USVI Public Records

Document text

GROUNDWATER GEOCHEMISTRY OF THE ST. CROIX CARBONATE AQUIFER SYSTEM Ivan P. Gill Dennis K. Hubbard August, 1986 Project No. 02 Agreement No. 14-08-0001-G1i050 i rs re es ar Technical Report No. 27 Caribbean Research Institute College of the Virgin Islands St. Thomas, U.S.V.1. 00802 = Technical Report No. MG-3 West Indies Laboratory Teague Bay, St. Croix U.S. Virgin Islands 00820 [ r r eats | GROUNDWATER GEOCHEMISTRY OF THE ST. CROIX CARBONATE AQUIFER SYSTEM Ivan P. Gil) Dennis K. Hubbard [ Project No. 02 Agreement No. 14-08-0001-G1050 Technical Report No. 27 Caribbean Research Institute College of the Virgin Is] ands St. Thomas, U.S.V.1. 00802 f is based was financed The research on which this report in part by the United States Department of the Interior, Geological) Survey, through the Virgin Islands Water Resources Research Center. Contents of this publication do not necessarily reflect the views and policies of the U. S. Department of the Interior, nor does mention of trade names or commercial products constitute their endorsement by the the United States Government. ome kiana: ne or ie oes a [ [ [ ABSTRACT Samples from public and private wells representing a variety of locations across the carbonate plain on St. Croix were collected and analyzed according to U. S. Geological Survey specifications. Almost a)] samples analyzed exceeded US EPA recommended limits for chloride and dissolved solids content. Several of the samples exceeded EPA limits for sulfate. However, limits for these constituents are based on esthetic and taste considerations rather than considerations of health. Mean levels of sodium in the groundwater, however, ranged from 99 to 937 ppm and are a potential health risk to those on sodium restricted diets. Chemical modeling calculations show that the groundwater is generally saturated with respect to quartz, and in several cases with kaolinite and gibbsite. Coastal wells, but not inland wells, were calculated to be saturated with respect to calcite, probably as a result of seawater influence. One wel] sample was calculated to be simultaneously undersaturated with respect to calcite and super- saturated with respect to dolomite. The major sources of groundwater mineralization on St. Croix are seawater mixing and diagenetic interactions. Concentration of rainwater through evapo-transpiration, though widely cited, is at best of minor importance. Diagenetic interactions were of greatest importance to groundwater mineralization in the inland portions of the aquifer. Toward the coasts, seawater mixing tended to overprint any diagenetic effects in the groundwater. All major elements tested except for potassium and magnesium plotted above the mixing curve for rainwater and seawater. In general, silicate phases play an unexpectedly large role in the chemistry of this dominantly carbonate aquifer. In addition to the dissolution of carbonate minerals, cation exchange and silicate mineral transformations probably contribute to the chemical make-up of the groundwater. ACKNOWLEDGEMENTS This report is the result of the cooperation and assistance of numerous people, many of whom are listed below. We appreciate the help of those who may have been omitted. Wanda LeBlanc and Ron Snelling were generous in their assistance and discussion of laboratory techniques and Dr. Ron Snelling performed the ICP analyses. Tom Oswald of Coastal Studies Institute and Dr. Gambrel! and his coworkers in the Wetlands Soils and Sediments Laboratory were generous with their time and equipment during the analytical phase of the project. Discussions with Dr. Don Thorstenson on field sampling and analytical procedures materially aided the project. Numerous members of the faculty at Louisiana State University lent guidance and advice, including Drs. P. Aharon, L. Chan, R. Ferrell, J. Hanor, C. Moore, M. Simms and R. Snelling. Michael Simms supervised the geochemical modeling programs. Shelley Choy and Ellen Tye assisted with the data work-up and drafting preparation. The manuscript was significantly improved by suggestions and discussion from Rick Huff, Mike Simms and Dr. H. Smith. Assistance and cooperation from branches of the Virgin Islands Government is appreciated, in particular Mr. E. Hansen and Mr. M. Restovic of the Department of Public Works and Ms. V. Springer of the Virgin Islands Planning Office. In addition, numerous wel! owners allowed us access to their property and well systems. Field and hydrologic information was provided by Tom Sedgwick, Lowell Schuster, Ken Eastman and Louie Hewlett, and we thank them for volunteering their time. We are appreciative of the help of the staff of the West Indies Laboratory, particularly H. Gieben, A. Lang, E. Phillips, J. Runge and R. Vatcher. Fernando Gomez-Gomez and Angel Roman of the U. S. Geological Survey, Puerto Rico provided helpful advice and equipment throughout the sampling phase of the project. Financial support for this project came from the U. S. Department of the Interior through the Virgin Islands Water Resources Research Center, the Applied Carbonate Research Program and Department of Geology of Louisiana State University, and field work grants from SOHIO, Dr. D. Eby, Champ!in O11, Shel! O11, Chevron Oil, American Association of Petroleum Geologists, and the Geological Society of America. One of the authors (Gill) was supported by a fellowship through the Louisiana State University Alumni Federation. ii [ TABLE OF CONTENTS ee . Abstract eererereeer eee eee ee ee eoee . Acknowledgements eoocee ° eee e ° e . . ee e e oe e e ii Table of Contents eee e eoeocereeeews . ° e e iii List of Figures eoeee eoeeoe ee e . e . List of Tables @oeoeverereerwreoe eee e e vi Introduction cooeowreoer oes ecoeoeevee oe ° e e e r Study Site eoeeoeeeee eoeoeee e e e e eroeeee [ Previous Work eoroerreo ere eer ere eoeee e e . . Summary eoee . eoeoeoee e e e 14 ( Methods e eoeoerererw eer eee ° eeceoeeeve ° . e 15 Sample Collection . eeooee ° e e e e e 15 Well Selection eee ee eoeoee e e ° . e ° . ° 17 Analysis eee ° ° 17 eoeoeeevereeerer eevee ° e e 18 Modeling Results eeoevoevee oe eeoeoereoeeoeee e e e 19 Concentration Levels e ° . ee e ° e e 19 [ Geographic Trends eoeoverewevevee severe eee eee ee wm vere 20 Mixing Curves eeoeoeeeeeere eee e ee e . 23 Chloride Content vs. Depth eeceoeeeeoweeoe 29 Trace Elements eoeoeoereeeoe e e e ee ° e e e 31 Chemica] Modeling eoeeeeeveve e ° ee 32 iii [ TABLE OF CONTENTS Discussion 35 eoeereoeereereere eee eer ee ee ewe er ee eevee 35 Water Quality Mineralization of Groundwater eeoeoererew eer eee ee eve 37 Chemica) Modeling eooeoeeove 46 Summary eooeoevrerereo eer eow eee eee eoeovoeeeoe oe eooereeeeoe 48 References 50 ooerere eee ee eee eee eee ee 53 Appendix: Chemical Analyses r iv ) Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure ® N & 0 @O VN AN UO 10 11 12 13 14 15 16 St. Croix General iz LIST OF FIGURES location map and study area.... ed geologic map of St. Croix.... Well sample locationS .......eeeees Plot of water column altitude vs. (From Robison, 1972) wee. chloride Chioride concentration isopleth map Sodium concentration isopleth map Sodium:Chloride ratio isopleth map Seawater Seawater Seawater from published reports) Seawater Seawater Seawater Seawater Seawater Chloride mixing curve: Na vs. Cl mixing curve: Salinity vs Cl mixing curve: Na vs. Cl (data mixing curve: K vs. Cl eee wees mixing curve: K vs. Na .. mixing curve: Ca vs. Cl mixing curve: HCOS vs. C} mixing curve: Mg vs. Cl vs. water column altitude eooerwowo wo eoe eee eoeve coocee eoeeee 11 21 22 22 24 25 25 26 27 28 28 29 30 Table 1- LIST OF TABLES Modeling results: saturated phases vi oeeceeve INTRODUCTION St. Croix draws a significant and increasing fraction of its water from the limestone aquifers of the central plains region. Despite this, there is a major lack of information regarding the geology of the aquifer system, and more to the point of this report, a major lack of information on the quality and chemistry of the groundwater itself. Past information on St. Croix groundwater is adequate for general description of groundwater trends, but was collected more than a decade ago (Cederstrom, 1950; Robison, 1972; and Jordan, 1975; among others). Little information is available on the techniques used in the collection and analysis of the samples, and few published sources include sufficient parameters to allow modeling of the various water and mineral reactions in the aquifer. These last two points are important in determining the sources of mineralization of the groundwater, and in understanding the possible chemical reactions occurring in the aquifer system. Geochemical information may be used as an independent check on hydrologic models of the aquifer and will aid in understanding the diagenetic reactions within the carbonate section. [ [ [ [ l [ [ [ [ [ [ [ ' The objectives of this report are to: 1) provide reliable, self-consistent data on St. Croix groundwater adequate for geochemical modeling and proposed hydrologic analysis; 2> produce information on both private and public wells throughout the limestone aquifer system; 3> briefly tie-in the chemistry of the groundwater to the on-going research on the subsurface geology of the Kingshil! Limestone. Water samples were taken from wells in a variety of locations in the St. Croix limestone region and analyzed at Louisiana State University and the West Indies Laboratory. This report deals with the initial results of the groundwater analyses. This project is part of an on-going doctoral research effort and refinements and additional results will no doubt be made in the near future. STUDY AREA St. Croix lilies at the northwestern edge of the Lesser Antilles arc, approximately 176km (95mi> southeast of Puerto Rico (Fig. 1). At its widest points, the island is 39 km (21 mi> long, 9 km 6 mid wide and covers a total of 207 sq. km (84 square miles). St. Croix is tectonically and geologically distinct from the rest of the primarily igneous islands of the Lesser Antilles. The mountainous eastern and western ends of the island are formed by well-lithified siliciclastic rocks of Cretaceous age (Fig. 2). These rocks are sedimentary, and are composed of tuffaceous and volcanoclastic material deposited in deep water (Whetten, 1966). Diorite and gabbro intrusives cut this sedimentary material at several points on the island. The central plain of the island is formed by deposits of alluvium and exposures of the underlying carbonate (limestone) rocks (Fig. 2). The carbonate units supply the majority of groundwater, range from 0 to S00 feet thick, and are underlain by dark, low-permeabillty Jealousy Formation clays that exceed 1400 feet in thickness (Cederstrom, 1950). 66 64 19 BINT ow #coas Ny e060 ateamtic Octan eco > at. my oe* o cemsmar cacis ougasca VIRGIN ISLANDS canregeae / . LY ot e.@ . 7} a? av woe Kv <i PUERTO AICO ~ 20° ce Pay or 8T.CROIX 90° P0050 BASIN ae °° e® JA LY i) } SOuNomOS tn wo 9o° lane + ¢ + + + 6 + & +7 + + e + +7 Rg av® +4 SS “N04 or eo +° ov 4¢ ++ of” +7 +7 14 \¢ + + tLe Noe t+ + CORES: B4 B7 B16A Villa La Reine (Type Section, Deep Basin) Qo 2 3 4 Evans Highway (Basins) Hess Ol! (Reef / Near Reef MILES +t Cretaceous Voicenoclastics and tntrusives 0 1 2 Figure 1. St. Croix location map and study area. N{ woz fea (vyZ6 1) UBNAUM JBWe *xJOI9 °3yS JO dew S{6oO[oaH pazy|esauag °Z ain6lg JOO} JO SUB} | BIOAES 0} SSOUHIDU) BIQBUBA — S}JUN BJBUOGIS SOjjJeAO (SBe18 sNOeoRes UY UMOYS JOU) WNIANTTV AYVNYSLVND 11 00¥ Lb < SSOUXDIYY [B30 (3N3S001IW-3N3909110) SAV19 WS ASNOTVSr (SNOSOVLAYO) SAAISNYLNI t+ + + SLINN SILSVIDONVOTOA GNY SNOZOVAANL SNOFSVLAYD Zo "13.009 Ssauxdlyy wnwyxew (YSODNNOA GNV SNSOOIW) | L. SSLVNOSHVYO YSDNNOA + SNOLSAWIT TIIHSONI I I L SONV1dN SNOSOVLSYNO GNYV Nivid SNOLSSWITNSSML3E AYVGNNOS #75 Q,°7 uw) The impermeable nature of the Jealousy Formation allows it to function as an aquitard, and it is probably safe to assume that it limits the incursion of salt water from below (Jordan, 1975). Because of the underlying clays of the Jealousy Formation and the layered nature of the Kingshill] Limestone the Ghyben-Herzberg mode) of island hydrololgy does not apply to St. Croix. The lithology and depositional characteristics of the carbonate units are variable. It should be noted that most of the limestone contains significant quantities of lithic material and clay minerals. Gerhard et al. (1978) found acetic acid insoluble residue ranges from 0 to almost 99 percent in beds of the Kingshill] Limestone. Percentages of illite, kaolinite and montmorillonite also vary throughout the gection. Lithology and depositional characteristics of the limestone units are described in more detail in Multer et al. €1977); Gerhard et al. (1978); Gill and Hubbard (1985); and Gill and Hubbard (1986). Wells sampled in this report are scattered throughout the central plain region (Fig. 3), and draw their water primarily from the Kingshil] Limestone and younger carbonate units. Some, such as the public wel] field at Fairplain and the private wells in the Glynn and Concordia estates receive groundwater from overlying alluvium as well. Two wells were sampled in ‘Estate Solitude, on the eastern end of the island to contrast water from a siliciclastic aquifer with water from the carbonate-dominated central plain. 4a a ee | 4 *SUO]}eDO| aldwes [| [aM “Ee 2INnB{gJ 96 -aM OOT-u a 6s Hm a - Se-HWN N | ——| e'y-dy ge-GN cory? Ss wie tO Y 9-d4 1-117 ee nm 7 u } _oQ 8-dd_ mbMd-99 steno + O17 i Vveé-3ad 98 SB a ea ° o~, i a 1. a} H VE-SE Wy 1-90 woyfio \ PREVIOUS WORK Data on chemical groundwater quality are found in Cederstrom (1950), with analyses dating back to 1919. Detailed analyses are contained within several tables divided by aquifer type; other tables contain chloride content data along with brief lithologic logs of the wells in question. Information on pH, temperature, and method of analysis is not available. Cederstrom (1950) found that the chemistry of the groundwaters showed far higher degrees of mineralization than did surface waters. He attributed this to several processes: leaching of salts from the Kingshill Limestone, organic contamination, precipitation and re-solution of soi 1-borne alkali salts, and cation exchange. It should be noted that cation exchange should not result ina change in solution mineral content. Hendrickson (1963) contains a history of development of the public well fields, including hydrologic and construction detalls of many of the public supply wells installed before 1963. Many of these wells are still in use. Water-quality information is limited to chloride content of the wel] water. Robison (1972) presents a general! overview of hydrologic conditions in the central plains region, including recommendations on secondary treatment of groundwater, water supply and water quality. Regarding water quality, the chemistry of individual wel] analyses is not shown. Instead, groundwater analyses are grouped into "normal" and “high-saline" waters, and averages for the groups are tabulated. Parameters include: Ca, Mg, Na, HCOg + COg, SO4, Cl, F, TDS, and hardness. Robison (1972) shows that in his “normal” water grouping, chloride content Increases sharply with depth, the inflection point on the curve corresponding to mean sea level (Fig. 4). Robison (1972) explains the high degree of mineralization in the water by: hypersaline "connate" water from dissolution of evaporites, dewatering of Jealousy Formation clays or fault-restricted groundwater circulation. Robison (1972) suggests that ionic diffusion and mixing with seawater may also be responsible for many of the groundwater characteristics. He estimates that shallow groundwater has been concentrated at least 30 times by evapo-transpiration, accounting for the Increase in chloride content from that of Virgin Island rainwater. 10 AVERAGE WATER-COLUMN ALTITUDE, FEET IN EACH WELL 120 l | l | | | 100 - ° = so - e = 60 - — 40 -- = 0 100 300 500 700 900 1,100 CHLORIDE CONCENTRATION, MILLIGRAMS PER LITER Figure 4. Water column altitude vs. chloride content Cafter Robison, 1972). Jordan (1975) produced a comprehensive report dealing with many aspects of supply, treatment, development and quality of St. Croix groundwater. His discussion is not limited to the central plains region, and summarizes many of the previous reports on St. Croix groundwater. Groundwater and surface water chemical data are tabulated, including: SiQ5, Fe, Mn, 11 Total Dissolved Solids ‘TDS), hardness, specific conductivity, pH, and temperature. No information on collection or analytical) techniques Is supplied, and the pH values given are anomalously high, in some cases approaching or exceeding seawater values even on samples of low ionic strength. These high values may be a result of sample degassing before pH measurement. Jordan (1975) explains the mineralization of the groundwater as a combination of processes starting with the evaporative concentration of rainwater as suggested by Robison (1972). After concentration, groundwaters are modified by mineral dissolution, ion exchange and mixing with "connate" water. Black, Crow and Eidsness (1976), a hydrologic consulting company, produced a groundwater management plan that includes specific recommendations for St. Croix. Some chemical groundwater data are included, but the parameters listed are averages of public wel] fields only, and include Cl, conductivity, hardness, Ca, Mg, COz and HCO3. No values for pH, temperature or Na are included. Black, Crow and Eidsness (1976) state that the factors that most strongly affect groundwater guality are solution, 12 concentration and evapo-transpiration. No information on collection or analytical techniques is supplied. Buros (1976), then affiliated with Black, Crow and Eidsness, produced a study of the hydrology of the Golden Grove area during an artificial groundwater recharge project. This report contains detailed information on the geologic setting of the Golden Grove/River Gut area and contains extensive data listings of the groundwater and surface water quality in this particular area. Parameters listed include Cl, specific conductivity, hardness, Ca, Mg, COs, HCO3, NO3, NH3, P, Chemical Oxygen Demand (COD), Biological Oxygen Demand (BOD), Total Organic Carbon (TOC) and coliforms. The most recent general work on groundwater in the Virgin Islands was produced by Geraughty and Miller Inc. (1983 a and b»). The report on groundwater conditions (Geraughty and Miller, 1983a) contains information on several of the public well fields on St. Croix; the appendix contains detailed analytical data on samples from both public and private wells on St. Croix. Temperature and pH data are not included on these tables; however, tests on trace heavy metals and organic pollutants were run on several public wel}! field waters, and the results are listed in the report. 13 i ne | The groundwater management plan (Geraughty and Miller, 1983b> summarizes the findings of the first report and presents suggestions to the Virgin Islands Government on water management. The most recent and comprehensive chemical data are those from Garcia and Canoy (1984). Published by the U. S. Geological Survey, the report contains complete information on major and minor elements and on organic pollutants. Collection methods used for the groundwaters are cited, and follow established U. S. Geological Survey guidelines. Nineteen wells were sampled in the U. S. Virgin Islands, eight of them on St. Croix. Summary Most of the existing information on St. Croix groundwater is usable for general purposes such as regional trends and health regulations. However, due to lack of information on several important geochemical parameters, many of the reports do not contain sufficient information for geochemical modeling. In addition, many of the data are from reports more than a decade old, and the collection and analytical} procedures used are not cited. 14 3 METHODS The methods of collection and preservation of samples can critically affect the usefulness and reliability of the analytical data. This report follows U. S. Geological Survey guidelines, for the most part those of Wood (1976) and Claasen (1982). Sampje Collection Samples were collected from wells only after the water chemistry had stabilized according to repetitve tests for temperature, pH and specific conductivity. Buffers for the pH determination were kept at ambient groundwater temperature through the use of a water flow bath. All samples were taken as close as practicable to the well head, and were collected through inert plastic tubing and fittings. All samples were filtered through 0.45 micron filters and preserved in accordance with the type of subsequent analysis. Aliquots for major and trace elements were acidified and stored in sealed, tightly capped polyethylene bottles. Aliquots for sulfate and nitrate analysis were preserved with mercuric chloride in polyethylene bottles. Aliquots for alkalinity and stable isotope analysis were stored in sealed glass 15 4 4 ee 3 ~" “yg "9 rs rn rr ee ce P| jars. All storage and delivery vessels were chemically cleaned and dried in the laboratory, and repeatedly precontaminated with the sampled water before final collection. Alkalinity was either analyzed immediately in the field, or within 24 hours at the West Indies Laboratory. Duplicate to quadruplicate runs were made on each sample following a Gran-type titrimetric procedure (Gieskies and Rogers, 1973) and the end-point calculated from a linear regression curve. Temperature to the nearest 0.1 degree C, pH to the nearest 0.01 unit and specific conductivity to the nearest 10 micromhos/cm were determined in the field. Specific conductivity was measured at ambient water temperature in micromhos/cm2, and was not temperature compensated. Field-determined salinity was read froma temperature-compensating specific conductance/salinity meter. Specific conductivity corrected to 25 degrees C and the sum of dissolved constituents were calculated later and are listed in the appendix. For this report, we consider the terms ‘salinity’ and ‘dissolved solids’ synonymous, and are referring to the conductivity- determined salinity. 16 aa | G | eo re rs | Well Selection Both public and private wells were sampled for this project. In general, public well fields utilize steel-cased wells and are equipped with oij1l-lubricated pumps. Private wells were almost invariably equipped with submersible pumps and PVC casing. Private wells were selected on the basis of geographic distribution and accessability. Analysis Major and minor elements were analyzed at Louisiana State University on an ICP spectrophotometer. Chlorides were determined titrimetrically by the Mohr procedure, or on a laboratory chloridometer. Sulfate was measured turbidimetrically or by ion chromatography. Alkalinity was analyzed in the field or within 24 hours of collection after storage in tightly sealed glass bottles. Wet chemical techniques generally followed Skougstad et al. (1979) or American Public Health Association (1971). Alkalinity titrations followed the Gran method as given in Gieskes and Rogers (1973). End points for this titration were determined by a linear regression of the Gran plot. Data points in the linear regression 17 re were selected objectively by a simple numerical technique developed for this project. This numerical technique determines the first-derivative slope of the Gran function and eliminates those points that occur prior to where the function becomes linear. Quality control procedures for analytical data followed guidelines set forth in Skougstad et al. ©1979) and Friedman and Erdmann (1982). Modeling Data sets selected on the basis of electrical neutrality were numerically modeled for thermodynamic speciation by the PHREEQE computer mode} (Parkhurst et al. 1980). Assumptions used in the model] were: 1> The presence of a solid phase was assumed in the saturation calculations. 2) A pe of 12 was used in all calculations. This value assumes mildly oxidizing and near-neutral pH conditions in the groundwater. 18 RESULTS ce [e} Vv A total of 34 samples were taken from 27 wells, rainwater and seawater. The well] locations represent a cross-section of public and private wells on the island, and are distributed across the central] plain CFlg. 3). Well locations were chosen on the basis of geographic distribution and accessibility. Water samples were analysed for Ca, Mg, K, Fe, Mn, Total alkalinity, Cl, Vs Ba, Sr; Si, SO4, Al Co; Gd, Cu, Zn, Co, and Ni. Temperature, pH, specific conductance and salinity (by conductance) were measured In the field. In general, the values obtained for this report correspond well to those of earlier publications. Although the correspondence between measured dissolved solids and salinity measured by conductance is not exact, it is precise enough for the purposes of this report. In this report we will use the terms dissolved solids and salinity interchangeably, and the terms will refer to the conductively determined salinity listed in the Appendix. Of the wells sampled, only five met federal] standards for chloride content, and only one met federal standards for dissolved solids (Appendix). 19 Federal drinking water standards are listed In the Appendix along with the analytical results. Organic compounds were not analyzed for this report. Of the 23 samples analyzed for sulfate, four exceeded the recommended maximum |]!mit of 250 mg/L. These wells were from the Fairplain and Barren Spot well fields. Sodium was present in quantities from less than 99 ma/L to 937 mg/L, or a range of almost an order of magnitude. Although no federal] standards exist for sodium, these levels are quite high and create a potential risk for those on sodium-restricted diets. Geographic Trends The geographic distribution of chloride in the groundwater closely matches that of Geraughty and Miller €1983>. Chloride values Increase markedly near the coast-lines and decrease inland (Fig. 5). No zones of anomalously high chloride or dissolved solids values were found, although such zones were discussed in Robison €1972) and Jordan (1975). As mentioned in Geraughty and Miller (1983a), this type of diagram should be interpreted with caution. Well] data displayed in this manner do not take Into account depth, lithology or information on pumping and usage. 20 “oe 2a) OV ee 98 io” 189 + 547 ” gi2 694 of Cx ee at on <169° 248 “G &, Meee, 3 896 Na | <150 av® 402 1 00' ave S00 408 ZA 1 2mi eS C. 1. = 250 ppm 2km _- LINE OF EQUAL CHLORIDE CONCENTRATION (mg/!) NOV 65-MAR 86 a WELL LOCATION, CHLORIDE CONCENTRATION (mo/I) Figure 5. Chloride concentration isopleth map. However, such diagrams can be useful for broad r geographic trends. The geographic distribution of sodium closely follows that of chloride, increasing rapidly toward the r coast (Fig. 6). However, the ratio of sodium to chloride ¢Na/Cl> in groundwater shows the reverse geographic trend (Fig. 7, increasing rapidly away from the coast. This implies that the proportion of sodium relative to chloride increases inland. 2i [ i Pore “2714201 6 {a a” oon? 9° Ka ¢v 188 509 739 8937 a Pid @810 $059? 235,°° | ed "wee 2304 —~Ba4s\s 121 ai} Zd 2000 1165 §10 1 ~— 69 5099 QO 2mi " C.l.= 250 ppm 2km LINE OF EQUAL SODIUM CONCENTRATION NOV 85-MAR 86 @ WELL LOCATION, SODIUM CONCENTRATION Figure 6. Sodium concentration isopleth map. [ “ue of Firs \e a Ka oon” a!-7 Pid “ % Pad ov [ a/ 1.1 Pid 3.2 : 1.6 > “1.2 —_— - -2 “ee 0 Ge7- s’ / ss A 1.5 7 8s oo 1 2mi “As "A C.t.- 0.5 UNITS 9 1 2km [ LINE OF EQUAL Na/C!I (ppm) RATIO a NOV 85-MAR 86 a WELL LOCATION, Na/Cl RATIO Figure 7. Sodium/Chloride (Na/C!l)> ratio isopleth map. [ 22 Mixing Curves Mixing curves were plotted for several elements using values for rainwater and seawater as the end members of the curve. Chloride was generally used as the independent variable, and was assumed to behave conservatively. For waters of low ionic strength, this assumption should be quite accurate. For the sake of simplicity, literature values of seawater constituents were used (Drever, 1982), and the rainwater was assumed to contain no dissolved solids. No change in the conclusions are caused by the inclusion of actual analyses of St. Croix rainwater and seawater CAppendix). Sodium plotted consistently In excess of values expected if the chemistry of the groundwater were a simple function of mixing fresh water and seawater. In most cases, the excess averaged more than 100 ppm Na greater than would be predicted for a mixed water of comparable chloride content (Fig. 8). 23 Chis iy e 140% 2» ww an BE-3] y 0 gon apree FF =4 ans lJ Hind PECSH FP HS bh Cl] RUl- r tS ee ¢ Coc eal fii sa oa —_ Sti OL -14e * _ F SNE 2 OGG-PHL GL- “phe {i-82 o4 ‘a - Hoge - | i] i t 1 a 7 | | | t ai due Gai Bist else 19/4 4 diss 1ebg CHLORIGE, PPK Figure 8. Freshwater mixing curve sodium vs chloride Salinity (dissolved solids) plots consistently in excess of the mixing curve, but with a high correlation to total chloride content (Fig. 9) When combined with the behavior of the dissolved constituents mentioned below, this suggests a net contribution of dissolved constituents to the groundwater through interaction with aquifer materials Previously published data showed similar trends (Fig 10> 24 ¢ - F ¢ ¢ a + 4 aa ¢ 15064 { + a ili) tty Ae F th” . 50h ee seauater data: Sal = 35073 prs Cl = 19352 pew br a ee ee eee eee re | | cc CHLORIDE, PPH Figure 9. Seawater mixing curve: dissolved solids vs chloride. | T 4 ) 200 460 688 886 1660 1208 140 160 CHLORIDE, PPH Figure 10. Seawater/freshwater mixing curve: sodium vs. chloride (data from published reports). 25 In contrast potassium consistently fel! below theoretical mixing values when plotted against both chloride and sodium (Fig. 11 12>. Both sodium and potassium are highly mobile and form extremely soluble minerals on evaporation. It is unlikely that concentrations of either element would be modified by evaporative precipitation as in the Hardie-Eugster mode } CHardie and Eugster, 1970), by evaporative precipitation and re-solution by rainwater, or by biological processes. ri oe Seauzter eoncentrations 215 we Oe ” eo (! em a’ R 2 ro oe eo” HR 1G Hi on inke ti - -" - Ee ned us Patt aie Hd bg . Ye oe we Poel Ch-82 li Ce i Oe, = ‘ Tuer?’ a AFF e v FF-4 & Fre “Hes by 4 wes FReé (87853 * saat | I ‘ asi ; ae pH “34 ade aaa {Gui Lene 1404 fei é me FR CHL OR fir, dom?! PPI Figure ii. Seawater mixing curve: potassium vs. chloride. 26 —j3 ~3 “F "9 73 “yo 3 ZF FR “9 Lit 6 gf Seauater concentrations: Ne=lared, Kegg3 - roe on i 2 a 7 a A rie Po si . _ § bts a i a > l! sits oa : K Pell F Hi ; | foo} et LS SODIUM. PP Figure 12. Seawater mixing curve: potassium vs. sodium. Both calcium and alkalinity show up In the groundwater well in excess of a rainwater/seawater mixing curve (Figs. 13, 14). This is to be expected in a carbonate aquifer. Magnesium values form a scatter plot around the mixing line when plotted against chloride (Fig. 15), and shows no consistent trend. 27 [ cul Literature ceauzter dats CleSs5e Cuedih pee. tt. Gh fe as 186 C. tree fA L 146 Alot = Ge ’ 1 1205 q ty Kt the Stel ihe Rie: a) , Chie bis Fies re ft as Nene HE: Pg Kew rt q ie “bjt Feed He ith —" _ of ro bbe —— ee ! ‘. eed Rie fae — ———— —— am —o: ] ! E ot bie . CHLORIDE » PPE Figure 13. Seawater mixing curve: calcium vs. chioride. 16GE- S6E- 8Ge- fae e 4 § 6 oe C ggg G Q att l Stl 3h Zin 166 HT} The ratte Chie vo aude 4 | J 1] 1 ' 4h: a tt ‘ehael CHL ry E PPE Figure 14. Seawater mixing curve: alkalinity vs. chloride. 28 rr rr rr Sn ne cs Si Ser | re eee eee en rs i ee re | t ‘ I} | t t ' Lis i RE if FFE ; ¢ in eet f {his Fred — ¥ = _ 74 a v (f-22 | haa Seawater wixite Cuve — GL-{é wo Was FE-3i ve ape fo | t Hee Lec el ee ey feds ro ! — cre i | i | ! . ts Mite 4h bal: bist 16bi leek 14iti: Ir CHUCRIDE, PP Figure 15. Seawater mixing curve: magnesium vs. chloride. oride Content vs. t Robison (1972) showed a marked relationship between the altitude of the water column and the chioride content of the groundwater. The plot showed an inflection point close to sea level where chloride contents increased rapidly with depth (Fig. 4). This type of plot assumes that groundwater enters the wel] column uniformly between the water table and the bottom of the wel! CRobison, 1972). A similar plot from our data does not show nearly as clear a relationship (Fig. 16). 29 chit ¢ CTU. MEAN SEA LEVEL Lats {He \| {i ats ve f- a“ “1 € 4 fis 4 $ & “100 | ! ! | U ath fab eee Lest Lie 170k 2008 50h CHLORIDE, PPH Figure 16. Well altitude vs. chloride (data from this report). There are several] differences between the two plots: our data were plotted as elevations of wel] bottoms rather than as average water-column heights, and our data base is smaller than that of Robison €1972) and does not include as many wells in the O to +100 ft msl range. The different procedure used in measuring water-column altitude is not significant since this would only change the relative position of the curve’s Inflection point. The smaller size of our data set, however, may be responsible for the less 30 clear correlation between well altitude and chioride content, and the lack of connection between the wells with bottom depth below sea level! and the wells significantly above sea level. Based on our data alone, the correlation between chloride content and depth is not nearly as strong as Robison (1972) implies; there is a wide range of chloride values in groundwater from any given wel} altitude. The majority of wells in this study have bottom elevations between mean sea level and -100 feet. Regardless of the interpretation, high dissolved salt content should be expected with well altitudes close to, or below sea level. Trace Elements The trace elements analyzed were al! below recommended concentration limits as set by the federa) government, with the exception of an iron analysis from the Fairplain well field and manganese values from two private wells CAppendix). The rest of the trace meta) analyses show very low values and in many cases are below the detection limits of the analytical technique CAppendix>. No anomalously high values for barlum were recorded, despite values in excess of the EPA 31 ~~" regulations found by Geraughty and Miller (1983b) in the Barren Spot well field. It should be noted that the recommended concentration limits for iron and manganese were set to minimize problems associated with stains and precipitates in household use rather than problems associated with human health (Freeze and Cherry, 1979). Both iron and manganese are necessary minerals in human nutrition. The possibility exists that at least part of the high iron levels in the FP-6 well were due to the steel casing of the well. However, the private wells showing elevated manganese levels were cased with PVC. Chemi odelin Seven samples were modeled for speciation by the PHREEGE program (Parkhurst et al., 1980), showing a range of saturated phases in the well waters tested. The groundwater was calculated to be oversaturated with respect to quartz in al! samples modeled, and chalcedony was calculated to be oversaturated in all wells tested except well PE-3A (Table 1)». The groundwater was calculated to be supersaturated with respect to calcite in only three of the wel] waters modeled: BS-31, CO-52, and FP-6. Similarly, the 32 groundwater was calculated to be supersaturated with respect to dolomite in NB-6, CO-S2 and FP-6. It is interesting to note that in well NB-6, the groundwater is simultaneously oversaturated with respect to dolomite and undersaturated with respect to calcite. Other phases calculated to be supersaturated in the groundwaters modeled include gibbsite, kaolinite, barite, hematite, goethite and Fe(OH)3 (Table 1). 33 [ TABLE 1 CHEMICAL MODELING RESULTS WELL COLL. DATE SUPERSATURATED PHASES ewww eee eK Re ee eee ee = ee ee ee ae ae ae a a ww wwe we ee BS-31 11/25/85 CALCITE, BARITE, CHALCEDONY, QUARTZ CO-52 03/14/86 CALCITE, DOLOMITE, CHALCEDONY, QUARTZ, GIBBSITE, KAOLINITE FP-6 11/26/85 CALCITE, DOLOMITE, [ BARITE, CHALCEDONY, QUARTZ, HEMATITE, GOETHITE, Fe(OH) GL-148 03/20/86 CHALCEDONY, QUARTZ, GIBBSITE, KAOLINITE NB-6 12/07/85 DOLOMITE, CHALCEDONY, QUARTZ, HEMATITE, GOETHITE, FeCOH)3 PE-3A 03/18/86 QUARTZ, GIBBSITE, KAOLINITE SO-R1i 03/16/86 CHALCEDONY, QUARTZ [ 34 on DISCUSSION Water Qualit Most of the groundwaters sampled exceeded EPA limits for Cl and dissolved solids, and several} exceeded the EPA limits for sulfate. The Recommended Concentration Limits for these constituents, however, are primarily based on taste and aesthetic considerations, rather than direct problems with human health (Freeze and Cherry, 1979). Sodium, however ranged from 99 to 2445 mo/L in the samples collected. No EPA limits have been set for sodium presumably because of the wide concentration ranges of the element in public water supplies (M. Simms, Louisiana State University, pers. comm., 1986). However several] states have established advisory levels of 20 mg/L for persons with cardiac- and blood pressure-related diet restrictions (Geraughty and Miller, 1983b). The average concentration of sodium in waters sampled for this report was 578 mg/L, or more than 20 times the 20 mg/L advisory level mentioned above. For the sake of comparison, some breakfast cereals contain approximately 300 mg of sodium per serving, and an ounce of potato chips might contain around 200 mg. A 35 more detailed discussion of health-related problems can be found in Geraughty and Miller Inc. (1983a and b>. The distribution of the more highly mineralized groundwaters is shown in Figures 5 and 6; these zones are concentrated along the coast-lines of the island. The areas of high mineralization are found on the south coast industrial zone near Martin Marietta, in the river basin in the Salt River and Concordia areas, and in the estates close to the shoreline south of Fredericksted. For the most part, these areas correspond closely to the areas of high chloride and dissolved solids discussed by Geraughty and Miller €1983a). Both Jordan (1975) and Robison (1972) discuss a zone of highly mineralized groundwater extending inland parallel to the mountains of the northside range and in areas close to the carbonate highlands. These highly mineralized areas were not found in this study, nor were they found by Geraughty and Miller (1983a). Jordan (1975) mentions dissolved solids in excess of 20,000 mg/L in these areas and suggested that they were the result of highly saline connate waters from the Kingshil}] Limestone. Although large areas were not sampled in this study, it is difficult to believe that such areas would not have been found in either the 36 Geraughty and Miller (1983a) study or this one. Assuming that no analytical errors account for the high dissolved solid groundwaters, it is difficult to believe that mixing with highly saline formation waters would be restricted to discrete areas of the maris, and that those areas would be so short-lived. We feel a better explanation lies in seawater contamination due to over pumpage. i t undwate Analyses of St. Croix groundwater show a consistent excess in dissolved solids relative toa mixing curve of fresh water and seawater (Fig. 9). This excess of dissolved solids is produced by an increase in most of the major groundwater constituents relative to chloride (Figs. 8 - 15) except potassium. In particular, the ratio of sodium to chloride increases rapidly inland despite the fact that absolute quantities of both elements decrease in the same direction (Figs. 5, 6, 7). Potassium is the only major constituent that falls below the seawater mixing curve when plotted against both chloride and sodium. 37 Several explanations of the chemistry of St. Croix -groundwater have been suggested: 1> Concentration of rainwater through evapo- transpiration. 2) Precipitation and re-solution of salts in the soi] zone. The first two explanations, concentration of rainwater and the precipitation/dissolution of salts in the soil zone are somewhat related, and difficult to distinguish chemically. Dissolved constituents of seawater are taken into the atmosphere as aerosols or salt crystals, and are directly deposited in the soils or accompany rainfall. The dissolved constituents are concentrated by evaporation and transpiration, leaving the remaining groundwater more concentrated with respect to the minerals in solution, or alternately, by precipitating salts which are then re-dissolved by the next influx of rainfall. The net result of these processes is an increase in the dissolved constituents in shallow groundwater relative to rainwater. Both Jordan €1975) and Robison (1972) cite these processes as 38 [ [ [ [ [ ; i [ [ [ f [ [ r i rr er as | a eee re rn ae ~ > “yj "3 OQ TY | being important sources of the mineralization of shallow groundwater and estimate that the rainwater has been concentrated 30 times to contain the amount of chloride found in shallow groundwater. Although this process probably does occur on a highly evaporative island such as St. Croix, it does not explain the change in ionic ratios relative to seawater that are found in St. Croix groundwater. 3) Solution of aquifer materials by rain and groundwater. Dissolution of aquifer minerals is a process by which groundwater reacts with the minerals of the aquifer, dissolving the rock and altering the composition of both the groundwater and the rock. In the case of carbonate aquifers, much dissolution of the rock material can occur because of the undersaturated nature of the rainwater ana the rapid dissolution kinetics of carbonate minerals. In a similar manner, groundwater oversaturated with respect to mineral phases can further alter its chemistry by the precipitation of minerals along its flow path. 39 — oJ 3 39 7 F 3° 9 7S SO SO SS r 3 4) Cation exchange. In cation exchange, clay minerals will take up cations from solution in exchange for bound cations already occupying exchange sites (Drever, 1982>. This is an equilibrium process and is affected by the concentration of ions in solution, including the pH, the radius of hydration of cations in solution, and the types of minerals in the aquifer. By this process, a groundwater could change the relative quantities of ions in solution by reaction with aquifer materials. 5) Ionic diffusion. Ionic diffusion is the process of migration of ions through a fluid in response to a chemical concentration gradient. This process does not include mass transport of materials by processes such as convection and water flow. 6) Mixing with "connate" waters. The term "connate water" refers to the water trapped in the aquifer during deposition of the sediments. As such, the water would reflect the chemistry of the seawater at the time the rocks were deposited. In actual usage, the term is 40 confusing and somewhat ambiguous, since it Is difficult to distinguish waters that represent original seawater from those that have been heavily altered through interaction with rock material (Drever, 1982). We would prefer the use of the term "formation water". By mixing relatively fresh groundwater with highly mineralized formation water, the resulting water would be intermediate in composition. 7> Mixing with seawater. The process of mixing seawater with fresher ground waters is an obvious one on smal} oceanic islands such as St. Croix, and would result ina groundwater of intermediate composition. If this process takes place with no competing processes, the ionic ratios of the groundwater should fal! on the mixing curve between the two end-member waters. Regarding the most likely mechanisms for the formation of St. Croix groundwater, we offer the following observations: 1> The chemistry of the groundwaters is complex, and is not strictly the result of simple mixing of fresh water and seawater. 41 “4a 73 7959 TS TT 7 FI 73 TF FI TSO Sa “3 "] 7S 2) The chloride content and dissolved solids content increase rapidly toward the coast, indicating that Interaction between groundwater and seawater is important. 3) Despite the increase in chloride content toward the coast, the increase in Na/C] ratios away from the coast indicates that the groundwater gains significant quantities of dissolved material from the aquifer. We suggest that the chemistry of the groundwater is best explained by a combination of several processes, of which reaction with the aquifer and mixing with seawater are the most important on a regional scale. With the present information we cannot eliminate any single process, but we feel that there is adequate information to indicate the relative importance of several. The concentration of rainwater through evapo- transpiration and the precipitation and re-dissolution of salts derived from sea spray are discussed in detail in Robison (1972), Jordan (1975) and in Black, Crow and Eidsness (1976). We feel that this is a less important process in the production of the bulk chemistry of the groundwater than the reports listed above imply. 42 re nar | er er “3a 79 “4S 7°59 “SO 77 73 7F 735 7S a re The concentration of rainwater to the extent mentioned in Robison (1972) could explain an increase in chloride, but would not alter the ratio between sodium and chloride. Sodium, potassium and chloride salts are extremely soluble, and do not precipitate until salinities well in excess of seawater are reached. Similarly, the salts formed through the evaporation of sea spray would re-dissolve with added rainfall, and would not alter the sodium/chloride or potassium’chloride ratios. The concentration of rainwater to the extent suggested by Robison (1972) produces a groundwater with a chloride concentration of 210 mg/L. Less than 12 percent of the samples collected for this report had chloride contents that could be explained by concentration only to this extent. Reaction with the minerals in the Kingsh!1} Limestone aquifer is an important source of dissolved solids in St. Croix groundwater, and is understated in Cederstrom (1950), Robison (1972) and Jordan (1975), among others. Mineral reactions in this case may include cation exchange, mineral alteration and minera) dissolution. These processes could Include the replacement of sodium in exchange sites on clays by potassium and calcium; mineral transitions such as 43 smectite to illite, or albite to smectite; and the dissolution of volcanoclastics and carbonates. All] the components mentioned above exist in significant if not abundant quantities in the Kingshil]] Limestone. Gerhard et al. (1978) measured insoluble residues in Kingshill Limestone strata ranging from less than 5% to over 90%. Common non-carbonate constituents in the Kingshil) Limestone include feldspars, clay minerals, hornblende, quartz and lithic clasts. It is not possible at this point to prove the existence of specific reactions in the aquifer/groundwater system. However, the previously mentioned processes provide the best explanation for the loss of potassium and the uptake of sodium, silica, calcium and bicarbonate by the groundwater, as well] as provide the best explanation for the increase in the sodium to chloride ratio away from the shoreline. This latter point indicates a change in dominance of diagenetically controlled groundwater inland as opposed to the seawater-controlled groundwater chemistry close to the shoreline. The increase in chloride, sodium, and salinity values close to the coast Indicates an increase in the contro) of seawater chemistry on the groundwater. Similarly, the ratio of sodium to chloride approaches 44 ee | [ [ [ [ that of seawater in wells close to the coastline. We interpret this to mean that mixing with seawater progressively overprints diagenetic effects in the more highly mineralized coastal! wells. The effect of lonic diffusion in this process is difficult to assess due to lack of information on the flow rates of St. Croix groundwater and the difficulty of securing reliable information of salinity changes with depth. However, because the St. Croix carbonate aquifers are layered, presumably restricting vertical flow (Black, Crow and Eidsness, 1976; Gill and Hubbard, 1986), and the Jealousy Formation clays presumably restrict seawater incursion from below, the effects of vertical ionic diffusion are probably minimal. The presence of "connate" fluids can neither be proven nor disproven at this point. However, since the Jealousy formation clays are over 1400 feet thick (Cederstrom, 1950), and presumably overlie noncompressable Cretaceous basement rock, the suggestion of formation waters derived from compacting clays Robison, 1972) is not without merit. At this point, our data give no suggestion of anomalous zones of higher salinities caused by mixing with highly altered formation waters. In addition, the layered nature of the aquifer would presumably restrict the 45 upward migration of Jealousy Formation fluids in the same manner that they would restrict vertical seawater intrusion. odelin Minerals calculated to be oversaturated with respect to St. Croix groundwaters at various locations include calcite, dolomite, barite, quartz, chalcedony, kaolinite, hematite, goethite, FeCOH)3 and gibbsite. This assemblage of minerals is further evidence of extensive interaction of silicate minerals in the aquifer system. The preponderance of quartz and chalcedony indicate the possibility of silica precipitation in aquifer rocks. Oversaturation with respect to aluminosilicates such as kaolinite and gibbsite in three of the seven samples modeled corresponds with wells in two areas of high siliciclastic content (CO-52, GL-148) and one we}] in an area where the mineralogy is less known ¢PE-3A). Authigenic clay minerals are found in the Kingshil} Limestone (Gill and Hubbard, 1986). The geochemical] modeling calculations indicate that St. Croix groundwater is capable of producing certain clay minerals under the present chemical conditions. 46 An interesting observation is that only three of the seven samples modeled were calculated to be oversaturated with respect to calcite. In this instance, St. Croix groundwaters are apparently reaching saturation with sillcate phases before saturation is reached with carbonates. In part, this 1s no doubt due to the high solubilities of the carbonates and the very low solubilities of quartz and its polymorphs in neutral to slightly acidic waters. However, the samples of water saturated with respect to calcite all lie close to the coast in areas of higher Salinity. The saturation with respect to calcite is probably explained by mixing with seawater. The sample modeled from the Negro Bay well] field was saturated with respect to dolomite, but undersaturated with respect to calcite, conditions considered to be ideal for the formation of dolomitic rock. Dolomite is present in the aquifer system within three kilometers of the sampled wel! (Gil) and Hubbard, 1986), at roughly the same distance from the coast. More modeling Is necessary to determine whether mixing-zone dolomitization is viable on a regional scale on St. Croix today. Mixing-zone dolomitization is one of several explanations for the formation of dolomite in the carbonate strata of St. Croix. 47 ~y The samples calculated to be saturated with respect to iron-bearing phases such as hematite and goethite both came from public wells equipped with steel casing (Wells NB-6 and FP-6). We feel that contamination from the steel casing is possible in this instance. Summary The major sources of groundwater mineralization on St. Croix are seawater mixing and diagenetic interactions in the aquifer. In this case, diagenetic interactions include cation exchange, mineral transformation and dissolution of aquifer minerals. Concentration of rainwater through evapo-transpiration has been cited as an important source of groundwater mineralization by several authors. However, the levels of mineralization produced by this process could also be explained by incorporation of aerosols and salt spray into the groundwater, and in any case do not explain the generally higher levels of dissolved constituents found in St. Croix groundwaters nor the relative proportions of those constituents. Chemical modeling of the groundwaters shows a Gominance of saturated silicate phases rather than carbonate phases in the groundwater. The majority of 48 wells sampled yield groundwater still] undersaturated with respect to calcite. Those wells oversaturated with respect to calcite lie near the coastline, suggesting the influence of seawater mixing rather than dissolution of aquifer carbonates. One wel! sample showed simultaneous undersaturation with calcite and supersaturation with respect to dolomite. Simultaneous supersatuation of dolomite and undersaturation of calcite has been invoked as an important prerequisite of mixed-water dolomitization. More samples need to be modeled to see whether these conditions are regionally extensive. Almost all] the well water samples exceeded EPA recommended limits for chloride and dissolved solids. Several of those analyzed for sulfate also exceeded EPA limits. Since the limits for chloride, dissolved solids and sulfate are set primarily on the basis of water taste and esthetics, whether these levels are objectionable depends on intended water use. The levels of sodium in the groundwater were very high, with groundwater sodium evidently being produced by water-rock reactions in levels above those expected from strict mixing processes. The sodium levels are a point of potential concern from a health standpoint. 49 +4 a ee re | REFERENCES CITED American Public Health Association, 1971, Standard Methods for the Examination of Water and Wastewater, 13th ed. Washington, D.C., 874 pp. Black, Crow and Eidsness, Inc., 1976, A water management plan for St. Croix, U. S. Virgin Islands, Gainesville, Florida: Black, Crow and Eidsness, Inc. Buros, O. K., 1976, Wastewater reclamation Project, St. Croix, U. S. Virgin Islands, Report No. EPA-600/2-76-134, Cincinnati: Environmental Protection Agency, 243 pp. Cederstrom, D. J., 1950, Geology and groundwater resources of St. Croix, U. S. Virgin Islands: U. S. Geological Survey Water Supply Paper 1067, 117 pp. Claasen, H. C., 1982, Guidelines and techniques to obtain valid groundwater quality samples: U. S. Geological Survey Open File Report 82-1024, 54 pp. Drever, J. I., 1982, The Geochemistry of Natural Waters. Englewood Cliffs, N. J.: Prentice Hall, Inc., 388 pp. Freeze, R. A. and Cherry, J. A., 1979, Groundwater. Englewood Cliffs, New Jersey: Prentice Hall, Inc., 604 pp. Friedman, L. C. and Erdmann, D. E., 1982, Quality Assurance Practices for the Chemical and Biological Analyses of Water and Fluvial Sediments, Techniques of Water Resources Investigations, Bk. 5, Ch. A6. Washington, D. C.: U. S. Geological Survey, 181 pp. Garcia, R. and Canoy, M., 1984, Reconnaissance of ground-water quality in the U.S. Virgin Islands, July, 1984: Geological Survey Open File Data Report 84-807. 50 Geraughty and Miller, Inc, 1983a, Report on current groundwater conditions in the U. S. Virgin Islands. Syosset, New York: Geraughty and Miller Inc., 89 pp. lanai » 1983b, Groundwater management plan for the U. S. Virgin Islands. Syosset, New York: Geraughty and Miller, Inc., 86 pp. Gerhard, L. C., Frost, S. H., and Curth, P. J., 1978, Stratigraphy and depositional setting, Kingshil] Limestone, Miocene, St. Croix, U. S. Virgin Islands: Amer. Assoc. Petrol. Geol. Bull., v. 62, no. 3, p. 403-418. Gieskes, J. M. and Rogers, W. C., 1973, Alkalinity determination in interstitial waters of marine sediments: Jour. of Sed. Petrol., v. 43, no. 1, p. 272-277. Gill, I. P. and Hubbard, D. K., 1985, Subsurface sedimentology of the Miocene-Pliocene Kingshil} Limestone, St. Croix, U.S.V.I., in P. D. Crevello and P. M. Harris, eds., Deep Water Carbonates: Buildups, Turbidites, Debris Flows and Chalks, Tulsa, OK: Soc. Econ. Paleon. Mineral. Core Workshop No. 6, p. 431-460. Gill, I. P. and Hubbard, D. K., 1986, Subsurface geology of the St. Croix carbonate rock system: Technical Report No. 26, Water Resources Research Center, College of the Virgin Islands, St. Thomas, U. S. Virgin Islands, 87 pp. Hardie, L. A. and Eugster, H. P., 1970, The evolution of closed-basin brines: Mineral. Soc. Am. Spec. Pub]. No. 3, p. 273-290. Hendrickson, G. E., 1963, Ground water for public supply in St. Croix, Virgin Islands: U. S. Geological Survey Water-Supply Paper 1663-D, 27 pp. Jordan, D. G., 1975, A survey of the water resources of St. Croix, Virgin Islands: U.S. Geological Survey Open-File Report, Caribbean District, San Juan, S51 pp. 51 Multer, H. G., Frost, S. H. and Gerhard, L. C., 1977, Miocene "Kingshill Seaway" - a dynamic carbonate basin and shelf model, St. Croix, U. S. Virgin Islands: jn Frost, S. H., Weiss, M. P. and Saunders, J. B. Ceds.), Reefs and Related - Carbonates--Ecology and Sedimentology: Amer. Assoc. Petrol. Geol. Studies in Geology No. 4, p. 329-352. Parkhurst, D. L., Thorstenson, D. C. and Plummer, L. N., 1980, PHREEGE - a computer program for geochemical calculations: U S Geological Survey Water Resources Investigations, Report WRI-80-96. 210 pp. Robison, T. M., 1972, Ground water in central] St. Croix, U. S. Virgin Islands: U. S. Geol. Survey Open-File Report, Caribbean District, 18 pp. Skougstad, M. W., Fishman, M. J., Friedman, L. C., Erdmann, D. E. and Duncan, S. S., eds., 1979, Methods for Determination of Inorganic Substances in Water and Fluvial Sediments, Techniques of Water Resources Investigations, Bk. 5, Ch. Al. Washington, D. C.: U. S. Geological! Survey, 626 pp. Whetten, J. T., 1966, Geology of St. Croix, U. S. Virgin Islands: Geol. Soc. Am. Mem. 98, p. 177-239. Wood, W. W., 1976, Guidelines for collection and field analysis of groundwater samples for selected instable constituents: Geologic Survey Techniques of Water Resources Investigations, Report No. TWl 01-D2, 24 pp. 52 APPENDIX RESULTS OF CHEMICAL ANALYSES e o data ta u uent EPA M. P. C.: Maximum Permissible Concentration EPA R. C. L.: Recommended Concentration Limit -- : Data not available <DL : Below detection limit >RNG : Over range of calibration 53 APPENDIX-CHEMICAL ANALYSES WELL NAME EPA M. P. C.#* R. C. L.#* Detection Limit Rainwater RU-149 Seawater Seawater-avg. O-R1i SO-R2 WD-94 WH-5S9 COLL. DATE Nov 25 85 May 19 85 May 19 85 Apr 4 84 Mar 8 86 Mar 12 86 Mar 21 86 Apr 4 84 Mar 14 86 Mar 12 86 Nov 27 85 Nov 27 85 Apr 4 84 May 15 85 May 19 85 Nov 26 85 Nov 27 85 Apr 4 84 Dec 9 85 Mar 20 86 Mar 14 86 Mar 10 86 Apr 4 84 Mar 10 66 Nov 29 85 Dec 7 85 Mar 18 86 Mar 13 86 Mar 18 86 Mar 86 Mar 16 86 Mar 16 86 Mar 7 86 Mar 12 86 LOCATION Barren Spot wel) Barren Spot wel} Barren Spot wel] Barren Spot wel} Carlton, Plot 10 Carlton, Plot 15 Castle Coakley, Plot Concordia wellitield, Concordia, Plot 52 College of the VI Fairplain wellfield, Fairplain wellfield, Fairplain welifield, Fairplain wel] Fairplain Fairplain Fairplain Golden Grove wel | Golden Grove wel! Glynn, Plot 148 Glynn, Plot 246C Hannahs Rest, Plot 100 La Grange wellfield, #1 Mars Hill, Plot 35 Negro Bay wellfield, #3 Negro Bay wellfield, #6 Pearl, Plot 3A... East End, St. Croix Ruby, Plot 149 Tague Bay, St. Croix Orheneh eh Solitude Remainder, Wel! Solitude Remainder, Wel] Williams Delight, Plot 94 Whim, Plot 5 * Maximum Permissible Concentration, US EPA *% Recommended Concentration Limit, US EPA #PW1 #PW1 1 2 A ep) ° < ct Satomayor Govt Mal loy Govt Govt Govt Govt Govt Govt Govt Govt Govt Williams - Williams . George qQn<rwu<c<c<c< tte ew rs] © N @ J joy rr) <e _ Dat ed eoed med mt Bnd Pd Domed et Fe id Laundromat ovt Govt Maneilly Smith fo mm mt . Roebuck - Roebuck McLean Jackson Jr. WomMwA! INI wz<<cag<oqor<<<<<<c<<c< SOURCE i ee ee ee ee ee ee: ee ee: ne ee APPENDIX-CHEMICAL ANALYSES Camb. temp) (conduct .) as CaCO3 Cumho/cm WELL NAME COLL. DATE TEMP (C) pH SPEC COND SALING TY ‘sum 6 TBS HARDNESS wee eee eee —_ee wm nme - em we mmm wee RP ee wwe eee Be ee eww ee ee ew ee ee wwe wm wm wee ee eee eee EPA M. P C.¥% -—— — -— -— -— -— EPA R. C. L He -— —— — —— — -<- -— Detection Limit —— ~-— BS-31 Nov 29 85 26.7 13 4110 2200 2100 2492 266 BS-31 May 19 19 8S 85 27.4 27.5 2? 05 3330 4150 1900 -— BS-3A BS-8 May 84 -— —— -— —-—— — CA-109 Apr Mar 86 29.1 95 822 290 265 CA-15 Mac 12 86 27.3 20 1327 1270 -- 175 CC-23 Mar 21 86 27.3 80 9500 2900 — 679 O-i Apr 84 -—— -— -— ~~ CO-52 Mar 14 86 32.3 91 1558 650 B56 367 CVI Mar 12 86 29.0 97 2110 1000 FP-4 Nov 27 85 28.0 01 4610 2200 2547 763 421 FP-5 Nov 27 85 29.0 — 96 -— 3550 — 1800 — 2027 FP-6 FP-6 Apr 15 85 84 28.2 87 4690 -— a -— FP-6 May ay 19 85 27.3 88 4650 29500 FP-6 Nov 26 85 28.1 96 5100 2750 866 FP-8 Nov 2? 85 27.4 06 3280 -—— 407 GG-PWi Apr 84 -—— -_— -—— o- -<- GG-PW1 lec 85 27.3 67 1903 800 175 GL-148 Mac 20 86 27.4 20 1805 800 138 GL-246C Mar 14 86 27.4 72 2102 2700 1000 12it 531 150 HR-100 Mar 10 86 27.5 — 38 —— 5100 — es ~~ Mar Apr 10 86 84 30.8 32 12750 6500 600 NB-3 MH-35 Nov 29 85 27.3 2203 1000 1376 285 B-6 Dec 85 27.1 99 2103 967 266 PE-3A Mar 18 86 30.4 12 3075 1500 1288 150 Mac 13 86 — 13 -— sd ~— Rainwater ar 18 86 26.8 68 2107 1033 -—— -— —— 3? Seawater Mar 86 -— — — — -— -— 6657 — 15 35016 6322 Seawater-avg. Mar 16 86 28.4 76 2655 1250 1504 564 426 SO-R2 Mar 16 86 28.2 71 2300 3420 1100 1700 162 WD-94 Mar 86 86 28.2 29 09 16 2997 1450 309 WH-5S9 Mar 12 * Maximum Permissible Concentration, US EPA %* Recommended Concentration Limit, US EP. APPENDIX-CHEMICAL ANALYSES MAJORS --- in order of decreasing concentration In seawater WELL NAME COLL. DATE wwmm mmm mmm es em mmm meme nw nm Hem em mw ewe ee Cl (mg/l) Na (ppm) S04 (mg/1) Mg (ppm) Ca (ppm) K (ppm) -— -— —e — EPA M. P C.# -—< 250 250 EPA R. C. L.## Detection Limit — 25 002 002 1.0 BS-31 Nov 29 85 894 896 272 23.9 66.7 6.7 BS-31 May 19 85 853 825 266 5.6 BS-3A May 19 85 654 739 208 4.0 pr 84 846 860 268 BR f09 Mar 86 <1S0 121 37 30 .6 55.6 CA-15 Mar 12 86 <150 235 61 15.1 <DL CC-23 Mar 21 86 1263 937 c6. 193°6 Apr ac 14 86 84 507 199 201 401 155 39.0 CVI CO-52 Mar 12 86 162 522 3.7 82.7 FP-4 Nov 2? 85 996 683 248 94.9 149.5 FP-5 Nov 2? 85 714 598 207 50.5 85.5 FP-6 Apr 84 1173 703 180 FP-6 May iS 85 1146 689 294 FP-6 May 19 85 1157 697 228 -— FP-6 Ov 26 85 1402 744 208 112.1 FP-8 Nov 27 85 604 21 GG-PW1 Apr 84 248 269 304 302 102 99 >RNG 69.6 <DL GG-PW1 GL-148 Mac Dec 20 86 85 199 367 17.1 27.0 GL-246C Mar 14 86 42 efi 50 52.6 126.1 2.1 HR-100 Mar 10 86 1052 1165 24.2 20.2 13.6 LG-1 Apr 84 99 -— <DL MH-35 Mar 10 86 2445 -—— 104 68.6 51.0 NB- Nov 29 85 395 403 103 40 47.7 NB-6 Dec 85 319 413 78 36 45.9 PE-3A Mar i8 86 575 -— 610 16 5.8 Rainwater Mar 13 86 <DL <DL ar 18 86 282 486 1370 414. 0 402.0 3.0 Seawater Mar — 86 19350 10760 12029 2710 1290 399.0 Seawater-avg. Marc 16 86 447 119 48 90. 8 6.4 SO-R2 Mar 16 86 669 960 -=— 69 110. 9 12.1 WD-94 Mac 406 299 154 14 40. 2 1.3 WH-5S9 Mac 12 86 $75 510 37 62. 2.6 * Maximum Permissible Concentration, US EPA *% Recommended Concentration Limit, US EP. APPENDIX-CHEMICAL ANALYSES Alkalinity WELL NAME COLL. DATE 4HCO3 (ppm) CO3 (ppm) EPA M. P. C.% 3 -- -- -- EPA R. C. Lie -- -- -- Detection Limit -- -- -- BS-31 Nov 25 85 617 -- BS-31 May 19 85 626 -- BS-3A May 19 85 693 -- BS-8 Apr 4 84 -- -- CA-109 Mar 8 86 532 -- CA-15 Mar 12 86 534 -- CC-23 Mar 21 86 567 -- CO-1 Apr 4 84 -- -- CO-52 Mar 14 86 551 -- CVI Mar 12 86 866 -- FP-4 Nov 27 85 690 -- FP-5 Nov 27 85 676 -- FP-6 Apr 4 84 -- -- FP-6 May 15 85 636 -- FP-6 May 19 85 -- -- FP-6 Nov 26 85 609 -- FP-8 Nov 27 85 828 -- GG-PW1 Apr 4 84 -- -- GG-PW1 Dec 9 85 690 -- GL-148 Mar 20 86 688 -- GL-246C Mar 14 86 504 -- HR-100 Mar 10 86 949 -- LG~1 Apr 4 84 -- -- MH-35 Mar 10 86 -- -- NB-3 Nov 29 85 693 -- NB-6 Dec 7 85 710 -- PE-3A Mar 18 86 582 -- Rainwater Mar 13 86 <DL -- RU-149 Mar 18 86 595 -- Seawater Mar 86 -- -- Seawater-avg. -- 142 18 SO-R1 Mar 16 86 586 -- SO-R2 Mar 16 86 535 -- WD-94 Mar 7 86 841 -- WH-59 Mar 12 86 528 -- * Maximum Permissible Concentration, US EPA *% Recommended Concentration Limit, US EPA Sr (ppm) ae pe ee pe ee — WN SI (ppm) S102 Cppm) 14 12.9 >RNG ios Nw w:° MNQAN 1 OW! et pe —=[— eee yy WOT 14 © OOOW \ Wn QACOpsA! 1 —NQae Oo ! .096 27.6 >RNG 36.4 34.4 APPENDIX-CHEMICAL ANALYSES MINORS -- In alphabetical order by chemical symbol] WELL NAME COLL. DATE Al (ppm) Ba (ppm) Cd (ppm) Co (ppm) Cr (ppm) Cu (ppm) EPA M. P. C.% -- -- 1.00 01 -- .05 -- EPA R. C. L,¥# -- -- -- -- -- -- 1 Detection Limit -- ol 001 -O1 .03 .025 01 BS-31 Nov 25 85 <DL .048 <DL <DL <DL <DL BS-31 May 19 85 -- -- -- -- -- -- BS-3A May 19 85 -- -- -- -- -- -- BS-8 Apr 4 84 -- -- -- -- -- -- CA-109 Mar 8 86 <DL .035 <DL <DL <DL <DL CA-15 Mar 12 86 <DL <DL <DL <DL <DL <DL CC-23 Mar 21 86 <DL .036 <DL <DL <DL <DL CO-1 Apr 4 84 -- -- -- -- -- -- CO-52 Mar 14 86 <DL 042 <DL <DL <DL <DL CVI Mar 12 86 <DL <DL <DL <DL <DL <DL FP-4 Nov 27 85 <DL 067 <DL <DL <DL <DL FP-5 Nov 27 85 <DL 050 <DL <DL <DL <DL FP-6 Apr 4 84 -- -- -- -- -- -- FP-6 May 15 85 -- -- -- -- -- -- FP-6 May 19 85 -- -- -- -- -- -- FP-6 Nov 26 8S <DL 064 <DL <DL <DL <DL FP-8 Nov 27 85 <DL 042 <DL <DL <DL <DL GG-PW1i Apr 4 84 -- -- -- -- -- -- 5 GG-PW1 Dec 9 85 <DL 044 <DL <DL <DL <DL GL-148 Mar 20 86 <DL 040 <DL <DL <DL <DL GL-246C Mar 14 86 <DL 038 <DL <DL <DL <DL HR-100 ac 10 86 <DL 033 <DL <DL <DL <DL LG-1 Apr 4 84 -- -- -- -- -- -- MH-35 Mar 10 86 038 <DL <DL <DL 064 NB-3 Nov 29 85 <DL 041 <DL <DL <DL <DL -6 Dec 7 85 <DL 037 <DL <DL <DL <DL PE-3A Mar 18 86 <DL 040 <DL <DL <DL <DL Rainwater Mar 13 86 <DL <DL <DL <DL <DL <DL RU-149 Mar 18 86 <DL .031 <DL <DL <DL <DL Seawater Mar 86 -- <DL <DL <DL <DL <DL Seawater-avg. -- .002 002 0.05 .00005 -0003 .0005 SO-R1 Mar 16 86 <DL O61 <DL <DL <DL <DL SO-R2 Mar 16 86 <DL 060 <DL <DL <DL <DL WD-94 Mar 7 86 <DL 040 <DL <DL <DL <DL WH-59 Mar 12 86 <DL 036 <DL <DL <DL <DL % Maximum Permissible Concentration, US EPA ** Recommended Concentration Limit, US EPA APPENDIX-CHEMICAL ANALYSES (Cat/An) WELL NAME COLL. DATE Fe (ppm) Mn (ppm) Ni (ppm) V (ppm) 2n (ppm) CHG.BAL. EPA M. P. C.*% 3 -- -- -- -- -- -- -- EPA R. C. L.#* = -- 3 .05 -- -- 5 -- Detection Limit -- 015 -0025 05 025 005 -- BS-31 Nov 25 85 <DL 004 <DL <DL <DL 1.08 BS-31 May 19 85 <DL <DL -- -- -- -- BS-3A May 19 85 <DL <DL -- -- -- -- BS-8 Apr 4 84 <DL <DL -- -- -- -- CA-109 Mar 8 86 -- -- <DL <DL <DL -- CA-15 Mar 12 86 <DL <DL <DL <DL <DL -- CC-23 Mar 21 86 <DL 057 <DL <DL .009 -- cO-1 Apr 4 84 <DL <DL -- -- -- -- CO-52 Mar 14 86 -- -- <DL <DL <DL 1.01 CVI Mar 12 86 <DL <DL <DL -358 <DL -- FP-4 Nov 27 85 -~ -- <DL <DL -013 1.00 FP-5 Nov 27 85 -- -- <DL .028 <DL 97 FP-6 Apr 4 84 .030 <DL -- -- -- -- FP-6 May 15 85 .040 <DL -- -- <DL -- FP-6 May 19 85 <DL <DL -- -- <DL -- FP-6 Nov 26 85 2.216 061 <DL <DL <DL 92 FP-8 Nov 27 85 <DL .010 <DL 035 <DL -- GG-PWi Apr 4 84 <DL <DL -- -- => -- GG-PW1 Dec 9 85 -- -- <DL <DL <DL 81 GL-148 Mar 20 86 <DL <DL <DL <DL <DL -- GL-246C Mar 14 86 <DL <DL <DL <DL <DL 95 HR-100 Mar 10 86 <DL 002 <DL 070 025 -- a | Apr 4 84 <DL <DL -- -- -- -- MH-35 Mar 10 86 <DL 016 <DL <DL <DL -- NB-3 Nov 29 85 <DL <DL <DL 033 007 93 B-6 Dec 7 85 -154 008 <DL <DL 051 1.04 PE-3A Mar 18 86 <DL <DL <DL <DL <DL -- Rainwater Mar 13 86 -- -- <DL .028 073 -- RU-149 Mar 18 86 <DL <DL <DL .040 <DL -- Seawater Mar 86 <DL <DL <DL <DL <DL -- Seawater-avg. -- .002 0002 0005 -- 002 1.00 SO-R1 Mar 16 86 <DL <DL <DL <DL <DL 1.09 SO-R2 Mar 16 86 <DL <DL <DL <DL 451 -- WD-94 Mac 7 86 <DL <DL <DL <DL 007 1.03 WH-59 Mar 12 86 027 009 <DL <DL <DL -- * Maximum Permissible Concentration, US EPA ##* Recommended Concentration Limit, US EPA