrations of

the ground water during sampling of the intermediate stages were similar between the

two sampling periods (Figs. 3-7b, 3-8b, and 3-9b). Well 2 had the highest concentrations

of C1 and Na+. The C1 concentrations were higher in Well 7 than Wells 1,3,4, and 6, but

the SO42- concentrations were similar in these 5 wells. The alkalinity concentrations









were lower in Well 2 than the other wells. The highest concentrations of alkalinity for

both sample periods were in Well 7. During both January and April the ground water

from each well was near saturation with respect to calcite, however, Well 7 was

supersaturated with calcite in April (Fig. 3-10b).

The isotopic compositions for ground water during both intermediate stage

periods were also similar. The 6180 values were light in all of the wells in January and

April and the values remained relatively unchanged (Fig. 3-1 lb). After several attempts,

the ratios of 87Sr/86Sr could not be calculated for Well 2 in January but were calculated

fro wells 1 and 7 (Fig. 3-12b). The ratio in Well 1 was less radiogenic in January than in

April. In addition, Well 1 had the most radiogenic 87Sr/86Sr ratio in April as compared to

the other wells.

Flood Stage

During the high water levels of the Santa Fe River at O'Leno State Park, all of the

karst windows were sampled with the exception of Twohole sink and Treehouse Spring.

These sites were inaccessible because of flooded roads and high waters and swift currents

in the river. All of the wells were sampled at this time.

Surface Water

Field measurements

Field measurements of the karst windows during the flood stage were less variable

than during the drought conditions. Conductivity was approximately 80 [[S/cm at most

locations, and the temperature of the water was 17C with the exception of Hornsby

Spring, which was 18.80C. The surface water pH levels were approximately 5 and DO

levels were 6 mg/L, although Hawg was 2.88 mg/L.









Chemical and isotopic compositions

The solute concentrations in March are the lowest of the four sample periods.

Unlike the low flow period, the concentrations at Ogden Lake were not anomalously

higher than the other karst windows and, for the most part, the solute concentrations were

similar at each location along the flow path (Figs. 3-3a, 3-4a, 3-5a, and 3-6a). Hawg

Sink, however, had higher concentrations of Ca2+, Mg2+, and Na+ than the other karst

windows. As with the cations, the anion concentrations fell within a narrow range and

were similar along the flow path (Figs. 3-7a, 3-8a, and 3-9a). Again, Hawg Sink had

higher concentrations than the other karst windows. Based on the PHREEQC

calculations all of the samples were undersaturated with respect to calcite (Fig. 3-10a).

The 6180 values were the lightest during this sample period (Fig. 3-1 la), with the

exception of the Rise, Treehouse Spring, and Hornsby Spring. The 87Sr/86Sr ratios were

the most radiogenic and the Sink and Ogden Lake had the highest ratios (Fig. 3-12a). The

low ratio at the River Rise may be due to error during Sr separation and analysis.

Ground Water

Field measurements

Field measurements of the physical parameters of the ground water in the wells

were higher than those measured in the surface water. Conductivity of the wells was

408-488 [[S/cm and the temperature was 21C except for Well 2 (25.80C). The pH levels

during flooding were close to 7 and the DO content of most of the wells was 0.13-0.33

mg/L. Well 3 and Well 4 had higher DO concentrations of 1.64 and 4.06 mg/L,

respectively.









Chemical and isotopic compositions

All of the wells had similar chemical compositions except for Well 2, which had

higher concentrations of some of the cations (Figs. 3-3b, 3-4b, 3-5b, and 3-6b), although

the concentrations of these solutes in Well 2 were lower than during the intermediate

sampling periods. The Ca2+ and Sr2+ concentrations were lowest in Well 2 during this

sample period, however; the Mg2+ and Na+ concentrations were higher in Well 2 than in

the other wells. The C1 and S042- concentrations were highest in Well 2 (Figs. 3-7b and

3-8b). These concentrations were lower than those from the intermediate sampling

periods. The alkalinity concentrations were similar in Wells 3,4,6 and 7 with Well 1

having the highest concentration and Well 2 the lowest (Fig. 3-9b). All of the ground

water samples were at or near saturation with respect to calcite (Fig. 3-10b).

The isotopic data of the ground water were similar to the intermediate sample

periods. The 6180 values were lightest in Wells 3,4,6, and 7 (Fig. 3-11b). The ratios of

87Sr/86Sr were less radiogenic than the surface water (Fig. 3-12b). Wells 2,3, and 4 had

the highest ratios and Wells 1 and 7 had the lowest.
















Ca Concentrations (Sinks)


2000 4000 6000 8000


Distance (m)





Ca Concentration (Wells)


; Jan.(10.45masl)
[ March(13masl)
I April(10.49masl)



















Well 2 Well 3 Well 4 Well 6 Well 7


Location

Figure 3-3 a. Ca2+ concentrations of karst windows vs. distance from Vinzants Landing
for each sample set and b.Ca2+ concentrations of the wells for sample sets 2-4.


120


Well 1
















Sr Concentration (Sinks)


4000


6000


Distance (m)


Sr Concentration (Wells)


Well 1 Well 2 Well 3 Well 4 Well 7


Location

Figure 3-4a. The Sr2+ concentrations of the karst windows vs. distance from Vinzants
Landing and b. the concentrations of Sr2+ in the wells for each sample period.


0 2000


8000

















Mg Concentrations (Sinks)


Ogden
a.
In ants ( Ravine
Paraners Twohole
Sink \ Ig Hawg Sweetwate
SJim Jug



: : y


0 2000


4000


6000


SMay(9.8-9.6masl)
-- Jan.(10.75-10.45masl)
March(11.65-13masl)
- April(10.49masl)

Hornsby
Rise re
V Treehouse
y 'V
y


8000


Distance (m)


Mg Concentration (Wells)


25



20
-J
E
) 15
15


10


Well 1 Well 2


Well 3 Well 4


Well 6 Well 7


Location

Figure 3-5a. The Mg2+ concentrations in the karst windows vs. distance from Vinzants
Landing and b. the concentrations of Mg2+ of each well for each sample
period.


i Jan.(10.45masl)
/ March(13masl)
SApril(10.49masl)














Na Concentrations (Sinks)

e May(9.8-9.6masl)
Jan.(10.74-10.45masl)
March(11.65-1 3masl)
vine -A April(10.49masl)
vine


Sweetwater
Rise

Twohole
awg

V .
vy: y


0 2000


Treehouse


Hornst


6000


Distance (m)


Na Concentration (Wells)


25



20



15
z

10



5


Jan.(10.45masl)
March(13masl)
H April(10.49masl)


Well #1 Well #2 Well #3 Well #4 Well #6 Well #7


Location




Figure 3.6a. The Na concentrations in the karst windows vs. distance from Vinzants
Landing and b. Na concentrations of the wells for each sample period.
















CI Concentrations (Sinks)
Ogden


May (9.8-9.6masl)
--- Jan.(10.74-10.45masl)
March(11.65-13masl)
A--April(10.49masl)


Rise
Treeh use

THorns]

V


6000


8000


Distance (m)


CI Concentration (Wells)


30



20



10


Well 1 Well 2


SJan.(10.45masl)
M March(13masl)
^ April(10.49masl)



















Well 3 Well 4 Well 6 Well 7


Location


Figure 3-7a. The CFl concentrations in the karst wiidows vs. distance from Vinzants
Landing and b. CF concentrations in each well for each sample period.


nz-ants Sink




^'t 4


0 2000


4000

















SO Concentrations (Sinks)
4


0 2000


4000


6000


8000


Distance (m)






SO Concentration (Wells)
4
350

0 : : March(13masl)
300 *. 1 1.,, 1 n .-y .,a iI


250


200


150


100


50


Well#1 Well #2
Well #1 Well #2


Well #3 Well #4 Well #6 Well #7

Location


Figure 3.8a.The S042- concentrations in the karst windows vs. distance from Vinzants
Landing and b. S042- concentrations of each well for each sample period.

















Alkalinity (Sinks)


,Bi Jug Hawg
infants Ogden jys aners ; Sweetwater
liznsSink V aie Jim
a TT i W Twohole
Q \ v y





a.
^^ ^^ ^ ^ ^^ ^' IA'I, ^ ^
~A-^a


May(9.8-9.8masl)
Jan.(10.74-10.45masl)
March(11.65-13masl)
April(10.49masl)


Hornsby^
Rise Hrnsby
S Treehouse


6000


Distance (m)


Alkalinity (Wells)


100


b.


H Jan.(10.45masl)
March(13masl)
H April(10.49masl)


Well #1 Well #2 Well #3 Well #4 Well #6 Well #7


Location

Figure 3-9a. The alkalinity concentrations of the karst windows vs. distance from
Vinzants Landing and b. alkalinity concentrations of the wells for each sample
period.
















Saturation Index (Sinks)


Linjants Sink Paraners
Ogden Bigy
R Jim

--- - R ine
A-&1-_
^A-^-- ^A '^^ ^


TwoholeSweetw" ter
Jug; a
Hawg
V 7 ^^


SMay(9.8-9.6masl)
-- Jan.(10.74-10.45masl)
-- March(11.64-13masl)
-A-April(10.49masl)


Treehouse

El


Hornsby


6000


Distance (m)





Saturation Index (Wells)






: April(10.49masl)










|iiii r I | I


Well #1 Well #2


Well #3 Well #4 Well #6 Well #7


Location
b.
Figure 3-10a. The Saturation Index ratios of the karst windows vs. distance from
Vinzants Landing and b. the Saturation Index ratios of the wells for each
sample period.


0 2000


I _




















8 0 (Sinks)


Sink
a.
gden
Rivine

^

Vin ants Paraners
SA Jim


May(9.8-9.6masl)
Jan.(10.74-10.45masl)
March(11.65-13masl)
April(10.49masl)


Treehouse


Twohoe -

Hawg

Jug \ Sweetwpter


Rise


Hornsb


0 2000


4000


Distance (m)


8 0 (Wells)


Well 1 Well 2 Well 3 Well 4 Well 6 Well 7

Location

Figure 3-1 la. 6180 values of karst windows vs. distance from Vinzants Landing and b.

The 6180 values in each well for each sample period.


-1.5





-2





O
-2.5





-3


6000


1 -








41








8Sr/86Sr Ratios


Ravine
a. \
an \


Sink


infantss




4 4.


Sweetwater


- -<


May(9.8-9.6masl)
Jan.(10.74-10.45masl)
March(11.65-13masl)
April(10.49masl)


Twohole
Jim H! awg

Jug y


0.708


0.7075


0 2000


Distance (m)


8Sr86Sr Ratios


0.7086



0.7084



0.7082



0.708



0.7078



0.7076


Well 1 Well 2 Well 3 Well 4 Well 7

Location

Figure 3-12a. The 87Sr/86Sr ratios of the karst windows vs. distance from Vinzants
Landing for each sample period and b. 87Sr/86Sr ratios of each well for each
sample period.


0.7095





0.709


0.7085


Hornsby

Treehous









8000


::::::~sssss~/~~i














CHAPTER 4
DISCUSSION

Variation in Chemical and Isotopic Composition

The water in the matrix porosity can dissolve the limestone of the Floridan Aquifer

increasing concentrations of Ca2+, Mg2+, and Sr2+ in the ground water. In contrast,

conduit water may have low Ca2+, Mg2+, and Sr2+ concentrations if it comes directly from

surface water. Consequently, the flow of water from the matrix to the conduit may

increase Ca2+, Mg2+, and Sr2+ concentrations of the karst windows. Conversely water

from the conduit flowing to the matrix will decrease these solute concentrations in the

matrix. Therefore high solute concentrations in the karst windows may reflect flow of

ground water from the matrix to the conduit while low solute concentrations may reflect

flow of surface water through the conduit to the karst windows.

Chloride and sodium are chemically conservative during reactions with the

carbonate rocks of karst aquifers and thus changes in their concentrations should be

decoupled from changes in Ca2+, Mg2+, and Sr2+ concentrations. The main sources of C1-

and Na+ in the natural waters of Florida (Maddox, 1992) include 1) introduction from

seawater along the coastal transition zone and 2) from marine aerosols in precipitation.

Because the study area is located inland, introduction of seawater is negligible as a source

of C1 and Na so their primary source in the study area is from marine aerosols in

precipitation.

The concentrations of C1- and Na+ can increase through evaporation or decrease by

dilution from precipitation. Precipitation and evaporation should increase or decrease









concentrations of both solutes proportionately so that the Na/Cl ratio would not change.

Sodium concentration may change through ion exchange with clays from the Hawthorn

Formation. Chloride concentrations will not be effected by these reactions and thus

changes in Na+ by this process could be observed as variations in the Na/Cl ratio. The

effects of cation exchange are likely to be much smaller than those caused by evaporative

effects or dilution by precipitation.

The concentration of S042- can also be used as a tracer; however, like the Ca2

Mg2+, and Sr2+ it is a non-conservative tracer involved in chemical reactions with the

matrix rocks. The sources of S042- in Florida include 1) dissolution of gypsum and

anhydrite 2) oxidation of rocks with sulfide-bearing minerals such as pyrite and 3) marine

aerosols and acidic precipitation from airborne sulfur oxides (Maddox, 1992).

Chemical Variations in Karst Windows and Wells

The high concentrations of C1-, Na, Ca2+, Mg2+, Sr2+, and S042- in the karst

windows in May 2002 may be due to introduction of water from the matrix that has

higher solute concentrations, evaporation from the karst windows because of low

precipitation, or a combination of the two. The high concentrations of Ca2+, Sr2+, Mg2+

and S04- in the karst windows are expected for water that has reacted with aquifer rocks

and flowed from the matrix to the conduit. However, the concentrations of Cl- and Na

are higher than the concentrations in the ground water in the wells at baseflow indicating

evaporation must also be occurring to increase the concentrations of these non-reactive

solutes.

Another possible explanation for high concentrations of Ca2+ and S042- during this

time may be from upwelling of water that has dissolved gypsum and anhydrite in the

lower portions of the Floridan Aquifer. In Katz et al (1999) it was suggested that calcite










dissolution yields HCO3- and Ca2+ concentrations in a 2:1 ratio in the Upper Floridan

Aquifer water according to the equation

CaCO3 + H2CO3 = Ca2+ + 2HC03- (1)

Katz et al (1999) suggested that HCO3/Ca2+ ratios less than 2 were attributed to an

alternative source of Ca2+ and SO42-, such as gypsum. The HCO3/Ca2+ ratios in May are

greater than two, however, ranging from 2.08 to 2.75 in the karst windows (Table 4-1)

and 3.38 and 3.33 at Vinzants Landing and the River Sink, respectively. During this time

all of the karst windows were at saturation or slightly supersaturated with respect to

calcite (Fig. 3-10a). Extended periods of evaporation with little to no precipitation during

the drought will concentrate the solutes in the water. Over time, these solutes can reach

or exceed their saturation points, resulting in mineral precipitation. Calcite precipitation

(e.g. reverse of reaction 1) would increase the HCO3/Ca2+ ratio and may cause the

elevated HCO3/Ca2+ ratios at Vinzant's Landing and the Sink. Based on the

concentrations of the reactive and non-reactive solutes, it appears that during the drought

there was loss of water from the matrix to the conduit as well as evaporation in the karst

windows resulting in the increased solute concentrations.

In January the increased concentrations in the karst windows from the Sink to the

Rise suggest an input of water with higher solute concentrations downgradient from the

River Sink. Dean (1999) found that at river stages of 10.74, 10.5 and 10.44 masl, S042-

concentrations increased by 53.2%, 50.2% and 49.3% between the Sink and Sweetwater

Lake, which he attributed to an input of water from the matrix to the conduit. The

HCO3/Ca2+ ratios in January were, for the most part, less than two (Table 4-1) and could

reflect upwelling of water that has dissolved gypsum. The SO42- concentrations are low,










however, suggesting the ratios probably reflect mixing with dilute surface water (e.g.

Katz et al., 1999).

Table 4-1. The HCO3/Ca2+ ratios in the karst windows and the wells.
Location May January March April
Vinzants 3.38 1.07 1.44 2.28
Sink 3.33 1.08 1.05 2.25
Ogden 2.08 1.01 2.11 2.26
Ravine 0.99 1.26 1.91
Paraners 2.29 1.07 1.23 2.26
Jim 2.62 1.00 0.96 2.25
Jug 2.75 1.00 1.42 2.27
Hawg 2.46 2.27 1.74 2.05
Twohole 2.2 1.81 2.04
Sweetwater 2.14 1.25 1.05 2.16
Rise 2.26 1.29 0.99 2.89
Hornsby 2.16 1.72 1.40 2.07
Treehouse 2.49 1.26 1.91
Well 1 2.36 2.14 2.40
Well 2 1.36 1.29 1.27
Well 3 2.32 2.36
Well 4 2.26 2.20
Well 6 2.12 2.31
Well 7 2.59 2.25 2.31

In contrast to the other solute data, the C1- concentrations are not diluted in the

karst windows in January and it is unclear at this time why the C1- concentrations remain

elevated following the increased precipitation in the fall of 2002. Unlike the C1-

concentrations, the Na+ concentrations decrease by 16-30% between May 2002 and

January 2003. The change in the Na+ concentration compared to no change in C1-

suggests that during drought conditions there may have been another source ofNa+ to the

water other than the marine aerosols in precipitation. One possible source could be from

cation exchange with the clay minerals of the Hawthorn Formation, which, in central

Alachua County, have cation exchange capacities ranging between 6 and 46 meq/100 g

(Rose, 1989). A possible explanation for the disproportionate decrease in the C1- and Na











concentrations in January is that during drought conditions Na in the smectite may

exchange with the high Ca2+ concentrations in the Santa Fe River, which would result in

a high Na/Cl ratio (Fig. 4-1). With increased precipitation, the Ca2+ concentration in the

river would be diluted, releasing the Ca2+ previously held in the exchange sites and

uptaking Na+ into the clays. This would result in a decrease in the Na/Cl ratio. The

lower Na/Cl ratios in January further support cation exchange with the Hawthorn Group

as a possible source ofNa+ during low flow.




Na/CI Ratios (Sinks)
12
May
[ January
11 1 March
a. April


09

08



0 6

05

Location
Figure 4-1. Graph of the Na/Cl ratios of the karst windows for each sample period.

Although the river stages were approximately the same in April and January the

concentrations of Ca2+, Mg2+, Sr2+, and S042- are higher in April than in January but the

Na+ and Cl- concentrations are lower. Flooding in March may cause these differences in

solute concentrations in April. During the flooding, water is lost from the conduit to the

matrix and may be held in the matrix porosity, allowing the water to react with the

carbonate rocks of the aquifer. As the floodwaters recede, the hydraulic head in the

conduit becomes lower than the matrix (Martin, 2003) allowing the reacted water to flow


o flow









into the conduit. The lower concentrations of C1 and Na in April may be a reflection of

dilute floodwaters that were held in the matrix, with elevated Ca2+, Mg2+, and Sr2

concentrations from dissolution reactions.

Solute concentrations do not change as much from one sample time to the next in

water collected from all wells, as compared to the karst windows except Well 2 (Figs. 3-

lb 3-9b). Some of the wells, however, do have some small changes in concentrations

between low and high flow with some solutes becoming more diluted and other solutes

more concentrated. The changes in concentrations of the solutes are different for each

well, suggesting that loss of water from the conduit is heterogeneous and varies at the

local scale.

The HCO3/Ca2+ ratio of Well 2 ranges between 1.27 and 1.36 for the study period

(Table 4-1), suggesting upwelling of deeper groundwater according to the model of Katz

et al (1999). However, the lower concentrations of all of the solutes in March suggest the

low HCO3/Ca2+ ratio reflects mixing with dilute surface water that has an HCO3-/Ca2

ratio of 1.05. The low solute concentrations in Well 2 in March suggest that at high

discharge rates at the River Sink, surface water may leave the conduit and flow to the

well, possibly along a zone of high permeability.

In January and April the HCO3/Ca2+ ratios in Well 2 are less than two and

correspond with high S042- concentrations of 305 and 243 mg/L, respectively and

indicate an alternative source of Ca2+ and S042-, such as gypsum, from water deep in the

Floridan Aquifer. During intermediate, baseflow conditions, the water in Well 2 appears

to represent ground water from the Floridan Aquifer. The solute concentrations in Well 2

are higher than the concentrations in the other wells and may reflect a deeper water









source. The differences in the ground water chemistry throughout the study area

demonstrate the complex nature of karst aquifers in that the composition of ground water

in karst terrains can vary on the local scale.

7Sr/8 Sr ratios as tracers in carbonate aquifer studies

Sr isotopes are particularly useful in studies of aquifers in marine carbonates

because their 87Sr/86Sr reflects the87Sr/86Sr ratio of the seawater at the time of deposition

(DePaolo and Ingram, 1985; Hess et al., 1986). The 87Sr/86Sr ratio of seawater at the

beginning of the Cretaceous was 0.7072. This value increased to 0.70775 at the

Cretaceous-Tertiary boundary (Burke et al.; DePaolo and Ingram, 1985) and then

decreased into the late Eocene to a value of 0.7076. From the late Eocene into the

Quaternary the 87Sr/86Sr ratio in seawater increased to the modern ratio of 0.70907 (Burke

et al., 1982; DePaolo and Ingram, 1985; Hess et al., 1986). During the Oligocene and

Eocene Periods, when many of the carbonate rocks of the Floridan Aquifer were

deposited, the ratio remained relatively constant at 0.7077-0.7078. Thus, dissolution of

this material should cause groundwater ratios to approach these low values.

Isotopic Variations in the Karst Windows and Wells

Water from the karst windows and wells appear to reflect mixing between Sr from

the carbonates, Sr from rainfall and Sr from the Hawthorn Group rocks. The 87Sr/86Sr

ratios in the wells are 0.7078 and 0.7079 prior to the flooding, which suggest that the 87Sr

in the water in the wells is enriched from dissolution of the calcite of the matrix rocks and

the ratios are representative of ground water in the matrix. During low flow, the values

of the 87Sr/86Sr ratios in the karst windows are 0.7078, similar to those of

Eocene/Oligocene seawater and presumably the carbonate rocks of the Floridan Aquifer









(Fig. 3-12a). These values suggest that water in the karst windows originates from the

matrix as has been suggested based on the solute concentrations.

During high flow conditions, the Sr isotope ratios in the karst windows reach the

highest values of all the samples of 0.7091 to 0.7094. At the Sink and Vinzant's Landing

the ratio is higher than the modem seawater value (0.70907) and may reflect the

radiogenic ratios associated with the clays of the Hawthorn, which are present upstream

from the River Sink. The 87Sr/86Sr ratio in Well 1 remains within the Eocene/Oligocene

seawater curve during high flow (Fig. 3-12b); however the 87Sr/86Sr ratios of Wells 3,4,

and 7 have slightly more radiogenic suggesting some mixing with water that has been in

contact with radiogenic 8Sr/86Sr ratios such as the clays of the Hawthorn Group.

During the two times of sampling during intermediate conditions, samples from the

karst windows collected in January have more radiogenic 87Sr/86Sr ratios than in April

2003. The differences in values suggest that the January samples are more influenced by

the clays of the Hawthorn Formation than the April samples, which have ratios that

reflect influence by the Floridan Aquifer carbonates. The karst windows in January 2003

are undersaturated with respect to calcite (Fig. 3-10a). This undersaturation implies there

has been little dissolution of the carbonates of the Floridan Aquifer, thereby limiting the

decrease in the 87Sr/86Sr ratios and increase in Sr2+ concentrations. In April 2003, the

87Sr/86Sr ratios in the karst windows are closer to the values of Eocene/Oligocene

seawater than in January, consistent with reactions with the carbonates of the Floridan

Aquifer. These reactions may have occurred during flooding in March when water from

the conduit would react with the high Sr2+ concentrations and low 87Sr/86Sr ratios of the

aquifer rocks. As the floodwater recedes, the reacted water in the matrix would flow to








the conduit, causing the higher Sr2+ concentrations and less radiogenic 7Sr/86Sr ratios

observed in April than in January. The slightly more radiogenic 7Sr/86Sr ratio of Well 1

in April may be a reflection of the floodwater held in the matrix that did not appear until

sampling in April.

Sr Mixing Model

The Sr mixing model assumes two end-member mixing using the Sr2

concentration and the s7Sr/s6Sr ratios of selected end-members. One end member used

for this model is the value of the River Sink sample during the flooding in March 2003.

This value is assumed to represent the most pristine surface water sample and thus would

be the closest to average precipitation in the region. The other end member used in the

model is water from Well 1 in January 2003, due to its high solute concentrations and

low radiogenic 87Sr/86Sr ratio, suggesting this water represents the ground water in the

region. The strontium mixing model used is based on Faure (1986) and Woods et al

(2000), assumes mixing of the end members of surface water and ground water following

an equation in the form:

(87 Sr/86Sr)M = a +b (2)
[Sr](2)

where: (s7Sr/s6Sr)M and [Sr]M are the 87Sr/86Sr ratio and concentration of the mixture,

respectively, a is the slope of the line from the equation:

[Sr]A [Sr]B (87 Sr/86Sr) (87 Sr/86 Sr)A
[Sr]A [Sr]B

and b is the y-intercept from the equation:

b= [Sr]A (87Sr/86Sr)A [SrB (87Sr/86Sr)B (4)
[SrLA [Sr]B


[Sr]B









where [Sr]A and [Sr]B and (87Sr/86Sr)A and (87Sr/86Sr)B are the concentrations and ratios of

end members A (the River Sink in March) and B (Well 1 in January) respectively.

The ratio of 87Sr/86Sr is plotted versus Sr and versus 1/Sr (Fig. 4-2). These figures

suggest there is two end member mixing, but the extreme isotope values occur during

extreme flood conditions. From the Sr isotope mixing model a quantitative percentage of

surface water in a given sample can be determined by calculating thefparameter for

mixing assuming a composition for the end members. In this model,f defines a mixture

of two solutions A and B. The equation forf based on Faure (1986), assumes two-

component mixing using the formula:


(XA X,)

where XM, XA, and XB are the concentrations of Sr2+ for the sample and end members A

(River Sink) and B (Well 1), respectively.

These calculations indicate that water with the lowest fraction sourced from the

surface water occurred in the karst windows in May 2002 (Table 4-2) and averaged 32%.

The highest proportion of water originating from the River Sink in the karst windows

occurred in January, March, and April 2003 (Table 4-2). During the March flooding, the

model suggests nearly 100% of the water in the karst windows is from the Sink. At river

stages of 11.75 or greater, the travel time between the Sink and Rise is less than one day

(Dean, 1999; Martin, 2003) meaning there is rapid flushing of water through the conduit.

This rapid transport of water would prevent water from leaving the matrix and entering

the conduit. Thus the low concentrations in the karst windows probably reflect dilution

by precipitation.









Thef parameter calculations indicate approximately 90% of water in the karst

windows in January 2003 originated from the Sink surface water end member but in

April 2003 52-71% originated from the Sink. Although the river stages were

approximately equal during these sample times the karst windows in January contain

19%-38% more water from the Sink than in April. As the river level began to rise with

increased precipitation in Fall 2002, water entered the Sink and flowed through the

conduit. The hydraulic head in the conduit was higher than the matrix, allowing water to

flow from the conduit to the matrix and result in water that is more dilute than in April.

In contrast, the water in the karst windows in April reflects flow from the matrix to the

conduit as the flooding receded. These results suggest that the antecedent conditions

along with river stage are important for water chemistry.

From thef parameter calculation, the estimated proportion of water originating

from surface water in Well 7 is similar between January, March and April 2003 (Table,

4-3) with only 25-30% of the water originating from ground water suggesting that surface

water may reach this well. The well, however, is saturated with respect to calcite for

each sample event and suggests that undersaturated surface water dissolves calcite as it

moves through the matrix toward the well. Wells 3 and 4 also appear to have high

percentages of water from the Sink end member for both March and April, which would

lead to the more radiogenic 87Sr/86Sr ratios and low Sr2+ concentrations in the wells at


















--- Sink


Mixing Model
SMay (Sinks)
n Jan.(Sinks)
March(Sinks)
April (Sinks)
* Jan.(Wells)
* March(Wells)
A April(Wells)


Well 1


------ ----


Ogden Lake









River Rise


10 20 30 40 50 60

1/Sr (ppm)


Figure 4-2. 87Sr/6Sr vs. Sr and 87Sr/86Sr vs. 1/Sr models for all karst windows and wells. The arrows indicate the end members.


0.71


0.7095


0.709


0.7085


0.708


0.7075


Sr (ppm)









n A7 r 7


'1 tvLl" '


those times. The mixing calculation suggests that Well 2 is composed of 95% surface

water in March. This is supported by low solute concentrations at that time and suggests

that at times of high flow in the river, water can be flushed toward the well. In April, the

Sr2+ concentration in Well 2 is greater than the Well 1 end member, resulting in the

negativefvalue at that time. Well 2 has higher concentrations of all of the solutes than

Well 1 and suggests that water from deeper portions of the aquifer may be upwelling into

Well 2. This possible source water for Well 2 is also evidenced by the low HCO3/Ca2+

ratios of less than two.

Saturation and Mixing Calculations

To better understand the composition of the water in Well 2, which appears to be

more affected by flooding than the other wells, calculations of calcite dissolution and

mixing


Tahl p a_3 Thp ral n~l at; nna nf thp nprrpntarrp nf a~~rC;rp ~riatpr ;n thp Larat


Location May January March April

Vinzants 59
Sink 76 72 100 85
Ogden -6 97 100
Ravine 90 99 52
Big 27 91
Paraners 32
Jim 37 90 99 71
Jug 37
Hawg 40 91 68
Twohole 41 80 63
Sweetwater 5 90 99 54
Rise 30 90 100
Treehouse 14
Hornsby 8


90 100
Treehouse 14
Hornsby 8









Table 4-3. The calculations of the percentage of surface water of the wells
Location January March April

Well 1 0 46 95
Well 2 95 -33
Well 3 92
Well 4 92 94
Well 7 71 75 70

were done using PHREEQC. The program was used to simulate the reaction of one liter

of a known solution of water with a given amount of calcite in order to determine how

much calcite needs to dissolve to reach equilibrium. The saturation index of Well 2 in

March is -0.40, which is close to saturation. For the simulation, water from the Sink in

March with a saturation index of -4.2 was reacted with calcite until equilibrium was

reached, which required dissolution of 1.4 x 10-3 moles of Ca2+. In March, the water in

Well 2 is greater byl.7 x 10-3 moles of Ca2+ than the water in the Sink, similar to the

amount calculated. This similarity suggests that Well 2 is a mixture of water from the

Sink that has dissolved calcite and ground water from the matrix

The large decreases in solute concentrations in Well 2 in March were not seen at

the other wells. Although the high solute concentrations in the well appear to reflect

ground water from deep within the aquifer at baseflow conditions, the large changes

during flooding suggest loss of water from the conduit to the matrix. For this reason,

PHREEQC was used to check the fractions of surface water and ground water in Well 2

in March determined using the Sr2+ concentrations. The 87Sr/86Sr model suggests that the

composition of Well 2 in March is approximately 95% surface water. Based on this, the

same end-members used in the Sr model were mixed at varying proportions. The results

show that a solution of 25-30% Sink water and 70-75% Well 1 water was closest to the









measured data, which is less than the 95% surface water estimated in the Sr model. This

discrepancy may be due to the elements used in the calculation. The PHREEQC model

takes into account the concentrations of all of the major elements in the solutions while

the Sr model is based only on the concentrations of Sr2+ in the end-members and the

mixture. The ground water in Well 2 differs from Well 1 and, therefore, the composition

of Well 1 may not be an accurate representation of the waters that are mixing at Well 2.

Because the Sr2+ concentrations of Well 1 and Well 2 are similar at baseflow conditions,

the percentage of surface water in Well 2 calculated from the Sr model may be more

accurate than the fractions determined using PHREEQC.

Nutrient Loading

The discharge at the Sink and Rise were calculated using water level data collected

by Martin (2003) (Table 3-1) and, along with the NO3- and P04- concentrations, were

used to determine the amount of nutrient loading to the system (Tables 4-4 & 4-5). The

discharge at the Sink in March (45.41 m3/s) is greater than the Rise (40.2 m3/s) indicating

that water is lost to the matrix. The NO3- and P04- loading at the Sink are 0.75 x 103

mg/s and 11.54 x 103 mg/s, respectively and are 0.48 x 103 mg/s and 9.4 x 103 mg/s at the

Rise. Less NO3- and P04- discharges at the Rise than enters the Sink during the flooding,

which is consistent with loss of water to the matrix. In January the discharge at the Rise

is slightly less than the Sink (11.2 m3/s vs. 11.6 m3/s) so there is some loss of water to the

matrix. This loss of water is reflected in the lower nutrient loading rates at the Rise

(Tables 4-4 & 4-5). In April, however, the discharge at the Rise (7.71 m3/s) is greater

than the Sink (3.70 m3/s) indicating water flows from the matrix to the conduit. If this

additional water has higher NO3 and P04 concentrations it would be expected that

nutrient loading to the Rise is greater than at the Sink. In April, the nutrient loading at









the Rise (2.78 x 103 mg/s NO3- and 1.78 x 103 mg/s P043-) is nearly two times the loading

at the Sink (1.27 x 103 mg/s NO3- & 0.67 x 103mg/s P043-) even though the

concentrations of the nutrients are similar between the two locations (Tables 4-4 and 4-5).

There is a decrease in the P04- concentration at both the Sink and the Rise in April

compared to the concentrations in January and March. The decrease in PO43-

concentrations coupled with an increase in loading of both nutrients suggest that as water

flows from the conduit with increasing discharge it is held in the intergranular porosity of

the matrix. As the system returns to baseflow this water flows back into the conduit,

increasing the nutrient loading to the Rise.

Table 4-4. The discharge and NO32- loading calculations at the Sink and Rise.

Location Date Water Q NO3 NO3- Loading
Level

(m) (m3/s) (mg/L) (mg/s x 1000)

5/14/02 9.76 -11.42 0.009 -0.103

1/15/03 10.81 11.61 0.031 0.36

Sink 3/3/03 11.95 45.41 0.016 0.73

4/28/03 10.49 3.70 0.344 1.27

5/14/02 9.31 -1.39 0.077 -0.107

1/17/03 10.11 11.2 0.024 0.27
Rise
3/5/03 11.11 40.2 0.012 0.48

5/1/03 9.94 7.71 0.360 2.78









Table 4-5. The discharge and P04- loading calculations of the Sink and Rise.
Water P04-Loading
Location Date Level Q P04-
(m) (m3/s) (mg/L) (mg/s x 1000)
5/14/02 9.76 -11.42 0.132 -1.5

1/15/03 10.81 11.61 0.105 1.22
Sink
3/3/03 11.95 45.41 0.254 11.54

4/28/03 10.49 3.7 0.181 0.67

5/14/02 9.31 -1.39 0.116 -0.16

Rise 1/17/03 10.11 11.2 0.103 1.15

3/5/03 11.11 40.2 0.234 9.4

5/1/03 9.94 7.71 0.137 1.06














CHAPTER 5
CONCLUSIONS

The heterogeneous characteristics of karst aquifer systems can leave ground water

vulnerable to contamination from surface water. This heterogeneity allows water to move

between the large solution channels of the conduit porosity and the smaller pore spaces of the

intergranular porosity of the matrix. As water enters a conduit through a sink or swallet

some of it may be lost to the matrix if the hydraulic head in the conduit is greater than the

matrix. If, however, the hydraulic head is higher in the matrix than the conduit, water can

flow from the matrix to the conduit.

Because surface water may contaminate ground water supplies it is important to

understand the processes and circumstances that can allow surface water to enter the ground

water matrix. Chemical analysis of surface water, in the form of sinks and springs, and

ground water from wells can be used to evaluate the exchange of water from the two

systems. In the Santa Fe River Sink/Rise system it appears that both the water level of the

river as well as the antecedent conditions play a role in the movement of water between the

conduit and the matrix.

Under low river levels and drought conditions, such as May 2002, no water enters the

River Sink and flows through the conduit or out of the River Rise. As a result, the water

chemistry in the karst windows reflects ground water from the matrix. As the river level

rises with increased rainfall, more water enters the Sink and flows through the conduit. The

hydraulic head in the conduit becomes higher than the matrix and surface water flows to the

matrix. The increased amount of surface water in the subsurface is reflected in dilute solute









concentrations of the karst windows. During flooding in March 2003, water in the karst

windows reflects a composition of nearly 100% surface water. As the flooding subsides, the

system returns to baseflow conditions. Following flooding, water from the matrix enters the

conduit and the karst windows. Because of the different circumstances preceding baseflow,

the chemistry of the water in the karst windows is different. After prolonged periods of

drought conditions an increased precipitation allows the conduit to quickly fill with water,

increasing its hydraulic head, possibly over that in the matrix. However, following periods

of high flow and flood conditions, the conduit quickly loses some water as precipitation

decreases and discharge at the River Sink decreases. The response of the conduit and matrix

to changes in precipitation and flow rates are important in understanding the hydrologic

characteristics of karst aquifer systems.

The matrix appears less responsive to changes in the flow conditions and precipitation

amounts than the conduit. The chemistry of ground water collected in the wells, for the most

part, did not change much from one sample time to the next. However, the significant

difference in the chemistry of Well 2 from baseflow to flood conditions indicates the

complex, heterogeneous nature of karst aquifers. In addition, the Sr model indicates high

percentages of surface water in the wells during the study period. Dissolution of matrix

material can increase the solute concentrations in the ground water and making it appear that

no surface water is present. Although there has been more extensive chemical analyses of

the karst windows in the Sink/Rise system, it is now important to better understand the

effects of flow conditions on the matrix with more sample collection at higher temporal

resolutions.















APPENDIX A
DAILY PRECIPITATION AND STAGE RECORDS











Daily Precipitation and Stage for 2002
Month ppt(cm) masl
1-Jan 0 9.83 19-Feb 0 10.22 9-Apr 0 10.25
2-Jan 0.508 9.82 20-Feb 0 10.22 10-Apr 0 10.24
3-Jan 0.305 9.82 21-Feb 0 10.22 11-Apr 1.22 10.22
4-Jan 0 9.81 22-Feb 0.508 10.22 12-Apr 0.36 10.22
5-Jan 0 9.81 23-Feb 1.473 10.21 13-Apr 0.10 10.22
6-Jan 0.508 9.81 24-Feb 0 10.20 14-Apr 0.41 10.22
7-Jan 0 9.80 25-Feb 0 10.22 15-Apr 0 10.22
8-Jan 0 9.80 26-Feb 0 10.24 16-Apr 0 10.22
9-Jan 0 9.80 27-Feb 0 10.23 17-Apr 0 10.23
10-Jan 0 9.80 28-Feb 0 10.22 18-Apr 0 10.23
11-Jan 0 9.79 1-Mar 0 10.22 19-Apr 0 10.23
12-Jan 0.813 9.79 2-Mar 1.19 10.23 20-Apr 0 10.22
13-Jan 0 9.79 3-Mar 5.03 10.24 21-Apr 0 10.21
14-Jan 5.461 9.79 4-Mar 5.00 10.32 22-Apr 0 10.20
15-Jan 0 9.88 5-Mar 0 10.43 23-Apr 0 10.17
16-Jan 0 9.87 6-Mar 0 10.49 24-Apr 0 10.15
17-Jan 0 10.28 7-Mar 0 10.52 25-Apr 10.13
18-Jan 0 10.29 8-Mar 0 10.53 26-Apr 0 10.11
19-Jan 0.432 10.30 9-Mar 0 10.53 27-Apr 0 10.05
20-Jan 0.076 10.30 10-Mar 0 10.53 28-Apr 10.01
21-Jan 0.000 10.30 11-Mar 0 10.51 29-Apr 0 9.95
22-Jan 1.219 10.30 12-Mar 0 10.50 30-Apr 9.92
23-Jan 0.076 10.32 13-Mar 0.610 10.47 1-May 9.89
24-Jan 0 10.33 14-Mar 0.051 10.45 2-May 9.87
25-Jan 0 10.34 15-Mar 0 10.43 3-May 0 9.84
26-Jan 0 10.33 16-Mar 0 10.42 4-May 9.81
27-Jan 0 10.33 17-Mar 0 10.40 5-May 9.75
28-Jan 0 10.34 18-Mar 0 10.40 6-May 9.74
29-Jan 0.737 10.34 19-Mar 0 10.39 7-May 0 9.73
30-Jan 0 10.40 20-Mar 0 10.39 8-May 9.72
31-Jan 0.305 10.43 21-Mar 0 10.39 9-May 9.71
1-Feb 0 10.40 22-Mar 0.356 10.36 10-May 9.70
2-Feb 0 10.39 23-Mar 0 10.34 11-May 9.69
3-Feb 0 10.39 24-Mar 0 10.34 12-May 9.68
4-Feb 0 10.38 25-Mar 0 10.33 13-May 9.67
5-Feb 0 10.37 26-Mar 0.279 10.33 14-May 1.07 9.69
6-Feb 0 10.37 27-Mar 0 10.32 15-May 9.67
7-Feb 1.219 10.30 28-Mar 0 10.31 16-May 9.66
8-Feb 0.076 10.28 29-Mar 0 10.30 17-May 9.65
9-Feb 0 10.28 30-Mar 0 10.29 18-May 9.64
10-Feb 0 10.28 31-Mar 0 10.28 19-May 0.152 9.65
11-Feb 0 10.28 1-Apr 0 10.23 20-May 0.051 9.63
12-Feb 0 10.26 2-Apr 0.356 10.26 21-May 0 9.60
13-Feb 0 10.26 3-Apr 0.483 10.26 22-May 0 9.62
14-Feb 0.127 10.26 4-Apr 0 10.26 23-May 0 9.62
15-Feb 0 10.25 5-Apr 0 10.26 24-May 0 9.62
16-Feb 0 10.25 6-Apr 0 10.26 25-May 0 9.59
17-Feb 0 10.25 7-Apr 0 10.26 26-May 0 9.59
18-Feb 0 10.22 8-Apr 0 10.25 27-May 0.051 9.57










28-May 0 9.57 18-Jul 0.127 9.45 7-Sep 0 10.37
29-May 0 9.57 19-Jul 0 9.45 8-Sep 0 10.41
30-May 0.152 9.57 20-Jul 2.01 9.48 9-Sep 0 10.37
31-May 1.27 9.55 21-Jul 0.254 9.50 10-Sep 0 10.35
1-Jun 0.152 9.56 22-Jul 0 9.51 11-Sep 0 10.32
2-Jun 9.55 23-Jul 0.051 9.48 12-Sep 0 10.29
3-Jun 9.54 24-Jul 0.076 9.48 13-Sep 0 10.09
4-Jun 1.016 9.54 25-Jul 0.076 9.45 14-Sep 2.26 10.26
5-Jun 9.54 26-Jul 0.025 9.45 15-Sep 0.508 10.28
6-Jun 0.305 9.54 27-Jul 0.254 9.48 16-Sep 0 10.29
7-Jun 0.178 9.54 28-Jul 0 9.45 17-Sep 0 10.32
8-Jun 0.254 9.52 29-Jul 0 9.48 18-Sep 0 10.33
9-Jun 0 30-Jul 0 9.46 19-Sep 0 10.36
10-Jun 0 9.53 31-Jul 0 9.42 20-Sep 0 10.43
11-Jun 0 9.51 1-Aug 0 9.42 21-Sep 0 10.47
12-Jun 0 9.51 2-Aug 0 9.45 22-Sep 0 10.53
13-Jun 0.686 9.50 3-Aug 0 9.84 23-Sep 0 10.57
14-Jun 0.025 9.50 4-Aug 0 10.16 24-Sep 0 10.54
15-Jun 0.025 9.50 5-Aug 0.0254 10.14 25-Sep 0.91 10.51
16-Jun 0 9.48 6-Aug 0 10.16 26-Sep 0.30 10.48
17-Jun 0 9.47 7-Aug 0.1524 10.14 27-Sep 0.53 10.45
18-Jun 0.305 9.47 8-Aug 0 10.23 28-Sep 0 10.43
19-Jun 0.025 9.47 9-Aug 0 10.23 29-Sep 0 10.42
20-Jun 1.499 9.45 10-Aug 0 10.24 30-Sep 0 10.42
21-Jun 0.203 9.48 11-Aug 0 10.24 1-Oct 0 10.41
22-Jun 0.864 9.48 12-Aug 1.93 10.23 2-Oct 0 10.42
23-Jun 0 9.48 13-Aug 0.254 10.23 3-Oct 0 10.40
24-Jun 0 9.45 14-Aug 0 10.20 4-Oct 0 10.40
25-Jun 2.515 9.47 15-Aug 0 10.20 5-Oct 0
26-Jun 0.025 9.48 16-Aug 0 10.17 6-Oct 0
27-Jun 0 9.50 17-Aug 0 10.17 7-Oct 0 10.36
28-Jun 0.533 9.51 18-Aug 0 10.19 8-Oct 0 10.33
29-Jun 2.210 9.51 19-Aug 0 10.19 9-Oct 0.23 10.33
30-Jun 0.254 9.52 20-Aug 0 10.19 10-Oct 0 10.31
1-Jul 0 9.51 21-Aug 0 10.17 11-Oct 0 10.31
2-Jul 0 9.50 22-Aug 0.025 10.14 12-Oct 0 10.33
3-Jul 0.660 9.50 23-Aug 0 10.14 13-Oct 0.41 10.33
4-Jul 1.727 9.50 24-Aug 0 10.13 14-Oct 0 10.30
5-Jul 0.102 9.50 25-Aug 0 10.13 15-Oct 0 10.30
6-Jul 0 9.51 26-Aug 0 10.10 16-Oct 0.23 10.30
7-Jul 0 9.50 27-Aug 0.43 10.07 17-Oct 0 10.29
8-Jul 0 9.50 28-Aug 5.21 10.14 18-Oct 0 10.10
9-Jul 0 9.49 29-Aug 0 10.05 19-Oct 0 10.29
10-Jul 0 9.51 30-Aug 2.39 10.05 20-Oct 0 10.29
11-Jul 0 9.48 31-Aug 0.025 10.16 21-Oct 0 10.28
12-Jul 0.051 9.46 1-Sep 0 10.16 22-Oct 0 10.28
13-Jul 2.69 9.46 2-Sep 1.65 10.20 23-Oct 0 10.26
14-Jul 1.65 9.51 3-Sep 0 10.20 24-Oct 0 10.26
15-Jul 0.305 9.51 4-Sep 0 10.23 25-Oct 0 10.25
16-Jul 0 9.51 5-Sep 0.56 10.33 26-Oct 0 10.23
17-Jul 0 9.45 6-Sep 1.6002 10.36 27-Oct 0 10.23










31-Oct 0 10.20 21-Dec 0 10.62
1-Nov 0 10.20 22-Dec 0 10.61
2-Nov 0 10.19 23-Dec 0 10.58
3-Nov 0 10.19 24-Dec 4.37 10.58
4-Nov 0 10.18 25-Dec 0.025 10.60
5-Nov 0 10.17 26-Dec 0 10.63
6-Nov 0.279 10.16 27-Dec 0 10.72
7-Nov 0 10.14 28-Dec 0 10.80
8-Nov 0 10.13 29-Dec 0 10.90
9-Nov 0 10.14 30-Dec 0 10.96
10-Nov 2.210 10.14 31-Dec 1.80 10.99


11-Nov


10.14


12-Nov 0.051 10.14
13-Nov 1.02 10.17
14-Nov 0.025 10.19
15-Nov 0.000 10.23
16-Nov 8.407 10.25
17-Nov 0.737 10.30
18-Nov 0 10.35
19-Nov 0 10.43
20-Nov 0 10.49
21-Nov 0.279 10.54
22-Nov 10.57
23-Nov 10.57
24-Nov 10.57
25-Nov 10.57
26-Nov 10.55
27-Nov 10.53
28-Nov 10.50
29-Nov 10.48
30-Nov 10.47
1-Dec 10.45
2-Dec 10.43
3-Dec 10.42
4-Dec 10.41
5-Dec 1.68 10.40
6-Dec 0.432 10.39
7-Dec 0 10.38
8-Dec 0 10.38
9-Dec 1.24 10.38
10-Dec 3.12 10.39
11-Dec 0.025 10.40
12-Dec 0.000 10.42
13-Dec 2.032 10.48
14-Dec 10.50
15-Dec 10.53
16-Dec 10.57
17-Dec 10.60
18-Dec 10.62
19-Dec 10.63
20-Dec 0.635 10.63










Daily precipitation and stage records for 2003
Date ppt (cm) masl
1-Jan 2.4384 11.02 19-Feb 0 11.42 9-Apr 10.85
2-Jan 0 11.04 20-Feb 0 11.72 10-Apr 10.86
3-Jan 0.0254 11.05 21-Feb 0 11.83 11-Apr 10.86
4-Jan 0 11.09 22-Feb 1.6002 11.84 12-Apr 10.85
5-Jan 0 11.16 23-Feb 0 11.78 13-Apr 10.80
6-Jan 0 11.18 24-Feb 0 11.63 14-Apr 10.75
7-Jan 0 11.17 25-Feb 0 11.54 15-Apr 10.71
8-Jan 0 11.13 26-Feb 0 11.48 16-Apr 10.68
9-Jan 0 11.07 27-Feb 0.1016 11.43 17-Apr 10.65
10-Jan 0.1524 28-Feb 0.1778 11.40 18-Apr 10.62
11-Jan 0 10.90 1-Mar 1.4478 11.45 19-Apr 10.59
12-Jan 0 10.84 2-Mar 0.4826 11.65 20-Apr 10.56
13-Jan 0.1016 10.81 3-Mar 0.1524 11.87 21-Apr 10.55
14-Jan 0.1524 10.77 4-Mar 1.0414 22-Apr 10.54
15-Jan 0 10.74 5-Mar 0.0508 23-Apr 10.51
16-Jan 0 10.71 6-Mar 0 24-Apr 10.50
17-Jan 0 10.68 7-Mar 3.2258 13.11 25-Apr 10.50
18-Jan 0 10.65 8-Mar 0 26-Apr 10.48
19-Jan 0 9-Mar 3.4544 27-Apr 10.48
20-Jan 0 10.60 10-Mar 0.0254 13.41 28-Apr 10.49
21-Jan 0 10.59 11-Mar 14.01 29-Apr 10.49
22-Jan 0.5334 10.57 12-Mar 14.22 30-Apr 10.49
23-Jan 0.0254 10.56 13-Mar 14.34 1-May 0.4826 10.48
24-Jan 0 10.54 14-Mar 14.16 2-May 10.45
25-Jan 0 10.53 15-Mar 13.86 3-May
26-Jan 0 10.53 16-Mar 13.58 4-May 10.43
27-Jan 0 10.49 17-Mar 1.27 13.25 5-May 10.43
28-Jan 0 10.50 18-Mar 0.0508 6-May 10.41
29-Jan 0 10.48 19-Mar 7-May 10.40
30-Jan 0 10.48 20-Mar 0.0762 8-May 10.39
31-Jan 0 10.46 21-Mar 0.0254 12.19 9-May 10.39
1-Feb 0 10.46 22-Mar 11.99 10-May
2-Feb 0 10.46 23-Mar 0.1016 11.87 11-May
3-Feb 0 10.45 24-Mar 11.69 12-May 0.2794
4-Feb 1.27 10.45 25-Mar 11.55 13-May
5-Feb 0 10.43 26-Mar 11.42 14-May
6-Feb 0 10.43 27-Mar 2.54 11.32 15-May
7-Feb 4.826 10.46 28-Mar 11.27 16-May
8-Feb 0.0762 10.53 29-Mar 11.29 17-May
9-Feb 1.1684 10.60 30-Mar 11.32 18-May 1.143
10-Feb 0.3556 10.78 31-Mar 1.27 11.29 19-May 3.6322
11-Feb 0 10.78 1-Apr 11.28 20-May 0.254


12-Feb


10.83


2-Apr


11.25


13-Feb 0 10.86 3-Apr __11.25
14-Feb 0 10.87 4-Apr 11.19
15-Feb 0 10.88 5-Apr 11.12
16-Feb 7.3914 10.86 6-Apr 11.05
17-Feb 0.1778 11.04 7-Apr 10.97
18-Feb 0 11.16 8-Apr 10.90















APPENDIX B
WATER CHEMISTRY DATA












Data collected between May 8,2002 to June 5, 2002


Location Cl SO4 Ca Na Mg Alk. Si02 6180 Sr 87Sr/86Sr NO3 P04 Temp. pH Cond. DO

(mg/L) (mg/L) (mg/L) (mg/L) mg/L) (mg/L) (mg/L) (%o) (mg/L) (mg/L) (mg/L) (oC)
Vinzants 12.7 22.7 46.6 7.5 12.4 157.5 14.0 -2.3 0.7 0.707942 0.018 0.120 24.6 7.72 358 4.98

Sink 23.8 16.7 38.0 11.5 10.7 126.5 5.3 -1.7 0.4 0.708141 0.009 0.132 26.6 7.37 319 1.72

Ogden 58.9 139.5 72.6 28.8 20.8 151.3 8.4 -1.9 1.7 0.707892 0.009 0.132 26.6 7.21 645 0.1

Big 22.1 65.3 60.7 12.6 15.0 163.4 13.2 -2.6 1.2 0.707882 0 0.122 25.2 7.42 463 2.78

Paraners 22.8 64.2 59.4 12.7 14.7 136.0 12.7 -2.6 1.1 0.707858 0.02 0.125 26 7.61 456 1.33

Jim 23.1 63.2 59.0 12.8 14.7 154.6 13.1 -2.6 1.0 0.707890 0 0.127 24.7 7.13 454 1.7

Jug 25.3 58.9 56.1 13.5 14.1 154.1 12.3 -2.4 1.0 0.707877 0 0.122 24.4 7.14 440 2.64

Hawg 25.6 57.9 55.1 13.5 13.9 135.7 12.4 -2.3 1.0 0.707907 0.009 0.127 24.3 7.03 437 0.51

Twohole 27.2 55.4 53.9 14.1 13.7 118.7 11.7 -2.1 1.0 0.707888 0.105 0.131 24.6 7.03 429 0.23

Sweetwater 22.9 95.8 69.1 13.1 16.4 148.3 14.7 -2.9 1.5 0.707859 0.119 0.103 24.1 7.65 512 0.72

Rise 30.3 73.2 57.8 15.1 14.1 130.5 11.4 -2.3 1.1 0.707899 0.077 0.116 22.8 7.33 458 2.04

Treehouse 64.8 12.8 14.1 140.3 -1.9 1.4 0.707812 27.5 7.5 475 5.44

Hornsby 12.8 79.2 74.4 9.4 13.5 184.9 12.4 -3.1 1.5 0.707849 0.067 0.052 23 7.54 498 6.46
















Data collected between January 15,2003 and February 5,2003


Location Cl SO4 Ca Na Mg K Alk. SiO2 180 Sr 87Sr/86Sr NO3 P04 Temp. pH Cond. DO
(mg/L) (mg/L) (mg/L) (mg/L) mg/L) (mg/L) (mg/L) (mg/L) (%o) (mg/L) (mg/L) (mg/L) (0C)
Vinzants 23.0 6.8 11.5 10.70 3.45 1.25 12 7.8 -2.8 0.029 0.14 10.4 5.2 131.1 8.7
Sink 22.8 6.5 11.1 10.50 3.37 1.23 12 7.8 -2.885 0.468 0.70792 0.031 0.14 10 6.1 131.1 8.8
Ogden 23.1 7.8 11.9 10.70 3.56 1.25 12 7.6 -2.825 0.070 0.70883 0.031 0.15 10.2 6.36 136 8.6
Big 23.6 17.7 14.9 10.60 4.04 1.19 16 8.2 -2.9 0.031 0.15 10 6.03 160.7 8.2
Ravine 23.8 17.6 16.1 11.20 4.33 1.23 16 8.6 -2.825 0.03 0.15 11 5.23 164.9 8.3
Paraners 23.3 17.4 15.0 10.70 4.09 1.20 16 7.8 -2.77 0.027 0.14 10 6.2 160.7 8.3
Jim 23.6 16.4 16.0 11.00 4.29 1.23 16 8.7 -2.73 0.188 0.70821 0.031 0.15 10.6 6.29 164.8 8.1
Jug 23.1 18.8 16.0 11.10 4.30 1.24 16 8.8 -2.725 0.028 0.15 10.9 6.02 163.9 5.6
Hawg 16.7 24.0 60.0 9.17 6.26 0.94 136 13.3 -2.485 0.174 0.70819 -0.001 0.16 15.5 6.66 368 1.6
Two Hole 26.3 28.7 28.7 12.10 5.57 1.43 52 10.5 -2.28 0.338 0.70805 0.004 0.16 12 6.66 235 3.1
Sweetwater 23.5 18.1 16.0 10.70 4.22 1.21 20 8.5 -2.69 0.188 0.70811 0.028 0.14 10 6.2 165.4 4.6
Rise 22.8 15.3 15.5 10.40 4.08 1.21 20 8.6 -2.83 0.180 0.70819 0.024 0.15 11 6.19 161.9 4.3
Treehouse 26.2 20.1 19.1 11.00 4.62 1.26 24 8.8 -2.8 0.041 0.15 11.1 6.48 182.4 7.3
Hornsby 19.7 35.7 32.6 9.93 5.97 1.24 56 10 -2.81 0.03 0.13 13.2 6.54 252 2.5
Well #1 8.4 2.0 107.0 3.79 3.25 0.24 252 10.5 -3.38 1.626 0.70784 0 0.09 21.5 6.7 485 0.2
Well #2 46.0 305.0 152.0 26.90 27.40 1.71 206 15 -3 0.065 0.59 26.3 6.76 1009 2.3
Well#7 13.4 14.6 101.0 7.06 5.41 0.62 262 9.1 -3.58 0.488 0.70793 0.002 0.14 20.4 6.51 530 0.2












Data collected March 3, 2003 to March 19,2003


Location Cl SO4 Ca Na Mg K Alk. SiO2 180 Sr 87Sr/86Sr NO3 P04 Temp. pH Cond. DO
(mg/L) (mg/L) (mg/L) (mg/L) mg/L) (mg/L) (mg/L) (mg/L) (%o) (mg/L) (mg/L) (mg/L) (oC)
Vinzants 14.3 2 6.96 5.5 2.15 1.34 10 3.3 -3.32 0.008 0.217 17.2 5.98 84.5 6.08
Sink 12.7 2 7.64 6.29 2.41 1.72 8 3.4 -2.46 0.025 0.709481 0.016 0.286 17.4 5.35 79.2 5.77
Ogden 12.8 2 7.59 5.76 2.32 1.55 16 3 -2.53 0.0273 0.709434 0.015 0.294 17.2 4.98 79 5.7
Ravine 12.8 2 7.92 5.82 2.36 1.53 10 3.1 -2.67 0.0379 0.709099 0.02 0.294 17.3 4.75 81.9 5.52
Paraners 12.9 2 8.14 6.29 2.46 1.68 10 3 -2.49 0.017 0.294 17.1 5.04 82 5.54
Jim 13.1 2 8.31 6.16 2.48 1.69 8 2.6 -2.36 0.0336 0.709091 0.012 0.300 17.4 5.05 82.6 5.43
Jug 13.2 2 8.48 6.32 2.48 1.68 12 2.7 -2.37 0.013 0.292 17.6 4.9 84.3 5.5
Hawg 15.7 7.2 16.1 7.5 2.93 1.57 28 3 -2.54 0.022 0.253 18.4 5.58 123.2 2.88
Sweetwater 11.2 2 7.6 5.34 2.22 1.44 8 2.9 -2.45 0.0391 0.709101 0.008 0.305 16.5 4.78 71.5 5.56
Rise 11.3 2 8.12 5.71 2.32 1.6 8 2.8 -2.35 0.0301 0.707725 0.012 0.338 17.1 4.67 72.5 5.54
Hornsby 11.6 2 11.4 5.74 2.45 1.7 16 3.1 -2.42 0.021 0.292 18.8 5.11 86.8 4.35
Well#1 8.3 2 113 4.33 2.21 0.09 242 7.8 -3.34 0.8899 0.7078735 0.001 0.075 21.9 6.8 484 0.33
Well#2 21.3 114 79.1 14.1 13.8 1.43 102 1.4 -2.76 0.1103 0.7080629 0 0.142 25.8 7.08 488 0.14
Well#3 5.3 3 91.4 3.68 1.82 0.09 212 10 -4.11 0.148 0.708141 0.057 0.063 21.8 6.93 409 1.64
Well#4 8.2 4.4 91 5.08 2.18 0.24 206 8.2 -4.16 0.148 0.708127 0.034 0.063 21.4 6.97 408 4.06
Well #6 7.2 2.5 100 3.87 1.89 0.33 212 6.1 -4.33 0 0.078 21.1 6.8 435 0.22
Well #7 13.1 17.7 87.2 7.53 5.78 0.57 196 6.9 -3.71 0.004 0.151 20.7 6.98 422 0.16














Data collected April 28, 2003 to May 1, 2003


Location Cl SO4 Ca Na Mg K Alk. SiO2 Sr 87Sr/86Sr 6180 NO3 PO4 Temp. pH Cond. DO

(mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (%o) (mg/L) (mg/L) (0C)
Vinzants 15.1 13.1 33.7 7.27 6.79 0.94 77 11.8 -2.59 0.301 0.255 20.8 7.06 230 3.24
Sink 15.5 12.8 35 7.38 6.8 0.94 78.9 12.2 0.267 0.708033 -2.33 0.344 0.244 21.4 7.07 235 3.8
Ogden 16.3 16.5 36 8.24 7.36 1 81.5 12 -2.21 0.383 0.233 22.1 6.99 242 3.74
Ravine 19.2 50 52 9.96 9.68 0.96 99.3 13.1 0.796 0.707894 -1.96 0.329 0.211 22.7 6.99 225 3.74
Paraners 17.5 27.9 43.1 8.57 8.23 0.96 97.2 12.5 -2.38 0.336 0.238 22 7.04 300 2.57
Jim 17.1 23.8 42.3 8.45 8.07 0.97 95.1 12.6 0.493 0.707955 -2.58 0.322 0.241 22.1 7.02 293 2.45
Jug 17.3 27.7 40 8.31 7.64 0.96 90.7 12.4 -2.68 0.395 0.227 22.1 7.02 287 2.1
Hawg 18.7 37.5 43.8 8.85 8.23 0.96 90 11.9 0.535 0.707937 -2.73 0.307 0.227 21.8 6.67 295 2.1
TwoHole 19.7 40.9 47.1 9.29 8.76 1.01 96 12.2 0.615 0.707927 -2.72 0.366 0.238 22.1 6.85 315 2.04
Sweetwater 17.8 49.3 51.9 8.85 9.27 0.96 112 12.9 0.764 0.707887 -2.71 0.367 0.189 22.1 6.76 338 1.69
Rise 18.2 62.9 57.1 9.36 9.82 0.96 108 12.6 -2.96 0.36 0.172 21.8 6.84 377 1.14
Treehouse 14.4 64 64.8 8.59 10.1 1.03 124 12.8 -3.05 0.423 0.180 22.4 7.06 400 2
Hornsby 12.2 64.6 69.5 7.72 10.1 1.05 144 13.2 -3.11 0.349 0.153 23.1 6.99 420 0.31
Well #1 7.3 20 97.6 3.26 1.39 0.15 234 9.4 0.108 0.708194 -3.34 0.014 0.095 26 6.92 907 0.22
Well #2 40.7 242 148 25.2 27.8 1.87 188 13.5 2.16 0.707843 -2.83 0.022 0.111 22 6.87 448 0.13
Well #3 5.6 10 88 3.31 1.67 0.41 208 10.5 -4 0.07 0.078 21.8 6.9 416 1.27
Well #4 8.2 10 94.5 4.41 1.51 0.31 208 8.1 0.129 0.708112 -4.25 0.054 0.078 21.7 6.92 423 2.08
Well #6 6 20 90.9 2.95 1.22 0.37 210 5.8 -4.39 0.001 0.045 21.2 6.94 416 0.17
Well#7 14.9 11.6 111 4.41 4.61 0.62 256 10.2 0.508 0.7079129 -3.68 0.035 0.149 20.8 6.79 538 0.14















LIST OF REFERENCES


Atkinson, T.C., 1977, Diffuse flow and conduit flow in limestone terrain in the Mendip
Hills, Somerset (Great Britain), J. Hydrol., 35: 93-110.

Burke, W.J., Denison, R.E., Heatherington, E.A., Koepnik, R.B., Nelson, H.F., Otto, J.B.
Variation of seawater 87Sr/86Sr throughout Phanerozoic time, Geology, 10: 516-
519.

Cao, H., Cowart, J.B., Osmond, J.K., 1999, Uranium and strontium isotopic geochemistry
of karst waters, Leon Sinks Geological Area, Leon County, Florida, Cave and Karst
Science, 26: 101-106.

Dean, R.W., 1999, Surface and groundwater mixing in a karst aquifer: An example from
the Floridan Aquifer: Gainesville, Florida, University of Florida, MS Thesis 74p.

DePaolo, D.J. and Ingram, B.L., 1985, High-resolution stratigraphy with strontium
isotopes, Science, 227: 938-941.

Desmarais, K. and Rojstaczer, S., 2002, Inferring source waters from measurements of
carbonate spring response to storms, J. Hydrol., 260: 118-134.

Dogramaci, S.S. and Herczeg, A.L., 2002, Strontium and carbon isotope constraints on
carbonate-solution interactions and inter-aquifer mixing in groundwaters of the
Semi-arid Murray Basin, Australia, J. Hydrol., 262: 50-67.

Faure, G., 1986, Principles of sotope Geology, John Wiley & Sons, New York, 464p.

Frederickson, G.C. and Criss, R.E., 1999, Isotope hydrology and residence times of the
unimpounded Meramec River Basin, Missouri, Chem. Geol., 157: 303-317.

Ginn, B., 2002, Using temperature and water elevation measurements to model conduit
properties in karst aquifers: An example for the Santa Fe River Sink/Rise System,
Florida, University of Florida, 23p.

Greene, E.A., 1997, Tracing recharge from sinking streams over spatial dimensions of
kilometers in a karst aquifer, Ground Water, 35: 898-904.

Grozos, M., Ceryak, R., Allison, D., Cooper, R., Weinberg, M., Macesich, M., Enright,
M.M, and Rupert, F., 1992, Carbonate Units of the intermediate aquifer system in
the Suwannee River water Management District, Florida. Florida Geological
Survey Open File Report No. 54.









Grubbs, J.W., 1998, Recharge rates to the Upper Floridan Aquifer in the Suwannee River
Water Management District: Florida, U.S. Geological Survey, 30 p.

Halihan, T., Wicks, C.M., and Engelin, J.F., 1998, Physical response of a karst drainage
Basin to flood pulses: example of the Devil's Icebox cave system (MO, USA), J.
Hydrol., 204: 24-36.

Hess, J., Bender, M.L., Schilling, J.G., 1986, Evolution of the ratios of strontium-87 to
Strontium-86 in seawater from Cretaceous to Present, Science, 231: 979-984.

Hisert, R.A., 1994, A multiple tracer approach to determine the ground and surface water
relationships in the western Santa Fe River, Columbia County, Florida: Ph.D.
Dissertation, University of Florida, 212p.

Hunn, J.D., and Slack, L.J., 1983, Water resources of the Santa Fe River Basin, Florida,
U.S. Geological Survey, Water-Resources Investigations Report 83-4075, 105 p.

Johnson, T.M., Roback, R.C., McLing, T., Bullen, T.D., DePaolo, D.J., Doughty, C.,
Hunt, R.J., Smith, R.W., Cecil, L.D., Murrell, M.T., 2000, Ground water "fast
paths" in the Snake River Plain aquifer: Radiogenic isotope ratios as natural
groundwater tracers, Geology, 28: 871-874.

Karst Waters Institute, Karst Water Institute Home Page, 2001, 24 June 2002

Katz, B.G. and Bullen, T.D., 1996, The combined use of 87Sr/86Sr and carbon and water
Isotopes to study the hydrochemical interaction between groundwater and
Lakewater in mantled karst, Geochim. et Cosmo. Acta, 60: 5075-5087.

Katz, B.G., Coplen, T.B., Bullen, T.D, and Davis, J.H., 1997, Use of chemical and
isotopic tracers to characterize the interaction between ground water and surface
water in mantled karst, Ground Water, 35: 1014-1028.

Katz, B.G., Catches, J.S., Bullen, T. D., and Michel, R.L., 1998, Changes in the isotopic
and chemical composition of ground water resulting from a recharge pulse from a
sinking stream, J. Hydrol., 211: 178-207.

Katz, B.G., Hornsby, H.D., Bohlke, J.F., and Mokray, M.F., 1999, Sources and
chronology of nitrate contamination in spring water, Suwannee River Basin,
Florida, U.S. Geological Survey, Water Resources Investigation Report 99-4252,
54 p.

Lakey, B. and Krothe, N.C., 1996, Stable isotopic variation of storm discharge from a
perennial karst spring, Indiana, Water Resource Research, 32: 721-731.

Maddox, G.L., Lloyd, J.M., Scott, T.M., Upchurch, S.B., and Copeland, R., 1992,
Florida's ground water quality monitoring program- background
hydrogeochemistry. Florida Geological Survey, Special Pub. No. 34.









Martin, J.B., and Dean, R.W., 1999, Temperature as a natural tracer of short residence
times for ground water in karst aquifers: In: A.N. Palmer, M.V. Palmer and I.D.
Sasowsky (Eds), Karst Modeling, Karst Waters Institute Special Publication,
Charlestown, West Virginia, No. 5: p. 236-242.

Martin J.B., and Dean, R.W., 2001, Exchange of water between conduits and matrix in
the Floridan Aquifer, Chem. Geol., 179: 145-155.

Martin, J.B. and Screaton, E.J., 2000, Exchange of matrix and conduit water with
examples from the Floridan Aquifer: U.S. Geological Survey Water-Resources
Investigations Report 01-4011, p. 38-44.

Martin, J.M., 2003, Quantification of the matrix hydraulic conductivity in the Santa Fe
River Sink/Rise system with implication on the exchange of water between the
matrix and conduits: Gainesville, Fl, University of Florida, MS Thesis, 80p.

McNutt, R.H., Frape, S.K., Fritz, P., Jones, M.G., and MacDonald, I.M., 1990, The
87Sr/86Sr values of Canadian Shield brines and fracture minerals with applications
to groundwater mixing, fracture history, and geochronology, Geochim. et Cosmo.
Acta., 54: 205-215.

Meyer, F.W., 1963, Reconnaissance of the geology and ground water resources of
C\Columbia, County, Florida, Florida Geological Survey, Report of Investigations
No. 30,

Padilla, A., Pulido-Bosch, A., and Mangin, A., 1994, Relative importance ofbaseflow
and quickflow from hydrographs of karst spring, Ground Water, 32: 267-277.

Parkhurst, D.L. and Appelo, C.A., 1999, User's guide to PHREEQC (version 2); a
computer program for speciation, batch-reaction, one-dimensional transport and
inverse geochemical calculations, U.S. Geological Survey, Water Resources
Investigations Report 99-4259, 312 p.

Redwine, J.C. and Howell, J.R., 2002, Geochemical methods for distinguishing surface
water from groundwater in the Knox Aquifer System, Env. Geo., 42: 485-491.

Rose, S., 1989, The heavy-metal adsorption characteristics of Hawthorn Formation (F1,
USA) sediments, Chem. Geol. 74: 365-370.

Scanlon, R.R. and Thrailkill, J., 1987, Chemical similarities among physically distinct
spring types in a karst terrain, J. Hydrol., 89: 259-279.

Scott, T.M., 1992, A geological overview of Florida. Florida Geological Survey, Open
File Report No. 50.

Screaton, E.J., Martin, J.B., Ginn, B., and Smith, L.A., 2004, Conduit properties and
karstification in the Santa Fe River Sink-Rise system of the Floridan Aquifer,
Ground Water, 42: 338-346.









Shuster, E.T., and White, W.B., 1971, Seasonal fluctuations in the chemistry of limestone
Springs: A possible means for characterizing carbonate aquifers, J. Hydrol., 14: 93-
128.

Skirvin, R.T., 1962, The underground course of the Santa Fe River near High Springs,
Florida: Gainesville, Fl, University of Florida, MS Thesis, 55p.

Smart, C.C. and Ford, D.C., 1986, Structure and function of a conduit aquifer, Can. J. of
Earth Science, 23: 919-929.

Smith, L.A., Martin, J.B., and Screaton, E.J., 2001, Surface water control of gradients in
the Floridan Aquifer: Observations from the Santa Fe River Sink-Rise system,: In:
J.B. Martin, C.M. Wicks, and I.D. Sasowsky (Eds), Hydrogeology andBiology of
Post-Paleozoic Carbonate Aquifers, Karst Waters Institute Special Publication,
Charlestown, West Virginia, No. 7: p. 44-48.

Smith, L.A., 2004, Using 222Rn as a tracer of mixing between surface and ground water in
the Santa Fe River Sink/Rise system: Gainesville, Fl, University of Florida, MS
Thesis, 63p.

Thrailkill, J., 1985, Flow in a limestone aquifer as determined from water tracing and
water level in wells, J. Hydrol., 78: 123-136.

Thrailkill, J., Sullivan, S.B., and Gouzie, D.R., 1991, Flow parameters in shallow
conduit-flow carbonate aquifer, Inner Bluegrass Karst Region, Kentucky, USA, J.
Hydrol., 129: 87-108.

Vervier, P., 1990, Hydrochemical characterization of the water dynamics of a karstic
system, J. Hydrol., 121: 103-117.

White, W.A., 1970, The geomorphology of the Florida Peninsula, Geological Bulletin
No. 51, Florida Geological Survey.

White, W.B., 1999, Conceptual models for karstic aquifers: In: A.N. Palmer, M.V.
Palmer and I.D. Sasowsky (Eds), Karst Modeling, Karst Waters Institute Special
Publication, Charlestown, West Virginia, No 5: 11-16.

Woods, T.L., Fullagar, P.D., Spruill, R.K., Sutton, L.C., 2000, Strontium isotopes and
Major elements as tracers of ground water evolution: Example from the Upper
Castle Hayne Aquifer of North Carolina, Ground Water, 38: 762-771.

Worthington, S.R.H., 1999, A comprehensive strategy for understanding flow in
carbonate aquifers: In: A.N. Palmer, M.V. Palmer and I.D. Sasowsky (Eds), Karst
Modeling, Karst Waters Institute Special Publication, Charlestown, West Virginia,
No 5: 11-16.















BIOGRAPHICAL SKETCH

Brooke Sprouse was born in Huntington, West Virginia, in October 1977. After

graduating from Huntington East H.S. in 1995 she attended Furman University in

Greenville, South Carolina. She graduated with a B.S. in earth and environmental

sciences in 1999. Following a two-year hiatus from school, she began graduate studies in

geology at the University of Florida in the fall of 2001. She is currently seeking

employment in the environmental/geological sciences fields in Charlotte, North Carolina.