Stable isotopic investigation of the hydrological cycle of West- Central Florida

Stable isotopic investigation of the hydrological cycle of West- Central Florida

Material Information

Stable isotopic investigation of the hydrological cycle of West- Central Florida
Netratanawong, Toedsit
Place of Publication:
Tampa, Florida
University of South Florida
Publication Date:
Physical Description:
ix, 120 leaves : ill. ; 29 cm


Subjects / Keywords:
Hydrologic cycle ( lcsh )
Stable isotopes ( lcsh )
Dissertations, Academic -- Marine Science -- Doctoral -- USF ( lcsh )


General Note:
Thesis (Ph. D.)--University of South Florida, 1995. Includes bibliographical references (leaves 105-113).

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
021540725 ( ALEPH )
33835070 ( OCLC )
F51-00192 ( USFLDC DOI )
f51.192 ( USFLDC Handle )

Postcard Information



This item is only available as the following downloads:

Full Text


STABLE ISOTOPIC INVESTIGATION OF THE HYDROLOGICAL CYCLE OF WEST-CENTRAL FLORIDA by TOEDSIT NETRATANAWONG A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marine Science University of South Florida August 1995 Major Profe ssor: William M Sackett, Ph.D


Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Ph.D. Dissertation This is to certify that the Ph.D. Dissertation of TOEDSIT NETRATANAWONG with a major in Marine Science has been approved by the Examining Committee on March 29, 1995 as satisfactory for the Dissertation requirement for the Doctor of Philosophy degree Examining Committee: Maior Professor: Willtam M. Sackett. Ph .D. Member: Sam B. Upchur c h, Ph.D. Member: Robert H Byrne, Ph D. Member: Kent A. Fanning, Ph.D. Member: EdwardS. VanVleet, Ph D.


TABLE OF CONTENTS LIST OF TABLES m LIST OF FIGURES IV ABSTRACT vii CHAPTER 1 GENERAL INTRODUCTION 1 Objectives of this Dissertation 6 CHAPTER 2. METHODS OF STUDY 8 Sampling, Storage, Treatment, Handling Techniques, and Study Areas 8 Gener a l Sampling Storage 8 Rainwat e r 9 Lake Water 9 E st uarine Water 10 Atmospheric-Water Vapor and Tran s pired Water 10 Ground Water 13 Carbonate Aquifer Samples 15 Pan Experiments 16 Analytical Techni ques 17 Standards 17 o180 of Water Samples 18 oD of Water Samples 23 o13C of Total Di sso lv e d Inorganic Carbon 23 o180 and o13C of Carbonate Rock Samples 24 Salinity Analysis 24 Statistical Analyses 25 CHAPTER 3. WATER CYCLE, RAINWATER, ATMOSPHERIC WATER VAPOR, AND EVAPORATION 26 Rain fall in We s t-Central Florida 30 Atmospheric W ate r Vapor in West-Central Florida 37 West-Central Florida's Evaporation 42 Chapter Di sc u ss ion 45 Rainfall 45 Atmospheric Water Vapor 46 Evaporation 47 Chapter Summary 49 CHAPTER 4 RIVER, AND ESTUARINE WATERS IN THE TAMPA BAY 50 o180 and o13CD!C in Tampa-Bay Waters 53 Chapter DiscussiOn 53 Chapter Summary 61 i


CHAPTER 5. GROUND WATER AND MIRROR LAKE WATER 63 Ground Water in West-Central Florida 63 o180 and o13Cmc in Ground Waters and Carbonate Aquifers 68 Behavior along the Flow Paths 68 A Time-Series Study of o180 and o13C in Ground Waters at the Roy Haynes and Bamn-Wimauma Sites 79 Major Ion Chemistry in Ground Waters of West-Central Florida 85 Mirror Lake 88 Chapter Discussion 93 Ground Water and Carbonate Aquifers 93 Mirror Lake 95 Chapter Summary 95 CHAPTER 6. GENERAL DISCUSSION 97 Methodology and Standard 97 Implications on the Modelling of the Hydrological Cycle 97 CHAPTER 7. CONCLUSIONS 102 REFERENCES 105 APPENDICES 114 APPENDIX 1. THE CALCULATION OF SPECIFIC HUMIDITY (q) 115 APPENDIX 2. DESCRIPTIVE MINERALOGY OF THE CARBONATE SAMPLES OBTAINED FOR THE o180 AND o13C ANALYSES 116 APPENDIX 3. THE PREPARATION OF 100% H 3P04 FROM 85% H 3P04 AND P20s 117 APPENDIX 4. REGRESSED VALUES OF m AND o s FROM PAN EXPERIMENTS 118 ii


Table 1. Table 2. Table 3. Table 4. Table 5. Table 6 Table 7. Table 8. Table 9. Table 10. LIST OF TABLES Effects of Numbers of Well-Volume Purges on the o180 of Ground Waters (Uncorrected Raw Data), and Stratigraphic Depths of Wells and Depths at Which Rock Samples were Taken 14 Regression Equations for Rainfall, Its Amount, Temperature, o18"'0 and oD o180 in Average Atmospheric Water Vapor, Transpired Water from Leaves, and Evaporative Water from Pan Experiments Evaporative Pan Experiments, and the Inferred Fractionation Factor, ex+ 37 38 43 Comparison of the Calculated oE from Craig and Gordon (1965; See Equation 8 Here) and Allison and Leaney (1982)'s ConstantFeeding Pans 48 Estimated Freshwater Flow Rate into Tampa Bay (After Zarbock, 1991) 51 Calculated V5 Vb, b. V ( = V 5 V b) and Replacement Time of Tampa-Bay Water 61 Estimated Freshwater Uses for the Year 1990 in West-Central Florida (Units in Million Gallons per Day, After Marella 1992) 64 Averages of o180 and o13CDic of Ground Waters in the Roy Haynes and Baum-Wimauma Sites 84 A Summary of Concentrations (mM) of Major Constituents in Upper Floridan Water along Northern and Southern Flow Paths, and of Major Ions in Rainwater of Central Florida (After SWFWMD, unpublished data, and Hendry and Brezonik, 1980, and Junge and Werby 1958). 89 iii


LIST OF FIGURES Figure 1. A Schematic of the Hydrological Cycle of West-Central Florida 2 Figure 2. Diagrams of Funnel (A) and Rain Gauge (B) Used in the Study 10 Figure 3. Sampling Locations for Estuarine Waters, Ground Waters, and Rainwater. The Big Circles are Ground-Water Sites (See Chapter 5 for Details). The Small Circles and Squares in Tampa Bay and Adjacent Rivers are Estuarine-Water Sites during March, 1990 and January, 1993, Respectively. Ground-Water Flow Paths from the Green Swamp to Coastal West-Central Florida are Represented as 2 Arrows 11 Figure 4a. Single Trap Used to Collect Atmospheric Water Vapor 12 Figure 4b. Reaction Vessel for Release of C02 from Ground Carbonate Samples 12 Figure 5. Cross Section of Aquifers from the Recharge Area in Green Swamp to Coastal Discharge Area of West-Central Florida. The Thin Vertical Bar Represents the Casing Depths; the Thick Vertical Bar is the Depth from the Open Casing to the Bottom of Well. The Total Well Depths are the Addition of the Thin and Thick Vertical Bars Carbonate Aquifer Samples were Taken from the X Marks 15 Figure 6. Diagram of the Plastic Syringe with Plastic Stopcock and a Special 0-ring Seal Used to Equilibrate C02(g) with 20 Water Samples Figure 7 Plot of the Change of the o180 of Residual C02(g) w1th T1me 21 Figure 8. Equilibration Time Needed to Reach Isotopic equilibrium between Water Samp l e and Added C02(g) The o18o of Equ i librated C02 is Shown on the Ordinate. 22 Figure 9a. Fifty-Year Variation of the Average Rainwater's Amounts in West-Central Florida The Average Amounts for Tampa and St. Petersburg are 119.8 and 132.7 em, Respectively (47.2 and 52.2 Inches, Respectively; NOAA 1941-1992) 31 Figure 9b FiftyYear Variation of the Average Air Temperature in WestCentral Florida The Average Temperatures of Tampa and St. Petersburg are 22.5 and 23.2 oc, Respectively (72.4 and 73 8 F Re s pectively; NOAA 1941-1992). 32 iv


Figure 10. 5180 Variation of Rainwater during the Year 1991 and 1992 in St. Petersburg, Florida The Weighted Mean is -4.4 o/oo 34 Figure lla. Linear Relationship between the Amount of Rainwater and Its o18o in St. Peter s burg Florida 35 Figure 11b. A Scatter Diagram Showing the Relationship between the Surface Air Temperature and o18o in Precipitation in St. Petersburg, Florida 36 Figure 12. Linear Correlation between the Specific Humidity and the 5180 of Transpired Water in Boyd Hill's Environmental Research Area 40 Figure 13a Linear Correlation between the Salinity and o180 in Tampa Bay, Florida 54 Figure 13b Linear Correlation between the Salinity and o13C in Tampa Bay Florida 55 Figure 14a. Variation of Total Disso lved C02 and Its o13C in TampaBay Estuarine Waters 56 Figure 14b. Variation of Total D isso lved C02 along a Salinity Gradient of Tampa Bay 56 Figure 15. A Box Model of Inputs and Outputs of Waters from Tampa Bay, Florida 60 Figure 16. Hydrogeological Sections of Ground-Water Aquifers in West-Central Florida (After Miller, 1986; Swancar and Hutchinson, 1992) 65 Figure 17. Schematic of Flow Paths of Ground Waters from the Recharge Area Close to the Green Swamp to the Coastal Discharge Area in the Tampa Bay 67 Figure 18a o180 Va ria tion during the Semi-Annual Studies of Ground Waters from the Recharge to Discharge Areas 69 Figure 18b. Distribution of o180 Along the Northern and Southern Flow Paths from the Green Swamp to Coastal West-Central Florida 70 Figure 19a. o13C Variation during the Semi-Annual Studies of Ground Water s from the Recharge to Discharge Areas 72 Figure 19b Distrib ution of o13C Along the Northern and Southern Flow Paths from the Green Swamp to Coastal W es t-Central Florida 73 Figure 20. A Scatter Diagram of o13C and o180 for Aquifer Carbonates from We s t-Central Florida 75 Figure 21. The Linear Relation s hip between o13C of Carbonates and Their Percentages from We s t -Ce ntral Florida 76 v


Figure 22. o180 and oD of Rainwater (This Study) and Local Ground Water (from Swancar and Hutchinson, 1992). The Intercept between the Two Best Fitted Lines is the Original Source of Ground Water in West-Central Florida. 78 Figure 23. Temporal Variation o f o18o i n Ground Waters at Roy Haynes (R) and Baum-Wimauma (W) Sites 80 Figure 24. Temporal Variation of o13C in Ground Waters at Roy Haynes (R) and Baum-Wimauma (W) Sites 81 Figure 25 EC02 Changes during the Sampling Intervals for the Roy Haynes and Baum-Wimauma Sites 82 Figure 26. Linear Correlation between the EC02 and the o13Cmc in Roy Haynes's Surfici a l Water 83 Figure 27a. Major Ion Concentrations of Ground Waters in the Northern Flow Path of W e st-Central Florida 86 Figure 27b. Major Ion Concentration s of Ground Waters in the Southern Flow Path of West-Central Florida 87 Figure 28. o180 V a ri a tions in Mirror-Lake Water during the Year 1991 90 Figure 29. A Box Diagram for Various Inputs and Outputs of Waters in M irror Lake St. P e tersburg Florida 92 Figure 30 A Linear Correlation between the Adjusted USF's o180 and the USGS's o1 80 data 98 Figure 31. A Schematic Summary of o1 80 of Various Water Types in Wes t-Centra l Florida 101 vi


STABLE ISOTOPIC INVESTIGATION OF THE HYDROLOGICAL CYCLE OF WEST-CENTRAL FLORIDA by TOEDSIT NETRATANA WONG An Abstract Of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Dep a rtment of Marine Science University of South Florida August 1995 Major Profes s or: William M Sackett, Ph.D. vii


Stable isotopic compositions of oxygen, carbon, and hydrogen were used to identify the sources of water and of organic matter, pathway reactions, evaporation processes, and transfer mechanisms among various water types as well as the extent of water-rock reactions. An attempt to understand the isotopic compositions of the hydrological cycle of west-central Florida, their variations, baseline levels and controlling mechanisms constitutes parts of this dissertation. A comparison of o18o in rainwater (weighted mean = -4.4 o/oo) and surficial ground water (mean = -4.1 o/oo) indicates that rainwater recharges the surficial aquifer with only a minor isotopic change o180 vs oD plots of rainwater and ground waters suggest that recently recharged water in the upper Floridan aquifer is similar to today's rainwater. As shown by calculations, carbonate aquifers react extensively with the ground waters in the upper Floridan aquifer through time to cause a shift of o18o from -4.4 to -2 8 o/oo Changes in o18o and o13Cmc in ground waters of the upper Floridan aquifer in two flow paths suggest that the confined nature of the aquifer causes the difference. Parts of the upper Floridan aquifer supply freshwater to rivers discharging into Tampa Bay as evidenced by similarities in o180 and o13Cmc of both freshwater types Evaporation may play a role in shifting the o180 of rivers from ground-water source of -2 8 o/oo to -2.3 o/oo. Precipitation compositions obey the global Rayleigh distillation model. Atmospheric water vapor is kinetically separated, isotopically, from water bodies with a mixed signature from transpired and evaporative waters. Trees act as conduits for ground-water transfer into the atmosphere without changes in isotopic compositions. Transpiration seems to contribute more to the total evaporation than free-water evaporation based on 0-18 compositions of surficial aquifer water and of rainfall. Pan viii


evaporation experiments provide support for the theoretical and observed isotopic behavior of atmospheric water vapor o180 and o13C in Tampa-Bay water correlate linearly with salinity. Based on isotopic mass balance calculations, the replacement time of Tampa-Bay water is about 67 days Mirror Lake has invariant o18o during a one-year study and has a calculated replacement time for its water of 43 days. Abstract Approved : Major Professor: William M Sackett, Ph.D. Distinguished Professor, Department of Marine Science Date Approved : ix


ACKNOWLEDGMENTS The author would like to express gratitude to his parents for their continuing support, affection, and sacritice. They provided the author another best gift of life through higher education. Dr. William Sackett kindly introduced the author to the stable isotopic studies, and provided partial financial support, encouragement, and understanding. All committee members namely Drs Sam Upchurch, Kent Fanning, Robert Byrne, and Edward VanVleet are acknow l edged for their intuitve comments, and assistance. Dr. Paula Coble assisted the author during the defense process. The author is obliged to the following: Gregg Jones, Eric DeHaven, and staff of the Southwest Florida Water Management District for ground-water samplings and arrangements, Ken Yancey and staff of the Boyd Hill Nature Trail for water-vapor samplings, Beth Holme s for collecting water samples in March, 1990, Dr. John Arthur of the Florida Geological Survey for carbonate aquifer samples, Kelly Hammer and Trina Kavula for parti a l sample handling and processing, Drs. John Welhan Carol Kendall, Ray Wilson and Ann Tihan sky, Amy Swancar, Dick Fletcher and Bob Balfour for their kind assistance. Mr. and Mrs. Joseph Vargo, undoubt e dly, kept encouraging the author towards a balanced life of education and provided a warm hospitality. For long-term friends and colleagues, Jose, Linda Joey, Gabe and Sandy who showed the good parts of Americans. The inter-library loan service, through the hard work of Tina, Deanna, and others who never refus e d the seemingly-endless requests from the author. The Graduate School and the federal and s tate offices are acknowledged for their time and partial financial support.


CHAPTER 1 GENERAL INTRODUCTION 1 The hydrological cycle covers movement of water from the oceans into the atmosphere, precipitation onto the land and further runoff of surface and ground waters into the oceans. The cycle is perpetuated by the heat received from the sun. The latitudinal heat gradients further drive various atmospheric-circulation cells which result in different weather patterns, both seasonally and perennially. Physico-chemical processes govern the amount of water types of clouds and humidity in concordance with the reservoirs' geologic al parameters and evapo-transpiration processes An attempt to better under stan d the hydrological cycle of west -c entral Florida (Figure 1) based on stable oxygen and carbon isotopes is the goal of this dissertation. Direct mea s urements of different types of water and their contributions to different reservoirs are difficult and often rely on meteorological data. For example the amount of water vapor in the atmosphere is calculated from relative humidity, air temperature, and saturation water pressure (Mcilveen, 1992 ; Appendix 1); evaporation is estimated from calculation of the heat or energy mass balance (Knapp, 1985). An insight into identification of source, sink and re cycled waters obtained using the above methods is limited. However, with the introdu ctio n of chemical and isotopic tracers, detailed mechanisms of water transfer in the hydrological cycle began to be understood (Epstein and Mayeda, 1953; Dansgaard, 1953) Natural chemical tr ac ers, such as oxygen-18 or hydrogen-2 (deuterium), are intrin sic in nature because they constitute parts of the water molecules themselves Oxygen-18 comprises about 0 .20 % of the total abundance of all stable oxygen atoms (oxygen-16 = 99.76 %; oxygen-17 = 0.04%) in the Earth's crust (summarized in


Atmospheric water vapor Gulf of Mexico Seawater flow into the freshwater aquifers Note: Not-toscale diagram Surficial Aquifer Intermediate Aquifer Upper Floridan Aquifer Figure L A Schematic of the Hydrological Cycle of West-Central Florida


3 Hoefs, 1987) It is difficult to analyze the amount of oxygen-18 directly, as the bulk of water is in the oxygen-16 form Fortunately, with the invention of the isotope-ratio mass spectrometer, IRMS (McKinney et al., 1950), and an understanding of the factors which regulate the isotopic fractionation (discussed later), as well as the evolution of measurement techniques for oxygen-18 in samples, precise determinations of oxygen isotopic ratios in samples were made possible (Urey, 1947; Epstein and Mayeda, 1953). The oxygen-18 determination is performed by means of a gaseous carbon-dioxide (C02g) surrogate. The dried C02g is proce sse d from samples using one of several methodologies (see last section in Standards section of Methods of Study) The C02(g) carries the stable isotopic signature of oxygen (oxygen -16, -17, and -18) obtained from the water sample. The 180 (oxygen 18) of the sample gas is ratioed to the 160 (oxygen-16) of the sample gas compared to tho s e of the standard gas, with minor correction for ox ygen-17 (Craig, 1957). The comparison is made with a great precision on the IRMS. This C02(g) also carries the 13C17o16o, which is corrected for during the calculation of oxygen isotopic ratios on the IRMS. The small deviation of these ratios from unity is the delta oxygen-18 (8180) value in parts per thousand (oloo) of the sample A representative equation is (18ol160)sample I (18ol16o) standard = 1 + 8180/1000 8180 = [{(18ol1 60)sample I (18oll60)standard}1 ]1000 (1a) or (lb). The 8180 of the standard is usu ally se t to zero by definition. The standards for 0-18 determination, whether they are carbonate standards or seawater standards, have well defined absolute concentrations of both oxygen -16, -17, and -18 (McCrea 1950; Epstein and Mayeda, 1953 ; Craig, 1957; Baert sc hi, 1976; Gonfiantini, 1978)


4 The isotopic fractionation factor (a) is defined similarly to a chemical equilibrium, e g 2H2180(1iquid) + C160160(gaseous) ---2H2160(liquid) + C180 1 8 0(gaseous) (2) and K = [H 2160 ] 2 [C18o1 8.Ql = equilibrium constant 1H2 t s0prct6ot601 = 1H218U]2T[H2160f If a (fractionation factor) = (180/160)gaseou s coi (180 / 1 60)liquid watep where a = K 110; where n = 2 = the maximum number of exchangeable oxygens in any one of the molecules for the above reaction, then (3) (4) (Urey, 1947; Epstein, 1959). The rate of the above reaction is controlled by temperature (Urey, 1947). The basic isotope-ratio measurement for 0-18 composition in water is the comparison of 18o ;16o compositions in the isotopically equilibrated C02 gas with the water samples to those of the standard water. The 8180 of sam pl e is usually reported in parts per thousand (o / oo) deviation from the standard. For stable carbon (carbon-13) isotopic stud i es on total dissolved inorganic carbonate in ground water and e s tuarine water as well as on aquifer carbonate minerals the isotopic fractionation factors govern the isotopic behavior and resulting co2(g) composition (e g. Urey, 1947 ; Craig, 1953; McCrea, 1950 ; Sackett and Moore, 1966). Carbon-13 comprises about 1.10 % of the total a bund ance of stable carbon isotopes, while carbon-12 comprises 98.9 % of the total abundance (Hoefs, 1987). Biological


5 metabolism and biochemical pathways affect, to a great extent the 13ct12C distribution in natural organic tissues (Craig, 1953; O'Leary, 1981; Sackett, 1989). Since the C02(g) is used also in the determination of stable carbon isotopes (on an IRMS), the carbon and oxygen stable isotopes are interrelated instrumentally (e.g Craig, 1957) The 180/1 6 0 of sample can affect the 13ct12c signal, and vice versabecause of the mass overlapping of 12C18o16o and 13c16o17o (mass 46), and 13c16o16o and 12c16o17o (mass 45) Under the isotope-ratio mass spectrometer, the corrected o18o and o13C are whe r e co rr = corrected value mea s = measured value (5) (6) A, B-factors involving 13c1 6o17o and 1 2c160170, respectively, c d factors involving 17 0 species. The final delta values on mod e rn IRMS instruments (1980 onwards) have been corrected and adjusted for oxygen-17 In order to improve coherence and c ontinuity of the dissertation because of the wealth of information in the hydrological cycle, the main text is arranged as follows : Chapter 3: Water cycle, rainwater atmo s pher i c water vapor, and evaporation, Chapter 4 : River, and estuarine waters in Tampa Bay, Chapter 5 : Ground water, and Mirror-Lake water. Each chapter (3 through 5) contain s previous and current studies, as well as data generated by the author and chapter d is cus s ion and brief summary. Each is almost complete and independent of other chapters General discussion based on the author's works on both oxygen and carbon i s otopes follow as Chapter 6. Conclusions appear in Chapter 7


6 Objectives of This Dissertation Since there are no background data on rainwater o18o in west-central Florida (IAEA, 1981; Rozanski et al., 1993) nor is there basic oxygen isotopic data on the evaporation and estuarine processes, the acquisition of these baseline data is important. In addition, temporal distribution of o18o in ground water is not known (Swancar and Hutchinson, 1992). The goals of this dissertation can be categorized as follows: 1 To monitor o18o of rainwater in St. Petersburg, Florida for two years 2. To determine the causal factors (temperature, amount, etc.) which regulate the 0-18 composition of rainwater. 3. To establish the relationship between the o180 of rainwater, atmospheric water vapor and ground water in wes t-central Florida 4. To characterize Floridan ground water from recharge in the Green Swamp area to the discharge area in southwest Hillsborough County of west-central Florida in terms of its 0-18 and C-13 (inorganic) compositions, and the effects of confined aquifers upon its 0-18 and C-13 (inorganic). 5. To examine the temporal and s easonal changes in o180 and o13C (inorganic) in 2 ground-water sites (Roy Haynes, Tampa and Baum-Wimauma, Wimauma) for surficial, intermediate, and Floridan aquifer waters 6. To estimate the degree of exchange of oxygen-18 between aquifer carbonate minerals and upper Floridan aquifer water 7 To determine the o18o of water in Mirror Lake, a small enclosed-basin lake in St. Petersburg, Florida, for pos s ible co nnections between rainwater, surficial ground water, and evaporation process during weekly sampling for one year.


7 8 To estimate the evaporation parameters from bodies of water in nature, and from pan experiments, and their rel ations wit h relative humidity and o18o in water and atmospheric water vapor. 9. To establish the relationship between the o18o of transpired water from tree leaves and soil water (ground water). 10. To establish the relationships among salinity, o13C and o180 of estuarine water in Tampa Bay. 11. To estimate the degree of exchange of surface waters with ground waters and to model the distribution of water types in different reservoirs of the hydrological cycle in west-central Florida


CHAPTER 2 METHODS OF STUDY Sampling, Storage, Treatm ent, Hand ling Techniques, and Study Areas General Sampling Storage oi8o 8 All water samples for o180 analyses were stored and sealed in glass jars or small glass vials at room temperature (23-26 C). The glass containers help reduce loss of H2160 through evaporation or "breathing" of water through the surface of the containers. As lon g as there was no loss of water through evaporation, the water samples could be stored for months without significant changes in o18o (Stewart, 1981) Preservative chemicals were not added to any of the water samples. For ground water and Mirror Lake samples the g la ss jars were completely filled, and capped tightly. This storage applied to the oD analyses. o13c For water samp les analyzed for o13C of t ota l dissolved inorganic carbon, the storage of samples at 5-8 o c was neces sary. The cool temperature was thought to reduce ba cterial oxidation of dissolved organic matter, which could alter the o13C (inorganic) of samples. However, there has been no laboratory study to determine the range of temperature of optimal storage for o13C analysis. Mercuric chloride or other preservatives was not added to samples for o13C analyses. Samples were analyzed as soon as possible (usually within one week of samp lings).


9 Rainwater Samples of rainwater were collected in a 25-cm plastic funnel (with small end closed, Figure 2a). The collecting funnel was placed on top of the roof of Department of Marine Science's Laboratory building (MSL), University of South Florida (St. Petersburg, Florida; latitude: 27.76 N; longitude : 82.63 W). The rainwater samples were analyzed for o180 within one week Most samples, which were collected over the weekends, were discarded because there were water droplets condensed inside the rain gauge. The condensation suggested that evaporation of water took place All samples were analyzed for o180 in duplicate The o18o study of rainwater was performed over a two-year period ( 1991-1992). The amount of rainwater was measured using a standard meteorological rain gauge (see Figure 2b). Average air temperatures were measured at nearby Albert Whitted Airport (Federal Aviation Administration's Control Tower; approximately 1 km northeast of the MSL building). The daily temperature data for the rainy days were obtained from the National Weather Service (Ruskin, Florida). These temperature readings were used to represent the temperatures at the MSL building Lake Water Mirror Lake in downtown St. Petersburg was monitor ed weekly for 0-18 composition during a one-year period (1991). Surface samp l es were taken from the southern side of the Lake by scooping lake water into glass jars. The jars were completely filled with water. All samples were analyzed for o180 in duplicate. The Lake water was assumed to be well mixed during 1991. Unfortunately, the average water temperature reading was not taken at the time of sampling, but it was expected to be within a few degrees of air temperatur e.


10 B Figure 2. Diagrams of the Funnel (A) and Rain Gauge (B) U s ed in the Study Estuarine Water Estuarine-water samples from Tampa Bay were collected at 2 diff erent periods: March, 1990 (mainly by Beth Holmes), and January, 1993 (by the author) One sample collected from each station was left at room temperature for 5180 analy s i s, while another was kept cold in a cold box (:::::: 5 C) for o13C analy s is. Locations of the sampling sites are dep i cted in Figure 3 Salinity was also determined for thes e samples us i ng a micro-chlorinity titration (to be described in the Analytical Technique s section). Atmospheric-Water Vapor, and Transpired Water from Plants' Leaves Either open moist air or moisture from the enclosur e bag on a branch of leave s wa s pumped at a rate of 1 to 1 1 / 2 litef / minute through a singles tep glass trap which wa s maintained at -85 o c using a slush of isopropanol-liquid nitrogen (Figure 4a). The slush was prepared by adding enough liquid nitrogen to isopropanol until the alcohol became a cold slush Schoch-Fi s cher et al. (1984) showed that both temper a ture and


11 km 0 20 40 50 28N 82W Figure 3. Sampling Locations for Estuarine Waters, Ground Waters, and Rainwater. The Big Circles are Ground-Water Sites (See Chapter 5 for Details). The Small Circles and Squares in the Tampa Bay and Adjacent Rivers are Estuarine-Water Sites during March, 1990 and January, 1993, Respectively. Ground-Water Flow Paths from the Green Swamp to Coastal West-Central Florida are Represented as 2 Arrows. flow rate of this moisture-trapping method affected the total amount of water vapor collected, and the subsequent degree of 0-18 fractionation. At an air temperature of I 0 C, relative humidity of 70 %, flow rate of 0.5 liters / minute, and a trapping temperature of -80 oc, the fractionation of o180 caused by incomplete condensation is negligible ( < 0 05 o/oo vs SMOW) The time required to trap atmospheric water vapor was dependent on humidity and air temperature levels on the sampling days and sites. Typically, the pump was run at a rate of 11122 Llminute for about 5 hours to condense enough water vapor ( 3 mL) for the analysis (see Appendix 1 for example)


Glass trap I To rotary pump ? To Atmosphere A ...__Alcohol Slush Bath -85 oc Figure 4a. Single Trap Used to Collect Atmospheric Water Vapor Joint B Side-armed acid port Figure 4b. Reaction Vessel for Release of C02 from Ground Carbonate Samples Relative humidity and temperature were measured using an electrical hygrometer (thermo-resistor cell) 12 Atmospheric-water vapor from outside of the MSL building was collected in late 1991, and during summer, 1993 Transpired water was collected from oak trees at the Boyd Hill Nature Trail (St. Petersburg, Florida) during late 1993. Because of the small amount of sample that could be obtained in a rea so nable working time (see total amount of water vapor in Appendix 1), only one o180 determination per sample was performed The samples were not analyzed for oD nor o13CDic


13 Ground Water Ground water wells were purged for 3-well volumes (of the entire well length) before collecting samples. This step was necessary to replace stagnant ground water lying in the well casing A few tests on the number of purges required for ground water were conducted using 0-18 composition of selected wells. This test would provide a minimum purge which was necessary to yield representative samples for the stable oxygen analyses Test results showed that for 8180 only 1'/2 well volume purges were sufficient for replacing the stagnant ground water (Table 1). However, in order to be compatible with other published chemical data, all ground water samples were purged for 3 well volumes before collecting samples for 8180 and o13Cmc analyses. Two sites, Roy Haynes in north Tampa and Baum-Wimauma near Sun City Center (Florida), were monitored for both o18o and o13Cmc in surficial and the upper Floridan aquifer waters on a bimonthly basis from February, 1992 through October, 1993. Another transect study of ground waters from the recharge area in the Green Swamp to a discharge area in southwest Hillsborough County (Figures 3 and 5) was performed over 3 periods: March, 1992; August, 1992; February, 1993 All ground water samples were analyzed for o180 in duplicate. However, only one 813Cmc analysis per sample was performed.


14 Table 1. Effects of Numbers of Well Volume Purges on the o18o of G roun d Waters (Uncorrected Raw D ata), and Stratig r aphic Depths o f Well s and Depth s at Which Rock Samples were Taken 3 well volume purges 1112 well volume p u rges Green Swamp (Floridan) aquife r 4.04 -3 9 5 Roy Haynes (Floridan) aquifer -2.69 -2.62 ----------------------------------------------------------------------------------------------------Wells Stratigraphi c Unit Stratigraphic Unit Stratigrap hic Unit and Total Depth and Casing Dep t h and Carbonate Rock Sample Depth (depths in feet) (depths in feet) (depths in feet) Green Swamp A von Park (285) Ocala (80) Oca l a (11 0) 88 Avon Park (385) Avon Park ( 1 95) A von Park (220) Romp 87 A vo n Park (380) A von Park (300) A von Park (330) Romp 68-2 Ocala (221) Tampa (120) Ocala (150) Roy Haynes Tampa (80) Tampa (70) Unknow n (?) Romp TR 1 0-2 Tampa (125) Tampa (115) No Sample Loughman A von Park (24 7) Ocala (85) Ocala (110) Lake Alfred 1 A von Park ( 425) Ocala (102) Ocal a (120) Lake Alfred 2 A von Park (?) Oca l a (102) Peace River (90) Lakela n d T ampa (127) Tampa? (98) No Samp l e Romp D V-2 Suwannee ( 130) Suwannee (1 08) Suwannee (115) Baum-Wimauma Hawt horn ( 1 00) Hawthorn (80) Hawthorn (80) Romp TR 9-2 Hawthorn ( 148) Hawthorn ( 1 1 8) Hawthorn (130) Romp TR 9 1 Tampa (288) Hawthorn ( 124) Haw t horn (150)


m 50 0 -50 -100 -150 -200-100 75 TR9-2 68-2 Suwannee Limestone 50 km 25 20 10 5 0 Lakeland 87 Avon Park Formation I I I I I ft 200 88 .... 't:S 0 i --f -400 c:l e c:J-I-600 I I I I I I I mites I 70 60 50 40 30 20 10 0 Figure 5. Cross Section of Aquifers from the Recharge Area in Green Swamp to Coastal Discharge Area of West-Central Florida. The Thin Vertical Bar Represents the Casing Depths; the Thick Vertical Bar is the Depth from the Open Casing to the Bottom of Well. The Total Well Depths are the Addition of the Thin and Thick Vertical Bars. Carbonate Aquifer Samples were Taken from the X Marks. I-' U1


16 Carbonate Aquifer Samples Carbonate rock samples from aquifers in the above transect (Figure 5 and Table 1; mostly from the upper Floridan aquifers) were obtained from a depository of the Florida Geological Survey (Tallahassee, Florida). The whole samples were ground into fine powder before roasting at 400 o c (McCrea, 1950) for 2 hours under vacuum to drive off volatile organic matter. This volatile organic matter could interfere with the liberated C02 from the sample during processing. The roasting temperature was below the decomposition temperature of the carbonate structure, so there was no loss of carbon or oxygen isotopes. There was no attempt to separate the size fractions of the ground samples The s amples were allowed to cool under vacuum, and were packed in aluminum foil. All carbonate samples were analyzed in duplicate for o18o and o13Cmc by reacting with 100 % phosphoric acid, H3P04 All ground carbonate rock samples were determined for their mineralogy (see Appendix 2) Pan Experiments Free evaporation of water in 2 eleven-liter pans was conducted on the roof of the MSL building The water in these two pans was evaporated to dryness. Evaporated water vapor and pan water were collected periodically during the course of evaporation (before complete dryness), and analyzed for o180. A "constant-feeding pan" (where the water level was maintained at a constant level by adding water with a known isotopic ratio) was set up in parallel with the free evaporation pans, and monitored for changes in o18o. Relative humidity and air temperature were recorded for a set of data generated from all these pans Ten milliliter sub-sample was taken periodically from the pan water for o18o analysis. This subsample was assumed not to affect the bulk oxygen isotopes of remaining water in the pan However, because of the small sample


17 size of water sample (for both pan water and evaporated water), only one analysis of o180 per sample was conducted. The oD analyses were not performed on water samples from this pan experiment. Analytical Techniques Standards 0-18 Standard 5180 of water samples was analyzed by equilibrating a known amo unt of C02 with a known amount of water in a closed container (Epstein and Mayeda, 1953 ; Craig, 1961; Dansgaard, 1 964) The isotopically equilibrated gaseous C02 from the sample was then dried, and run against the gaseous C02 from the sta ndard, generated in a similar fashion, on the same isotope ratio mass spectrometer (IRMS). The gaseous C02 carried the 0-18 composition of water through the partition (fractionat ion) chemistry of the isotope during the equilibration. Epstein and Mayeda (1953) used the gaseous C02 produced by re a cting a reference carbonate material with 100 % phosphoric acid, H 3P04 at 25 o c (McCrea, 1950) as a standard. Dansgaard (1964) used local tap water in Denmark as a standard water in his l aboratory during earlier works. However Craig (1961) preferred using equilibrated C02 prepared from Standard Mean Ocean Water (SMOW) as a reference gas for the 0-18 determination At 25 C, Craig's standard (gaseous C02 ) is 0.20 o / oo lighter than the gaseous C02 prepared from PDB standard (Belemnite from the Pee Dee Formation, South Carolina, USA). These surprisingly close values (0.20 o / oo difference) of the two standards may, at first, lead to an erroneous idea that bot h standards are interchangeable within analytical error. In fact, the reaction of carbonate with H3P04 at 25 o c causes a


18 fractionation of 0-18 (by =:: + 10.3 o/oo) in the evolved gaseous C02 with respect to carbonate in the system CaC03(s)-C02(g) (Sharma and Clayton, 1965). But the water C02 equilibration at 25 o c causes a fractionation of 0-18 (by =:: +41.2 o/oo) in the final residual gaseous C02 with respect to water in the system H 20(l)-C02(g) (O'Neil et al., 1975). Friedman and O Neil (1977) clarified this misunderstanding in some detail and gave a conver s ion equation for the PDB and SMOW standards for o18o at 25 C. Since the measurements of isotopic ratios on the IRMS involve comparison between C02 of the sample and the standard, the resultant isotopic data are always reported in relative abundance (in o/oo) with respect to a standard at a given temperature. Temperature adjustment on the o180 is needed if the experiments for the samples are conducted at different temper a tures than the standard's. The working SMOW standard wa s obtained from deep water of the Gulf of Mexico (depth =:: 1,000 m; salinity =:: 34.3 o loo) during lhe Septembers of 1990 and 1991, and treated as 0 oloo vs international SMOW standard. The 5180 of these 2 standards were identical within analytical error. The USF working SMOW standard was within 0 3 oloo to the working SMOW standard used in the USGS's Menlo Park laboratory (Carol Kendall's Lab.; see further dis cussion in Chapter 6). C-13 Standard o13C of both carbonate samples and total dis s olved inorganic carbon (in water samples) were measured against a Solenhofen standard (Bundesanstalt fur Geowissenschaften und Rohstoffe Germany). This standard has o13C of -0.15 o/oo and o18o of -4.99 oloo vs PDB at 25 o c


19 o180 of Water Sampl es (Rainwa t e r Atmospheric Water Vapor, Ground Water, Lake Water, and Transpired Water) A modification of the Epstein and Mayeda ( 1953)' s te ch nique was adopted for determination of the 0-18 composition of water. In practice, a 20 mL w ate r sample was introduced first into a plastic syringe, followed by 20 mL of high-grade C02 (Figure 6). The plastic syringe was equipped with a two-way stopcock. The syringe was agitated vigorously at 2.5 Hz for 11/2 hours on a paint shaker at room temperature. The ratio of atomic oxygen in water compared to that in gaseous C02 initially, was approximately 600: 1, that is [20 mL of water sam ple][ I g/mL] : [20 mL of co0][2 mole/1 mole C02]. [18 g/mole] [22.4 xiOO mL/mole] However, the actual ratio was slight l y lower than 600 :1, but not s ignificantly l ow for the 0-18 composition of gaseous C02 to control the 0-18 composition of the final equilibrated residual gaseous C02 (see Craig [ 1957] for further discussion). In other words, the o18o of the water primarily controlled the o180 of the final re sidual gaseous C02 The equilibrated C02 should be proces sed as soon as possible (preferably within 5 hour s ) to avoid loss of 12c16o16o through the wall of the syringe (Figure 7; see also Yo sh ida and Mizutani, 1986). The syringe was put on an adapter coupling with a plastic threaded joint and an 0-ring (Figure 6). A septum technique was not recommended becau se a serious leak could occur. The residual gas was then withdrawn and cryogenically trapped at liquid-nitrogen temperature (-185 C), while water vapor was separately trapped at -85 oc (using a cold slush of isopropanol). The dried C02 was later run on an isotope-ratio ma ss spec trometer, Finnigan MAT 250 with triple collectors, using a working SMOW standard as a reference gas (0 o / oo).


20 Glass joint 0-ring--adapter Plastic stopcock Water sample Figure 6 Diagram of the Plastic Syringe with Plastic Stopcock and a Special 0 ring Seal Used to Equilibrate C02(g) with Water Samples The agitation time (90 minutes) used in thi s study (Figure 8) was determined experimentally, and was close to the 100 minute equilibration time of Roether (1970; 18 oc and 2.4 Hz agitation). Since the agitations were conducted at a room temperature other than at exactly 25 o C, the temperatur e correction of -0.2 o / oo per degr ee Celsius (Vogel et a!. 1970) was adopted. By giving


-4 CQ fl.) i>-0 -5 00 ...c <.0 -6 cS18Q = -6.06(. 08) + 0.025(.00l)leak time, r2=0.989 0 50 100 150 200 Leak Time (hours) Figure 7. A Linear Plot of the Change of the a180 of Res idual C02(g) with Time


0 0 0 0 0 0 0 -3 Et 0 Cl.) {I) ;a. -4 0 00 c.o -5 tfolso = 0 .28(.06)(1/{time+ 1})0 .44(. 02), r2=0. 760 > 0 10 20 30 40 so 60 70 80 Equilibration time (minutes) Figure 8. Equilibration Time Needed to Reach Isotopic equilibrium between Water Sample and Added C02(g) The o18o of Equilibrated C02 is Shown on the Ordinate


23 A = the adjusted (corrected) o18o of sample (in o/oo), M = the measured o18o of sample (in o/oo), t(std) = the temperature (0C) at which the standard was processed, t(samp) = the temperature (0C) at which the sample was processed, A = M [( t(samp) t(std) )( -0.20 )] (7) The precision of the analysis was estimated from one degree Celsius error in temperature difference plus the error in the repeated analyses of the same C02 gas on both inlets of the mass spectrometer. The one degree difference translated into 0 20 oloo (Vogel et al., 1970), while the instrument error was 0 .10 o loo (-0.034 0 067, n = 147) Therefore the applicable uncertainty for o18o analysis was 0.30 o /oo (0 20 + 0 .10). This uncerta i nty was calculated by assuming that there was a relationship between the instrument variables and the temperature of the experiment in which they were not independent from another. The total error was then, a summation of each error. oD of Water Samples Five rainwater samples were sent to analyzed for oD at USGS laboratory in Menlo Park, California (which is managed by Carol Kendall). Water samples were converted to hydrogen gas using Zn( s ) reduction technique at 450 oc (Coleman et al., 1982). The purified H2( g) was run for oD on a mas s spectrometer against a SMOW standard with a correction of H3 +. Analytical error wa s 1 5 oloo. o13C of Total Dissolved Inorganic Carbon of Water Samples Following techniques that were suggested by Sackett and Moore (1966), each 60 mL water sample was reacted with 85 % H3P04 at room temperature (23-26 oc) in a special, side-armed glass fla s k Mook et al. (197 4) showed that temperature effect on


24 o13C exchange in the system C02 (g) -HC03(aq) was 0.30 o!oo in the 23 -26 o c temperature range ( = 0.10 o /oo per degree Celsius). The gaseous C02 evolved from the reaction was cryogenically separated from water vapor similarly to that for the 01 8 o described previously A Solenhofen limestone sample was used to generate a standard gas for o13C. For comparable precision, each reaction vessel was left for at least 15 minutes at 25 o c in a water bath. Analytical uncertainty was estimated to be 0.15 o loo ( 0.10 o /oo instrument error plus 0.05 o loo for the technique; assuming the variables are not independent). o180 and o13C of Carbonate Rock Samples Using the McCrea (1950) technique, powdered carbonate rock sample was reacted with 100 % H3P04 under partial vacuum at 25 o c in a special side armed flask for 12 hours or longer (see Figure 4b). The 100 % H3P04 was prepared by react ing enough solid phosphorus pentoxide (P 2 05 ) with 85 % H3P 04 stoichiometrically (see Appendix 3 for detailed calculation). Since the reaction, which generated C02 involved only two-third of the total oxygen atoms from carbonate (McCrea, 1950) temperature control of the reaction was critical. McCr ea ( 1950) suggested using a prolonged reaction time at 25 o c for 24 hours. Based on temperature effect on o180 of evolved gaseous C02 I would suggest that the reaction processes should be done at room temperature (see the di sc u ssion in the section of a nalysis of o180 in water). If the experiment were performed at 50, or 70, or 90 C, there is a possibility that the evolved gaseous C02 mi g ht reeq uilibrate with the equal molar amount of water at a different temperature during c ryo gen ic sepa ration ste ps. A Sol e nhofen limestone sample was used as a sta ndard for both o13C arid o180 in this carbonate study. Analytical uncertainty was estimated to b e 0 .10 o/oo for o13C and 0 .30 o loo for o18 o (assuming the variables are not indep ende nt). The amount of evolved C02 from


25 the reaction was measured manometrically Percentage of carbonate in the sample was calculated from the measured C02 amount and the sample's dried weight. Salinity Analysis A micro-chlorinity titration of water samples with silver nitrate AgN03 was adopted for salinity determination (Grasshoff, 1983) The chloride ions in water samples were titrated with silver ions in a flask which was periodically swirled manually. The white silver chloride precipitated and sank to the bottom of flask The end point was determined by observing the change of color of white precipitate (AgCI) to light pink using a sodium-fluoresceinate indicator. Analytical uncertainty of the technique was within 0 3 o/oo Statistical Analyses All linear regression lines performed in this study were at 95 % confidence level. Computer programmes, based on Excel and StatGraphics were used to assist the calculation


Chapter 3 WATER CYCLE, RAINWATER ATMOSPHERIC WATER VAPOR, AND EVAPORATION 26 As mentioned in the Introduction, the hydrological cycle is driven mainly by the heat received from the sun, and a variety of physical processes acting upon water molecules. During evaporation, a kinetic control of exchange of water molecules between liquid and vapor phases occurs (Penman, 1948; Craig and Gordon, 1965; Knapp, 1985; Steinhorn, 1991) This process is mainly dependent on the vapor pressure gradient, mass-transfer coefficient, and wind speed The evaporation also affects the 1 80/160 distribution between liquid water and water vapor (Craig and Gordon, 1965; Dansgaard, 1953, 1964). Under i sotopic equilibrium, any evaporation from su rface water can be explained simp ly by a Rayleigh distillation process1 (Dansgaard, 1953). This di stillation mechanism depends on the partitioning of isotopes between liquid water and water vapor at a given temperature, Q'(H201)-Q'(H20v) (Majoube, 1971). For example, at 25 C, liquid water with o180 of 0.0 oloo evaporating into the atmosphere yields a theoretical water vapor with o180 of -9.3 o/oo. In fact, early studies (e.g. Craig and Gordon, 1965) have suggested that th e kinetic factor involved during evaporation plays a great role in shifting the observed o18o of atmospheric water vapor to 1 3.0 o/oo at 75 % relative humidity instead of the predicted value of -9 o/oo. Under wet atmospheric conditions (100 % relative 1. Rayleigh proce s s i s a fractionation pro c e ss whi c h d e pend s on the initia l fra c tion of liquid pha s e and the rem a ining vap o r pha s e during the evaporati o n or c ond e n s ation of wat e r mol ec ule s. The representative equation i s R / R0= f{aIJ, where R= 1801160 of remaining vapor, Ro= 18Qfl6Q of original vapor f = fra ction of rem a ining water vapor, a = i s otopic f r a c tionation fac tor.


27 humidity) the observed o1 80 of water vapor will then be equal to the theoretical value at a given temperature (Craig et al. 1963; Craig and Gordon, 1965). D . H t6o. unng evaporation, 2 IS preferentially released into the atmosphere (Dansgaard, 1964; Merlivat, 1978). The condensation process reverses the trend-the H2 1 80 is preferen t ially concentrated in the precipitation. However, since not all of H2180 is initially evaporated into the air, the first condensate (as rain) will have o18o lighter than the original water (e.g. rainwater's o18o = -4 oloo from the source water of o180 water = 0 o /oo). This condensation process is described by a similar Rayleigh process (Dansgaard, 1953, 1964). As evaporation of surface ocean water occurs mostly in the equatorial zone (the warmest part of the oceans), the precipitation will separate H2160 from the H2 1 80 through atmospheric circulation cells from the tropics to the subtropics, and, finally, reaching the polar regions. Globally, the o18o in precipitation has distribution patterns parallel to the latitudes, and become s more ne ga tive (lighter) polewards (Yurtsever and Gat, 1981). In addition, seasonal effects on the o18o of precipitation have been documented (Lawrence and White, 1991). In general, the heavier the precipitation, the lighter is its o180. This process was coined "the amount effect" by Dans gaar d (1964) The amount effect may be used to estimate o18o of precipitation from the average amount of observed precipitation in a geographica l area. The calculation, depends on a relationship equation, such as o18Q = constant1 + slope(amount o f rainf a ll). The o18o of precipitation seems to vary linearly with surface air temperature which is less than 15 o c (confer Dansgaard, 1964 ; Jacob and Sonntag, 1991; Joussaume and Jouzel, 1993 ; Rozanski et al., 1993). This relationship caused enthusiasm within scientific communities as a possible means of estimating the surface air temperature in the pas t based on palaeo-water studies and information from precipitation (e.g. s now). However, the air temperature-o1 8o (precipitation) relation bre aks down at temp e rature


28 greater than 15 oc (e.g. Jacob and Sonntag 1991; St. Petersburg data below) In addition, the history of rain-cloud formation and the travel paths of precipitation may complicate the usefulness of this type of relationship (Jouzel, 1986) In water molecules, hydrogen comprises twice the molar ratio to oxygen atoms Natural abundances of stable hydrogen isotopes are: 1H = 99 985 %, 2H (or D) = 0.015 % (Hoefs, 1987). In natural precipitation the observed o18o varies linearly with the oD as exemplified by the equation : oD = 8o180 + 10, the equation for Global Meteoric Water Line (Craig, 1961) The intercept(= 10) here is called the deuterium excess. The excess is obtained from linear regression of oD vs o180 data. This deuterium excess value is dependent upon the relative humidity and temperature of the evaporating body of water (Merlivat and Jouzel 1979; Petit et al., 1991). Petit and co-workers (1991) have shown, based on the deuterium excess of surface snow in Antarctica and on modelling that the source of water vapor for snow in Antarctica is water vapor from the ocean around 30-40 S This idea may have major implications for palaeo-climatic studies in the near future. The slope of 8 in the Meteoric Water Line is valid for precipitation throughout the world (Dansgaard, 1964; Rozanski et al 1993). During evaporation, the slope value is lowered to around 3-5 depending on the humidity, air temperature and degree of rainout (Gat and Bowser, 1991) The degree of rainout depends upon the Rayleigh process and the remaining water vapor compared with the fraction of original water vapor. By examining the slope of the oD and o18o alone in a water type for evidence of evaporation may not be correct as processes such as interaction of meteoric water with carbonates on ground and exchange of precipitation with atmospheric water vapor could alter the slope as if evaporation takes place Evaporative parameters e.g. o s and m (equations 9-11), are better for prediction the degree of evaporation (the fraction of water which evaporates into the atmosphere from a body of water). the changing o values and extent of fractionation factor of water in a syst e m


29 During evaporation the equilibrium fractionation factor can be denoted as a* (e. g for HzOI --HzOg, a* = 0 .99071 at 25 oc; Majoube, 1971). Craig and Gordon (1965) introduced another term called E*, which under evaporation is equal to (lex )1000. TheE* for the previous a* is equal to 9 2866 TheE* term is often taken from (a+ 1)1000 for convenience, which equals 9 3736 at 25 C. (The a+ is the fractionation factor for the reaction H20g --H 20b and is 1.0093736 at 25 oc; a* = 1/a+) Within analytical error, thee* and (a+ 1)1000 terms are very close, and often are used interchangeably b y Craig and other workers. The evaporation process at the air-water interface seems to be complicated by an inclusion of kinetic effects on equilibrium fractionation. The contributio n from this kinetic factor is called .1. E. This kinetic factor depends upon the relative humidity of the overlying atmosphere ( .1E = 16(1-h)/1000; h=relative humidity; Gat, 1970). The total enrichment factor during evaporation of liquid water into atmosphere is E; E = E + .1. E. Based on studies by Craig and Gordon (1965) and Merlivat and Coantic (1975) there is a laminar layer of air overlying the water interface during the evaporation from water to the bulk, more turbulent atmosphere. The laminar layer resists the evaporation occurring at the water-air interface. The isotopic .1. E and P (isotopic transport resistance in the water/air interface) are the responsible resistance terms in the isotopic evaporation equations. Craig and Gordon (1965) derived an equation for the isotopic content of oxygen or hydrogen in evaporative water vapor, oE, as (8) where E = evaporation rate P\ = isotopic transport resistance in the liquid h = (air) relative humidity normalized to surface water temperature oa = isotopic content of atmospheric water vapor.


Welhan and Fritz (1977) further expanded the above equation to a steady-state condition (of oJ, in which the fraction of water that remained is close to zero as o5 = (hoa + e)/(h-e) (9) (o-oJI(o0-oJ = rn (10) and m = (oE-o)/(o-o5 ) where o5 = steady-state isotopic content ( = limitf....a o) oo = initial isotopic content of liquid at f = 1.0 o = isotopic content of liquid at f < 1.0 (11). 30 f = fraction of remaining liquid = V /V0 (volume at time t / volume at time 0) m = (h-e)/(1-h+ !J. e+(a*)Ep*J (12 ). o5 and m are unique for evapor a tion under various pan experimental condition s. The parameters obtained from the evaporative pan experiments can be used to estimate oE of lakes or other land-locked bodies of water. The estimated oE is comparable with the evaporation data obtained from energ y budget and mass-transfer methods (Welhan and Fritz, 1977) Rainfall in West-Central Florida Florida receives rainwater throughout the year Rainfall is common during the summertime, and generally appears as thunderstorms, showers, and heavy rainfall. Convective rain cloud formation is responsible for the Florida's summer rainfall (Winsberg, 1990) During the w i ntertime frontal precip i tation is predominant but the amount of rainwater is smaller compared to the s ummer rainfall. The rainfall in wes t central Florida occurs more frequently dur i ng the afternoon (Schwartz and Bosart, 1979). Fiftyyear averages of rainfall amount and surface air temperature for th e Tampa and St. Petersburg areas are shown in Figures 9a and 9b. The 1991-1992 rainfall amounts (:::::: 1 130 mm) were below the 50-year average ( = I ,3 00 mm) while


-e Cj .._, -= = 0 e = 250 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 200 150 100 a .. a St. Petersburg 1\ il 'i i \ iTj il Ill !II! p : t :r' n Ill il 'ill il I I fl II 11 1 I f'i11 I .'J n \ 1 r Ill 11; ; '' K b 1 --tc--Tampa I /II I I I li ,;. "' Ill lo /!i I I f I I II 0 1 '! l .. 1 1 L i d , .. u fi I 1 li .11.1 \ : l ll!; 1 = : 1 :: a i 1 K: r"' K I /v ,, .. r:= .I I I I T I I 1 !J I I ; 'i' I I I Iii; 1 \ f tl : 1 rl i t \: 1 tll I \ f \ 1 r i 1\ ;: \i:J li\ : I I i t F./ I I I Jl! l\ ' I j i \ I f\ ' 1\ fl .. i ,\ I I ll I \ H J I t I tJ 1 I I I II 1 I I I I r I I I' i \ ,\J,j, \ \ltl) li I I' \\tf,lll II) I, \'" \/11 :1(\ I a 1\) I a\ t I il W I 1 k I v I I lll + 'It 1 I a 1 'i':l 1 a 111 1 r II 1 il 1c it i u v l II :Jc year 50 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 1940 1950 1960 1970 1980 1990 Figure 9a. Fifty-Year Variation of the Average Rainwater's Amounts in West-Central Florida . The Average Amounts for w Tampa and St. Petersburg are 119.8 and 132.7 em, Respectively (47.2 and 52.2 Inches, Respectively; NOAA 1941 1-' -1992).


-u <= '-' 1-o = .... eo: 1-o c.. s .... I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 24.5 I I I I I I I I I I I I I I I I I I I 24.0. 23.5 23.0 22.5 22.0 21.5 21.0 /i j h I\ I\ J \ JW .111 f \ 1= \ f \ n i \ I \\ "'l I m B,;,.! ii i .!J r i I I El r : r: i t ,. I I I i "$ :: i tl i : i I t:l il '\ 1 \ 1 r: 1: i 1n i : i : 1 1 1 .r I 'Ill :I :: !P i: I I I: : : I v '\ I :1 : I : : I : I'!' : I "' I ,H 1 n i i i\ "' i i 1 1 i i i 1!1 ..,. 1 : i li "'f .. : I . I I I I I' \ i I I 1\ j i f \ / \ j i ... I i i '\ j i i I I \ i I ; : I I 1 h 1:1 : :: : : r.t : : : I f' I : a : r : i 1 I .: 1 IQ : 1 : :: : : :: : i !" 1 : 1 I ,. 1 : 1 t' ;: I \ I .J. I I /' I : : : I I,;, iL : I, tiJ'll5\ I: ::at.:: l:l;t!J.,. .:. B' \1.' \ : 'f'-"'! I \ J, \ I{ \i I \1 \ i I \ : : \ v : \ : f' I I I 1'1\; i r:( i1 11 \ f \: i i I .J I 1 1 1 I I !a I Ill \i l f I I I I \ j Q i: "" :111: I I 1 )_ 1 I lie ll5 I J El )It li I I I I :i a r I )f I .Jr I I I t1 II I; I I I I I II I I r 'lc \ I I I I 1 II I I I \ I 1 X.* I I \' lll: I I '-I I \ I r I I I 1 1 \ I I 1 I rl I 1 I I 1 I k jc I )tlf I I I I Ji: I lr lie \ I II I \1 \1 'lc II I '* \I I i I i I I lr I I I "' Jr I I a-St. Petersburg 1 1 II -tc-Tampa ... 1-... fyear 20511 I I I I I I I I f I I I I I I I I I f j I I I I I I I 1 I I I I I I I I I I I I I I I I I I I 1 I I I I t 1940 1950 1960 1970 1980 1990 Figure 9b. FiftyYear Variation of the Average Air Temperature in West-Central Florida. The Average Temperatures of Tampa and St. Petersburg are 22.5 and 23.2 C, Respectively (72.4 and 73.8 F, Respectively; NOAA 1941-1992).


33 the 1991-1992 temperatures were close to the long-term average ( = 23 C) Figure 10 illustrates variations of o180 of individual rainfall in St. Petersburg collected in the present study. The sample most depleted in 180 (o18o = -11.6 o/oo) was obtained during a tornado in 1992 and repre s ented one of the heaviest rainfall events (4.5" or 114 mm). A weighted mean2 for the o1 80 in rainwater during 1991-1992 is -4.4 o/oo; a normal mean3 of o18o for the same rainwater is -3.3 o/oo Table 2 describes regression equations among rainwater o180, amount, and surface air temperature. A linear relationship between the amount of rainwater (independent variable) and its o180 (dependent variable) was not high equation 14; Figure lla). Multiple correlation b e tween the amount of rainwater (independent variable) and its o180, and air temperature (dependent variables) is weak equation 15). There is relationship between o180 in rainwater and surface air temperature (Figure 11 b). Carol Kendall (USGS Menlo Park) performed analyses of five rainwater samples for oD. These oD data were regres s ed with the o180 data to construct a "St. Petersburg" meteoric water line [oD = 7 65(.20)5180 + 10.57 ( 1.14) ; = 0 998, n = 5]. This relationship (Table 2) is close to the one that Craig (1961) has found (oD = 8ol8o + 10, a Global Meteoric Water Line; see also Figure 22 in Chapter 5). The similarities in the oD and o1 80 of rainwater in St. Petersburg and Global Meteoric Water Line suggest that similar evaporation-condensation processes contro l the r ainfall pattern. 2 weighted m e an = Ei{(am o unti)(Oi)} / {E i(amo unt)} IAEA(l981) (13a) 3. normal mean = Ei(Oi)/ E i ( 13b)


fl) > 0 0.0 -2.5-5. 0 -7.5 10.0 -12.5 p I p i o i ,f..... r .. D iL OD I 0 D I ... \ -9 ;, .._ _g_ ol: ; :! ?':. h \ p6 :: \ \ : .l I P t" ./ I 0 ;1 1 I f h 0 ci\: : : 00! j c.,. .'i P. f eli!" fa f I l bl ?-. _.d?\ 9 i J.J b I j do j! ...... j i f n .. ,.1 .:f1! 1 1 P b f .. l .c,( !D -o o \ .. n : I j \ 1 !\ . l! l / 1,1\ f:1 : 1 'fl. : ., L.., l \,',,,'=.. ::_/t .. _f \0 :. __ Yt wci i \ t 1 U 0 V v ....................... ... J W.i j () b/ .. & ; j ; b b \{ I I I ; ; ........ O 1 99 1 o -1 992 -15.0 0 6 jloo 0 z 1 2 F i g ure 10. 151 80 Varia t io n of Rai n water d u ri n g t h e Year 1991 an d 1 992 in St. Petersburg, Florida. The Weighted Mean is 4 .4 o/oo.


::: 0 rJ'J Cl) ..... 0 QO 1"""'4 c.o ol8o = -2.48(.20)-0.0587(.0097)amount, r2=0.252 + Oi* + + -5 -10 + 0 2 0 4 0 Amount of 6 0 Rainwater 8 0 (mm) 100 120 Figure lla. Linear Relationship between the Amount of Rainwater and Its o180 in St. Petersburg, Florida w U1


0 Oi 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 OoO :: 00 0 000 0 0 0o o 0 0 0 Oo o0o 0 8 o or,8 o o 0 0 0 00 0 00 "8o I'JJ 0 0 0 0 0 Ooo fl) 0 0 0 0 j;loo -5 0 0 0 0 0 0 0 00 0 0 00 0 0 0 0 00 -0 c.o 0 0 -10 l 0 10 15 20 25 30 35 Surface air temperature ( oC) Figure lib. A Scatter Diagram Showing the Relationship between the Surface Air Temperafure and o18o in Precipitation at St. Petersburg, Florida w 0\


Table 2. Regression Equations for Rainfall Its Amount Temperature and oD ' u o180 = -2.48(.20)0.0587(.0097)amount (in mm), y-2=0.252, n= 110 o18Q = 3.54(.81 8) 0.0597(.009)amount(mm) + 0.0445(.033)t(0C) r2 = 0. 251 n = 110 oD = 7 .649(.201)5180 + 10.569( 1.14) y-2=0.998, n=5 Atmospheric Water Vapor in Wes t-Central Florida 37 (14) (15) (16) Average o180 of atmospheric water vapor, which was collected outside USF's Marine Science Bui l ding, is -13.5 ( 1.9) o/oo (n=22, Table 3). This value (-13.5 o/oo) is lighter than the transpired water (o1 80=-11.8 [ 1.4] o/oo, n=5) from leaves in Boyd Hill's Environmental Res ea r c h Area, ERA. However, the o18o of atmospheric water vapor (-13.5 o / oo) is slightly heavier than the evaporative water vapor, obtained from pan experiments, (o180=-13.9 [ 1.1] o / oo; n=9). Assuming the water vapor in the atmosphere is a mixture of transpired water and evaporative water, contributions from each water vapor compartment can be derived. Calculation4 suggests that the transpired water accounts for about 20 % while the evaporative water vapor contributes about 80 % of the total average atmospheric water vapor for coastal west-central F l orida during the s ummer and fall seasons The predominant contribution of evaporative water vapor was expected since the sampling location is located near the water of Bayboro Harbor. Salt contents in seawater did not affect the oE of evaporative 4. By settin g up equation: + = 13.5 and letting Ocvap = -13.9 oloo and Otransp = -11.8 o/oo ---> x ""'0.80


Table 3. o18o in Average Atmospheric Water Vapor, Transpired Water from Leaves, and Evaporative Water from Pan Experiments 3. I Auno>phcric Water Vapor 11-Scp-91 -13.79 13-Scp-91 13.05 17-Sep-91 -14.54 IS-Sep-91 -12.59 I S-Sep-91 -16.51 23-Sep-91 -12.84 2-0ct-91 -13.74 22-0cl-91 -13. 52 29-0cl-91 13.28 310cl-91 -13.56 n mean std. deviation 5 Apr -93 13-May-93 18-May-93 25-May-93 15-Jun-93 16-Jun-93 17Jun-93 12-Jul-93 13-Jul-93 21-Jul-93 24-Aug-93 26-Aug-93 22 -13.55 0.96 -13.16 -11.95 -12.92 -13.04 -13.72 -13.01 -13.39 -12.99 -13.44 -14.57 -15. 2 -13.24 In a+= 1.137i(fn-o.4156(f-2.0667/IOOO (Majoube, 1971) T= 273.15 + 1 (oC) Boyd Hill %rei hum oC sal. wal vapor a+ o 18 0 (obs) atm press 59 9 28.36 23.13 1.()()91 -10.72 1021.8 60.54 28. 76 23.92 1 00907 -10.12 1018.5 61.51 26 98 21.91 1.00921 -12.31 1016. 1 71.88 25.31 23.2 1.00935 -13.43 1010 6 66 7 24. 09 20 02 1 00945 -12.64 1016. 1 18 Given 64. 1 %relative humidity and 26 7 degree C and relationship 0 0=-18 87+0.4818q lhe calculated o 180 is equallo -12.23 o/oo q (glkg) a c & eq.8 eq. 20 0.01408 0 99097 0 0090211 0 006416 -0. 02525 0 00622 0 .01461 0 .991 0.0089901 0.0063136 -0 .0251 0 00603 0.01341 0.99087 0 009129 0 0061584 -0.02535 0 .00591 0.01428 0 99073 0.0092617 0 0044992 -0 02429 0.00327 0.01225 0.99063 0.0093601 0.005328 -0 02539 0.00475 pans rei hum% oC 34. 3 40 6 50.6 35. 9 40.5 38. 4 48 36 38 6 3H 35. 8 38. 7 46 34 42.6 35.9 42.5 36.2 average hum average de& 42 37 a+ 1.00819 1 00853 1.00835 1 00852 1.00842 1.00833 1.00867 1 00853 1.00851 1'o 1.00345 18 o 0 (obs) -15.71 -14.21 -12.82 -12.86 -14. 92 -12 .86 -14.81 -13 .88 -13.07 -13.9 a 0.99187 0 .9915 0 .9917 0 99155 0 99165 0.99174 0.9914 0 99154 0 99156 0 99162 t 0 0081257 0 0084568 0 0082789 0 0084496 0.0083495 0 0082578 0 0085951 O.

39 water vapor unless the ionic strengths of water samples were higher than 2.0 (Craig and Gordon, 1965; Sofer and Gat, 1972). During transpiration, the root water was found to travel through xylem and leaves' stems without isotopic fractionation (Zimmermann et al., 1967). Since the soil water and leaf "':ater have been not studied here, root water is assumed not to fractionate the oxygen isotopes. The evaporation should take place at the leaves' stomata and depends primarily on the water temperature inside leaves (a+, fractionation factor) and relative humidity (h, and The relative humidity, air temperature, and atmospheric pressure can be used to calculate a specific amount of water vapor called specific humidity (Appendix 1). A linear correlation between the specific humidity and o18o of transpired water (Figure 12) can be used to compute the unknown o180 of transpired water if the relative humidity and air temperature are known (see the calculation of specific humidity in Appendix 1). Using the average relative humidity (64.1%) and temperature (26. 7 oq readings from the bag enclosure around leaves of oak trees in the ERA area, and assuming that the water temperature inside the leaves during the evaporation equals those of air, o180 of transpired water vapor of -12.23 o/oo is calculated from an equation (Figure 12; Table 3) : o18o = -18 .9 + 0.482q, This value is close to the average "in bag" measured o18o of transpired water of -11.8 o /oo (Table 3). White and Gedzelman (1984) found that, in air monitoring at Lamont Doherty Geological Observatory, the oD and q correlated linearly and the extrapolation of q value to 100 % humidity results in a oD of lo ca l ground water. At 100 % relati ve humidity and 34.5 oc temperature the calculated q equals to 0.03330 (33.30 g of water per kg of air) and the computed o18o of wa ter vapor is -2.8 o/oo. This value is similar to the ground water's o18o ( = -2.8 o/oo) in the ERA. The simi lar ity is caused by a stationary state of evaporation of surface water as explained by Craig and Gordon


-JO o a.. -11 ; 'a a.. -Q. {I) -12 a.. .... c)l8Q = -18.9( I. I) + 0.482(.145)specific humidity, r2=0.581 0 Q I / 0 0 43 / J 0 .... 110 -14 0 15 1 9 r . I I I I 6 8 10 12 14 16 18 Specific humidity, q (g/kg) Figure 12. Linear Correlation between the Specific Humidity and the o180 of Transpired Water in Boyd Hill's Research Area o


41 (1965), i.e. OsE=osL (steady state oE of evaporative water vapor is equal to steady state oL of body of water at 100 % humidity after a long, complete i so topic exchange). During transpiration the loss of stem water through leaves can cause the water vapor to deplete in oxygen isotopic ratio up to a maximum of 28.5 o/oo relative stem water (under purely diffusion-controlled pro cess). The tran spirat ion process combines the effects of fractionation factor (e ), kinetic enrichment (ek), and the relative humidity. A maximum 180 enrichment (omax) of leaf water, relative to source stem water, Oj, can be calculated from: Omax = e + Ek + (oA-ek)h (Dongmann et al., 1974) (17) oA = o18o of measured atmospheric water vapor oi = o18o of uptake water by plant Other symbols are as defined earlier. ek = (28.5)(2/3) = kinetic enrichmen t (Dongmann et al., 1974; Merlivat 1978; F l anagan and Ehleringer, 1991, p. 273; the 2/3 is an average contr ibution from diffusion and turbulence conditions under normal transpiration) measured oA=13.5 o/oo vs SMOW; h = 64 %; t( oC)=26.5; oi= -2.8 o/oo vs SMOW calculated omax = +8.46 o/oo relative to sou rce stem water = +5.66 oloo vs SMOW. This value is close to the range of calculated oL (average +6.05 oloo vs SMOW for the first three data points; equation 21, d e scribed later; Tab l e 3). In addition the calculated 0E ( = -25.0 o/oo) explains that the diffusion process (Dongmann eta!., 1974) is predominant during the evaporation of leaf wat e r in the Boyd Hill area.


42 West-Central Florida's Evaporation Isotopically, the evaporation of surface water is governed by a small laminar layer just above the air/water interface, and the Rayleigh distillation process (Craig and Gordon, 1965; Merlivat and Coantic, 1975) It obeys the following equation (Craig et al. 1963): (c/ -1)ln(f) where A= ln(1 +o) a*, fractionation factor of H201 --H20g f = fraction of remaining liquid (surface water) o = o180 or oD. (18) (19) The slope, a* -1, can be used to calculate the fractionation factor for water during evaporation under physical conditions appropriate for a geographical area Using the data from 2 pan experiments and o0 = -2.42 o/oo the slope (=[a "'-1], Table 4) of the pan 3 (6.392 L) and pan 1 (3.428 L) equal -10.27 and -7 .01 o/oo, respectively (r2 = 0.955 and 0 834, respectively). Since the pan 3 has dimensions closer to a standard Class A evaporative pan (which has been u sed to e s timate evaporation rate by National Weather Service), in terms of surface area to depth ratio the slope of 10 27 o/oo is preferred. This slope is equal to -e *, so the experimental a+ is 1 01037 which is slightly heavier than an average value of 1 00837 in the temperature range of the pan experiments (average air temperature in the exposed sunlight = 38 o c ; see Table 3) Although the actual water temperature could be a few degrees cooler, there is still a s mall difference between the experimental a+ and theoretical a+. Craig et al (1963) suggested that this difference was caused by the inclusion of the kinetic effect acting upon the water body. In addition, they (Craig and co workers) found that the evaporation rate had no effect on the experimental a+.


Table 4. Evaporative Pan Experiments, and the Inferred Fractionation Factor, a+ Class A, evaporation pan pan 3 Vo=6.39 L pan 1 Vo=3.43 L diameter I depth 4.8 7.67 15.1 _____ pan3 an1 L_ytvo ____ _Q_18o V/V o18o 1 1 -2.42 1 -2.42 0.929 -1.06 0.900 -0.09 0.906 -0.49 0.867 -0.36 0.777 0.804 0.734 1.623 0.750 0.94 0.634 1.816 0.594 4.08 0.618 2.88 0.429 6.14 0.435 6.05 0.358 7.29 0.102 11.69 By using the equation: 'A-J..o= (ex* -l)In(f), the best fitted curves give: intercept slope=(a-1)= -E ex a+=IIcx pan3 0 0.9551 -0.01027 0.98973 1.01038 pan I 0 0.8338 0.007015 0.99299 1 .00706 4 ,j:l. w


44 Following Welhan and Fritz (1977) and Allison et al (1979), the parameters m and o5 (equation 10) were calculated from a linear equation by varying the initial guess of o5 and obtaining the calculated m, and the correlation coefficient (r). The process was repeated until the best rand smallest residual sum square of errors between the predicted o5 and observed o were reached. A QBasic program written for solving them and r constants is given in Appendix 4. For pan 3, the best-fit o5 is + 12.22 o/oo while them is 1.041. However the calculated m (0. 6771) from humidity (m ::::(h-e)/(1h+ .1 e); a modification of equation 12] seems to be much lower for the h=0.418 and t=38.3 oc (the first three data); at the same time the calculated o5 of +29. 79 o / oo is much higher than the+ 12.22 o/oo regression o5 The lower m value (=0. 67) could be a result of bias of humidity measurements towards the daytime averages rather than the full daily cycle. But the larger o5 value ( = 29 8) may be cau s ed by utilizing of air temperature instead of water temperature during the calculation of E If the average water temperature of 37 oc in pan and relative humidity of 42 % (n=9; pan 2-a constant-feeding pan) are used for calculation of oE using equation 8 the computed oE is -14 05 o/oo. This oE is very close to the measured value of -13 9 o/oo In addition, by applying an equation: (20) the calculated oL (the final o of the liquid) equals + 10.58 o / oo which is close to the above regression + 12.22 o/oo o5 The difference here is attributed to the fact that the + 10.58 o/oo oL is calculated from a constant-feeding pan while the + 12. 22 o / oo oL is obtained from evaporative pan experiments.


45 Chapter Discussion Rainfall Rainwater, obtained during a 1991-1992 period (""' 1130 mm), was below normal rainfall (1 ,327 mm; 50-year average). The weighted o18o (= -4.4 o/oo) of the 1991-1992 rainfall is thus assumed to be a mere average. St. Petersburg rainwater has a stable isotopic composition (8180 and oD) similar to the meteoric water line, and the 8180 is within a -4 o/oo isopleth of o180 contour lines depicted globally by Yurtsever and Gat (1981). Strangely, the St. Petersburg winter and summer rainfall's o18o does not fall on the another o180 contour lines generated from the IAEA database (Lawrence and White, 1991). These differences may be due to: 1 There are no IAEA sampling stations in the vicinity of central Florida, and/or 2. The o18o of rainwater does not correlate at all with surface air temperature, but only is correlated with the its amount, and/or 3. Other meteorological conditions predominate (e.g. cloud formation source of atmospheric water vapor which varie s from year to year), and/ or 4 Lawrence and White (1991) utilized a biased set of data It is likely that variations in the source of water vapor may control the rainwater's isotopic compositions, especially during the wintertime in Florida. As shown by Newell and Zhu (1994), there are filamentary structures within atmospheric water vapor pathways which travel around the globe and shift directions seaso nally. During the summer, a local source of water vapor with invariant surface water temperature from the Gulf of Mexico and the adjacent Gulf Stream is likely to contribute more to the rainfall's o18o in Florida. Jacob and Sonntag (1991) conducted an 8-year monitoring of 8180 and oDin precipitation in Heidelberg, Germany, and found that the correlation between the 8180


46 and surface air temperature did not exist for those samples collected where the annual air temperature exceeded 15 C Similar findings were reported by Rozan ski et al. (1993) for 30-year IAEA datab ase and by the data presented in this dissertation It is possible that the re-equilibration of precipitation droplets with the water vapor and/or evaporation while still aloft may override the temperature-o18o correlation in the vicinity oft > 15 oc and 30 N (e.g. Woodcock and Friedman, 1963; Stewart, 1975 ; Jou ssaume and Jouzel 1993; Gedzelman and Arnold, 1994) Coincidentally, from the o180-amount linear relationship (Table 2) an intercept of -2.5 (.2) o/oo was found for a very small amount of rainfall. (If the limit of rainwater amount is approaching zero there would be no o18o data-a paradox of the relationship ) This value is close to river endmember's o18o (-2.2 [0.3]) and suggests that the similar water source and alteration processes (e g. evaporation) operate on the freshwater end member of Tampa Bay (see Chapter 4) and during light precipita tion. There may be questions regarding the validity of weighted mean of rainwater especially on the effect of the most depleted rainwate r in 0-18 composition (o180 = -11.7 o/oo in one sample in 1992) on the 1991-1992 weighted mean Yearly weighted means of rainwater are -4.1 and -4.7 o/oo for 1991 and 1992, r espective ly These yearly weighted means indicate that the majority of rainfa ll events control the 0180weighted mean Atmospheric Water Vapor The 20/80 ratio for the plants' transpiration and surface water evaporatio n could become larger if followed inlan d since the studied sites are located near the open water of the Tampa Bay 01Bo contribution of inland transpiration, if analyzed, could become the largest comparing to free-water evaporation in the total evaporation budget


47 as shown in o180 data of surficial aquifer water in Roy Haynes site (see Chapter 5 for discussion). Only the local oak trees were studied for the transpired water. It is possible that minor variations could occur in other plants. At least the o18o of transpired water extrapolated to 100 % humidity helps suggest that the source of water used by the oak trees comes from the ground. Evaporation It can be seen that the evaporation of surface water fractionates 1 8o t16o more than the equilibrium conditions suggest. The extra enrichment (or kinetics) is implicated within thee factor (as de), which depends on the humidity and water temperature Generally, the liquid isotopic resistance, p \. is v ery s mall compared to the (1-h) factor, and is often ignored during the calculation of oE (Craig and Gordon, 1965; Welhan and Fritz, 1977). Merlivat and Coantic ( 1975) and Merlivat (1978) found that the turbulent surface of the water (up to 7 m /s wind) did not affect the oE extensively. For pan experiments, if the air temperature is assumed for the water temp e rature, the calculated oE could approach its minimum limit while the calculated oL should be close to its maximum (Table 3). From observed data, the oE (-13.9 o/oo) agreed well with the calculated oE (-14.1 o/oo) but the calculated oL ( + 10.6 o/oo) is li g hter than the measured ones ( + 12.2). The difference for the latter is caused by uncertainty in the calculation of oE (evaporative pan experiments) from the parameters m and o5 which depend indirectly upon the curve-fitting method. The exact agreement (Table 5) between the two calculated oE in the closesys tem equation (Craig and Gordon, 1965; see the above equation 8) and Allison and Leaney (1982) 's constant feed pan experiment (the pan no. 2 experiment in thi s s tudy) s ugge sts


Table 5. Comparison of the Calculated oE from Craig and Gordon (1965; See Equation 8 Here) and Allison and Leaney (1982)'s Constant-Feeding Pans Allison and Leaney (1982) extended the equations 8 to 11 (in this dissertation) with their constant-feeding pan experiments to give oE lake or pan = (m + 1)(olake or pan -K) + ol (21) oE = 5180 of evaporative water m = m factor of Welhan and Fritz (1977) = (h-e)/(1-h+ [see equation 12 in Chapter 3] o1 = 5180 of input water = -2.42 o/oo vs SMOW K = o1/(m + 1) + m(hoa+e)/{(m + 1)(h -e)} Oa = -13.5 o/oo (measured) o11an = +3.74 o/oo (observed) E = 16(1-h)*10 J h oc a+ otso * eq. 8 0' E 34.3 40.6 1.008192 -15 .71 0.99187 0.008126 0.01051 -0 01543 50.6 35.9 1.008529 -14.21 0.99154 0.008457 0.007904 -0 01160 40.5 38.4 1.008348 -12.82 0.99172 0.008279 0 00952 -0.01426 48 36 1.008522 -12.86 0.99155 0.008450 0.00832 -0.01246 38.6 37.4 1.008420 -14 92 0.99165 0.008350 0 .009824 -0.01483 35. 8 38.7 1.008327 -12.86 0.99174 0.008258 0.01027 -0.01531 46 34 1 008670 -14 .81 0 99140 0.008595 0.00864 -0.01334 42.6 35. 9 1.008529 -13.88 0 99154 0.008457 0.009184 -0.01403 42.5 36.2 1.008507 -13.07 0.99156 0.008435 0.00920 0.01402 eq. 21 ft 20 -0.01543 0 1242 -0.01160 0 008334 -0.01426 0.01089 -0.01246 0.009031 -0.01483 0.01148 -0.01531 0.01214 -0.01334 0.009718 -0.01403 0.01050 -0.01402 0.01051 00


that the observed air temperature does not differ much from the pan's water temperature and the correct m should be close to 0.6771. Summary 49 A 1991-1992 monitoring of rainwater's 8180 in St. Petersburg, Florida provided a weighted mean 8180 of -4.4 o/oo. The correlation between the surface air temperature and the 8180 does not exist. The relationship between the amount of rainfall and its o18o is not very well correlated. D180 of transpired water (-11.8 o/oo) is heavier than D18o of atmospheric water vapor (-13.9 o / oo). Calculations based on a relationship between 8180 of transpired water and specific humidity suggest that plant uses ground water without major isotopic fractionation within the xylem The calculated DE of the leaves of oak tree suggest that a diffusion process predominates during the evaporation process. Both constant-feeding pan and free pan evaporation experiments provide a total enrichment factor (E) of 1.01027 (-10.27 o/oo in terms of fractionation for evaporation of water) during the summertime. The linearly-regressed evaporative parameters, m and D8 are 1.041 and + 12.2 o/oo, respectively. A good agreement was found between the measured DE (-13.9 o/oo) and computed DE (-14 1 o/oo) in the pan experiments.


50 Chapter 4 RIVER, AND ESTUARINE WATERS IN TAMPA BAY Tampa Bay is the largest body of estuarine water along the west-Florida shelf (Figure 3). The bay area is 1,031 km2 with an average depth of3.7 m. The drainage area (watershed) is 4,600 km2 (NOAA, 1993). The estimated freshwater flow rate (runoff into the Bay) varies from 50 to 72 m3/s (Goetz and Goodwin, 1980; Hutchinson, 1983; Zarbock, 1991; Table 6). The Hillsborough river is the major contributor of freshwater to the Bay ( z 29 %) and is followed by the Alafia ( = 21 %), Manatee ( = 16%), and Little Manatee ( z 11 %) rivers (Table 6). The Tampa Bay Bypass Canal supplies 2.6 % of the freshwater inpu t to the Bay. Salinities in the Tampa Bay range from 20 o/oo in the northern area (Hillsborough Bay) to 30 o/oo in the southern area near the mouth of the Bay during January and June months (Tampa Bay National Estuary Program, 1992; NOAA, 1993). In S eptembe r months, the salinity range increases by 5 to 7 o/oo. Water tempera t ure vari es from 12 oc in the winter to 33 o c in the summer (NOAA, 1993). Strong horizontal salinity gradients develop seasonally in the Bay water. The flushing rate is unknown, but is estimated by the author to be on the order of 53-67 days (see the calculation below). The evaporation rate ranges from 40" (1.016 m) per year to 56" (1.414 m) per year (Cherry et al., 1970; Hastenroth an d Lamb, 1978; Fransworth and Thompson, 1982; Bush and Johnston, 1988). Ba se d on 15-yea r data of pan evaporation study, Fransworth and Thompson (1982) gave a 50" (1.27 m) estimation of annual evaporation r ate in westcentral Florida.


Table 6. Estimated Freshwater Flow Rate into Tampa Bay (After Zarbock, 1991) Current conditions Authors cited by Zarbock Dooris Goodwin/Hutchinson Flannery Dames and Moore Historical conditions Dames and Moore flow rate (m3/s) 50.7 63.6 56.9 71.6 64.6 Surface water discharges to Tampa Bay (after Hutchinson, 1983) flow rate (m3/s) Hillsborough River 18.0 Sulphur Spring 1.18 Alafia River 13.0 Little Manatee River 6.79 Manatee River 9.99 Ungauged areas 9.73 Rocky Creek 1.31 Sweetwater Creek 0.613 Lake Tarpon Canal 0.832 Tampa Bypass Canal 1.62 Total 63.08 51 Surface sediments of Tampa Bay consist mostly of quartz sand with minor amounts of siliceous and calcareous detritus (Doyle, 1985). The underlying sections are


52 Suwannee and Ocala Limestones (::::::: 100m below mean s ea level; Ryder 1985) Hutchinson (1983) showed that the leak a ge of fresh ground water from the Floridan aquifer into Tampa Bay could be as large as 4.4 m 3 /s. More recently, Brooks et al. (1993) gave estimates ranging from 3 6-3 9 m 3 /s in the dry season to 3.9-4.3 m 3/s in the wet season. However, the extent and the direction of this underground water leakage or the reversal process (salt-water intrusion) depends on the ground water flow paths, potentiometric surfaces (heights of water head), water consumption by well pumping, and changes in the rates of recharge and discharge of ground water, as well as climatic modification. Freshwater drainage as well as ground water feed the rivers. In the Green Swamp area, the Floridan ground water contributes up to 11 % by volume of the Hillsborough river (Pride e t al., 1966). The underlying geology will be described in detail later in the Ground Water Section The geology and geochemistry of dra i nage basins and river beds control the chemistry and the o1 8o of the water and o13C of the total dissolved inorganic carbon (DIC) Sackett and Moore (1966) and Mook (1986) suggested that in riverwater the reaction between a limestone of marine origin (o13C = +2 o/oo) and C02 derived from organic matter (o13C = -27 o / oo) should produce a o13Cmc of -12.5 o/oo (={[+2][1]+[-27][1]}/{[1+1]}). This calcul a tion was based on an equal contribution of C02 from organic matter and lime s tone. A representative equation is CaC03{ s ) + C02( g) + H20(1) ---> 2HC03 -( a q ) + Ca 2+ (aq) (22). The o18o the river endmember depends upon the contributions from rainwater, evaporation and groundwater (e g Gat and Dansgaard, 1972). Previous USGS s tudi e s in the 197 0s show e d that in Tampa Bay the average total dissolved organic carbon w a s 4.5 x w-4 M, and the pH ranged from 7.7 in Hillsborough Bay to 8 5 in s outhern part of the Bay (Goetz and Goodwin 1980).


53 o180 and o13C (DIC) in Tampa Bay Waters Sackett et al. (1991) conducted a pilot isotopic study during March, 1990, for which the results are included here (in Figures 13a and 13b). Figures 13a and 13b illustrate the distribution of 8180 and o13C (DIC) in Tampa Bay estuarine waters. The results of two sampling periods (March 1990 and January, 1993) give linear correlations between the salinity and o180 as well as between salinity and o13C. For o13C, the slopes and intercepts of the two equations are similar within errors despite the spreading of the March, 1990 data. The samples in March, 1990 were collected nearshore, while the samples for January 1993 were taken from a boat. The extrapolated o180 for the river endmember, based on averaging of the equations for the two periods, is -2.3 o/oo vs SMOW while that of surface water (salinity = 32.02 o/oo during January) of the Gulf of Mexico is + 1.00 o/oo (both with 0.3 o/oo, n=38). The extrapolated river source of o13Cmc is -10 5 o/oo vs PDB, and the surface water's o13Cmc of the Gulf of Mexico is -2. 6 o/oo vs PDB (.9 o/oo; n=34). A plot of the concentration of the total dissolved inorganic carbon (EC02 ) against its o13C shows some scattering for the combined 1990 and 1993 data Only the 1993 data shows a trend displaying a moderate linearity (r2=0.214; Figure 14a) In addition, a weak correlation between the EC02 and its salinity is found (r2=0. 182; Figure 14b) Chapter Discussion 18 13 d 1 h The good correlations between the o 0 and o C (DIC) an sa 1mty suggest t at both o13Cmc and o18o behave conservatively along salinity gradient. The 8180 of water in Tampa Bay is a result of mixing between the freshwater and saline waters For 0180 the lower correlation coefficient for the March, 1990 sampling is attributed


00 -C;Q 2 1 o--11 I / / 0 March, 1990 (_ ___ __) D January, 1993 ( ___ _j 0 / / 0 "' / (Q / / / / 0 0 / ,"" / ,-/ \"'/ p/,; ./0 ""/ D ,/ / / / 0 o180 = -2. 19(.22) + 0 108(0 .010)sa linit y, r2=0.841 0 ,/0 / / 9 / // 0 0 / / o18Q = -2.43(. 24) + 0.099( 0 010)salinit y, r2=0.913 "" 0 5 10 15 20 25 30 35 40 Salinity (o/oo) Figure 13a. Linear Correlati o n between Salinity and o18o in Tampa Bay, Florida


0 o March, 1990 <-.--_J -2 0 January, 1993 < _ _ _) 0 0 -4 0 -6 0 -8 0 -10 oDe = -10.41(.79) + 0 .239(.039)salinit y, r2=0.660 -12 0 oDe= -10.64(.43) + 0 .254(.019)salinity, r2=0.945 0 5 10 15 20 25 30 35 40 Salinity (o/oo) Figure 13b. Linear Correlation between Salinity and o13C in T ampa Bay, Florida


56 -4 9 0 8 -6 0 January, 1993 -8 0 0 -10 0 ol3C = -9.73(. 65) + 4 9 6 (.86)l:C02 r2=0.215 0.4 0.6 0.8 1.0 1 2 1.4 :2: C02 (mM) Figure 14a Variation of Total D issolved C02 and Its o13C -in Tampa-Bay Estuarine Waters 0 1.4 +--___._ _._____.._.._____._ _._____.__..______.__.......____.__-+ 1.2 1.0 0.8 0.6 0 4 0 = 0.680( 0 .153) + 0 .010(.007)sa linity r2=0.182 D D 0 January, 1993 1 0 2 0 Salinity (o/oo) [lJ D [lJ D D D 3 0 Figure 14b Variation of Total Dissolved C02 along a Salinity Grad ie nt o f Tampa Bay


to storage which was not designed for 8180 analyses, and to the difference between shore and ship samplings The 813Cmc of shore samples in March, 1990 can be affected by atmospheric C02 exchange and stagnant waters 57 The river endmember's 8180 is a result of combination effects of natural evaporation, ground water seepage, and natural runoff The ground water seepage is likely to contribute mostly to the river endmember as the o18o of the upper Floridan water is around -2.8 o/oo (see Chapter 5). The 0.5 o!oo ( = [-2 .3 ] [-2 8]) positive shift of the river endmember represents a 4.7 % water loss from the ground water. (The calculation is performed using equations 18 and 19, described in Chapter 3, and parameters, 80 = -2.8 o/oo, b = -2.3 o / oo ) Most direct rainwater on the drainage basin is likely to evaporate into the atmosphere (evaporation rate :::::: 1.27 m/year ; precipitation rate :::::: 1.33 m / year); only a small fraction of water from the drainage basin contributes directly to the river heads If the EC02 of ground water is similar to that of seawater endmember, the correlation between the EC02 and salinity should not be found or be very weak (Figure 14b). The similarities in EC02 of the river endmember (:::::: 1 mM) and the ground water ( z 1 mM) support the idea that ground water feeds the river head The o13C (DIC) in the water is dependent on 1. The C02 sources depleted in C-13 which, generally, are predominantly from the oxidation of local organic matter (C-3 land plants), 2 The C02 sources enriched in C-13 which are atmospheric C02 and products of weathering of carbonate materials (see Mook, 1986; Mook and Tan, 1991 for further discussion). 3. The extent of isotopic exchange of carbon-13 between the atmospheric C02 and the water body (see e g Mook, 1986; Sackett et al., 1991) As shown earlier a simple mixing model for two sources produces 813Cmc of -12.5 o/oo vs PDB. The heavier b13C of river endmember can be caused by


58 1. Partial contribution of phytoplankton and terrestrial plants on the derived co2 2 A partial kinetic exchange with atmospheric C02 which tends to enrich the water body in carbon-13 through time. 3. As shown in Chapter 5, the ground water may be the only major source of river endmember in Tampa Bay. The o13Cmc of ground water is -10.4 o/oo vs PDB, which is almost exactly equal to the value of the river endmember (-10.5 o/oo). Carbonate species in the water, i.e. C020, CO?aq HCo3 -co3 2 have aq aq associated pair of fractionation factor, e.g. the HC03-(aq) --C02(aq) has a fractionation factor of 13c of -8.97 o/oo at 25 oc, the co2(aq) -co2(g) has a fractionation factor of 13c of + 1.06 o/oo at 25 oc (Mook et al., 1974). The o13C of the dissolved bicarbonate is not the same as the o13C of the total dissolved inorganic carbon since for the o13C1otal the 13C contribution of all carbonate species have to be accounted for A s pH of the e s tuarine water was not measured for the samples in either 1990 and 1993, the o13Cbicarbo n ate cannot be calculated directly (e.g. Mook and Koene, 1975; Mook, 1986). Further discussion on the exchange effect of the atmospheric C02 on the marine end member of Tampa Bay based on o13C is not possible. The slope of the linearly regres s ed line of the salinity and o13Cmc of Tampa Bay water is +0.247 (.044) which is similar to the Florida Bay water (slope= +0. 286 [.03]; Holmes, 1992). The freshwater o13Cmc intercepts are 10.5(.9) and -11.7(.8) o/oo for the waters in Tampa Bay and Florida Bay, respectively. The difference in the o13Cmc of river endmembers is attributed to variations in hydrology of the drainage areas and the differences in ground waters. The drainage areas in Tampa Bay consist of sandy s oils and mixtures of quartz s a nd, and carbonate fragments. The upper Floridan ground water seems to contribute mostly to Tampa Bay compared to the surficial and intermedi a te waters (Brooks eta!., 1993). In south


59 Florida, the Everglades and networks of canals (Klein et a!., 1975) keep the drainage water in longer contact with the organic-rich bogs, therefore the o13Cmc is slighter lighter. Surficial ground water is the important source of freshwater flow out onto Florida Bay (Klein and Hall, 1978). Using a simple one-dimensional box model proposed by Bowden ( 1967) and inclusions of the l eakage of the Floridan water into the Bay, the evaporation, and precipitation directly onto the Bay, the replacement time5 the surface outflow's volume, and the bottom inflow's volume can be calculated (Figure 15). The two equations, used in the model here, are for the conservation of volume and of water molecules (in terms of o1 80): VP + Yr + Yg + Vb = Y s + Vc V pop + V rOr + V gOg + V bob = V sOs + V ;:,De Upon manipulation : V s = (23a) (23b) [Vpop + Vror + Ygog-V/>c -ob(Vp + Vr + YgVJ]/(os -ob) (24) The replac ement time for the Tampa Bay is the volume of the Bay (3.483 km3 ; NOAA/EPA, 1989) divided by the rate of surface outflow at the Bay mouth. The chosen parameters are V e evaporation rate = 50" per year (41.5 m3/s) V P' rainfall = (2/3) of 50-year annual average = (2/ 3)(132.67 em / year) --28.93 m3/s (using Bay area= 1,031 km2)-for the 1991-1992 cond ition ; and during normal rainfall, V P = 43.4 m3/s V river runoff= 55 m3/s (low estimate) p V g ground water seepage into the Bay from the Floridan aquifer = 4.4 m3/ s ob, o18o of bottom inflow volume = +0.6 o/oo o o18o of rainw ater = -4.4 o / oo P 5. Time required to repla ce the entire vo lume of a stu died body of water In case is e_qual to the volume of T ampa Bay divided by flux of s urfa ce outflow wat e r fro m ampa ay mto Gulf of Mexico (V s)


60 EVAPORATION-Vet 8e Figure 15. A Box Model of Input s and Outputs of Waters from Tampa Ba y, Florida or> o18o of river = -2.2 o loo o5 o18o of marine surface water = +0.956 o loo 00 o18o of Floridan w ater = 2.8 o / oo (to be show n in details later in the next 0 Chapter) oe, o18o of evaporative water vapor = -13.9 o/oo (see Chapter 3) The unknown t e rm s are V 5 rate of surface outflow Vb, rate of bott om inflow The calculated flow volume of the V 5 Vb and the replacement time are shown in Table 7 below :


61 Table 7. Calculated V5 Vb, 11 V (=V5-Vb), and Replacement Time of Tampa Bay Water 2/3 rainfall normal rainfall V5 (m3/s) 762 7 606.2 vb (m3/s) 715.9 544.9 fl. V (m3/s) replacement time (days) 46.8 52.9 61.3 66.5 The largest controlling factor other than the surface outflow (V 5os) and bottom inflow of water at the mouth of the Bay is the evaporation effect (V eoJ. The ground-water leakage (directly into Tampa Bay, not from rivers) does not contribute significantly to the overall oxygen isotopic budget nor the lower-than-normal rainfall. From this information, the calculated replacement time of water in the Tampa Bay is about 67 days for average rainfall condition It should be noted that this is a minimum estimate Summary The o13Cmc and o18o in Tampa-Bay water correlate well with salin ity. The extrapolation of o13Cmc to zero salinity gives o13Cmc of -10 5 o / oo (.9) vs PDB, which is assumed to be equal to the freshwater endmember. In a similar fashion, the extrapolated o18o yields -2.3 oloo ( 0.3) vs SMOW. In Tampa Bay, the EC02 does not correlate with salinity nor o13Cmc The similarities of o13Cmc and o180 values of freshwater endmember and that of ground water (to be shown in Chapter 5) suggest that the source of river endmember is the upper Floridan water. Calculations based on a simple isotopic mass balance of all water inputs and outputs from the Tampa Bay show that the time required to replace all water in Tampa Bay is on the order of 67 days. Reduced rainfall does not affect the computed


replacement time much. Surprisingly, the direct ground water input into the Bay is negligibly small in the calculation. 62


63 Chapter 5 GROUND WATER AND MIRROR LAKE WATER Ground Water in West-Central Florida In the Southwest Florida Water Management District (SWFWMD), 83 % of the total water use for public sectors is obtained from ground water (Marella, 1992, Table 8). In Pinellas, Hillsborough Polk, Pasco, Hernando, Manatee and Sarasota the ground water accounts for up to 86 % of the total freshwater consumption Ninety-one percent of this ground water use comes from the upper Floridan aquifer (Marella, 1992) The upper Floridan aquifer is a part of a complex aquifer system, which acts as a water reservoir in the State of Florida In west-central Florida the main hydrological sections are surficial, intermediate, and the Floridan aquifers (Figures 5 and 16). Above the very thick igneous and metamorphic basements, gypsum (CaS04. 2H20), and anhydrite (CaS04), were laid down, and followed by a calcite-dolomite section which was overlain by a calcite section The Floridan aquifer is located in the calcite dolomite and calcite sections The lower Floridan water is saline, while the upper Floridan water is fresh and potable. In most locations a confining layer separates the upper Floridan and the lower Floridan sections. The overlying and confining strata are mixtures of dolomite, calcite, phosphatic and clay minerals which are called the Hawthorn Group, and act as confining beds, and intermediate aquifers. Closer to the surface, the remaining quartz -rich layers of sands and the mixtures of calcite, dolomite, aragonite, and shells constitute the surficial aquifer (Ryder, 1985; Scott, 1991).


Table 8. Estimated Freshwater Uses for the Year 1990 in West-Central Florida (Units in Million Gallons Per Day, After Marella, 1992) total withdrawals ground water surface water SWFWMD 1,572 1,302 (83%) 269 (17%) break-down categories of ground waters surface wafer ground waters Total Surficial Intermediate Upper Floridan r---------------------------Pinellas Hillsborough Polk Pasco Hernando Manatee Sarasota 48 1 0 2 0 179.7 0.35 2 2 353.7 7.2 10.7 138.7 0.21 0 45.5 0.02 0 96.5 0 6.7 59.5 1.2 29.2 fresh ground waters/ {fresh ground waters + surface fresh waters} Floridan water/{surficial + intermediate + Floridan waters} 47.9 177.2 335.9 138. 5 45.5 89. 8 29. 2 = 86% = 91% 1.4 88.9 83.8 2.4 0.35 45.3 3 2 0'1 ""'


Sy,tem S enes unu -----r------H o locene Tern c e Surfic1JI QuJtemary and depo s n s "''u1fer Pletscocene system Caloosahatchee Marl and Tam1am1 Pliocene Formauon I I I I I : Intermediate Peace >qu1fer Rl\.C f system F o rmJllon or tncenne d e3te confin1ng M1ocene c unu! i .E. i Arcadi a F onnauon Tampa Member SuwaMee L1mesrone I Ocal a l1mesrone : : F l ondan >qu1f er Eocene I s y stem' I Avon Parit F orma11on1 j I I Oldsmar and Ctdar Ke, s Pale ocene F orm allons 'Based on n()C'nenclacurr of S cou 11988) = eased on norncnct:uure o f Southe.utcm Gcologtc a l Soctcc,..ll986 t 'B:ncd on of M alter c l9Mbl B a>Cd on nomenclature o r Hutchnson t n prcsst. 'Bued on nomenclature or W ola.nsky al983 \. 65 -_ unn \ o nh o r:--E;,, o l South u r rT >mpa Ba) ___ T .. mp.s o r Surf t c tal SurfictJI c.urfic tll >qu1f er >qu 1 f er Jllulit:r t:pper Sem1confinrn$ confin1ng un n I I Tamiarm-I l U pper H a"th o m :>qu1fer' Absenror confining Sem1conf1 n 1n$ unu= >qu1 fer' untr' I l.o\o\Cf Ha"th o m -L"pper Tampa a q u tfer' Lo"er T>mpa conf1nmg semtconfintng un111 Suv.annce permeabl e zone I I Lower I Suw&MCC Ocala semiconf1n1ng U pper l.ipper unir' Flondan I Flo ndan aqui ferl :>qu1fer' Ocala -Avon Parlt I I mode.-.1cly i permeable wne Avon Park : highly permeable zone M 1ddl e confinang confinmg conf1n1ng unu' I uni t unu' L ower Flondan Lower Flondan Lower 3<1U1fer' :>qu1fer' Flondan aqu t f e r Figure 16. Hydrogeological Sections of Ground-Water Aquifers in West-Central Florida (After Miller, 1986; Swancar and Hutchinson, 1992)


66 The Floridan aquifer is located near the land surface in the northern Polk, Pasco and Hernando Counties, and dips down in the Pinellas and Hillsborough areas (Ryder, 1985). The surficial and intermediate aquifers are absent beneath most of the Tampa/St. Petersburg and Tampa Bay (Campbell, 1983; Ryder, 1985). The upper Floridan aquifer underlies the intermediate aquifers in other parts of west-central Florida and supplies freshwater (in addition to feeding the rivers) into Tampa Bay (Hutchinson, 1983; Brooks et al. 1993) The hydraulic head of ground water (potentiometric highs) and the hydraulic conductivity of the aquifer structures as well as their thickness govern the transmissivity6 of the upper Floridan water. The higher the transmissitivity, the higher the potential water withdrawal can be. In the recharge area the transmissitivity is lowest, while the potentiometric surface is the highest (e.g. Stringfield, 1936; Ryder, 1985; Bush and Johnston 1988). The transmissivities of the upper Floridan in the west-central Florida range from 20,000 to 250 000 ft2/day (0.0215 0.269 m2/s; Bush and Johnston, 1988) Rainwater recharges the aquifers through the surfic ia l aquifer sinkholes, and rock fractures which, subs equently feed into the upper Floridan aquifer. Since the Green Swamp is located at the highe s t elevation (:::: 42 m above the national geodetic vertical datum [NGVD]; highest hydraulic potential), it is considered the most important recharge area (Stringfield, 1936 ; Pride et al., 1966; Figure 17). However in a recent study based on tritium in ground water, Swancar and Hutchinson (1992) suggested that the perimeter of the Green Swamp was the major recharge area. This freshwater reacts with the host rocks which alter the water chemistry along the flow paths. The bicarbonate-calcium rich phase (ground water enriched in Ca2+ and HC03 -) is the dominant ground water type in the region, except near the west coasts where the 6. Tran s missivity is the rate at whi c h the ground water tlow s through a unit thickness of T _he flow is cause d by the hydr a ulic conductivity (head of water /distance o f flow). l m2/ s transmiSSIVIty I S equivalent to 1 m 3/s/ m or 6.96 x 106 gal/day/ft.


67 km 0 20 40 50 820W Figure 17. Schematic of Flow Paths of Ground Waters from the Recharge Area Close to the Green Swamp to the Coastal Discharge Area in the Tampa Bay sodium chloride-rich phase is more important (Back and Hanshaw 1970 ; Upchurch 1992). The upper Floridan water is in equilibrium or slightly supersaturated with solid calcium carbonate in south Hillsborough Counties (Jones et al., 1993) The average pH of the upper Floridan water is 7.5 while the average temperature is 25 o c (Upchurch 1992). Total dissolved inorganic solids increase along the flow paths in the confined aquifer from the recharge area (Green Swamp 175 mg / L) to the coa stal discharge ( = 600 mg/L) in west central Florida (Sprinkle 1989 ; Upchurch 1992 ; Jones et al. 1993; unpublished SWFWMD data) According to Jones et al. (1993), the calcul a ted ion fluxes a lon g stream tubes of the flow paths of ground waters in west-central Florida did not show a significant supersaturation with respe c t to the calcite and dolomite. This finding was contradi c tory to previous studies, especially to the Plummer group (Plummer, 1977 ; Plummer et al. 1983)


68 o180 and o13C (Inorganic) in Ground Water s and Carbonate Aquifers Behavior Along the Flow Paths The upper Floridan aquifer flow paths a re arbitrarily divided into northern and southern directions according to its confined Hawthorn boundary (Scott, 1991; F igure 17). The northern flow path is located in the area where the Hawthorn Group was not well developed or absent. In contra s t, the southern flow path is bounded by the Hawthorn confining unit. The northern flow paths are supposed to follow the potentiometric highs to lows (Figure 17), be g inning at the Green Swamp (location 1) and flowing through the 88 Rockridge (4), the 87 (5 Forestry Tower), the 68-2 (8 north Tampa), and coastal Riverview (9TRI0-2) wells The southern flow paths begin in the Green Swamp (1) and flow through the Lake Alfred (3), Lakeland (6) Valrico (7-DV-2), Apollo Beach (10TR9-2) and coastal Ruskin areas (11, TR9-l). The TR91 site is an artesian well. The Loughman (2) is not considered for ground water chemistry because of the unusual he a vier o180 and o1 3Corc in its ground water. The Loughman's Flori dan well is likely to receive w a ter from the surface intermediate and Floridan aquifers as a result of collapse of well ca s ing even though the well was drilled to Avon Park Limestone depths (Table 1 Chapter 2). However, its carbonate aquifer samples, obtained at the time of well drilling, are thought to be unaffected. For o18o (vs SMOW) in the upper Floridan water the major recharge areas are located in the Green Swamp (-4.5 o / oo), Lake Alfred (-4.4 o / oo) and Lakeland (-4 2 o / oo) areas (Figures 18a and 18b). This as s umes that the recharge water has a sim i lar o18o to today's rainwater (weighted mean = -4.4 o / oo) A l ong the flow paths the upper Floridan waters have o18o ranging from 2.43 to-3.36 o/oo with an average of -2.83 o/oo (by ignoring th e Green Swamp, Lake Alfred, and Lakeland sites; Figure


0 -0.5 -1 -1.5 -2 0-2.5 C'-l -3 0 -3.5 1.0 -4 -4.5 -5 -==:;r--Northern---___ Northern B Mar-92 !Til Aug-92 Green Swamp Loughman Lake Allred 88 Rockridge 87,deep Lakeland Utilities DV-2 68-2 TR 10-2 TR9-2 TR 9-1 (Valrico) (Riverview) (ApoUo (free flow) Beach) Figure 18a. c518o Variation during the Semi-Annual Studies of Ground Waters from the Recharge to Discharge Areas 0'1 IJ)


X >< >< >< -2.8 TRl0-2 68-2 87 88 otso Green Swamp Northern Not-to-scale diagram -4.4 TR9-1 TR9-2 -2.8 Lakeland 8ts0 Green Swamp Southern -4.4 Not-to-scale Figure 18b. Distribution of o180 Along the Northern and Southern Flow Paths from the Green Swamp to Coastal West-Central Florida -..J 0


18a). This average o180 is similar to the regional -2.82 o/oo (Swancar and Hutchinson, 1992; for the Pinellas, Pasco, Hillsborough, and Polk counties). 71 The o180 of upper Floridan water, if followed along northern and southern flow paths, has 2 endmembers: the recharge area (""" -4.4 o/oo), and characteristic upper Floridan signature (-2.8 o/oo). The unconfined nature of the northern flow path could allow (today's) rainwater o180 (-4.4 o/oo) to modify the o18o of ground water along the path. The rather invariant o180 in the northern flow path (Figure 18b; other than the Green Swamp) indicates that there are extensive water-rock reactions through time or the source water feeding into these aquifers was different from today's rainwater. For the southern flow path (Figure 18b), the confining nature of aquifers define and control the evolution of o180 in the upper Floridan water. The recharge and discharge areas must constrain two major o18o endmembers in this southern flow path The expected linear trend of mixing of these two end members was found only from Lake Alfred to TR 9-2 (Figures 18a and 18b). It is possible that the TR 9-1 and TR 9-2 wells are located within the same ground-water circulation cell. o13Cmc (vs PDB) of the upper Floridan water is dependent on the nature of water rock interaction and changes in source of organic-derived C02 In general, there are two extreme o13Corc values in the two flow paths with values of""" -10.5 and -15 o/oo (Figures 19a and 19b). In the northern flow path, where the confining layer is absent, there is a decreasing trend of o13Cmc along the travel path from the Green Swamp to TR 10-2 (Figure 19b). This trend may be attributable to immediate input of organic derived C02 from shallower aquifers. On the contrary, the confining aquifers of the southern flow path has the o13Corc close to isotopic equilibrium value that there is no trend in o13Corc along the path. Figure 19b illustrates the deviation of o13Corc at Lake Alfred, Lakeland, and DV-2 sites which is probably caused by excessive water pumping in DV -2 and Lakeland sites and a large extent of oxidation of dissolved organic carbon in Lake Alfred site.


0 -2 -4 = Q -6 g.. u -8 Northernt:=:j \ J !,() -10 Southern -12 +--S Mar-92 Northern -----NorthernSouthern -14 -16 -----B % II Northern _________ ------------1m Feb-93 I ____________________ _ Sol!!h ern ____________________________ Green Swamp Figure 19a. Loughman Lake Alrrect 88 Rockridge 87,deep Southern Lakeland Utilities DV-2 68-2 TRl0-2 TR9-2 TR 9-1 (Valrico) (Riverview) (Apollo (rree now) Beach) o13C Variation during the Semi-Annual Studies of Ground Waters from the Recharge to Discharge Areas -....) 1\J


Northern 88 87 68-2 Not-to-scale diagram TR9-l TR9-2 Southern Not-to-scale diamro Green Swamp -10 oBc -13 -10 0. I 8Bc c:: Q) e 0 -15 Figure 19b Distribution of o13C Along the Northern and Southern Flow Paths from the Green Swamp to Coastal West-Central Florida -..J w


74 Figure 20 illustrates th e variation of o13C and o18o of carbonate aquifer samples collected according to their casing depths (Table 1 and Figure 5, Chapter 2). The exception is Lake Alfred 2 site which the carbona te sample was taken from shallower depths. The carbonates close to the recharge area have o18o ranging from -1.46 to -1.17 o/oo vs PDB and o13C ranging from +1.91 to +0. 70 o/oo vs PDB. The upper values of these carbonates are close to the unaltered (marine) carbonate platform (e.g. Gross, 1964) The carbonates in the discharge area (Ruskin TR9 1 ; Hawthorn group) have o180 and o13C equal to -2 09 and -5.97 o/oo vs PDB, re spective ly. The Roy Haynes (A; unknown carbonate depths), Valrico (7), and Apollo Beach (10) wells are not considered by the author to be representative aquifers in the discharge area since they fall on differ ent endmembers of the o13Cmc vs o180 plot, and could be a result of alteration by meteoric water. The alteration process causes a negative shift of the o 180 of carbonates. Mineralogically, the carbonates in the selected sectio ns of aqu ifers are pure calcite, with only minor trac es of quartz and dolomite. The exception is the Wimauma site, which contains mostly dolomite (Appendix 2). A linear correlation is found between the o13C in the carbonate and percentage of carbonate (r 2 =0. 795; Figure 21). The percent carbonate was calculated from manometer-calibrated C02 obtained from the reaction between H3P04 and carbonate rock sample of known weight. The extrapolated o13C of carbonate samp l e to 100 % equivalent carbonate is + 1.8 o/oo vs PDB. Assuming, for the confined aquifers, the southern flow path of ground water star ts in the Green Swamp and flows to the coasta l Ruskin area, and applying a mass isotopic balance suggested by Taylor (1974):


= 0 fll .... 0 QO c.o -1 Loughman 0 -----l> Lake Alfred 2 -2 0 9 1 -3 -4 Roy Haynes 0 DV-2 9 2 0 0 -8 -6 13 0 88 -4 -2 C vs PDB Al< I O 0 t /' 87 Green Swamp 0 68-2 0 2 Figure 20. A Scatter Diagram of o13C and o180 for Aquifer Carbonates from West-Central Florida -.J U1


0 -2 Lake Alfred 2 0 -4 Green Swamp 0 68-2 o13c = -9.45( 1.93 ) + 0.113(.023)percentage carbonat e, r2=0.795 9-1 Louglunan 30 40 so 60 70 80 90 100 Percentage of carbonates Figure 21. The Linear Relationship between o13C of Carbonates and Their Percentages from West-Central Florida


wOi-water + ro,-rock = wOf + rs: -water Uf-rock (25a) where w =mass of water; r =mass of rock oi-water =initial o180 of water =-4.4 6o/oo vs SMOW oi-rock =initial o180 of carbonate rock = -1.43 o/oo vs PDB (= +29.39 o l oo vs SMOW) Of-water =final o1 80 of water Of-rock =final o 180 of carbonate rock Note: o180sMOW = 1.0308M180pos + 30 86 =-2 .70 o /oo vs SMOW =-2.09 o /oo vs PDB (= +28.71 o /oo v s SMOW) for the water equilibration and phosphoric-carbonat e reactions performed at 25 oc 77 the calculated water/rock ratio i s 0.375. If the pis the porosity of carbonate aquifer in volume/volume unit, the w/r = p(density of wat er)/[(1-p)density of rock] (25b). Assuming the densities of water and carbonate rocks are 1.0 and 2. 7 g /cm3 respectively, the calculated porosity of the rock is 0.503 The computed porosity is similar to the initial value suggested by Budd et al. ( 1993). It should be cautioned that thi s effective porosity is an average over a geolog ical time (to reach an isotopic equilibrium), and other geological and climatic condition s have not greatly altered the primary carbonates to secondary carbo n ates, or not modified the recharge water from tod ay' s rainwater. Figure 22 shows the intersection of two linearly-regressed lines from the rainwater (present study) and the ground water (Swancar and Hutchinson, 1992) with the crossing points of o18o of -4.23 ( 0.73) o /oo vs SMOW and oD of -21.8 ( 5.7) o /oo vs SMOW. Below the intersection coordinate, there is no ground-water data point. The intersection point thus, represents the possible minimum set of o180 ( =: -4.2 o /oo) a n d oD ( =: -22 o/oo) This confirms that the local ground water in west-central


0 ::t 00 Vl .... Q Ground Water: oD = 1.04(.46) + 5.39(. 16)ol8o, r2=0.935 Symbol + for ground water 0 -20 c.o -40 -60 \ Rainwater: oD = 10.57( 1.14) + 7.65(. 20)oi8o r2=0.998 Symbol "o" for rainwater 18 cS 0 vs SMOW -12 -10 -8 -6 -4 -2 0 2 Figure 22. o180 and oD of Rainwater (This Study) and Local Ground Water (from Swancar and Hutchinson 1992). The Intercept between the Two Best-Fitted Lines is the Original Source of Ground Water in West-Central Florida. -.J (X)


79 Florida is originated from the rainwater (o18o = -4.4 o/oo). If the mo s t depleted set of data (o180 = 11.6 o/oo, oD = -7 7 o/oo) was removed, the intersection of two lines would be at o180 of -4.8 o/oo and oD of -24.7 o/oo A slight change in the intersection does not preclude the hypothesis that ra i nwater recharges ground water. A Time-Series Study of o180 and o13C in Ground Water s at the Roy Haynes and B a um Wimauma Sites This time-series study was designed to test a nd monitor the temporal changes of o18o and o13C variations and their consistencies in ground waters, both at the unconfined and confined aquifer sites. The Roy Haynes (location A in Figure 17) is considered an unconfined aquifer well; while the Baum-Wimauma is located in the confined aquifer (location B in Figure 17) Figure 23 illustrates the consistent of o18o of the upper Floridan water in Roy Haynes (symbol 0 18RF1) with an average of -2 74 o/oo throu g h sampling intervals (Table 9). This Roy Haynes average is close to the average regional o180 (-2.8 o/oo) of the upper Floridan waters (Swancar and Hutchinson, 1992). The consistent stable isotopic chemistry of the upper Floridan (Roy Haynes) and the intermediate waters (Baum-Wimauma) is reflected also in o13C (inorganic, symbols C13RFl, Cl3Wint; Figure 24). However, it takes a long time for total dissolved C02 to stabilize even after monitoring for many months (may be up to 5 month s or longer based on Figure 25). The regular monitoring of ground waters is needed to purge enough water to achieve a representation of the chemistry of oxygen18 and carbon-13. The i ntermediate water's o180 in Baum-Wimauma site (symbol 0 18Wlnt) is between those of the upper Floridan and surficial waters of Roy Haynes (Figure 23) The total dissolved C02 in the Roy Haynes's surficial water correlates linearly with the o13C (inorganic; r 2=0.70 1 ; Figure 26).


0 -2 -2.5 tl) -3 fl) ;o.. c.o -3.5 -4 -4.5 --+--------1 t---,1-----D--018 R sur X -018 R Fl -::+::-018 W sur -fr--018 .,. -_ ----.J -----\.. --... / ----.I ---,_ ----'-----_ ,_-----_,_-------. ' --------,---------,----R=Roy Haynes __ .... -:-. __ .. _. ____ ___ .:. W=Baum Wimauma ' Surf=Surficial water lnt=lntermediate water Fl =Upper Floridan water I I I I I -----; --------, --------,--------------.--' ' ' ' ' ' ' ' ' ' , X ' "" I --ll: -----------.'.------.. ---' ' -----:----------:-'1.-----: lf( ' .J 1.. I .1. t ' 4 ' J I IS Feb92 25-Mar92 29-Apr92 10Jun92 5-Aug92 70ct92 9-Dec92 4-Feb-93 8-Apr93 2-Jun-93 3 Aug-93 60ct-93 Figure 23. Temporal Variation of o18o in Ground Waters at Roy Haynes (R) and Baum-Wimauma (W) Sites (X) 0


-9 -11 -13 -15 -21 -23 -25 ' o I I I t I I ,. -. . ,.- T . 1 .- I j r I ' )K ---)K ' I I I I I I 0 ....... X .. ... ... .... . ... .... ... . . . ............ .......... ... ........................ : .....______ : x x x x......___ : : : : ---------X , X ......___ , , . . . x . : : : : : : : x-----x : . ' ; ' -----------: ------_ : -----: --. --_ : --. ; -.. --. --. :-.. -. -. -----. ---->:-I I f I I I ' ----------------------------------------------------------------------.--.--------.. -.. -t o I ' t ' : R = Roy Haynes : W=Baum Wimauma , , , ..... -: ... -... Surf= Surfi c ial water ... : , ' : : Int =Intermediate water : : : ; ; Fl = Upper Floridan water : : : : :--------: : ........ _-xc.13RFI I . ------j --l ---'"' ----J ------1 ---.. ------.. -IS Feb-92 25Mar-92 29Apr-92 10Jun-92 5Aug-92 7-0ct92 9-Dec92 4Feb-93 8Apr-93 2-Jun-93 3Aug93 6-0ct93 Figure 24 Temporal Variations of o13C in Ground Waters at Roy Haynes (R) and Baum Wimauma (W) Sites


3.5 3 2 .5 = 2 .... 0 I.) 0 Cll Cll :a 1 -; 0 E-4 0.5 0 ----------------. --------------------------------------------------------------------.-----o I I I I I I I I R=Roy Haynes W=Baum Wimauma ........ .... -.... -...... . --.... "')( -...... . .... . ...... . -.... ........ ; ; ; ; ; ; Surf=Surficia_ l water . . . . lnt=lnt ermeda a te wat e r : ; ; ; ; ; ; ; Fl =Upper Fl o ridan wat e r I I I 0 o o I o I I I -o I t o -----:----------:-------; -----:--------:---------; --------:--------:-----------. -. 1--oC02Rsur x C02RFI I ' ---------------.... --------.------.... ---------------... -----------------------... ------. ---C02Wsu C02Winl] : ' ' I t I 0 t I o ----, ----.---. -i ---. -.----. -- ---i -------t --' ....... . i .... . : .... -.. : .... :-:-: -....... . . . . : .... . . : . . . . : ; ; X ; ; I I I t I I 0 I I I I I I + t I I : : : : : : : : ; : I I 0 0 t o I I + I < o I I I 1 t I o t o ' ' ' ' ;(. ' ' ' 0--' ' : : : : : : : I : : I I I o o t I I I o ' f---1 1825-29105-7-9-4-8-2 -36 FebMar-Apr-Jun-Aug-Oct -Dec-Feb-Apr-Jun-Aug Oct -92 92 92 92 92 92 92 93 93 93 93 93 Figure 25 ECO? Changes during the Sampling Intervals for th e Roy Haynes and B a um Wimauma sites


o13C = -26.4(.23) + 14.3(.8)l:C02. r2= 0 701 Roy Haynes's Surficial Aquifer u.2 0.4 0.6 0.8 1.0 1.2 Total C02 (mM) Figure 26. Linear Correlation between the EC02 and the o13c01c in Roy Haynes's Surficial Water


Table 9. Averages of o180 and o13Cmc of Ground Waters in the Roy Haynes and Baum-Wimauma sites o180 of ground waters (vs SMOW, 1 standard deviation) Surficial Intermediate Roy Haynes -4 09 ( 0.24) N/A Wimauma -3 .95 ( 0.21) -3.63 ( 0.27) o13Cmc of ground waters (vs PDB, 1 standard deviation) Roy Haynes -17.5 N/ A Wimauma -13 .0 ( 1.59) -12.2 ( 0.38) See Figure 24 for a decrea s ing trend the -17.5 o l oo oDe i s an average Floridan -2.74 ( 0.23) N/A -15.7 ( 0 50) N/A 84 The average o18o of the surficial aquifer water in the Roy Haynes and Baum Wimauma (Table 9) is -4.02 o/oo, which is almost 0.4 o/oo heavier than the average regional o180 of rainwater (Chapter 3). By using equations 18 and 19 described earlier in Chapter 3, o0 = -4.4 o/oo, Ome a s ur e d = -4.02 o/oo, and the (c/-1) factor of -10 27 o/oo, the fraction of liquid (of the infiltration) remaining in the reservoir is 96 35 % (volume/volume) The fraction of liquid which evaporates into the atmosphere is thus equal to 3.64% [=(10.9635)*100]. This 4% loss of water is very small and, as a first approximation, the infiltration w a ter (portion from rainwater) recharges almost 100 % into the surficial aquifer. Evaporation affects the oD as well, but the extent is, unfortunately, smaller than the o18o within analytical error. Assuming the equation: oD = 5 3935180 + 1 .037 (Figure 22) applies to the local ground water, the corresponding oD (for o18o = -4.02 o/oo) is -20.6 ( 5.4) o / oo. This oD deviates by only 0.46 o/oo from the computed oD from the local meteoric water line (oD = 7 6498180 + 10.569). This means that


the oD in water is not a good indicator for the small evaporation (normal analytical errors for the o180 and oD are 0.20 and 1.5 o/oo, respectively). Major Ion Chemistry in Ground Waters of West-Central Florida 85 Back and Hanshaw (1970) illustrated the usefulness of major-ion chemistry in delineating the chemical evolution and types of ground waters in central Florida. More recently, Sprinkle (1989), Upchurch (1992), and Katz (1992) summarize the trend and evolution of chemistry of the upper Floridan waters in west-central Florida. Generally, the major cations of ground water are calcium, magnesium, sodium and potassium The predominant anions are bicarbonate chloride and sulfate (Figures 27a and 27b). The concentrations of dissolved salts are in milli-molar levels By following the northern flow path (defined earlier in the study) of the upper Floridan water, there is no clear trend of dissolved salts. In contrast, the southern flow path shows an increasing trend of many dissolved constituents with the exception of K + in the Valrico (7-DV-2) and Ca2+ in the Green Swamp (1). This could be a result of the confining aquifer along which ground water travels from the recharge to discharge areas as illustrated in o180 data (Figure 18b). A summary of concentrations (in mM) of major constituents and their ratios is tabulated in Table 10 (see also Figures 27a and 27b). For the northern flow path, one can find : 1. ca2+ molar concentration is enriched in ground water by a factor of 4 over the Mg2+ (Figure 27a). 2 Total ca2+ molar concentration can account for the forming of (if any) calcite (CaC03), dolomite (Ca{Mg}{C03h), and anhydrite (CaS04). The trend is shown also in the southern flow path except for Lakeland, TR9-l, and TR9-2.


2 _1.5 8 .._ c 0 --f 1 c Q,> f;j c 0 u 0.5 1 2.49 1 3.87 1 5.25 -----.---------------. CJ Green Swamp 88 Rock Ridge m 68-2 TR 10-2 -----------------1 0 . Ca2+ Mg2+ Na+ K+ Cl-S04 Figure 27a Major Ion Concentrations in Ground Waters in the Northern Flow Path of West-Central Florida ():) m


2 E -c 0 --s... 1 -+-1. :"'' '''I-c QJ u c 0 u 0.5 ":3 49 1 2.22 2.19 ----------------------------------------------------------------------E.J Green Swamp Lakeland U1 TR 9-2 TR 9-1 -----------------------------------------------------------------------1 o Ca2+ Mg2+ Na+ K+ Cl-S04 Figure 27b. Major Ion Concentrations in Ground Waters in the Southern Flow Path of West-Central Florida ()) -...)

PAGE 100

3 The molar concentration ratios of Ca!Mg and Ca/S04 are higher in the northern flow path than in the southern path 88 As mentioned previously the o13CDrc (Figure 19b) of the ground water in the northern flow path decreases in a fashion which parallels the increase in Ca2+ concentration. There may be extra source s of isotopically light C02 which reacts with carbonate rocks, and subsequently releases more Ca2+ and lower the o13Cmc in the travel path. In southern flow path, the total Ca2+ molar concentration is slightly higher than the Mg2+ (Table 10, Figure 27b). The lower Ca!Mg ratio and almost invariant trend in o13Cmc (Figure 19b) al s o help s uggest that the upper Floridan water is more or less in equilibrium with aquifer carbonate s. Surprisingly, the S04 2 concentration in rainwater may account for S04 2 concentration in the upper Floridan water, with the exception of 68 -2, DV 2 TR10-2, TR9 -1, and TR9-2 (Table 10). Mirror Lake Mirror Lake i s located near downtown of St. Petersburg, and is a land-locked body of water used currently to retain storm water on a temporary basis Its area is 12 acres (4.856 x 104m2), while the volume is 55 x 106 gallons ( = 2.082 x 105 m3). The lake has a circular shape, and was 18 feet deep (5.6 m) during the excavation in the 1920s (1925 blue prints from the Engineering Department of the City of St. Petersburg; Dickinson et al., 1969) For over 10 years early in the history of Lake, the lake water was withdrawn for public use In the center of the lake, a water fountain was installed to help add oxygen to the water. The lake water level remained constant throughout the weeki y water samplings for o 180 analyses during 1991. The lake stores the storm water and re-directs it over a weir to another outflow system into the Tampa Bay. The average o18o of Mirror Lake water was 0 .85 o/oo ( 0.4; n=51) during the year 1991 (Figure 28). The temporal variations of o 180 in lake water were small

PAGE 101

Table 10. A Summary of Concentrations (mM) of Major Constituents in Upper Floridan Water along Northern and Southern A ow Paths, and of Major Ions in Rainwater of Central Florida (After SWFWMD [unpublished data], and Hendry and Brezonik [1980] and Junge and Werby [1958]) Y2 M&2 NA K SQ.4 OllM& NAlK Q.lSQi HAlQ. E12w Patb Green Swamp 1 .10E+OO 5 35E-02 1.26E-01 3.84E-02 1 27E-01 3 12E-03 21 3.3 41 0 .99 352 88 Rock Ridge 2.00E+OO 3 00E-01 7.39E-01 2 81E-02 5 92E-01 3.12E-03 6.7 26 190 1.2 639 Romp 87 1.81E+OO 2 47E-01 S .OOE-01 3 45E-02 2.61E-01 2 08E-03 7 3 15 125 1.9 869 68-2 1.60E-01 4.11E-03 6 96E-01 7.67E-02 4.51E-01 2.50E-02 39 9.1 18 1.5 6.4 TR 10-2 2.50E+OO 5 76E-01 3 .87E+OO 6.91E-02 5 .25E+OO 3.54E-01 4 3 56 15 0.74 7 S2111bm1 El2w Patb Green Swamp I.IOE+OO 5.35E-02 1.26E-01 3.84E-02 1.27E-01 3.12E 03 21 3.3 41 0 .99 352 Lakeland 2 50E -01 5 76E-01 2.87E-01 1.56E -02 2.60E-01 3.12E 03 0 43 18 83 1.1 80 DV2 9 48E-01 6 17E-01 4 35E-01 3 07E-OI 1 95E-01 3 64E-02 1.5 1.4 5 3 2.2 26 TR 9-2 1.92E+OO 1.28E+OO 6.09E-OI 4 35E-02 6 49E-01 1 .67E+OO 1.5 14 0.39 0.94 1.2 TR9-I 3 .49E+OO 2.22E+OO I.OIE+OO 6.39E-02 1 .27E+OO 2 .19E+OO 1.6 16 0 58 0.79 1.6 Riliowills;x Hendry and Bremnik (1980) wet 1.02E-02 4 94E-03 1.91 E-02 5 12E-03 2.76E-02 2 13E-02 2.1 3.7 1.3 0 .69 0 48 bulk 2.02E-02 7.41E-03 3 57E-02 6.65E-03 5.30E -02 2 36E-02 2.7 5.4 2.2 0.67 0 86 Junge and Werby (1958) wet 1.35E-02 N/A 2.74E-02 3.58E-03 2 26E-02 I.SSE 02 N/A 7 .7 1.5 1.2 0 87 00 \0

PAGE 102

0 00 f.l) 0 00 c.o 0.0 ___..__-t-0.5--1.0 -. A -1.5 A . .. .. .. 4 :: .. . .. : .:: '. 'i: A ,. ... ... ... . . . . .. .. .. .. .. .. . A . .. .. A A .. .. .. .. . . . . . . . . . . . .. .. .. 1 . . ,' \ . . "'' '-.. . . . 6 :, . . . . . . . .. .. .. A . f4.. . i . i .. 6 .. .. .. A + .. .. .. .. .. .. .. . .. . . . A . .. .. A . Jt. t ' . ' ' .. . .. :. . . . . . . . . .. . . .. . .. .. .. .. .. .. . .. A A . . . . . . .. . . . i ,. .. .. . . . ,. . . . . . . . . . . .. .. .. .. .. A ,A. A A JaD Mar May Jul Sep Nov Figure 28 o180 Variation in Mirror-Lake Water during the Year 1991

PAGE 103

91 (Figure 28) The calculated oE from equation 11 (in Chapter 3), using m = 1. 041, Os= + 12.22 o / oo and o1801ake = -0. 85 o / oo, is -14.46 o /oo. This computed oE is close to -13.9 o / oo obtained from the constant-feeding pan experiment. Similarity in isotopic compositions between o18o of calculated evaporative water from the Lake and from pan experiments suggests that the controlling parameter s should be almost identical for the pan experiments and Mirror Lake. An average replacement time varying from 36 to 43 days may be calculated assuming that the flow volume of s torm water into the lake (V st[incoming storm water]) equals the outflow of the storm water from the lake (V st[over the weir)) in order to remain level and the input water s spends the same time in the Lake as the output waters (Figure 29), two following equations are set up: Y st(over the weir) + Y E = Y p + Y g + Y st(incoming s torm water) (26) and Ys1=storm-water flow volume=unknown VE=evaporative flow volume=50"/ year on 4.856xl04 m2 = 1.956 X w -3 m3/s V p=precipitation flow volume= 1.33 m/year on 4.856xl04 m2 = 2.048 X 10-3 m3/s V g=ground-water flow volume= unknown ost=storm-water's o18o::::: -4.4 o / oo (assuming same as rainwater) oE=evaporation's o180= -13.9 o/oo (measured, Chapter 3) o =rainwater's o180= -4.4 o / oo (measured, Chapter 3) p o =ground water's o180= -2.8 o/oo (measured, and as s umed to be from g Floridan aquifer) oL =Mirror Lake's o180= -0.85 o/oo (average from weekly sampling for one year)

PAGE 104

EVAPORATIONVeo oe STORM WATEitV st. 88t PETERSBURG Figure 29. A Box Diagram for Various Inputs and Outputs ofWaters in Mirror Lake St. Petersburg, Florida TL =volume of Mirror Lake=(2/ 3)(2.08x i05)m3 = 1.39x105 m3 (2 /3 of the 1920's volume) The calculated Yst and Yg are 5 .19 X w-3 and -9.20 X w-5 m3!s, respectively The direction of V g is opposite the one shown in the diagram above (the lake water recharges into ground-water aquifer) The replacement time of water in Mirror Lake can be computed by dividing the oLT L by ( og V g + O s t V st + op V p) which is equal to 43 days At higher evaporation rate (60" / year or 2.347xlo-3 m3/s for the Mirror Lake) and keeping other parameters the same, the computed V st and V g are 6.42 x w-3 and 2.99 x w-4 m3 Is. The ground water seeps into the lake at higher evaporation rate as the sign of V g is positive. At higher evaporation rate the replacement time is only 36 days.

PAGE 105

93 Chapter Discussion Ground Water and Carbonate Aquifers Assuming the average soil-derived o13Cmc of -24 o/oo in central Florida (Rightmire and Hanshaw, 1973) and the o13C of starting calcitic aquifers of +2.0 o loo vs PDB (from this study), the resulting ground water's o13C should be -11.0 o/oo (={[-24][1] + [ +2.0][1]} / {[l + l]}; see equation 22, Chapter 4), which is close to the observed -10.4 o/oo value. The buffering capacity of calcite and dolomite in confined aquifers helps maintain both the pH ( = 7.5) and o13Cmc ( = -10.4 o / oo vs PDB) in ground waters along the southern flow paths that receive no additional inputs of organic-derived C02 The rather constant Ca/Mg ratio of the southern flow path also indicates that the carbonate aquifer is in near chemical equilibrium with the HC03 -(aq) This finding may not preclude the use of isotopic mass balance to compute the water-rock ratio and effective porosity. There is still an unanswered question as to why in the unconfined aquifer (the northern flow path) the o18o of upper Floridan water does not show a similar trend as in o13C (Figures l8b and l9b). Could this phenomenon mean that the surficial or intermediate ground waters play a role in those changes, or was there a more extensive water-rock interaction, or did the upper Floridan water in this area originate from rainwater much different from today's? One possibility is that the source of this upper Floridan water comes from 2 sources: the Green Swamp and the Pasco High areas. Further isotopic and major ion studies could help identify the source and processes involved in the northern flow path. For 0180 in ground waters of the confined aquifer (southern flow path), the progressive enrichment of o180 from the recharge to discharge areas depicts a simple

PAGE 106

94 mixing of younger and older (rock-altered) waters. By knowing the recharge's o18o in carbonates and the temperatur e at which the equilibration of dissolved carbonate takes place, the modified o180 of carbonate can be calculated. Assuming that 1. The temperature of recharge water of 23 C in the Green Swamp area (similar to the average air temperature), and 2. Accepting that the average temper a ture of upper Floridan water is 25 C, and 3. The palaeotemperature vs degree celsius relationship is -0.24 o /oo per o c [Ep stein et al 1953], the calcite that is reprecipitated in aquifers during the water / rock interaction should be 0.48 more depleted in o180 than the original calcite (o18o = -1.43 o / oo) in the recharge zone. The calculated o18 of calcite at 25 oc is -1.91 o /oo vs PDB (= [(-0.48)-1.43]) which is close to a value of -2 09 o /oo observed at the Ruskin area (TR9-l, Figure 20). This finding further supports the idea that the water/rock interaction is responsible for the alteration of o1 8o in both carbonate rocks and ground waters in the confined aquifers. However, the conclus ion would be correct only if the rates of dissolution and precipitation are fast enough to maintain chemical equilibrium, and the contact time is long enough for isotopic equi librium to take place. The infiltration portion of rainw ate r into surficial aquifer ( = 2" of water level) doe s not show evidence of excessive evapo ration at all (le ss than 5 % volume loss) based on o18o data. It i s possible t h at the transpiration process transfers the remainin g water of 50" high into the atmosphere annually (from 52" rainwate r). The isotopic data on transpiration over large sca l e of west central Florida is needed for further under standing of the effec t of t ranspiration on the water budget.

PAGE 107

95 Mirror Lake From the stable isotopic viewpoint, evaporation rate plays a major role in directing the flow of lake water out or into the ground water. However, the average flow volume of storm water, out of the lake, is 5 x w-3 m3/s over a one-year period based on model calculations given earlier. Summary o18o and o13C01c of the upper Floridan water were followed along 2 flow paths from the recharge to discharge areas. The northern flow path (along unconfined aquifers) has an invariant o180 in ground water, whereas the o13C01c decreases in a linear fashion. The southern flow path (along confined aquifers) shows an increasing o18o trend; however, the o13C01c seems to be constan t except for the DV-2 Lakeland, and Lake Alfred areas. High Ca/Mg ratios in the upper Floridan water of the northern flow path can be caused by excessive water-rock interaction. The reaction coupling with sources of light C02 results in the decreasing o13C01c a lon g the travel path. Time-series s tudy of ground water at Roy Haynes and Baum-Wimauma indicates the neces s i ty of adequate purging of wells before co llecting sa mpl es for isotopic analyses. The upper Floridan water for each site has consistent o13C01c and o180. These isotope data did not vary much during r epeated studies during a 15-month period. For surficial water the o13C01c can be affected by local shallow water input. The EC02 i s also dependent on the frequency of well purges. Calculations using isotopic mass balance of o1 80 in carbonate and ground water in the recharg e and discharge areas of the confined aquifers (southern flow path) gives a

PAGE 108

porosity (:::::: 0 .5) similar to a previous s tudy, and also effective water/rock ratio of 0 .38. This water/rock ratio assumes that air pockets are absent. 96 The intersection value (-4.23 0.73) of linear-regressed lines of oD and o180 of rainwater (St. Petersburg) and ground water (upper Floridan water) suggests that the water recharging upper Floridan water h as a similar o180 to today's rainwater (of -4.4 o /oo). Further computation using isotopic mass balance of o180 in Mirror Lake, downtown St. Petersburg, shows that the r e pla ceme nt time of Lake water is 43 days This replacement time depends upon the evaporation rate. The small loss of infiltration water before recharging surficial aquifer suggests that the entire evaporation process is predominantly the transpiration process of various plants for the inland areas.

PAGE 109

Methodology and Standard CHAPTER 6 GENERAL DISCUSSION 97 The inter-laboratory calibration with the USGS (Menlo Park)'s Kendall l aboratory revealed that the USF working o18o standard from the Gulf of Mexico was 0.3 o/oo heavier than the Kendall's standard. The slight difference was within the analytical uncertainties in the USF laboratory ( 0.3 o/oo vs SMOW). Figure 30 illustrates a plot of adjusted USF o18o data against the USGS o18o data. The adju st ment of USF data is done because the USF working s tandard is abou t 0.3 o/oo vs SMOW heavier than international SMOW standard. The correlation equation s uggests that the reproducibility of USF l aboratory is comparable to the USGS laboratory. For unknown reasons, the sealed IAEA sta ndard ampoule, which was stored on a shelf for 2 years before breaking open for cal ibr at ion with the workin g USF standard, was -1.35 o/oo relatively li g hter (in o1 80 t e rm) than the USF working SMOW standard. Implications on the Modelling of the Hydrological Cycle of West-Central Florida Attempt s to estimate the replacement time of the entire water volume of both Tampa Bay and Mirror Lake s u ggested that both water bodies had comparable replacem e nt times (43 67 days). For Tampa B ay, the dredging of shipping lanes may possibly alter the hydraulic prop e rtie s of the Bay (Galperin et al., 1991), and the replacement time ac co rdingly. In the n ea r future, the rep l acement time of Tampa Bay

PAGE 110

2 0 -4 (J) Vl ;;>-0 6 C-0 00 USGS o 18o = 0.07(.1 6) + 0 .99(.04)US F o 18Q, r2=0. 990 -(J) -8 C!) (J) ;:::> -10 -12 -10 -8 -6 -4 -2 0 US F 188 0 vs SMOW Figu r e 3 0 Li near Corr e lation b etwee n the Adj u s t ed USF' s o1 8o an d the USGS's o18o Data

PAGE 111

99 using the isotopic balance model will be validated by 3-dimensional modelling using the temperature, wind, salinity, and tides, which is sti ll in progress. As shown in the study of o18o of r ainwa ter, the surface air temperature did not covary at all with the o18o in rainwater (Figure llc, Chapter 3) Questions remain whether the assumption of linear correlation between the o18o of rainwater and air temperature which are normally used for palaeo-climatic studies are correct for the sub tropics Generally, the correlation breaks down at surface air temperature above 15 o c (Jacob and Sonntag, 1991). This temperature effect is secondary as the real condensation of precipitation occurs inside the rain clouds. The primary temperature effect involves water in liquid solid and vapor forms (Dansgaard, 1953; Jouzel, 1986). The surface air temperature (secondary effect) correlates with the o18o in precipitation as if it were primary effect only when 1. The exchange with atmospheric water vapor is minimal during falling down to the ground (Woodcock and Friedman 1963; Craig and Gordon, 1965) 2. There-evaporation of droplets of precipitation is small (Stewart, 1975), 3. Evaporation of precip i tation at ground surface is minimal. In subtropic and equatorial area s the secondary temperature effect would not strongly correlate with the primary temperature effect, so the correlation between the o180 in precipitation and surface air temperature is very weak or absent as exemplified in rainwater of west-central Florida. Precipitation (or re-precipitation) of ca rbon ate minerals in aquifers records the water temp erature at which the chemical and isotopic reactions take place (Epstein et al., 1953). Thus, the old carbonate aquifers in Florida may offer a potential to estimate the palaeo-temperature of the ground water (and of surface average air t emperature), and extent of water rock interactions The surface air temperature can then be estimated from the ground water temperatures derived from carbonate aquifers.

PAGE 112

]J[J)(Q) The agreement of the in ferred ( == 0.5) using m2SS ball:ance callculatioos o f the 5180 of ground water and carbo nat e aquifers with sttOOy {&!clld er aL. 1993) indicates that the confined so uth ern flow pa th (see Cl!napre r 5) o f upper Floridan water is more or less in equilibrium with th e carbonate rocks as suggestted b)Y Jones ett al. (1993). In addition, t h e confi nin g boundary of aquife rs influeDllces the evolution o f isotopic chemistry of ground water alo ng th e flo w p ath s Lack of 5180 data in soil w ate r p recl udes th e compleie coruuection of inrernctioos between soil water and surficial aquifer water. o18o s rudy of s urficial! aquifer watter here compared with rainwater's o1 8o s ug gests a minim al evaporation o f infiltration portion of rainwater before rechar gi ng aquifers. It m a y suggest thaa ttr.mspiranion would predominate over ev a poratio n o f open w ater bod y in t e rm s of total! e\rapman:Don rate. Until further st u dies of o1 80 p r ofiles in s oi l wate r are done. ili e e xttent of transpiration over free water e v aporation i n the tottal e vaporatio n lmdg ett remain s uncertain. There is still a room for further studies o f p artitionin g o f w ater in soi L attmospl!neri c water vapor, and direct transpiration for in l and statio ns compared w ith free water evaporation u s ing the i s otopi c metho d s s ince t his diss ertatio n focusses on the coastal environment. The known partitionin g of w ater flu xes an d isot opes can be fu rther used to calculate the chan ges in transpired 0 2 and up take o f by plants i n Florida. The heavy freshwater demand in the 2 1 s t c e ntu ry fro m th e wells in the Pasco High, and west of the Green Swamp will help emph as i ze t h e need s for a be tte r understanding of the ground water re si denc e time the recharge zones, and th e w nter sources other than in the upper Floridan w a ter. A dia g rammatic summ a ry of o1 80 in va ri o us water t y p e s w hic h i s the main theme of this paper is shown in Figure 31. It sh o uld be n o ted also th a t th e intermediat e ground water is s till not well under s tood in term s of both s table oxygen an d carbon isotopes.

PAGE 113

Atmospheric F Transpired water water vapor = -13.51 I Rain I from leaves -3.3 -11.8 Surface water = + 0.96 Gulf Saltwater: ? mean = -4.4 Surficial Aquifer =II4.1 Intermediate Aquifer = ? Upper Floridan Aquifer II= -2.8 Note : Not-to-scale diagram Figure 31. A Schematic Summary of o180 of Various Water Types in West-Central Florida ..... 0 .....

PAGE 114

CHAPTER 7 CONCLUSIONS 102 1. For rainwater in St. Petersburg Florid a, the mean and weighted mean o18o are -3.3 and -4.4 o /oo vs SMOW, respectively. The weighted mean takes into account the fact that the o180 of rainfall becomes more negative as it rains heavier. 2. For west-central Florida, the o18o of rainwater does not covary with the surface air temperature (Figure 11c). Therefore palaeo-temperature calculations, based on the o180 of surficial water and temperature relation s hip in Florida, without supporting data from isotopi c s tudies of aquifer rocks, should be approached with caution. 3. The west-central Florida's meteoric water line is o D = 7.65(.20) o1 80 + 10.6( 1.1) r2= 0.998, n=5. This equation is similar to the Global Meteoric Water Lin e of Craig (1961). 4 The o180 of atmo s pheric water vapor is -13.5 o/oo v s SMOW based on s hort-term studies, while the tran spi red water from oak trees is -11.8 o/oo. The evaporative water vapor from a pan experiment ha s a o18o of -13 9 o/oo. 5 The o180 of tran s pired water correlates linearly with the specific humidity in the air. The correlation equation is o180 = 18.9( 1.09) + 0.482(.145)q, where q is the specific humidity n= 10). 6. During transpiration of water from leaves, calculations show that the leaf water is kinetically in equilibrium with the source ste m water and the atmospheric water vapor. In addition, diffusion pr edo minates durin g the tran sp iration processes. 7. The observed o18o of evaporative water vapor (oE, 13.9 o/oo vs SMOW) is similar to the calculated one ( -14. 1 o/oo) from pan experiments; the calculated o180 ( + 10.5

PAGE 115

103 o/oo vs SMOW) of the remaining liquid (oJ is close to the regressed o18o ( + 12.2 o/oo) obtained from pan experiments. 8. The apparent fractionation factor of free water evaporation (-"' or approximately = -[(a+ -1)1000]) is equal to -10 .27 o / oo form = 1.04 The slightly heavier "' value relative to the equilibrium value is attributed to an additional kinetic fractionation factor ( ) This evaporation parameters can be used to calculate the fraction of remaining liquid during evaporation, oE, and oLin local and adjacent bodies of water. 9 o18o and o13C behave conservatively along salinity gradients. The behavior suggests there are 2 major water endmembers: freshwater and saline waters The correlation equations are March, 1990 January, 1993 March, 1990 o18o = -2.2(. 2) + 0.108(. 010)salinity, n=24, o18o = -2.4(.2) + 0 099(.010)salinity n= 14, o13c = 10.4( 0 79) + 0.239(.039)salinity n=24, January, 1993 o13C = -10.6(.43) + 0.254(. 019 )sal in ity n=l4. The river endmember has similar o13Cmc (-10.5 o / o o vs PDB) to the upper Floridan water (-10.4 o / oo vs PDB). However, the river endmember's o18o (-2.3 o / oo vs SMOW) is 0.5 o /oo heavier than the Floridan water's o18o (-2.8 o / oo). The differen ce in o18o in the river endmember may be due to a lo cal evaporation effect. Ground water is likely to contribute freshwater to Tampa Bay in terms of ri ve r flows. However, the freshwater volume of rivers discharging into Tampa Bay accounts for 56 % of total freshwater inputs. The rainwater that falls directl y on Tampa Bay contributes the remainin g 44 % of the fre s hwater volume in the B ay 10. The average o18o of the upper Floridan water is -2.8 o / oo vs SMOW, while the average o13Cmc of the Floridan water is -10.4 o / oo vs PDB. 11. Ba s ed on a mass isotopi c balan ce in the confined aquifers of southern flow path, the calculated effective upp e r Floridan water / carbonate roc k m ass ratio is 0 .37 5, and the effective porosity is 0.503.

PAGE 116

104 12. The inte rsec tion points of oD and o18o of ra i nwater and local Floridan water are for oD of -21.8 o/oo and o18o of 4.23 o /o o vs SMOW. The 4.23 o /oo intersection point represents the most depleted o18o composition that ground water can receive isotopic signature from rainwater. The ground water with o180 value of -4.2 o/oo must be located in the recharge area. 13. Calculations, base d on observed o180 in surficial aquifer, suggest that the infiltration water from rainwater seeps into the aquifer with lit tle evaporation. Addition o180 profile data of soil water will help confirm the idea that transpiration process predominates over free water evaporation in the total evaporation budget for the inland areas. 14. The purging time required before collecting ground-water samples in the unconfin ed and co nfined aquifers varies depending the type of stable isotopes. For o18o the upper Florid a n water seems to be unaffe c ted by the number and frequency of purges. Inputs of surface water and frequency of purges during ground-water sampling can affect the o13Cmc in surficial aquifer. 15. Based on similar physical conditions, pan evaporation experiments can be used to infer the isotopic parameters of nearby lakes (e.g Mirror Lake). 16. Based on simple mass-balance calculations using oxygen i so topes and flow rates of water inputs and outputs, the replacement time of Tampa-Bay water is on the order of 67 days, while that for the Mirror Lake is 43 days.

PAGE 117

105 REFERENCES Allison, G.B., & Leaney, F.W. (1982). Estimation of isotopic exchange parameters, using constant -feedin g pans. J. Hydro!. 55, 151-161. Allison G.B., Brown, R.M., & Fritz, P. (1979). Evaluation of water balance parameters from isotopic me as urements in evaporation pans. In: Isotop es in Lake Studies pp. 21-32. Vienna Au stria: IAEA. Back, W & Hanshaw B.B. (1970). Comparison of chemical hydrogeology of the carbonate peninsulas of Florida and Yucatan. J. Hydr o!. 10, 330-368. Baertschi, P. (1976). Absolute 1 80 co ntent of sta ndard mean ocean w ater. Earth Planet. Sci. Lett. .31, 341-344. Barr, G.L. (1992). P otentiometric surface of the upper Floridan aquifer in Florida. M a y 1990. Florida Geological Survey, map series# 138. Berner E.K., & Berner, R.A. (1987). The glo bal water cycle: Geochemistry and environment. New York: Prentice H all. 397 pp. Bolton, D (1980). The computation of equivalent potential te mperature. Mon. Weather Rev. 108, 1046-1053. Bowden, K.F. (1967). Circulation and diffusion. In Est uaries (edited by Lauff, G.H.). pp. 19-22. Ameri can Association for the Ad vancement of Science's publication # 83. Brook s, G.R., Dix T.L., & Do y l e, L.J. (1993). Groundwater / surfacewate r interactions in Tampa Bay: Impl ications for nutrient fluxes. Report submitted to the Tamp a Bay Nation al Estuary Program. 43 pp Budd, D A., Hamm es, U., & Vacher, H.L. (199 3 ). Calcite cementation in the upper Floridan aquifer: A modern example for confined aquifer cementation models? Geology 21, 33-36. Bush, P.W., & Johnston, R.H. (1988). Ground-water hydraulics. regional flow. and ground-water development of the Floridan Aquifer System in Florida and in parts of Georgia. South Carolina. and Alabama. USGS Professiona l Paper 1403-C. 80 pp. Campbell, K.M. (1983). Geology of Hill sbo rough County. Florida. Florida Geologi cal Survey' s Open Fil e Report# 6

PAGE 118

106 Cherry, R.N., Stewart, J.W., & Mann, J .A. (1970) General Hydrology of the Middle Gulf area. Florida. Florida Geological Survey's Report of Investigation# 56. 96 pp. Coleman, M.C., Shepard, T.J., Durham, J.J., Rouse, J.D., & Moore, G.R. (1982). Reduction of water with zinc for hydrogen i sotope analysis Anal. Chern. 54, 993995. Craig, H. (1953) The geochemistry of the stab l e carbon isotopes. Geochim. Cosmochim. Acta 3., 53-92. Craig, H. (1954). Carbon 13 in plants and the relationships between carbon 13 and carbon 14 variations in nature. J. Geol. 62, 115149. Craig, H. (1957) Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochim. Cosmochim. Acta 12, 133 149. Craig, H. (1961). Isotopic variations in meteoric waters. Science 133, 1702-1703 Craig, H., & Gordon, L.I. (1965). Deuterium and oxygen-18 variations in the ocean and the marine atmosphere In Stable Isotopes in Oceanographic Studies and Paleotemperatures (edited by Tongiorgi, E.). 122 pp. Pisa: Consiglio Nazionale delle Ricerche, Laboratorio di Geologia Nucleare. Craig, H., Gordon L.I., & Horibe, Y. (1963). Isotopic exchange effects in the evaporation of water. 1. Low temperature experimental re s ults. J. Geophys. Res. 68, 5079-5087. Dansgaard, W. (1953). The abundance of 018 in atmospheric water and water vapour. Tellus 5., 461-469. Dansgaard, W. (1954). The 018-abundance in fresh water. Geoch im Cosmochim Acta .Q, 241-260. Dansgaard W. (1964). Stable i s otopes in precipitation. Tellus l.Q, 436 468. Deines, P. (1980). The i s otopic composition of reduced organic carbon. In Handbook of Environmental Isotope Geochemistry. vol. 1A (edited by Fritz, P. & Fontes, J.Ch.). pp. 329-406. Ams terdam: Els evier. Dickinson, R.E., Brezonik, P.L., Huber, W.C., & Heaney, J.P. (1969). Florida lakes. Part III. Tallahassee: Gazetteer. Dongmann, G Nurnberg, H.W., Forstel, H., & Wagener, K. (1974). On the enrichment of H2 1 80 in the leaves of transpiring plants Rad Environm. Biophys. 11, 41-52. Doyle, L.J. ( 1985). A short summary of the geology of Tampa Bay. In Proceedings of the Tampa Bay Area Scientific Informat i on Symposium 1982. pp. 27-32 Epstein, S. (1959). The of th e ratio in and some geologic implicatiOns. In Researches 1n Geo c hemistry. vol. 1 (edited by Abelson, P.H.). pp. 217-240. New York: Wiley.

PAGE 119

Epstein, S. & Mayeda, T. (1953) Variation of 018 content of waters from natural sources. Geochim. Cosmochim. Acta 1, 213-224. 107 Epstein, S., Buchsbaum, R., Lowenstam, H.A., & Urey, H.C (1953) Revised carbonate-wa ter isotopic temperature sca l e Geol. Soc. Am. Bull. 64, 1315-1326. Flanagan, L.B. & Ehleringer, J.R. (1991). Stable isotope composition of s tem and leaf water: Applications to the study of plant water use. Functional Ecol. .5., 270-277. Fransworth, R.K. & Thompson, E.S (1982). Mean. monthly. seasonal and annual pan evaporation for the United States. National Weather Service Publication # 34. 87 pp. Friedman, I. & O'Neil, J.R. (1977). Data of geochemistry. 6th edition. Chapter KK. Compilation of stable isotope fractionation factors of geochemical interest. US Geological Survey' s Professional Paper 440-KK. 1 2 pp. +49 Figures. Galperin, B., Blumberg, A F. & Weisberg, R.H. (1991). The improtance of density driven circula tion in well mixed estuaries: The Tampa Bay Experience. In Estuarine and Coastal Modelling. 2nd International Conference. Tampa pp. 332343. Gat, J.R. (1970). Environmental isotope balance of Lake Tiberias. In I sotope Hydrology 1970 pp. 109-127. Vienna, Austria: IAEA. Gat, J.R. & Dansgaard, W. (1972). Stab l e isotope survey of the fresh water occurrences in Israel and the northern Jordan Rift Valley J. Hydro!. 16, 177-212. Gat, J.R. & Bowser, C. (1991). The heavy isotope enrichment of water in coupled evaporative system s. In Stable I sotope Geochemistry: A Tribute to Samuel Epstein (edited by Taylor, H.P., Jr., O'Neil, J.R., & Kaplan, I.R.). pp. 159-168. Geochemical Society Spec. Pub.# 3. Gedzelman, S.D. & Arnold, R. (1994). Modeling the isotopic composition of precipitation. J. Geophys. Res. 99CD5), 1045510471. Gonfiantiani, R. (1978). Standard s for stable isotope measurements in natural compounds Nature 271, 534-536. Goetz, C.L. & Goodwin, C.R. (1980). Water quality of Tampa Bay. Florida: June 1972 May 1976. US Geological Survey's Water-Resources Investi gations# 8012. 49 pp. Gross, M.G. (1964). Variations in the 018/016 and C13/ C12 ratios of d iagene tically altered limestones in the Bermuda islands. J. Geol. 72, 170 194 Hanshaw, B B., B ack, W., & Rubin M. (1965). Radiocarbon determinations for estimating groundwater flow velocities in central F l orida. Science 148, 494 495. Hastenroth S. & Lamb, P.J. (1978). H eat budget atlas of the tropic a l Atlantic and eastern Pacific Oceans. Wiscon sin: University of Wisconsin Presss. 98 pp.

PAGE 120

Hendr y, C D. & Brezonik, P.L. (1980). Chemistry of precipitation at Gainesville, Florida. Environ Sci. Tech. 14, 843-849. Hoefs, J. (19 87). Stable isotope geochemistry. 3 rd edition. New York : Springer Verlag. 241 pp. 108 Holmes, M E. (1992) Fluorescence and sta ble isotopes in near shore w ate rs of south Florida an d their rel a t io n to fluorescent banding in the co r a l Montastrea annularis. MS Thesis University of South Florida. 76 pp. Hutchinson, C.B. (1983) Assessm en t of the interconnect ion between T am pa Bay and the Florid an aquifer. Florida. US Geological Survey's Water-Resources Investigations # 82-54. 55 pp. International Atomic Energy Agency, IAEA (1981). Statistical treatment of environm en t a l i sotope data in precipitation. Tec hnical Report# 206. 255 pp. Vienna Austria. Jacob 8 H. & Sonntag, C. (1991) An 8 yea r record of the seaso n al variation of2H and 1 0 in atmospheric water vapour and precipitation at Heid elberg, Germany. Tellu s 4 3 B, 291-300. Jones, I. C., Vacher, H.L., & Budd, D.A. (1993). Transport of calcium, magnesium and S04 in the Floridan aquifer, west-central Florida : Imp licatio n s to cementation rates. J. Hydro!. 143, 4554 80. Joussaum e S. & Jou zel, J. (1993). Paleoclimatic trac ers: An investigation using an atmospheric general c ir culat i on model under ice age conditions. 2. Water isotopes. J. G eophys Res. 98CD2), 2807-2830 Jouzel, J (1986). Isotope s in cloud phy s i cs : Multiphase and multistage condensation processes. In Handbook of Environmental I sotope Geochemistry. vol. 2B (edited by Fritz, P. & Fonte s, J.Ch.), pp. 61-112. New York : Elsevier. Junge, C E. & Werby R.T. (1958) The concentration of ch lor i de, sodium, potassium, calcium, a nd sulfate in rain water over the United States. L Meteorology 15, 417-425. Katz B. G. (1992). Hydr ochemistry of the upper Floridan aquifer. Florida. US Geologi ca l Survey's W a t e r-R esources Investigations Report# 91-4196. 37 pp. Klein H. & H ull, J.E. (1978). Biscayne aquifer. southea st Florida. USG S Water Resource s Invest igation. 78-107. 52 pp. Klein, H., Ambruster, J.T., M c Ph erso n B.F., & Freiberger, H.J. (1975). W ate r and the south Florida environment. USGS Water-Resources Inve st igation 24-75 165 pp. Knapp, H V (1985) Evaporation an d tr a nspiration In H and book of applied Meteorol og y (edited by Hou ghton D D.) pp. 537-554. New York: Wiley. Kohout F.A. (1965). A hypothesis concerning cyc lic flo w of s alt water related to geothermal h eating in the Floridan aquifer. Trans New York Acad. Sci. series II 28, 2 4 9-271.

PAGE 121

109 Lawrence, J R. & White, J.W.C. (1991). The elusive climate signal in the isotopic composition of precipitation. In Stable Isotope Geochemi s try: A Tribute to Samuel Epstein (edited by Taylor H .P., Jr. O'Neil J.R., & Kaplan, I.R.) pp. 169185. Geochemical Society Spec. Pub.# 3. McCrea, J.M. (1950). On the isotopic chemistry of carbonates and a paleotemperature scale. J Chern Phys. _lli, 849-857. Mcilveen, R. (1992) Fundamentals of w ea ther and climate. London: Chapman and Hall. 497 pp. McKinney, C.R. McCrea, J .M., Epstein, S., Allen, H A., & Urey, H C. (1950). Improvements in mas s spectrometers for mea s urement of small differences in isotope a bundance ratios. Rev Sci. Ins trum. 724-730 Majoube M. (1971). Frac tionnement en oxygene 1 8 et en deuterium entre l'eau et sa vapeur. J Chim. Phys. 68, 1423-1436. Marella, R.L. (1992) Water withdrawals, u s e. and trends in Florida, 1990. US Geolog i cal Survey's Water Resource s Investigations Report 92 4140. 38 pp. Merlivat, L. (1978a). The dep e ndence of bulk evaporation coefficients on air water interfacial conditions as determined b y the isotopic method J. Geophys. Res. 83, 2977-2980. Merlivat, L. (1978b). M olecula r diffusivities of H2 1 60 HD1 60, and H218o in gases. J Chern. Phy s 69, 2864-2871. Merliv a t, L. & Coantic, M (19 75). Study of mass tran s fer at the air-water interface by an i sotopic method. J. Geophys. Res. 80, 3455-3464. Merlivat, L. & Jouzel J (1979). Global climatic interpretation of the deuterium oxygen-IS relation s hip for precipitation. J Geophys. R es 84, 50295033. Miller, J.A. (1986) Hydrogeologi c fram ew ork of the F loridan aquifer system i n Florida and part s of Georgi a South Ca rolina and Alabama. USGS Prof. Pap. no. 1403-B. 91 pp. Mook W.G. (1968) Geochemist ry of the s table ca rbon and oxygen i s otope s of natural w a t ers in the Neth erla nd s Unpubli s hed PhD Di sse rtation University of Groningen 157 pp. (English translation). Mook, W.G. (1986). 13c in atmospheric C02 Netherlands J. Sea Res. 20, 211 223 Mook W.G. & Koe ne, B.K.S. (1975). C h e mistry of disso lved inorganic carbon in estuarine and coa sta l brakish w aters. Est. Coastal Mar. Sci .3_, 325-336. Mook, W G. & Tan, F. (1991). Stabl e carbon i so topes in rivers and e s tuaries In Biogeoch emis try of Major World Riv e r s ( e dited by D ege n s, E.T., Kempe, S. & Richey J.E.). pp. 245-264. New York : Wiley.

PAGE 122

Mook, W.G., Bommerson, J.C., & Staverman, W.H. (1974). Carbon isotope fractionation between dissolved bicarbonate and gaseous carbon dioxide. Earth Planet. Sci. Lett. 22, 169-176 National Oceanic and Atmospheric Administration (NOAA) (1941-1992). Climatic data-Florida, vol. 45-95. Ashville, North Carolina. 110 Netratanawong, T. (1990). Mineralogy of the Monterey Shale in the Point Pedernales area. south central coastal California. Master's thesis, University of South Florida 102 pp. NOAA (1993) Tampa Bay oceanographic project: Physical oceanographic synthesis. Technical report NOS OES 002 184 pp. NOAA/EPA (1989). Su sce ptibility and s t a tus of Gulf of Mexico estuaries to nutrient discharges. Summary report. 37 pp. Newell, R .E. & Zhu, Y. (1994). Tropospheric rivers: A one-year record and a possible application to ice core data. Geophys. Res. Lett. 21, 113-116. O'Neil, J.R. & Epstein, S. (1966). A method for oxygen isotope analysis of milligram quantities of water and some of its applications. J. Geophys. Res. 11, 4955-4961. O'Neil, J R., Adami, L.H., & Epstein, S. (1975). Revised value for the 018 fractionation between C02 and H20 at 25 C J. Research USGS .;2, 623-624. O'Leary, M.H. (1981). Carbon i so tope fractionation in plants. Phytochemistry 20, 553-567. Penman H.L. (1948). Natural evaporation from open water, bare soil and grass. Phil. Trans Roy a l Soc. London l93A 120 145 Petit J.R., White, J.W.C., Youn g N.W., Jou ze l, J., & Korotdevich, Y.S. (1991). Deut eri um excess in rec e nt Ant a rcti c s now. J. Geophys. Res 96CD3), 5113-5122. Plummer, L.N. (1977). Defining re ac tions and mass transfer in part of the Floridan aquifer. Wat. Resources Res.}, 801-812. Plummer, L.N., Parkhurst, D.L., & Thorstenson, D.C. (1983). Development of reaction models for g round-water systems. Geo c him. Cosmochim. Acta 47, 665686. Pride, R.W., Meyer F.W. & Cherry, R.N. (19 66) Hydrology of Green Swamp area in central Florida. Florida Geological Survey's Report of Investigation # 42. 137 pp. Ri g htmi re, C.T. & Han shaw, B.B. (1973). Relationship between the carbo n isotope composition of soil CO? an d dissolved carbonate species in groundwater. Wat. Research Res. 2, 958-967. Roether, W. (1970). Water C02 exchange set-up for the routine 180xygen assay of natural waters. Int. J. Appl. Radi ation I so. L 379-387.

PAGE 123

llJ Rosenfeld, W.D. & Silverman, S R (1959). Carbon isotope fractionation in bacterial production of methane. Science 130, 165 8-1659. Rozanski, K. Araguas-Araguas, L., & Gonfiantini, R. (1993). Isotopic patterns in modern global precipitation. In Climate Change in Continental Isotopic Records (edited by Swart, P.K., Lohmann, K.C., M c Kenzie, J. & and Savin, S). pp 1-36. American Geophysical Union Monograph# 78. Ryder, P.D. (1985). Hydrology of the Floridan aquifer system in west-central Florida. USGS Prof. Pap. 1403-F. 63 pp. Sackett, W. M. ( 1989). Stable carbon isotope studies on organic matter in the marine environment. In Handbook of Environmental Isotope Geochemistry. vol. 3A (edited by Fritz, P. & Fontes, J.Ch .). pp. 139-169 Am sterdam: Elsevier. Sackett, W.M. & Moore, W.S. (1966). I so topic variations of dissolved inorganic carbon. Chern. Geol. 1. 323-328. Sackett, W .M., Netratanawong T., & Holmes, M (1991). Stable carbon and oxygen isotope variations in waters of the Tampa Bay estuary. In Proc eedi ngs of the Tampa Bay Area Scientific In formation Symposium 2 (edited by Treat, S.F. & Clark, P.A.). pp 137-142. Sackett, W., Brooks, G Conkright, M., Doyle L., & and Yarbro, L. (1986). Stable isotope compositions of sedimentary organic carbon in Tampa Bay, Florida, USA: Implications for evaluation oil contamination. Appl. Geochem. l, 131-137. Schoch -Fisc her H., Rozanski, K., Jacob, H., Sonntag, C., Jouzel, J. Ostlund, G., & Gexh, M A (1984). Hydrometeorological factors controlling the time variation of D, 180 and 3H in atmospheric water vapour and precip itat ion in the northern westwind belt. In Isotop e Hydrology. 1983. pp. 3-30. Vienna, Austria: IAEA Schwartz, B. E. & Bosart, L.F. (1979). The diurnal variability of Florida rainfall. Mon. W eather Rev. 107 1535-1545. Scott, T.M. (1991). A geological overview of Florida. In Florida's Ground Water Quality Monitoring Programs-Hydrogeological Framework (edited by Scott, T .M., Lloyd, J.M., & M ad dox, G.). pp. 5-92. Florida Geological Survey's Report of Investigation #32. Sharma, T. & Clayton, R.N. (1965) M eas urement of 018/ 016 ratios of total oxygen of carbonates. Geochim. Cosmochim. A cta 29, 1347-135 3 Sprinkle, C.L. (1989). Geochemistry of the Floridan aquifer system in F lorid a and in parts of Georgia. South Carolina. a n d Alabama. USGS Prof. Pap. no. 1403-I. 80 pp. Sofer, Z. & Gat, J .R. (1972). Activities and co nc ent rations of oxygen-18 in concentrated aqueous sa lt sol uti ons: Analytical and geophysical implications. Earth Planet. Sci. Let. .12, 232-238. Steinhorn, I. ( 1991). On the concept of evapora tion from fresh and saline water bodie s. W at. Re sources Res. 27, 645 648.

PAGE 124

Stewart, M.K. (1975). Stable isotope fractionation due to evaporation and isotopic exchange of falling waterdrops: Applications to atmospheric processes and evaporation of lakes. J Geophys. Res 80, 1133-1146 112 Stewart, M.K. (1981). 180 and D enrichment by evaporation from sample container. Int. J. Appl. Radiation Iso. 32, 159-163. Strain, P.M. & Tan F.C. (1979). Carbon and oxygen isotope ratios in the Saguenay Fjord and the St. Lawrence estuary and their implications for paleoenvironmental studies. Est. Coastal Mar. Sci. 119 126. Stringfield, V. T. (1936). Artesian water in the Florida peninsula. US Geological Survey's Water Supply Paper #773-C. pp. 115-195. Swancar, A. & Hutchinson, C.B. (1992). Chemical and isotopic composition and potential for contamination of water in the upper Floridan aquifer in west central Florida, 1986-1989. USGS Open File Report 92-47. Tampa Bay National Estuary Program ( 1 992). Review and synthesis of historical Tampa Bay water quality d ata. Technical Publication # 07-92. 177 pp. Taylor, H.P., Jr. (1974). The application of oxygen and hydrogen isotope studies to problems of hydrothermal alteration and ore deposition. Econ. Geol. 69, 843-883. Upchurch, S.B. (1992). Quality of water in Florida's aquifer systems. In Florida's Ground Water Quality Monitoring Programs Background Hydrogeochemi s try (edited by Maddox, G.L., Lloyd, J.M., Scott, T.M., Upchurch, S.B., & Copeland, R.). pp. 12-29 0. Florida Geological Survey's Report oflnvestigation #34. Urey, H.C. (1947). The thermodynamics properties of isotopic substances. J. C hern. Soc. London (1947), 562 581. Vogel J.C., Grootes, P.M., & Mo ok, W.G. (1970). I sotopic fractionation between gaseous and dissol ved carbo n dioxide. Z. Physik 230, 225 -23 8 Welhan, J.A. & Fritz, P. (1977). Evaporation p a n isotopi c behavior as an index of isotopic evaporation conditions. Geochim. Cosmochim. Acta 1.1, 682-686. White, J.W.C. & Gedzelman, S D. (1984). The isotopic composition of atmospheric water vapor and the concurrent m eteo rolo g ical conditions. J. Geophys. Res. 89CD3), 4937-4939. Win sberg, M D (1990). F lorida weather. Orlando: University of Central Florida Press. 171 pp. Woodco ck, A.H. & Friedman, I. (1963) The d e uterium content of raindrops L. Geophys. Res. 68, 4477-4483. Yoshida, N. & Mi z utani Y. (1986). Preparation of carbon dioxide for oxygen-18 determination of water by u se of a plastic sy ringe. Anal. Chem. 58, 1273-1275.

PAGE 125

113 Yurtsever Y. & Gat, J.R. (1981). Atmospheric waters. In Stable Isotope Hydrology. D euterium and Oxygen-18 in the Water Cycle (edited by Gat, 1 .R. & Gonfiantini, R.). pp 103-142. IAEA Techn ica l Reports Series no. 210. Zimmermann, U., Ehhalt, D., & Munni ch, K O. (1967). Soil-water movement and evapotranspiration: Changes in the i sotopic composition of the water. In : Iso to p es in Hydrology 1966. pp 567-585. Vienna, Austria: IAEA Zarbock, H W (1991). Pas t, present and future freshwater i nflow to Tampa Bayeffects of a changing watershed. In Proceedings of the Tampa Bay Area Scientific Inform ation Sym posium 2 (edited by Trea t, S.F. & Clark, P.A.). pp. 23-33.

PAGE 126


PAGE 127

APPENDIX 1 THE CALCULATION OF SPECIFIC HUMIDITY (q) q = Ee/p (Mcilveen, 1992) q = s pecific humidity (g o f water va p o r per kg of air) = 0.622 e = water vapor p ress ure (mbar) p = atmosph eric pressure (mbar) RH = lOOe/es (Mcll veen, 1992) RH = rela tive humidit y e = water vapor pressure (mbar) (28) (29) e5 = saturated water vapor pressure (mbar at a given temp e r atu r e) e s = 6 .11 107.5 t /( t +237.3) e5 = saturated wate r vapor pressure (mbar) t = degree Celsius (30) (Bolton, 1980) 115 The e5 i s calculated from the observed temp eratu re (eq. 30), and substituted into eq 29. The calculated e is then substituted into eq. 28. Exampl e : At 25 oc, and 60 % rel ative humidity ---> e5 = 3 1.79 mbar atmospheric press ure = I, 013 mbar (1 at m .) e = 19.01 mbar q = 0.0116 8 (or= 11.7 g of water per 1 kg of air) At pumping air rate of 1 .5 Llmin. for 5 hours, MW of air ""' 29 --> ma ss of air = 0.5826 kg, the conden s ed water = 6.8 g.

PAGE 128

116 APPENDIX 2. DESCRIPTIVE MINERALOGY OF THE CARBONATE SAMPLES OBTAINED FOR THE o180 AND o13C ANALYSES 4990 Lake Alfred 2: Calcite trace: quartz, aragonite, cholorapatite 5471 Loughman: Calcite (pure) 5473 Green Swamp: Calcite trace : quartz, fluorite 13052 Lake AI fred 2: Calcite tra ce : quartz dolomite, chlorapatite 13515 TR 9-1: Calcite trace: dolomite, quartz aragonite 14887 68-2: Calcite trace: quartz 14889 87: Calcite (pure) 15650 88: Calcite trace: quartz 16268 Baum Wimauma : Dolomite trace: quartz, analcime?, calcite 16428 Roy Haynes: Calcite tra ce: quartz 16577 DV-2 : Calcite trace: quartz 16618 TR9 2: Calcite trace: quartz The first 4 or 5 digit numbers are the Florida Geological Survey's ID numbers for the cores. The following name is for th e SWFWMD's ID. The tra ce mineral s are listed in the decreasing order of abundance. The major mineral comprises ;;::: 75 % of total abundances. The identifi ca tion was based on the peak readings from the X-ray diffra c togram gene rat ed f r om a diffractometer in the Dep artmen t of Marine Science (see Netratan awo n g [1990] for a d etailed discussion on the methodology).

PAGE 129

117 APPENDIX 3. THE PREPARATION OF 100% H 3P04 FROM 85% (w/w) H3P04 and P205 1. First, the following chemical equation is established : (30). Note that the maximum amount of water absorbed by P205 is 3 times the molar concentration of P20s. 2. Determine the given P205 size (mw = 141.94) 3 In 1 gram of 85% H 3P04 there is 15(1) / {(100)(18.015)} moles of water. 4. Calculate the amount of 85% H 3P04 needed for a given P205 e.g. Given 100 g P205 and 85% H 3P04 4 1 Maximum of water amount taken up by the P205 = 3{[(100gP205)(1 mole of P205)]/[(141.94 gP205)]} 4.2 Amount of H 3P04 required = 100(18 015)(3)(100) / {(15)(1)(141.94)} = 253.8 g of 85%H 3P04 5. If one needs to mea s ure the volume, use the density appropriate for the 85% H3P04 ( = 1.689 g/cm 3 ).

PAGE 130

118 APPENDIX 4. REGRESSED VALUES OF m AND o5 FROM PAN EXPERIMENTS The equation given by Welhan and Fritz (1977) is used: (Chapter 3) (10). Upon the manipulation mlnf = ln(o5-o) ln(o5-o0 ) lnf = (llm)ln(o5-o) (1/m)ln(o5o0 ) y = b*x +a y=lnf; b= lim; o5=o0+e-am; o0=-2.42 o / oo. The algorithm is to minimize the diffe r e nce between the guessed o5 and the observed o for all the data points. The s lope an d the intercep t are iterated until the best r is obtained. Xuewu Liu helped write the following BASIC code for used in the QBasic under the DOS computer. Since thi s QBasic cannot generate the final .EXE file, therefore the code will always be see n during the tran s lation by the QBasic compiler. pan3 panl PAN3 DIM x(8), y(8) xO = -2.42 n = 8 + 12.22 + 14.19 LPRINT "guessed ;calculated ;difference; m; r" FORi = I TOn READ y(i), mx(i) y(i) = LOG( y(i)) NEXTi FOR p = 0 TO 6 STEP 0 2 5 xi = 7.3 + p FORi = 1 TOn + 1.041 +0.8074 x(i) = x1 mx(i): x(i) = LOG(x(i)) NEXTi GOSUB 1000 m = 1 I b x 11 = xO + EXP( -a m) LPRINT x 1 x II, xI x 11, m, r

PAGE 131

APPENDIX 4. (Continued) NEXTp END 1000 XX = 0: X = 0: yy = 0: y = 0 : xy = 0 FORi= 1 TOn X = X + x(i) y = y + y(i) XX = XX + x(i) X(i): yy = yy + y(i) y(i): xy = xy + x(i) y(i) NEXTi xx = xx x x I n: yy = yy-y *yIn: xy = xy x y I n b = xy I xx a = (y b x) I n r = xy I SQR(xx yy) RETURN DATA 1 ,-2.42,0.9287,-1.06,0.9064,-0.49,0. 7771 ,0.804,0. 7503,0. 94,0.5943,4.08,0.4293,6.14,0.3579, 7.29 PAN 1 DIM x(8), y(8) xO = -2.42 n = 8 LPRINT "guessed;calculated;difference; m; r" FORi= 1 TOn READ y(i), mx(i) y(i) = LOG(y(i)) NEXTi FOR p = 0 TO 4 STEP .025 xl=11.7+p FORi= 1 TOn NEXTp END x(i) = x1 mx(i): x(i) = LOG(x(i)) NEXTi GOSUB 1000 m = 1 I b xl1 = xO + EXP( -a m) LPRINT xl, xll, x1-x l1 m r 1000 XX = 0: X = 0: yy = 0: y = 0 : xy = 0 FORi= 1 TOn X = X + X(i) y = y + y(i) XX = XX + X(i) X(i): yy = yy + y(i) y(i): xy = xy + x(i) y(i) NEXTi xx = xx x x I n : yy = yy y y I n : 119

PAGE 132

APPENDIX 4. (Continued) xy = xy-x y I n b = xy I xx a = (y -b x) I n r = xy I SQR(xx yy) RETURN DATA 1,-2.42,0.9002 ,0.09,0.8669, 0.36,0. 7339,1.623 ,0.6342, 1.816,0 6176,2 88,0.4347,6 .05 ,0.1021, 11.69 120


Download Options

No images are available for this item.
Cite this item close


Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.


Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.


Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.


Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.