This is to certify that the thesis entitled A Geochemical and Hydrological Investigation of a Modern Coastal Marine Sabkha presented by Robert Gudramovics has been accepted towards fulfillment of the requirements for Masters degree in Geology M Major professor Date June 4; 1981 0-7 639 AK\\ #"‘{\\‘ i 'J! u ‘ . “Mi/w “a“ 8 2‘ 04 Jilllfléfi 13% .5 ' Alli—4 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place 1n book return to remove charge from circulation records A GEOCHEMICAL AND HYDROLOGICAL INVESTIGATION OF A MODERN COASTAL MARINE SABKHA by Robert Gudramovics A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geology 1981 C;(’;7zz2:/ ABSTRACT A GEOCHEMICAL AND HYDROLOGICAL INVESTIGATION OF A MODERN COASTAL MARINE SABKHA by Robert Gudramovics The purpose of this study is to integrate the geochemistry and hydrology of a modern marine coastal sabkha and by doing so determine the origin and evolution of the subsurface brines. The study area is the Laguna Madre Tidal Flats, located approximately 90 km south of Corpus Christi, Texas. i The conclusions of this study are: l) the major subsurface brine is a Na-Cl solution that is marine in origin and depleted in calcium with respect to sea water; 2) a second water mass, continental in origin, has been observed at depth toward the continental side of the tidal flats; 3) input of water to the sabkha is by the process of Flood Recharge during which the Laguna Madre waters are driven across the surface of the sabkha by high winds; 4) recharge area is selective and is a function of wind direction, wind velocity, and topography; 5) both water masses evolved chemically by the process of Evaporative Concentration with some subsequent precipitation of calcium carbonate and calcium sulfate minerals. ACKNOWLEDGEMENTS I would like to thank my committee members Dr. David Long, Dr. Graham Larson, and Dr. John Wilband who were helpful in the development and completion of this thesis. I would like to especially thank Dave who, in addition to being a very helpful advisor, became a needed friend. We endured broken vehicles, scorpions, black widows, hurricane Allen and the temperature extremes of the study area, and I endured his fiddle playing; from all of this we learned 22325 to say "it can't get worse. I would like to thank the United States Geological Survey in Corpus Christi for providing me with a research assistantship for this study. I would also like to thank Steve, Mark, Tom and all the other people from the "Chateau" who provided friendship to the "loud" New Yorker. I would especially thank my sister Rita and her husband Steve who provided a place in which to escape from all of this madness. ii TABLE OF CONTENTS page LIST OF TABLES v LIST OF FIGURES vi INTRODUCTION 1 LAGUNA MADRE TIDAL FLATS 7 GEOLOGIC HISTORY 7 REGIONAL SETTING 9 SEDIMENT TYPE AND SOURCE REGIONS 12 CLIMATE 15 FIELD OBSERVATIONS 17 FIELD METHODS 20 ESTABLISHMENT OF WELL SITES 20 SAMPLING AND FIELD ANALYSES 23 GEOCHEMISTRY/HYDROLOGY 25 LABORATORY METHODS 29 GEOCHEMISTRY 33 EVAPORATIVE CONCENTRATION 33 SOLUTE FRACTIONATION 36 BRINE MIXING 38 OBSERVATIONS 40 DISCUSSION 47 HYDROLOGY 54 SEEPAGE REFLUX 55 iii CAPILLARY ACTION EVAPORATIVE PUMPING FLOOD RECHARGE OBSERVATIONS DISCUSSION BASIN AREA COMPUTER MODELING SUMMARY AND CONCLUSIONS BIBLIOGRAPHY APPENDICES SULFATE ANALYSIS BROMIDE ANALYSIS DENSITY MEASUREMENTS ANALYSES HYDROLOGY iv page 56 56 58 6O 66 69 73 83 85 91 91 93 96 103 106 Table l. 2. 3. 4. 5. 6. LIST OF TABLES Precision of Geochemical Analyses. Molar Ratios Key to letters used in Figure 8 Selected chemistries for the basin brines as a function of distance from gas well access channel. Mineral Saturation States Laboratory Density Measurements page 32 49 51 70 78 102 Figure 9. 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. LIST OF FIGURES Map of General Region Map of Sedimentary Provinces Map of Study Area Concentration Trends of Ions in Sea Water August Data August Data March Data March Data Trilinear plot of the Sample Waters Concentration Trends of Samples Reflux and Evaporative Pumping Models August Piezometric Potentials March Piezometric Potentials Basin Area August Piezometric Potentials Bromide Contour Maps Bromide vs. Transmittance Density vs. Temperature Density vs. Salinity Density vs. Temperature vs. Salinity vi Page 10 13 21 35 42 43 44 45 52 53 57 63 64 65 68 95 99 100 101 INTRODUCTION Geochemical and hydrological processes in ancient sabkha systems have been proposed for the generation of: a) certain evaporitic minerals (Kinsman, 1969), b) various stratiform metalliferous deposits (Renfro, 1974), and c) low temperature uranium deposits (Rawson, 1976). A better understanding of these processes and their controls will advance our knowledge about these systems and also aid us in the exploration for these ancient mineral deposits. This study will test two hypothesis: a) that evaporative pumping (Hsu and Siegenthaler, 1969) is the major hydrological mechanism, and b) that evaporative concentration (Carpenter, 1978) is the dominant geochemical process operating in an active coastal sabkha. The study area is the Wind Tidal Flat area of Laguna Madre, which lies approximately 90 km south of Corpus Christi, Texas (Fig. 1). Coastal sabkhas are flat (less than a 1/1000 seaward slope), salt encrusted, highly evaporitic environments which are occasionally inundated by flood waters (Kinsman, 1969). Due to their limited accessibility and inhospitable nature, only a few sabkhas have been studied in any detail in terms of the geochemical and hydrological processes effecting the associated brines. Framework sediments of coastal sabkhas may range from almost pure carbonate to noncarbonate, depending on local supply of terrigeneous detritus (Kinsman, 1969). Sabkhas have been recognized as an active environment for the -1- -2— precipitation of aragonite, gypsum, halite, dolomite, and various other evaporitic minerals only within the last two decades. The first sabkhas identified were located along the Persial Gulf, and therefore, are the type 1coalities. These include the Abu Dhabi coastal sabkha, various coastal and continental sabkhas of the Trucial Coast, and the coastal sabkhas of the Qatar peninsula. These coastal sabkhas are predominatnly carbonate, while the continental sabkhas are characteristically composed of noncarbonate sediments (fine to mediumrgrained quartz). The initial studies of the Persian Gulf sabkhas have dealt with general physiography, sedimentology, stratigraphy, fauna and geologic development Shearman, 1963; Kinsman, 1964,1966; Evans and Shearman, 1964; Illing et a1., 1965; Evans et a1., 1973). Later studies entailed the determination of the geochemisty and origin of the brines. From this various authors have proposed hydrological and geochemical models for the source and evolution of the brines, and the formation of the observed evaporitic minerals. Traditionally these models were based on the geochemcial variation of the brines, types of evaporitic minerals present, and the zonation of these evaporitic minerals (Kinsman, 1966; 1969; Friedman and Sanders, 1967; Butler, 1969; de Groot, 1973; Bush, 1973). These early studies did not consider the climatological , sedimentological and geomorphological parameters that influence both the geochemisty and hydrology of the sabkha. The early investigations of the Abu Dhavi sabkha did comment on the very prominent "shamal" winds in the region but failed to integrate this climatological parameter with controls on the geochemical and -3- hydrological processes of this system. It seems that traditionally the major emphasis for understanding the processes Operating within the sabkhas was that this was the only known environment where dolomite formed in situ (de Groot, 1973). Therefore, the main rationale behind hypothesizing hydrological and geochemical models for sabkhas was in the development of mechanisms to explain dolomite formation. Additional coastal and continental sabkhas have since been identified and studies. A quartz sand-based coastal marine sabkha located in Baja California was studied by Phleger (1969). He investigated the distribution of evaporitic mineral on the surface of the sabkha. In addition he observed that periodic flooding of the area by marine water driven onto the surface of the sabkha by winds could produce the observed distribution of these minerals. Phleger (1969) attributed the formation of these minerals and the origin of the brines to a flood recharge hydrological process. In the Bardawil coastal plain, northern Sinai, Levy (1977a,b) studied a quartz sand-based coastal marine sabkha. In this study two distinct water masses were observed within the sabkha. A NaCl and a Ca012 brine, both of which he attributed to a normal marine parent. Levy (1977a,b) points out that these brines are distinct because of the different geochemical processes effecting them. The NaCl brine was generated by evaporative concentration and subsequent solute fractionation of a marine water, while the CaClz brine was generated by solute fractionation, of the NaCl brine, as a result of the brine interacting with evaporitic minerals present within the sabkha. West and others (1979) investigated the presence of gypsum -4- nodules in a modern sabkha on the Mediterranean coast of Egypt. The sabkha was carbonate based and the origin of the brine within it were not determined, however West and others (1979) state that the brines are not solely sea water derived. Their investigation attributed the origin of the gypsum nodules to sulfate rich hypersaline groundwater (600/00) dissolving gypsum crystals below the water table and reprecipitating near the surface because of evaporation of this water. Migration is upward by capillary movement. Rouse and Sherif (1980) investigated the origin of brines along a quartz sand-based coastal marine sabkha along the Gulf of Sirte, Libya. A sulfur isotope analysis revealed that a major source of the brines is contenental water, and only by occasional flooding during the winter months does a minor amount of marine water enter the system. A major weakness of these past works was that none have integrated both a detailed hydrological and geochemical investigation of an active coastal sabkha. It has only been recently that this type of investigation of a modern coastal marine sabkha has been conducted. McKenzie and others (1980) investigated the Abu Dhabi sabkha, in the Persian Gulf. In this study an isotope analysis of oxygen and hydrogen was conducted to determine the origin of the brines and the hydrological mechanism operating in the area. Three distinct zones, based on brine composition, were observed, each having a distinct hydrological mechanism operating within them. The area of the sabkha subject to periodic flooding, near the coast, had flood recharge of marine derived water as the major hydrological process occurring. Toward the continent, in an area with occasional flooding, a zone of mixed waters exists. These waters are a mixture -5- of a marine and a continentally derived water. In this zone a transitory hydrostatic state (McKenzie et a1., 1980) exist, which is a function of the amount of flooding. Immediately after flooding this zone exhibits a flood recharge hydrostatic state; afterwards capillary action occurs and when flooding has not occurred for an extended period, evaporative pumping becomes the major hydrologic process. In the zone closest to the continent evaporative pumping of continentally derived waters is the dominant hydrological mechanism. This study demonstrated that the hydrology and geochemisty of the system are controlled by the climatological, stratigraphical and geomorphological parameter of the area. The major hydrological models that have been proposed in the past for sabkhas are: a) Reflux (Adams and Rhodes, 1960), b) Capillary action (Friedman and Sanders, 1967), c) Flood recharge (Butler, 1969), and d) Evaporative pumping (Hsu and Siegenthaler, 1969). The geochemcial models based on the behavior of ions within the interstitial waters are: a) Brine mixing (Raup, 1970), b) Evaporative concentration with subsequent solute fractionation (Carpenter, 1978), and c) Solute fractionation as a result of the interaction of brines with host sediment and/or present evaporitic minerals (Eugster and Jones, 1979). The geochemical and hydrological investigation of the Laguna Madre sabkha was initiated with the establishment of a well system in the area (Fig. 3) during July, 1979. Water depths were measured at each well site and water samples for chemical analysis were taken from each well, as well as from the trench dug at each well site, the Intracoastal Waterway, and the gas access canals. The data for this -6— study was obtained from two field expeditions, August, 1979 and March, 1980. The hydrology of the system was determined from the piezometric potentials of the interstitial waters as a function of locality and depth. The geochemisty of the waters was determined by the analysis of the major ions in the water samples obtained during each sample period. Previous studies of the Wind Tidal Flat of Laguna Madre include a sedimentological and stratigraphical study conducted by Fisk (1959), as well as a preliminary hydrological and geochemical investigation of the sabkha conducted by Amdurer (1978). The latter investigation, however, was limited to the number of well sites established to accurately define this system. Geologic systems are rarely simple or identical; each is unique because of the various minor or major factors effecting it. This study will further refine our knowledge about coastal sabkha systems. The Abu Dhabi sabkha has been extensively studies and we now have a thorough understanding of one system (McKenzie et a1., 1980). This study will determine the geochemical and hydrological processes operating within the Laguna Madre sabkha and how the climatological, sedimentological and geomorphological parameters of this area effect it. In doing so one can determine some common processes occurring in coastal sabkhas and what parameters effect them. By doing this one will be better able to evaluate which of these processes can have occurred in ancient sabkhas and determine if and how they generated ore deposits. LAGUNA MADRE TIDAL FLATS Geologic History A combination of factors have led to the origin and development of the Laguna Madre tidal flats. These include fluctuations in sea level, regional uplift, climatic conditions, and the amount and type of sediment input into the flats by streams and longshore currents. During the early Pleistocene the southern Gulf coastal region was a relatively stable barrier island and back lagoonal coastal plain (Price, 1933). The mainland margin north of the Rio Grande was the original shoreline, as indicated by the presence of deltaic, lagoonal and barrier island features in the coastal plain (Price, 1958). The development of major continental glaciers during the Wisconsinan ice age resulted in lowering of sea level by approximately 140 meters and subsequent migration of the shoreline 80 to 230 kilometers gulfward (LeBlanc and Hodgson, 1959). Streams then cut into the exposed Pleistocene surface and barrier islands formed gulfward of the new coastline. Stream valleys up to 38 meters in depth have been measured beneath the tidal flats (Fisk, 1959; Wilkinson, 1975). As the glaciers began to melt at the end of the Wisconsinan ice age a steady rise began in sea level accompanied by coastline and barrier island transgression. Sea level first rose at a rapid rate -8- until approximately 4000 years ago when the rate of rise decreased sharply (Amdurer et a1., 1979). Post-Pleistocene regional uplift of approximately a meter along the gulf coast also occurred (Price, 1933, 1947; LeBlanc and Hodgson, 1959; Fisk, 1959). This raised the gulf coastal plain sufficiently to develop the existing Texas coast and barrier island chain gulfward of its Pleistocene counterpart. About 6000 years ago when sea level reached within 7.5 - 9.0 meters of its present level the divides between the estuaries and bays were flooded and the group of widely separated barrier islands began to grow (Amdurer, 1978). Quaternary sediments which accumulated in estuarine, lagoonal and eolian environments, overlie the erosional surface of the Pleistocene sediments in the area. Sedimentation continued at a rate of approximately 3.1 meters per thousand years until, about 4000 years ago, the separated islands merged to form a continuous barrier island and an extensive lagoon west of the barrier islands resulted (Amdurer et a1., 1979). A sedimentation rate of 1 meter every 150 years has been recorded in situ by Fisk (1959), and indicates that the final closing of the Laguna Madre depression at the western edge of the Sand Bulge occurred approximately 150 years ago (Price, 1958; Fisk, 1959). Presently the Laguna Madre tidal flats are still being modified by extensive winds in the area. Sedimentation and deflation occur continuously by eolian processes; also, occasional storms temporarily open passes on Padre Island. Therefore, the surficial features of the Laguna Madre are constantly reequilibrating. Regional Setting The Laguna Madre tidal flat is located approximately 90 km south of Corpus Christi, Texas. Situated along the Gulf of Mexico coastline it is 22 km in width, 26 km in length and has a total area greater than 550 kmz. The tidal flat is bordered on the west by a remnant Pleistocene eolian coastal plain and on the east by a barrier island. Padre Island (Fig. l) is one of the major islands which make up the Texas coast barrier-island chain. It extends from just south of Corpus Christi Bay to the Brazos Santiago Pass which is just north of Mexico, a distance greater than 180 km. It extends almost parallel to the Texas coast and forms slightly gulfward arc at its southern end. The Laguna Madre itself is divided into two halves by the Laguna Madre tidal flat, each half is restricted by a mud flat which further separates the lagoon. The northern Laguna Madre is separated into two distinct basins by the ”Middle Ground", a mud flat that is about 3.2 km in length and located midway between the tidal flat and Baffin Bay. The more northern basin extends 64 km north of the Middle Ground and has a depth of less than 1.5 m. The southern basin of the northern Laguna Madre is called "The Hole". Its dimensions are 16 km in length, 3.2 km in width and has an average depth of 1.5 m. The southern Laguna Madre has a greater width than its northern counterpart. The southern lagoon is constricted by a mud flat that is located opposite the Arrayo Colorado. The more northern part is called Red Fish Bay, which has a maximum width of 9.7 km and an average depth of less than 3 m. It shoals to a depth of less than 1 m at the location where it curves westward into the tidal flat. The -10... 2.1 _ 2‘. man“ a.“ Figure 1. Map of general area (after Fisk, 1959) -11— more southern half has a length of 40 km, a width of 11 km and a depth of approximately 1.5 m. In 1949 an Intracoastal Waterway (ICWW) was dug through the Laguna Madre. This ship channel is 6.5 m wide and 3.7 m deep. Additional canals have since been dug within the flats (Fig. 3). These canals are used for gas well access and are usually less than 3.5 m in width and 2.1 m in depth. Due to the length of Padre Island, the nearest sea water input to the Laguna Madre are at Aransas Pass, 108 km north of the tidal flats and at Brazos Santiago Pass 87 km south of the flats. Therefore, the lagoonal waters, even though they are derived from the Gulf of Mexico, were chemically distinct because they can be modified by rain, river input, and/or high degrees of evaporation. For example, a minimum of 12.5 0/00 and a maximum of 108.6 0/00 have been measured for the surface waters of the lagoon (Fisk, 1959). The Laguna Madre tidal flats merge to the east with the hummocks and dunes of Padre Island and to the west with an erosional reentrant of the mainland margin (Fisk, 1959). A few islands consisting of mainland remnants also exist within the western section. The largest of these are the Mesquite Rincon and El Toro. The Basin (Fig. 3) is a topographically low region of the flats, which often had ponded water prior to the construction of the Intracoastal Waterway. The boundary of it is characerized by low topography and surface sediment consisting of soft, spongy gypsiferous clays (Fisk, 1959). The Sand Bulge is another prominent feature of the flats. This area extends west from Padre Island and is characteristically higher in elevation than the rest of the flats and is composed of loose sands low in mud content. -12— Sediment Type and Source Regions The Laguna Madre tidal flats consists predominantly of sand, with minor amounts of clays, gypsum, and algal micrite. The sand composition is approximately 90 percent quartz with feldspar, chert, shell fragments and heavy minerals comprising the remaining 10 percent (Amdurer, 1978). Clay composition is predominantly illite with minor amounts of fine grained quartz and calcite (Fisk, 1959). Sands from beach, barrier island, dune, tidal flat and washover fans are subangular, sorted to well sorted, and fine to very fine grained. Sediment samples from these regions have varying amounts of clays. Textural and compositional uniformity of these sands indicate a similar source region (Amdurer, 1978). Five depositional provinces have been defined for the northern Gulf of Mexico; a) Eastern Gulf, b) Mississippi, c) Western Gulf, d) Rio Grande and e) Texas Coast (Van Andel and Poole, 1960). Each is characterized by the heavy-mineral suite within the sediment. They are defined by the rivers providing the sediment to the Gulf and by the region in which this sediment is depositied. Two suites of heavy minerals have been identified within the Laguna Madre flats (Fig. 2)(Van Andel and Poole, 1960). The northwest corner of the tidal flat is characterized by heavy minerals consisting of tourmaline, zircon, kyanite and stauralite. This suite of minerals originated from the Colorado, Brazos and other central Texas rivers, and therefore occur in the Western Gulf province. The sediment was transported to the tidal flat from the input areas by southerly longshore currents along the Gulf coast. A much larger portion of the tidal flat sediment -13- Western 0 Gulf " Province ‘ “00 (I ) Rio Grande DO Province $ 0 5 KILOMETERS Figure 2. Map of sedimentary provinces. -14- contains green hornblende, pyroxenes (mostly augite), epidote and basaltic hornblende. This suite is characteristic of the Rio Grande province. A clay assemblage consisting of illite also indicates a Rio Grande source (Fisk, 1959). Evidently, the sediment of this province is transported to the Gulf of Mexico from the Rio Grande. This sediment as well as older reworked sediment located on the continental shelf and deposited during a lower sea level are then transported to the tidal flat by northerly longshore currents and deposited on Padre Island. The sediment is then transported to the flats by the dominant southeasterly winds of the area. Heavy mineral composition of sediment in Laguna Madre indicates two source regions for this terrigenous sediment. Central Texas rivers evidently provide sediment to the northwest region of the flats, but the major portion of the flats appear to have a Rio Grande sediment source region. Although there are two sedimentary provinces defined in the Laguna Madre tidal flats, all sediments within the study area are relatively uniform in texture and composition. The reasons for this are that both the central Texas rivers and the Rio Grande drain similar regions, both provinces have sedimentary input into the Gulf at similar rates, and both provinces have similar processes of sediment transport and deposition. The major variations found in the sediment are therefore a function of the distance between the sediment locale and Padre island, and the energy of the environment of deposition. -15... Climate The climate in the coastal region of south central Texas is semi-arid (N.O.A.A., 1979). Climatological data for the Laguna Madre tidal flats is, however, lacking; therefore, the climate has been inferred for the area from five surrounding meteorological stations. These are: Corpus Christi, Chapman Ranch and Kingsville to the north, and Port Mansfield and Raymondville to the south (Fig. 1). All stations are located within 90 km of the study area. Meteorological data from 1941 through 1970 indicates a normal yearly temperature of 22.6°C and a normal yearly precipitation of 70.0 cm (N.O.A.A., 1980). The normal summer temperature is 29.1°C and the precipitation is 6.1 cm. During May and September rainfall is greatest and during the winter months rainfall is the lowest. The period of June through November is the tropical storm season. During this time almost all precipitation can be attributed to these storms. Summer sunshine exists for eighty percent of the possible time. The months of September and October experience a climate that is a continuation of summer conditions. Prevailing winds are from the southeast, resulting in relatively mild summer temperatures. Occasional shifts in wind direction result in temperatures rising above 38.0°C during the summer. Relative humidity averages 50 to 60 percent, with summer morning highs greater than 90 percent. During the present study, 28.5°C was the average 1979 summer temperature and 9.43 cm the average monthly summer precipitation; temperature highs reached 37.0°C. In 1980 the average summer temperature and monthly precipitation was 29.6°C and 8.36 cm -16.. respectively, with a temperature high of 37.9°C. In July, 1979, the month prior to sampling, the average temperature and precipitation were 29.3°C and 7.37 cm respectively; while in August, 1979 the average temperature and precipitation were 29.1°C and 7.58 cm respectively. In February, 1980 average temperature and precipitation were l4.7°C and 3.16 cm respectively, and during the March, 1980 sampling period 19.6°C and 1.28 cm reflected the average temperature and precipitation. During August 9-10, 1980 hurricane Allen struck the south Texas coast. During the period prior to Allen a severe drought and heat spell affected the area. Only 0.03 cm of precipitation fell. As a result of Allen, 24.9 cm of rain fell during a two day period. The high winds and rains produced extensive flooding. Other hurricanes that occurred in the past and their effects on the area are: September, 1971, hurricane Ceila, 31 cm of rain; September, 1967, hurricane Beulah, 22 cm of rain in one day. One must note that even with these weather extremes the potential evapotranspiration rate of the area is 100 to 120 cm per year (Rusnak, 1960), while the annual precipitation rate is only 70.0 cm. This results in an annual average precipitation deficiency of 30 to 50 cm. _17_ Field Observations In the period of March, 1979 to August, 1980, numerous field expeditions were undertaken to the Laguna Madre tidal flats. This study was initiated in March, 1979 with a reconnaissance expedition to the area. At that time the area was relatively dry and extensively covered with desiccated algal mats. During the period of June, 1979 through August, 1979, field logistics and establishment of well sites were completed. During August, 1979 and March, 1980, water samples were obtained from the well sites and from surficial water bodies. At various times during the study period the tidal flats were flooded. This resulted from periods of intense rain, high winds and during and after hurricane Allen. Flooding also occurred in June, 1979 after a period of heavy spring rains and an aerial flight over the study area indicated that more than half of the tidal flats were covered with water. Standing water remained until the end of June and within a month of normal weather conditions, the once moist and extensive algal mats became desiccated. In March, 1980 during a five day sampling period, an extensive area between well site 13 and 12 (Fig. 3) was observed to undergo flooding. This was produced by the continuous high winds from the northwest which pushed waterway and canal waters onto the flats and spread them in a southeasterly direction. With the termination of these winds, flooding ceased on the following day and the once extensive flood sheet began to separate into numerous smaller ponds. At that time algal mats on the eastern side of the tidal flats were moist and extensive while those on the western side were dry and desiccated. In the summer of 1980 an -1 8.. extensive drought with high temperatures affected the region. This produced extensively desiccated algal mats throughout the area and concurrently the waters of the Intracoastal Waterway and gas access canals became more brackish than normal for this time of year. Effects of the hurricane on the area were extensive and widespread for the entire tidal flat was flooded and numerous storm passes were reOpened on Padre island. Water depths varied as a function of local topography with greater than nine feet of water covering some areas. Full analysis of flood waters has yet to be done, however the salinity of the water was recorded and was found to be less than that of normal sea water. This suggests that a significant portion of the flood water originated as continental runoff. Standing water remained on the flats for more than one week, with topographically higher areas, such as the Padre Island side, draining faster than the lower areas, such as the basin. Two weeks after the hurricane struck, a trench was dug at well site 13 and a water table measurement indicated a near normal water depth. The algal mats which had previously covered the study area were no longer present; instead rounded clasts of algal mats ripped up by the hurricane flood waters were found throughout the area. Wells protruded several additional inches from the ground indicating that some surficial sediment had been eroded. Fresh selenite rosettes were observed within the sediment at various well sites on the western side. Depth of occurrence varied for each well site. These crystals were distinct from selenite crystals dredged during Intracoastal Waterway construction. Spoil mound ctystals averaged 10 cm in diameter with some crystals greater _19_ than 1 m in diameter. These crystals contained varying amounts of impurities such as sand and shell fragments. A more detailed description of these crystals is given by Masson (1955). The selenite rosettes found at the well sites averaged 2 cm in diameter and contained little to no impurities. These crystals apparently are less mature than those found at the spoil mounds. In addition to the presence of gypsum rosettes, preliminary analysis of a few cores revealed fine grained crystals of iron sulfides. In July, 1980, before hurricane Allen struck, a brief expedition to the northwest section of the tidal flats showed that the algal mats were widespread but were not present thorughout the northwest section for westward, beyond this boundary, a very white, sun-bleached sand covers the flats. Springs were also found. Apparently animals within the area dug shallow holes which then filled up with interstitial waters. This water tasted less brackish than normal sea water, indicating possible continental input of interstitial waters. Within the sand pile from the dug hole gypsum crystals were found. These crystals, however, were not similar to the characteristic selenite rosettes found at the well sites; instead they fonmed yellow-green, clear, euhedral crystal. This crystal morphology has been described by Kinsman (1966) as typical of anhydrite which has rehydrated to gypsum due to its contact with less saline, continental groundwater from inland areas. Analysis of these crystals and the groundwater within this area has to be completed before this interpretation can be properly evaluated. FIELD METHODS Establishment of Well Sites During the period of June, 1979 through August, 1979 seventeen well sites were established in the Laguna Madre tidal flats (well sites 2-18; Fig. 3). At each well site a 1.9 m, 3.0 m and a 3.8 m length of 5 cm in diameter PVC (polyvinylchloride) pipe was pounded into the sediment with a Divercoring apparatus. This device consists of a 5 cm inside diameter pipe sleeve which fits over the PVC tube, a shaft connected to the sleeve which is used to guide the hammer, and a 18 kg cylindrical hammer which fits onto the shaft and is used to pound the PVC tube into the sediment. With this device 3.8 m is the maximum length of tubing that can be inserted practically into the sediment. After it is inserted into the ground a 5 cm plug is used to provide an air tight seal for the 3.0 m PVC pipe. A clamp is then placed around the pipe to provide a grip; the pipe is then pulled from the ground. Cores between 1.2 m and 2.1 m in length have been recovered with the coring device described above. Once out of the ground the pipe is then cut to the length of the sediment core, capped and sealed at both ends. The shallow and deep wells, 1.9 m and 3.8 m PVC tubes respectively, had the sediment removed from them with a modified hand auger. Additional augering was required for some wells because of -20- -21- o WELLS ESTABLISHED \ H A AMDURER (l978) A WELLS ESTABLISHED 5 THIS STUDY quHH o SPOIL BANKS \ J E3 ISLANDS AND MAINLAND P P \ BUOVW VNOSVW ms nu. \JAcczss 6 ‘0 CANAL‘ INTRACOASTAL WATERWAY LAGUNA KILOMETERS Figure 3. Map of study area _22.. sediment oozing into them. Upon removal of sediment the wells are labeled and PVC caps are placed on them. These caps prevent rain water, flood water and sediment from entering the well. Once established, these wells (sites 2 through 18, Fig. 3) along with well site A and well set B (Fig. 3) were then accurately mapped. Surface elevation was then determined for each well site by the use of an Auto-leveler. -23- Sampling and Field Analysis There were two sampling periods during this present study, August, 1979 and March, 1980. Water samples collected in August were taken from the Intracoastal Waterway, a gas well access canal located near well set B, well sites 2 through 18, well site A and well set B (Fig. 3). Water samples collected in March were taken from: the Intracoastal Waterway, gas well access canal located near well site A, surficial flood waters between well sites 12 and 13, well sites 2 through 9, well sites 11 through 18, and well site A (Fig. 3). Well site samples included trench water, and shallow and deep well waters. The water sampling procedure required two visits to each well site. At the first visit, water depths were measured within the shallow and deep wells, a trench was dug and sampled, water table depth was measured within the trench, and the wells were pumped dry. On the second visit water samples were only taken from the shallow and deep wells. To eliminate the possibility of water contamination a hand operated peristaltic pump with tygon tubing was used for the well sampling. Before the sample was collected distilled water was first run through the tygon tubing, followed by some well water. Each water sample was stored in a 500 m1 and a 125 ml polyethylene bottle. Five milliliters of 70% formaldehyde was also added to the 125 m1 bottle to suppress any organic activity. During sample collection, field analysis included temperature and pH determinations. Within 10 hours after the water samples were collected, alkalinity was determined by sulfuric acid titration (Brown et a1., 1970). Alkalinity and -24- chlorinity determinations were not performed during sample collection because of equipment limitations. -25.. Geochemistry/Hydrology Early investigations of the Laguna Madre Tidal Flats did not specifically deal with determining the geochemical and hydrological processes of this system; however, they do report on certain geochemical and hydrological characteristics of the area. Masson (1955) investigated the origin and distribution of gypsum within the Laguna Madre Tidal Flats. He found that selenite rosettes increase in size with depth. Masson also reports that flood waters on the surface of the tidal flats reach three times the concentration of sea water. From the distribution of the selenite crystals within the sediment, the analysis of this flood water, and the field observations of Laguna Madre waters being pushed onto the tidal flats, Masson (1955) postulates that the gypsum precipitates, within the sediment, from a highly saline water similar to sea water in composition. He states that the Laguna Madre waters are periodically moved as wind blown sheets across the surface of the tidal flats and then sink into the underlying sediment to provide a source for these highly saline brines. A detailed investigation of the Laguna Madre Tidal Flats was conducted by Fisk (1959). This study dealt mainly with the geomorphology and geologic hyistory of the area; however, some geochemical analysis of surficial Laguna Madre waters and tidal flat interstitial waters were reported. From field observations of surface flooding, Fish (1959) postulates that the highly saline interstitial waters are recharged by flood waters. These flood waters are blown onto the tidal flats from the Laguna Madre and seep -2 6.. downward. Fisk (1959) also points out that there is a high variability in Laguna Madre water salinity. It has been only recently that a detailed geochemical and hydrological investigation of the Laguna Madre sabkha has been conducted. Amdurer (1978) postulates that reflux and evaporative concentration control the hydrological and geochemical processes, respectively. He suggests that the Laguna Madre waters are pushed onto the sabkha surface by high winds, but only at a few sites does this water seep downward. These localities are where the algal mats are desiccated and cannot prevent the water from flowing downward. This selective recharge, he states, leads to the low variability in brine composition relative to the high variability of the source waters. This high variability of the source waters is attributed to these waters evaporating, becoming diluted by additional flood waters, and dilution by rain water. For the Sand Bulge area, Andurer (1978) postulates that reflux is occuring toward the Intracoastal Waterway. This is supported by the fact that for all his well sites the piezometric potential decreased with depth and the piezometric surface sloped toward the waterway. In addition, Amdurer (1978) suggests that the hydrology of the system was significantly changed with the introduction of the Intracoastal Waterway because it provides an escape route for the brines within the system. Evaporative concentration of this flood water is the geochemical process occurring; however, because calcium is fractionated from the water by the precipitation of algal micrite, gypsum does not precipitate (Amdurer, 1978). Amdurer (1978) states that this is supported by thermodynamic modeling of the waters which indicate that -27- gypsum is undersaturated, and the lack of field evidence for gypsum occurring.ip.§i£u. During a preliminary field trip to the Wind Tidal Flats in March of1979, fresh gypsum crystals were found in the area. This suggested that the geochemical processes postulated by Amdurer (1978) were not necessarily accurate for the entire system. His study was limited by the distribution of his well sites and their interaction with the fresh water lenses of nearby islands. Well site A, well set B and well set C (Fig. 3) are the well sites established by Amdurer (1978) to define the sabkhas hydrology and geochemisty. From figure 3 one can see that well site A could be affected by the freshwater lens of the small nearby island (this has been found to be true during the present study). Well set B (Fig. 3) found within the "Basin" area of the tidal flats has been found to have a typical brine geochemistry and, therefore, would not be an accurate representation of the average brine geochemistry. Well set C (Fig. 3) was used primarily by Amdurer (1978) to delineate sabkha hydrology and geochemistry. Except for the western most well, all the wells in well set C lie less than 1.5 km from Padre Island. Well site 17 (Fig. 3), established during this present study has been found to be influenced by the fresh water lens of Padre Island. Since well site 17 lies approximately 1.5 km from Padre Island this indicates that the wells used by Amdurer (1978) were similarly affected. In addition, Amdurer (1978) postulates a reflux of brine toward the Intracoastal Waterway; however, his western most well site is about 5 km from the geochemical and hydrological investigation of the Laguna Madre sabkha, the localities -2 8- and possible influences of the fresh water lenses of nearby islands influenced his geochemical and hydrological interpretations of the area. This study was designed to avoid the limitations of the past work in the area. The distribution of the well sites and the well depths were determined in order to avoid the potential influences of the freshwater lenses of nearby islands and to obtain a complete cross section of piezometric potentials from well sites westward of the Intracoastal Waterway fo Padre Island. In doing so this study will accuately define the geochemical and hydrological processes that are occurring within the Laguna Madre sabkha. LABORATORY METHODS Within 72 hours after collection in the field, the water samples were taken to a laboratory at Corpus Christi. The samples in the 500 m1 bottles were filtered using a 0.45 micron membrane filter. Chlorinity was then determined by Mohr titration (Brown et a1., 1970). After analysis the samples were acidified to a pH of less than 2.0, by adding concentrated HNO3. They were then packed and shipped to the geochemistry laboratory at Michigan State University for further analysis. The precision of the chemical analysis is shown in Table 1. In Michigan, the 125 ml, formaldehyde treated samples were analysed for sulfate. Sulfate analysis was performed by a slightly modified technique proposed by Dunk et a1. (1969) (Appendix A). The remainder of the analyses were done using the filtered and acidified water samples in the 500 m1 bottles. Potassium analysis was conducted on 1/10 dilutions of Intracoastal Waterway and canal samples; for the other samples a l/lOO dilution was required. Standards and blank solutions contained 1000 mg/l sodium to suppress ionization of K during analysis by A.A. (atomic absorpion). Samples already contained at least this amount of sodium; therefore, addition of sodium to them was not necessary. Calcium analysis was conducted on 1/100 dilutions of all samples. After a 10 g/l lanthinum matrix was added to the samples and standards to suppress ionic interferences (particularly -29- -30- from 804) and ionization during A.A. analysis. This was accomplished by mixing 1 ml of 50 g/l lanthinum to 4 ml of sample and standards prior to aspiration in the A.A. A 10 g/l lanthinum blank was used in this analysis. Sodium and magnesium analysis were performed on 1/1000 dilutions of samples. This dilution was necessary in order to have the Na and Mg concentrations of these samples within the measurable limits of the A.A. In addition, sodium analysis was conducted by Flame Emission with a rotated A.A. flame peak because of the high concentrations of Na in the samples. To suppress ionic interferences and ionization during both Na and Mg analysis a 10 g/l La matrix for both samples and standards was necessary. La was added to the samples during dilution. Strontium was analysed on a 1/10 dilution of the samples. A 10 g/l chloride matrix was prepared for the standards and blank. This was necessary to minimize error due to variations in viscosity and vaporizing temperatures between samples, standards and blank during A. A. analysis. Bromide analysis was performed colormetrically by a method used by Presely (1971). This method requires oxidizing the bromide with chloramine-T (sodium paratoluene-sulfonchloramine) and measuring the resultant color change spectrophotometrically at 595 millimicrons setting. This technique requires the use of bromide standards to generate a working curve for bromide concentrations and absorbance. The technique used has been modified from that proposed by Presely (1971) because of the greater Br concentrations encountered. A more detailed description of this procedure and modification is given in -31- Appendix B. Density measurements were done with the use of a pycnometer. Conversion of density values measured at laboratory temperatures to density values at field temperatures were required for the determination of the systems hydrology and use in the chemical modeling of the system by the WATEQ program (Truesdell and Jones, 1973). The conversion and formulas used are given in Appendix C. -32- Table 1 Precision of Geochemical Analysis Measurements Precision pH i 0.04 units Density :_0.0003 g/cm3 Alkalinity : 6.0% Br :_l.8% Ca 1: 3.07. CI _-t_- 6.0% K i 3.07. Ms 1 2.0% Na ::5.0% 804 :_5.0% Sr : 3.0% GEOCHEMISTRY Three geochemical mechanisms have been hypothesized for the chemical evolution of waters within sedimentary basins. These are: a) evaporitic concentration (Carpenter, 1978), b) solute fractionation (Eugster and Jones, 1979), and c) brine mixing (Raup, 1970). This investigation will test the hypothesis that evaporitic concentration is the dominant geochemical mechanism effecting the brines within the Laguna Madre sabkha. Evaporitic concentration - this mechanism attributes the behavior of solutes and the resultant chemical nature of the interstitial brines to the removal of water from that brine (Carpenter, 1978) by the process of evaporation. In evaporative concentration the ratios of ions within the solution remains constant until the onset of mineral precipitation. The constant relationship between ions is expressed as B/A-k, or B = kA. The logarithmic form of this equation is: log B = log A + log k. A plot of A and B on a log-log graph produces a straight line with a 1:1 slope regardless of the value of k; therefore any ion which deviates from this relationship is affected by some process other than that removal of water (Carpenter, 1978). During evaporative concentration solutes can exhibit three types of behavior (Eugster and Jones, 1979). Type I behavior is characterized by conservative ions which neither precipitates out -33- -34- during evaporation of the water nor participate in any diagenetic reactions with the subsequent mineralogical environments (Carpenter, 1978). Hence, a plot of a conservative ion concentration versus degree of evaporation would be a straight line. In this case, chloride and bromide are predominantly used as conservative ions when considering the evaporation of seawater. Chloride remains conservative until the precipitation of halite, and bromide is conservative until late stage bittern precipitation (Valyashko, 1956; Zherebtsova and Volkova, 1966). In evaporative concentration a second type of solute behavior also is exhibited by ions which are directly involved in mineral precipitation. This type of behavior is further subdivided into a Type Ila and 11b (Eugster and Jones, 1979). Group Ila ions are extracted from solution at the onset of mineral precipitation and continue to be depleted by additional mineral precipitation. Once precipitation has stopped, the solute then increases in concentration at a constant rate. Sulfate exhibits this type of behavior during the evaporation of seawater and gypsum precipitation. Group IIb solutes are ions that are removed from solution at the initiation of mineral precipitation and continue to be removed with additional evaporation until they are no longer present in solution. At that point mineral precipitation ceases. During seawater evaporation calcium exhibits this behavior from the onset of gypsum precipitation until gypsum precipitation ceases and no more calcium is present. The third type of solute behavior during evaporative concentration has been characterized as a Type V solute behavior (Eugster and Jones, 1979). The solute increases in concentration -35- 5.5 5.0‘ M9 N0 ’ 504 4.51 Log X (mg/l) s» C? 2.5" K,Co Co 2-0 r F I I a L5 2&3 215 310 3.5 4J3 Log Br(mg/|) Figure 4. Concentration trends of ions in sea water with evaporation (after Carpenter, 1978) -36- during the removal of water until it becomes saturated and begins to precipitate. Since this type of behavior is solubility controlled, the concentration of the solute is constant from the point of mineral precipitation regardless of how much additional water is removed from solution. Silica has been observed to display this type of behavior (Eugster and Jones, 1979). For example, in Lake Madadi basin silica never reached a concentration of greater than 60 ppm even though chloride increased by three orders of magnitude. Carpenter (1978) has developed a computer program to generate a plot of solute behavior during evaporative concentration of seawater. The plot shown in figure 4 represents the chemical evolution of the major solutes in seawater during evaporative concentration. The major points of deflection for the different ions are at the initiation of gypsum, halite, magnesium sulfate and sylvite precipitation. Solute fractionation - In this geochemical mechanism solute fractionation is a result of processes other than mineral precipitation. Fractionation of ions is a result of: a) selective dissolution of efflorescent crusts, b) selective dissolution of sediment coatings, d) ion sorption on active surfaces, d) degassing of the water, and 3) redox reactions (Eugster and Jones, 1979). Efflorescent crusts are produced as a result of capillary action drawing interstitial waters upward where it then evaporates and produces a precipitate. Selective dissolution of this crust occurs when surficial runoff seeps down into the sediment. The selective dissolution of the more soluble salts produces a pronounced -37- fractionation of some ions. This can be demonstrated by the dissolution of a small amount of halite. Since the concentration of sodium in halite is 390,000 ppm, the dissolution of halite and not the remaining precipitates will result in a solution that is strongly fractionated with respect to sodium and chloride. Dissolution of sedimentary coatings and diagenetic cements can also result in water with a significant fractionation of the solutes. Levy (1977b) proposed that the origin of a calcium chloride brine within the Bardawil sabkha was the result of a brine that was produced by evaporative concentration interacting with calcium carbonate within the sediment. The sorption of ions on active surfaces can significantly deplete the solution of certain ions. This fractionation of ions is characterized as a Type III or Type IV solute behavior (Eugster and Jones, 1979). Type III behavior is characteristic of solutes that are removed from the water when the removal mechanism or combination of mechanisms is not concentration dependent. Type IV behavior is characteristic of an ion removed only during a limited range of water evaporation. Potassium can exhibit these types of behavior. It can be selectively adsorbed by swelling clays within the sediment (Grim, 1968), and it can be adsorbed by aluminosilicate gels (Jones et a1., 1977). Cation anion exchange within clays can also produce a solute fractionation by extracting ions frmm solution and introducing a substitute ion into the solution. Redox processes, such as bacterial sulfate reduction can result in solute fractionation. Degassing of solution as a result of variations in temperatures can result in loss of C02. This effect on the -38- carbonate system can result in solute fractionation by the resultant variations in carbonate solubilities. Brine mixing - In this process the geochemical composition of the water is a direct result of a mixing of two chemically distinct waters. The mixing of two different brines can result in the precipitation of salts (Raup, 1970) and carbonates (Runnells, 1969; Badiozamani, 1973; Wigley and Plummer, 1976). Conservative properties such as solute concentrations, temperature, alkalinity, acidity and total carbon (Stumm and Morgan, 1970) will display a linear relationship. This linear relationship is a function of the value present in the end member parent solutions and the proportion of each parent in the mixture. The non-conservative properties such as pH, Eh and activities will display a non-linear relationship. This relationship means that the mixture can contain a value of the property that is greater or less than that of either parent. The non-linear behavior of these properties has been attributed to three processes; the algebraic effect, the Pcoz effect, and the ionic strength effect (Wigley and Plummer, 1976). The algebraic effect is produced as a result of the non-linear behavior of the ion activity products during brine mixing. This behavior has been proven experimentally (Wigley and Plummer, 1976). For example, a mixture, with both parents undersaturated with respect to a mineral, can become saturated with respect to that mineral. Raup (1970) produced the precipitation of salts by the mixture of artificial or seawater derived brines. In a mixture of two seawater derived brines a salt precipitated. This salt was predominantly halite with minor amounts of sylvite. In the dorag model -39- (Badiozamzni, 1973) the nonrlinear behavior of the ion activity product has been hypothesized as the mechanism for dolomite precipitation. The mixing of meteroic waters with 5 to 30% seawater will produce a solution that is saturated with respect to dolomite and undersaturated with respect to calcite. In the PCO effect, the most important in the carbonate system (Wigley and Plummer, 1976), the mixture of two brines produces a change in the C02 content of the system. This can be produced by the variation in water temperature that is produced by mixing two water bodies at different temperatures. A decrease in C02 of the waters can result in the precipitation of calcium carbonate. The ionic strength effect is produced by the non-linear dependence of activity coefficients on the ionic strength. In the extreme case, two solutions, both saturated with the same mineral, can become undersaturated with respect to the mineral. This can produce dissolution of the subsequent mineralogical environment. OBSERVATIONS Trends The waters of the Laguna Madre sabkha are essentially Na-Cl brines that are depleted in Ca, with respect to sea water. Using bromide as a conservative indicator, samples collected in August ranged between 3.5 to 8.9 times the concentration of sea water; likewise, samples collected in March ranged between 3.4 and 9.7 times the concentraton of sea water. Chemical analysis of the water samples from the basin, well set B and well site 10, indicate that these waters are distinct from those at sites 9 through 18; because of this the basin region will be discussed separately. The Intracoastal Waterway and gas well access canals have a negligible effect on all sample sites except site 2. This site is 20 m from the Intracoastal Waterway and appears to be effected by the wake of barges which churn the waters along the shores of the Waterway. The chloride, potassium, sodium, bromide, magnesium, and sulfate concentrations for well sites 9 through 18 all exhibit similar trends (Figs. 5-8). A concentration plateau exists for these solutes with the maximum values found at well sites 15 and 16. Concentrations decease sharply east of this plateau and gradually west of it. In general, chloride, potassium, sodium and bromide concentrations at each well site increase with depth, while magnesium and sulfate, -40- -41- sampling (Day 1) to the second day (Day 2) the strontium concentrations increased. Concentrations of calcium ranged from 420 mg/l to 1080 mg/l for August with similar values for March. Strontium concentration ranged from 11 mg/l to 50 mg/l for August with March values slightly lower. Alkalinity is mostly in the form of HCO3 at the pH levels found in these waters. Trench waters had the lowest alkalinity for each well site, with the shallow and deep samples having higher values. Alkalinity was highest for the westernmost well sites and decreased in an easterly direction. In general, the chemical trends of the major ions within the subsurface waters remained relatively constant between August, 1979 and March, 1980. Salinities increased with depth and the highest concentrations were found approximately 3 km west of the Gulf of Mexico. August Data -42- Cl' 75 1W 23 o ’95 ..J\u° A - ml 0 a i \ )‘9 0 ° °\ 3 L ‘ l o L U l‘ D m ,0, A 5‘ e i . 3 Q I C 7 I 1 112' I) ‘2 II ‘0 '5 II ‘7 |I I WELLS Ca " 1m 1130‘ \ W < 9 o E - t.— v”. 3 a a U 2 500 O 250 9 U A C 7 C 5 “N 13 2 ll 1‘ I! II N ll 0 WELLS . so. 20 I, < D E : >e } E 2; ‘° .3 (n 4 O 3 O . , D O , I A C 7 ‘ 5 ‘3" l3 ’2 II l‘ I! [.11 I. 5 Figure 5. B" All Plots: o-surfoce O- Trench A- Shallow 51- Deep 7001 SN $00 0 ‘ 4°04 § 9% ° 0 30m a (W ‘ Q A 100 I ’ '3 rent 7 ISM 13 lllllilfillt‘l II WELLS MqOO on u m u 5.0 u 1 ' . °. 951C 7 Isuzu l! \erM vale” re WELLS Alkalinity 700‘ m 9 500- too D 3001 _ zoo< \9 :‘g‘twg ‘. . 00° 3 | I“: 7 I 5 ‘32! l! 12 ‘I M I: v. I7 :0 -43- K’ weosr ALL PLOTS: e-suancc o-TRENCH A-SNALLow O-DEEP Ne'u/l) o b 4e 0 B e e .é 7 3333:- u (e u (e .e ref! :3 e WELLS "a. 3'0. 0 ee O-q 0“ a- “ < fl . ’1 "i e ‘ o no \ a :- \ ” w A o ee< . n4 C $ 0 eieireie'seene u'en'uueiiu' me e WILLS Figure 6. 1751 I291 -44- March Data CV A ‘M \\ a j" m ‘ o U n 29* o e e e I I AC 7 ;S ‘SZ‘ '3 :2 T: I. ['5 1'. .77 Iv. fie WELLS O. CO 'mo . "250‘ o. n. A-wi a E ’50! ‘ V e 3 c U 500‘ V e no 9 0:5 ;7 es u& I, a u u 3 Eur .e o WELLS a $0. m 9. I3 ’1‘ :4 a ‘ o - g/fl‘ <; p . 0 § R V ‘0 ' l/ 5 ll. 6 o . 07 w ‘ fl .- 9 e A c I e s 432w :3 :2 .. we -5 e .7 ‘e S WELLS Br' AN Plots: e-Sudoce O‘Trench A'Shollow O'Deep coo . a O 9 eAc r e Sou! a 2:..; 3 an m D WELLS MqOO am my : I0.0< 3 o 5 a» . q: -. 0 0 0 e eic 1 35 on us nzuc'e 6 NH 3 3 WELLS Alkalinity mm i com o E o V’wm » 2 «304 V -' G 5‘: m 3A " zm< EL firmA%/& a ' On ._ W. IOO1 9 c A C 7 7‘ 5 ‘32-. I: I2 II I. 15 l. l, I. 6 WELLS Figure 7. -45- N’ uencu ALL PLOTS: e-sunrAcE o-mencu 4-3NALLOw a-oecp Ne'le/H 8 I“ affine/I) PLEASE NOTE: PAGE NUMBER 46 IS MISSING IN NUMBER ONLY AS TEXT FOLLOWS. FILMED AS RECEIVED. UNIVERSITY MICROFILMS INTERNATIONAL DISCUSSION There are four possible source regions for the Laguna Madre sabkha brines: sea water from the Gulf of Mexico, water from the Laguna Madre, continental water from the Tertiary Goliad Sand aquifer, or groundwater from the Quaternary eolian plain deposits. The Goliad Sand is the principle fresh water aquifer for this region. It consists of fine to coarse calcareous sand interbedded with sandstone. It outcrops westward of the study area and dips eastward where it reaches a depth of greater than 300 m beneath the tidal flats. The molar ratios of the conservative ions within the waters of the Goliad Sand aquifer (Table 2) have been calculated from well samples within the area classified as fresh water (less than 1000 mg/l total dissolved solids) (Shaver and Baker, 1973). The Quaternary eolian plain deposits are the windblown sediments that completely cover the bedrock surface in the region. It consists of unfossiliferous, massive, fine to very fine sand, and highly calcareous clay. The eolian plain deposits contain a small quantity of slightly saline to very saline waters (1410 mg/l to 28000 mg/l chloride concentrations) (Shafer and Baker, 1973). The molar ratios of the sabkha brines were derived from sample waters taken from well sites along traverse 9 through 18. Well sites A, 17 and 18 were not considered because of their proximity to islands. Well site A is close to a small island while 17 and 18 is -47- -48- near Padre Island. Chemical trends (Figs. 5-8) indicate a possible influence on these well sites by the fresh water lenses of these islands. In order to more accurately define the average molar ratios of the brines, these well sites were not included in the determination of these values. In general, the molar ratios (Table 2) indicate that the dominant source region for the sabkha brines is the Laguna Madre. This conclusion is supported by a field observation in March, 1980 of surface flooding in the area produced by winds pushing Laguna Madre waters onto the tidal flats. Although the molar ratios of the Gulf of Mexico and the sabkha brines are similar, the Gulf of Mexico is not the direct parent of the sabkha brines. If this was the case, the chemical trend across the sabkha (Figs. 5-8) would show an increase in concentration with increasing distance from the Gulf. The opposite trend is observed. If fresh water was effecting these trends, then one would expect the molar ratios of the brines to be different from that of sea water. In addition, the Laguna Madre waters more clearly reflect the brine molar ratios. Variability in Laguna Madre water composition is seen in Figure 9. This variability is attributed to the isolation of the lagoon from the Gulf of Mexico. River input into Laguna Madre along with precipitation and high evaporation rates, produce this variability. This variability is also reflected in the molar ratios of the brines. Solute behavior within the sabkha brines is characteristic of a parent water body, similar to sea water in composition, but effected by the evaporative concentration process. Bromide, chloride, sodium and magnesium exhibit a Type I solute behavior (Eugster and Jones, 1979) while calcium, strontium, and sulfate exhibit Type II behavior. -49- MOLAR RATIOS Source Goliad aquifer Eolian plain seawate 1' August ICWW August canal March ICWW March canal August trench August shallow August deep March trench March shallow March deep August and March 17 & 18 August Well A K/Mg 0.970 0.057 0.175 0.153 0.143 0.137 0.145 0.145 0.151 0.139 0.125 0.131 0.129 0.166 0.163 c1/ Mg 35.4 11.7 9.7 16.7 12.6 12.6 16.8 11.1 11.5 10.2 8.5 9.4 9.0 9.9 13.1 Na/Mg 73.2 11.1 8.2 11.1 10.3 8.6 8.9 8.2 8.9 8.1 7.9 8.0 7.9 9.0 9.0 K/Cl 0.300 0.005 0.018 0.009 0.011 0.011 0.009 0.013 0.013 0.013 0.015 0.014 0.015 0.017 0.012 Na/Cl 21.290 0.944 0.852 0.662 0.816 0.685 0.531 0.741 0.782 0.805 0.949 0.849 0.885 0.925 0.689 Na/K 74.50 194.67 47.00 72.44 71.44 62.93 61.45 56.47 58.72 60.39 63.34 61.09 61.51 54.37 55.20 Table 2 -50- A plot of sulfate and calcium versus chloride (the conservative indicator of the degree of water removal)(Fig. 10) indicates that both calcium and sulfate fractionate concurrently. From this is it evident that calcium and sulfate fractionation is a result of the precipitation of gypsum. This geochemical behavior is supported by the field observation of selenite crystals occurring ip.si£p at various well sites. The molar ratios of the conservative ions are similar to those of sea water; however, those ratios involving potassium show a slight depletion in that ion. This indicates that a minor amount of solute fractionation of potassium occurs within the interstitial brines (Table 2). A possible mechanism for this is cation-exchange with clays. Ililite, which has been reported in the area (Fisk, 1959; Van Andel and Poole, 1960) has a cationrexchange capacity of 10 to 40 meq/lOOg of clay (Grim, 1968) and therefore could produce this observed fractionation of potassium. Calcium, magnesium and sodium can also be exchanged by illite; however, potassium is the most easily "fixed" ion in illites (Grim, 1968, p. 220). The other ions can be taken up by illites and with changing conditions be released again into the solution. However, this fixation of potassium by illites can permanently remove it from solution and, therefore, account for its slight deficiency. -51- .moum moamamm onu mom Asmaav xwam so voouooou mcofiumLuaooaoo adaaxwa one .esawaaa .ommuo>m ou muomom « seeps sees Amuse o.H ssse .umswae .sumes s scans sees Amuse o.H ssse .eoue: .momes x «asaexma "Asssev ewes asoexa: sese .somes so eases .H: e o «asaecea “Asmsev ewes masses: sesH .mumes so asses .e: e z Aswmcs>m “Asssev ewes gasses: sese .somes so asses .Hz.e z eaaasxsa ”Asssev ewes gasses: sese .moees so euuoz .ez m A Redefines “Asmsev ewes excuse: sese .mumes so escoz .Hz.m e «ammum>m "Asmsev ewes assess: sese .somes mo eucoz .ez m s Amssev Lacseae masses: ssse .momes m Amssev sensuous use Heme sesame m.ee ssse .mesu o ANAsH- Assesses see Hem: sesame m.m ssse .mumes s Amssev Assesses use Hem: assume s.o ssse .msmes m AmssH- censuses use Heme assume m.e case .somes a Asksev Lonesome mousse H.o mesa .somes o AsssHv museuumm sesame H.o NssH .mumes s Asssev use; mousse H.o ess- .mumes e mosses mess: muszem onseoou amass; .mmom sees Haves was: one AA ouwmr- mcswmd How mum flaw-u sewn-QC mmHEmzuo mwGHGD om 9:;th QH vmmfi mumuumH on .AQM um mHn-mm. -52- ’ Q % .8©o to ©© Q ..' . (9 \ H ®z©® . 0 N M © K JC) 9 L 2C) l5 IO 5 O a- Ca Figure 9. Trilinear plot of samples and Laguna Madre waters. \ ($3 4.0 ‘ g.» 0 L06 so.(mg/I) m b Hafite ppt s0< ::¢o« \\ O E O e) co Siww 1 3"?- a—rHalite ppt zo« l-Seawoter A-Loquna Madre o-This Study x'Amdurer Study O-Basin Area 3:0 410 510 6:0 LOG Cl (mg/I) Figure 10. Concentration trends of samples. HYDROLOGY The hydrology of an active sabkha has long been the subject of debate. Because of their inaccessibility and inhospitable environments, little hydrological data has yet been obtained for an active sabkha. This has resulted in the formulation of the various hydrological models mostly from theoretical considerations, laboratory experimentation, and the presence and zonation of the various evaporitic minerals (including dolomite) within the sabkha. The four major hydrological models that have been proposed are; a) seepage reflux (Adams and Rhodes, 1960), b) capillary action (Friedman and Sanders, 1967), c) evaporitic pumping (Hsu and Siegenthaler, 1967) and d) flood recharge (Butler, 1969). The validity of these hydrological models are based on their ability to generate highly concentrated brines and produce a high enough flow rate through the sabkhas sediments to explain the presence of recent dolomites, and the great thicknesses of ancient supratidal dolomites (Friedman and Sanders, 1967; Hsu and Siegenthaler, 1969). It has only been recently that actual hydrologic field data for an active sabkha has been obtained (Kinsman, 1969; Hsu and Schneider, 1973; Amdurer, 1978). Only Amdurer (1978) has integrated the hydrology of the system with the geochemistry of the brine. However, this study did not accurately define the system because of the limited area of sample sites. One objective of this thesis is to test the hypothesis that evaporitic -54- _55_ pumping is the major hydrologic process operating within the Laguna Madre sabkha. This has been done with hydrologic data obtained from manometers at various locations and depths within the sabkha. This section will first define each of the hydrological models, then present the field data, and from this evaluate the proposed hypotheses. Seepage Reflx - Reflux action is defined as the seaward escape of a dense brine (King, 1947). In the seepage reflux model the sabkha is sporadically flooded and this water collects into evaporitic lagoons (Adams and Rhodes, 1960; Deffeyes et a1, 1965). In these lagoons high evaporation results in an increase in surface water concentration and density. This increase in density causes the surface water to sink to the bottom of the lagoon. There the hypersaline brines, because of their high density, have a hydrodynamic head which results in these waters percolating downward through the sediment. Once within the sediment these hypersaline brines reflux seaward. The hydrodynamic head necessary for the waters to seep into the sediment and the resultant flow rate is governed by the sediment permeability. The generation of the hypersaline brines only occurs on the surface of the evaporiic lagoons and not within the sediment (Adams and Rhodes, 1960). According to this model the water table should be above the piezometric surface with both dipping seaward (Fig. 11). Therefore, net water motion is downward and seaward. If this process were operating within the sabkha one would expect to find the density and salinity of the waters to decrease with depth. A density inversion will occur where the more dense waters overlie less dense waters; however, this density inversion at depth is not necessary for this -56- hydrologic mechanism to operate. Capillary Action.- This model postulates that the vadose water ascends vertically by capillary action (Friedman and Sanders, 1967). The ascending water is concentrated by evaporation and forms a caliche like deposit near the sabkha surface. Because evaporation rate is greater than rainfall, these efflorescent crusts are only periodically dissolved by rainwater. The high evaporation rate could, with solution of the efflorescent crusts, result in an increase in concentration of the interstitial brines. A hydrostatic head is not necessary to drive the waters upward (Hsu and Schneider, 1973); instead this flow is accomplished by capillary action. The interstitial water adheres to the sediment within the vadose zone, the near surface waters within this zone evaporate and are replaced by additional water from below (much like a candle wick sucking up wax to replace that burned away). The piezometric surface is equivalent to the water table and the piezometric potential does not vary with depth. A continental source region for the interstitial water and a seaward sloping water table are implied for this model (Hsu and Schneider, 1973). Evaporative Pumping - In evaporative pumping a vertical hydraulic gradient is induced within the interstitial waters "by an upward decrease in hydrodynamic potential during evaporation" (Hsu and Siegenthaler, 1969, p. 11). The flow rate within the sediment is primarily a function of evaporation with sediment permeability having a negligible affect. According to this model the piezometric surface is higher than the water table (Fig. 11) and the piezometric potential increases with depth (Hsu and Schneider, 1973). This results in a -57- uglegféfiLE pg mnglC suRFACE REF LUX requflyc " SURFACE .— ----- /‘_‘_"_ évmunve mm Figure 11. Reflux and evaporative pumping models. -5 8.. vertically ascending brine (Amdurer, 1978). The source regions for the interstitial waters can be both continental and marine. If the source is continental, then a seaward dipping water table above high tide level should be present. If the water is marine in origin, then the water table, should be equal to or slightly below the high tide level (Hsu and Schneider, 1973). The development of the Evaporative pumping model was based on laboratory experiments by Ben and Siegenthaler (1969). It was designed to investigate the potential for this hydrologic mechanism to operate along a highly evaportic coast, to determine what governs the flow rate and, most importantly, to see if this mechanism can generate a flow rate high enough to account for the origin and thickness of recent and ancient dolomites. In a preliminary hydrologic investigation of the Abu Dhabi sabkha, Persian Gulf, Hsu and Schneider (1973) determined that the piezometric surface was higher than the water table and from this concluded that evaporative pumping was indeed the major hydrologic process operating. However, there were some inherent weaknesses in this preliminary investigation; the major one being some manometers penetrated a zone below an impermeable limestone layer. It was at these sites that the hydrologic head increased with depth. However, this increase could have been produced by a regional head that pushed the trapped waters upward once the impermeable limestone layer was penetrated (Amdurer, 1978). Additional hydrologic data has to be obtained before the hydrologic mechanism operating within the Abu Dhabi sabkha can be determined. Flood Recharge - Flood recharge occurs when storms and high winds -5 9— push sea water onto the sabkha surface and this water then seeps downward into the sediment (Butler, 1969). Flood recharge is not similar to seepage reflux because a density head is not necessary for the waters to descend vertically. The interstitial waters are concentrated by evaporation and solution of soluble salts. According to this model the piezometric potential should not change with depth. A direct correlation should exist between the frequency and extent of flooding, the evaporitic minerals present and the chemical composition of the hypersaline brines (Butler, 1969). The frequency and extent of flooding is related to wind velocity, wind direction, duration and topography. The periodic flooding that occurs with flood recharge rejuvenates the interstitial brines. This hydrologic model allows for local evaporative pumping to occur wherever the frequency and extent of flooding, and the amount of evapotranspiration will determine to what degree it occurs. The recharge water can either be fresh water or marine and movement of the subsurface waters is in direct response to variations in the water table. In the study area, if this were occurring, one would expect the piezometric potential at each well site to be near that of the water table and salinity should gradually increase with depth. -60- Observations A plot of the piezometric potentials as a function of depth for the August and March sampling periods is shown in Figures 12, 13 and 14. The equations used to determine the piezometric potentials and and example are given in Appendix E. During both August and March the piezometric potentials for the majority of the well sites were above that of the Intracoastal Waterway and the Gulf of Mexico In addition, the piezometric potentials for the well sites along traverse 9 through 18 (Fig.3) show no significant trend with increasing depth. The piezometric surface for well sites 14 through 17 form a plateau with decreasing potential towards the east and west. Since water flows from high to low piezometric potential, it is evident that the interstitial water flow on the west side of the plateau is toward the Intracoastal Waterway, and on the east side towards the gulf. The hydrologic data for the Laguna Madre sabkha is consistent with a flood recharge model and this is in agreement with the hydrological interpretations for the area by Masson (1955) and Fisk (1959). The data does not support evaporative pumping, seepage reflux or capillary action as the major mechanism operating within the area. The evaporative pumping model predicts that the area with the highest brine concentration (at the salinity plateau) would have the lowest piezometric potential because water movement should be toward these sites of highest evaporation. Instead, these sites have the highest potentials with water movement away from them. In addition, if evaporative pumping of marine water were occurring, as the geochemistry suggests, the piezometric potential for all the well -61- sites should be lower than than that of sea water. Instead, the majority of well sites have higher potential (Figs. 12 and 13). Evaporative pumping can operate to a minor degree at some well sites, but it is not the major process controlling interstitial water flow in the area. A seepage reflux process with a marine source region would require that the piezometric surface dip toward the Gulf. Also, a definite decrease in piezometric potential with increasing depth at all the well sites should occur. Figures 12 and 13 show that although the piezometric potential at most well sites are greater than that of the Gulf, the interstitial waters are not refluxing into the Gulf or into the Intracoastal Waterway. In addition, there is not clear decrease in piezometric potential with increasing depth at the well sites. Therefore, seepage reflux is not the major hyrologic process controlling interstitial water flow in the area. In capillary action a continental source region with a seaward dipping piezometric surface is required (Kinsman, 1969). The plots of the piezometric potentials for the various well sites indicate that major water flow is not from the continent. In addition, the geochemistry of the waters indicate that these brines are not continentally derived. Both of these observations strongly suggest that capillary action is not the major hydrologic process governing water flow within the study area. However, it is possible that capillary action is occurring to a limited extent near the surface of the study area. The interstitial waters of the Laguna Madre sabkha appear to respond to a hydrodynamic process in which interstitial water is -62- recharged by periodic flooding with floodwater seeping into the sediment. Major interstitial water movement is in direct response to fluctuations in water table. This interpretation is consistent with the geochemical predictions about the system discussed in the previous section. ~63— .304 AUGUST Piezometric Potemiol 160‘ _.’ l I I ,- \ I x \ l \ l \n1 ‘0" /Scovoler 1 O'Ttench A'Shollou - 0'00" 60‘ #— 6 £6 % 8 3 J§h 6 IE u m m I6I7 l8 Figure 12. -64- M A R C H 1801 Piezomemc Potennon /Soovoter 30‘ . O'Tnnch 50‘ A-Shono- ‘ 3'00" 404 9 a A c 7 6 5 452% I3 l‘2 I'l I3 (5 1'6 1'7 (8 ‘ WELLS Figure 13. ~65- Oosin Ana August Piezometric Potential o ............................................................ ‘ no ""0 o-Trencn A-Shollow 80* mo.» 6 AMZ Aims AIM AM: '0 Figure 14. -66.. Discussion The piezometric potentials shown in Figures 12 and 13 support flood recharge as the major hydrological process for the Laguna Madre sabkha. Bromide concentration is directly related to the degree of brine concentration because of its conservative nature. Therefore, bromide contour maps as a funcion of depth can be informative as to the hydrologic nature of the interstitial waters and can be used to relate the geochemisty of the system to the hydrology. The bromide contour maps for the August and March sampling periods (Fig. 15) indicate that there is an increase in bromide concentration with depth, but that the increase with depth is not continuous across the sabkha. This salinity change with depth is consistent with the flood recharge model. Figure 15 shows that the recharge waters move down as salt fingers rather than as a continuous salt sheet. This pattern of flood recharge was also observed by Butler (1969) for the Abu Dhabi sabkha with the use of chloride contour maps. The bromide contour maps (Fig. 15) also indicate that although actual bromide concentration does vary for the August and March sampling periods the pattern of recharge is similar. A rapid decrease in bromide concentration at well sites 17 and 18 and for the shallow and deep waters at well site A is seen in figure 14. A greater amount of flood recharge to these well sites cannot account for this decrease in bromide concentration because the molar ratios of the conservative ions within the waters of these sites are slightly different from those of the other sites (Table 2). In addition, the dilution effect at these sites is less for the drier -67- month of March. This data indicates that the decrease in bromide concentrations for waters at well sites 17 and 18 is due to a dilution effect by the fresh water lens of Padre Island, and that the decrease in the shallow and deep waters at well site A is due to the effect of a fresh water lens from a nearby island (Fig. 3). The decrease in brine concentration along the western side can be caused by two factors. The first is that this area is being recharged in part by continental groundwater (Amdurer, 1978). The bromide contour maps (Fig. 15) do not indicate dilution from below or westward. Therefore the mixing zone between the sabkha's marine groundwater and continental waters most likely occurs westward of well site 9 or at a depth below 3.8 m. The second reason for the dilution effect on these western sites is that they are being rejuvenated more often. Since flood recharge is the major hydrological mechanism in the area, one would expect that the areas that flooded most often would have interstitial waters with comparatively lower concentrations. The well located on the western side of Laguna Madre are at a lower elevation and also near the Intracoastal Waterway and gas well access canals (Fig. 3). Since these well sites have a higher potential for flooding than those on the eastern side, one would expect that their water concentrations would have lower values, this appears to be the situation. ELEVATION (cm) DEPTH FROM SURFACE (cm) ~68— Topogroph y o 3 =2 KILOMETERS 3 SP0". PADRE if“. 001 BANK ISLAN o . 2001 \ {I \\ I. \ Ioo« ’ \‘ GULF o I Br (mg/l) August 9 I3 I2 II l8 0 . . . . . . 5 e/\/§ g‘ I90‘ (6 \ o I’ C |\‘\ 3804 H \k“ Br (mg/I) March 9 8 A C 7 6 5 432W I3 II l6 0 i 1 L 1 L I UK; T, 49/ ,’ l ( / \/¢o Hg, 7'8; \ a v {300 (4so\ ) M 00$/ I /~<,o ,1 § . Iso- K > ) [1 WW 3 {Ii Figure 15. Bromide contour maps. _69_ Basin.Area The brines within the basin area, well site 10 and well set B, are chemically distinct from those found at the other well sites. Going from east to west, well set B (Fig. 3) includes AM2, AM3, AMA, and AM5. The wells within the basin are characteristically lower in sulfate content and, when considering cation percentages, the basin brines are higher in Ca and lower in Mg (Fig 8). In addition to this area containing distinct waters from those found in the other areas, the chemical trends and molar ratios of the well waters indicate that there are two distinct vertically stratified water masses within the basin. In comparison, the more shallow water is characteristically higher in concentration and has a greater Mg/Ca and lower Cl/Br molar ratio (Table 4). The ionic ratios of this upper water mass and the waters of the gas well access canal are similar; Table 4 demonstrates that they both have similar C1/Br ratios and indicates that the Laguna Madre water is the most likely parent for this upper water mass. The higher Cl/Br molar ratio of the deeper water mass cannot be explained by simple evaporative concentration of Laguna Madre water since the ratio between these conservative ions would not be affected by just the removal of water. Therefore, this deeper water mass is either a Laguna Madre water that has been extensively modified by solute fractionation or has a different source region, or is a mixture of Laguna Madre waters and groundwater of continental origin. A Pleistocene river valley underlies the basin area (Fisk, 1959). Since coarser sediment fills this valley continental water could be ssessv mumsssou sous «« oaaamm o: I 4 : 30700 a: C I.O| l 20., m 0. O LOOd—\ '07.. 0%. 0.99 T r 1 1 0 IO 20 30 40 Temp (° C) Figure 17. Density vs. Temperature. -100- |.O4‘l ér- 0°C £5 L037 20°C C) ‘\\ 40°C (filLOZ V >4 1: no: a: c: a: CD 11x33 0.99 T ' I ' 0 IO 20 30 40 Salinity ( g/kg) Figure 18. Density vs. Salinity. -101- m }.__..._.....___ I 04* IO3‘ __..__./____ IOI‘ Density (g/Cms) IOO' 0 99411— . , (9° 0 10 20 30 40 Temp (T) Figure 19. Density vs. Temperature vs. Salinity. -102- Table 6 Laboratory Density Measurements measured predicted measured - Sample Temp(°C) density(g/cm3) density(g/cm3) predicted 12D 25.5 1.17787 1.17789 -0.00002 28.7 1.17638 * -------- 30.5 1.17559 1.17552 +0.00007 32.2 1.17477 1.17470 +0.00007 33.2 1.17425 1.17421 -0.00004 98 27.5 1.11427 1.11431 -0.00004 30.0 1.11328 * -------- 31.7 1.11244 1.11256 -0.00012 32.8 1.11206 1.11209 -0.00003 33.9 1.11141 1.11161 -0.00020 ICWW 24.2 1.02571 1.02551 +0.00020 29.2 1.02399 * -------- 32.2 1.02299 1.02299 0.00000 35.7 1.02177 1.02173 +0.00004 38.0 1.02099 1.02085 +0.00014 * Measured density at this temperature was used to generate salinity. Appendix D Analysis Well 8T AMlT CAM2 7T 6T 5T 4T 3T 2T ICWW 13T 12T 11T 14T 15T 16T 17T 18T 9S AMlS 78 68 SS 48 138 128 118 148 158 168 Temp 29.0 32.0 32.0 34.0* 22.0 31.0 32.0 32.0 32.0 22.0 34.0* 33.0 33.0 34.0 32.0 32.5 32.5 30.0 31.0* 32.0* 32.0* 31.0* 33.0* 33.0* 33.p 33.0 32.0 33.0 34.0 34.0 34.0 34.0 32.0 pH 6.60 7.20 7.08 8.28 6.68 6.40 6.55 6.50 6.40 7.05 8.20 6.85 6.50 6.60 density Ca 1.1101 1.1225 1.1179 1.0230 1.1400 1.1309 1.1324 1.1239 1.1240 1.0202 1.1505 1.1634 1.1488 6.90 6.80 6.80 7.05 7.10 6.88 7.03 7.15 6.90 6.80 6.88 6.65 7.00 6.90 6.95 6.80 6.70 6.80 1.1654 1.1705 1.1367 1.0658 1.1147 1.1144 1.0630 1.1272 1.1353 1.1357 1.1252 1.1286 1.1437 1.1694 1.1692 1.1715 1.1750 1.1803 1080 970 970 560 990 850 810 930 820 1400 440 670 490 530 540 470 520 770 780 870 990 960 920 1010 740 1030 870 660 420 470 440 420 430 -103- “8 6300 7200 7000 1300 7600 6900 8200 7000 7000 1600 1100 9100 10400 9400 11100 10900 10600 6800 3700 7000 6900 3500 7000 6900 8400 7100 7600 8300 9900 10400 10300 9700 8700 Na 49300 56600 57700 1260 66700 61200 60600 53600 54100 13900 11500 66800 77100 70900 82300 75200 81700 72800 30000 53200 51600 29300 61300 68500 64800 60000 63800 73500 78000 77900 81500 80500 89500 August Analysis 1550 1630 1750 300 1870 1870 1930 1640 1570 370 270 2160 2230 2020 2282 2370 2410 2200 910 1580 1460 900 1860 1740 1890 1560 1640 2010 2410 2310 2570 2480 2680 C1 88662 101950 102605 22028 118638 108398 109222 96697 96333 25327 20209 119575 137666 124784 147919 135535 146812 127753 54117 96388 93360 51227 109988 120573 115766 108480 113631 128992 136833 137747 144075 142971 151867 SO 11952 11952 13475 3206 13142 12300 15141 12643 13842 4344 2281 17456 19473 19784 19274 20472 17807 10310 6806 11952 12472 8875 10972 11360 15621 10626 13221 15436 20485 20903 19883 16680 19883 HCO Sr 400 20 400 20 253 21 155 8 230 17 258 17 263 20 303 18 290 22 655 17 125 5 295 20 135 14 230 14 20 15 16 24 26 25 23 15 29 29 25 28 26 17 11 ll 11 11 11 157 95 160 205 365 355 220 300 285 265 255 325 270 230 365 138 240 70 Br 310 310 329 68.9 405 384 342 367 351 67.8 59.8 423 489 413 511 515 399 186 320 338 154 371 397 400 386 372 430 523 505 510 525 564 Well 173 188 9D 8D AMlD 7D 6D 5D 4D 2D 13D 12D 11D 14D 15D 16D 17D 18D 10T AM5T AM4T AM3T AMZT lOS AMSS AM4S AM3S AMZS 10D AMSD AM4D AM2D Temp 31.0 30.0 33.0* 33.0* 31.0* 33.0* 33.0 32.0 32.0 33.0 32.0 33.0 32.0 34.0 34.0 34.0* 31.0 30.0 30.0 31.0 31.5 31.0 31.0* 32.0* 31.0* 31.0* 31.0* 31.0* 33.0* 31.0* 31.0* 31.0* August pH density Ca 6.80 7.05 6.75 6.70 6.90 6.65 6.45 6.50 6.50 6.60 6.50 6.85 6.85 6.70 6.73 6.65 6.80 7.10 6.80 6.65 6.65 6.70 6.70 6.45 7.13 6.68 7.03 6.70 6.48 6.60 6.68 7.50 1.1495 1.0851 1.1134 1.1150 1.0834 1.1261 1.1265 1.1402 1.1352 1.0854 1.1423 1.1715 1.1697 1.1725 1.1764 1.1836 1.1674 1.0890 1.1004 1.0920 1.1135 1.0700 1.1031 1.1057 1.0328 1.0570 1.0303 1.0258 1.1018 1.0894 1.0769 1.0368 640 690 990 1460 750 1070 990 790 960 1080 920 520 550 510 550 490 600 790 1100 1120 1090 930 1040 1530 690 1010 940 740 1270 1660 1570 940 M8 8100 4300 6900 7300 4600 7600 8000 8100 8200 5500 8100 12500 11800 12100 12500 11600 10700 5100 5400 4600 6400 3400 6100 5700 1700 3100 1700 1400 5200 4500 3700 2100 -104- Analysis (continued) Na 76000 41500 58600 58400 39800 60600 58800 70700 65400 38800 72800 80700 79100 82100 85900 87200 79400 42900 46700 43900 57300 34900 49800 49800 17400 28000 15800 13400 49000 42300 36200 1880 2500 1310 1560 1400 1230 1780 1520 2050 1530 1050 2060 2560 2390 2590 2670 2640 2620 1330 1390 1220 1600 970 1470 1510 440 780 380 350 1460 1290 950 480 C1 134153 72628 104611 106496 67787 110321 108012 124871 118514 71019 130726 148721 144035 150166 158087 156630 143865 77492 82896 79559 102305 62733 89944 90599 30664 52355 27983 23769 87942 78119 68240 33384 SO 13500 8396 11958 11505 12642 11616 12010 14637 12301 10074 11491 20693 20639 20179 18973 20139 17782 8144 10434 5956 10643 4134 10245 9043 3449 2843 4134 3393 8311 5621 2151 4979 HCO Sr 18 50 21 19 12 25 25 25 28 20 17 13 13 12 12 11 12 47 28 23 23 18 21 30 11 16 13 11 28 35 31 13 183 210 380 368 200 395 450 400 450 486 305 260 460 150 390 120 168 225 645 425 425 820 438 385 345 450 511 175 385 335 735 268 Br 464 242 335 329 209 397 378 429 390 228 427 536 507 526 561 578 515 267 283 262 317 196 295 305 57.6 144 64.1 54.3 305 239 212 70.7 Well 8T AMlT CNAL 7T 6T 5T 4T 3T 2T ICWW 13Ts 13Td Dayl Day2 12T 11T 14T 15T 16T 17T 18T 98 8S AMlS 73 68 SS 48 3S 13S 128 118 Temp 21.0 21.0 23.0* 22.0 21.5 21.0 22.0 21.8 21.0 20.5 24.5 24.0 22.0 pH 6.35 6.55 6.70 8.15 6.40 6.40 6.65 6.20 6.10 7.08 8.18 6.60 6.80 290 5* “-- 29.5 20.5 21.5 21.5 21.5 21.9 21.0 21.0 23.0 23.0 23.0 23.0* 23.9 24.0 23.0 23.0 23.0 21.0 22.0 7.50 7.10 6.70 6.30 6.40 6.50 6.60 6.90 6.70 7.50 6.95 7.20 6.80 7.00 6.60 6.60 7.20 7.35 6.98 density Ca 1.1116 1.1128 1.1203 1.0259 1.1411 1.1382 1.1394 1.1322 1.1356 1.0308 1.0259 1.1568 1.1513 1.1048 1.1248 1.1696 1.1526 1.1704 1.1713 1.1733 1.1417 1.0792 1.1161 1.1156 1.1122 1.1152 1.1343 1.1438 1.1277 1.1309 1.1513 1100 1190 950 680 1080 1130 910 1120 1000 1270 440 720 720 1450 1170 710 670 570 540 560 990 940 1040 1080 960 920 1110 880 1050 960 750 -105- M8 6560 6820 8000 1330 8060 7760 8430 7880 8720 1820 1366 10130 9740 6470 8600 11750 10650 12200 12280 11940 7390 4460 6970 6920 7210 6450 7770 9150 7470 7970 9100 1.1791 520 12660 1.1725 530 11840 Na 47700 54500 55700 11200 69400 66900 68100 65700 63800 12600 11100 77300 77700 48400 59500 79300 74700 77100 76100 81600 72800 36100 54700 53700 49500 55500 66600 71200 60000 60000 73700 82400 78100 March Analysis 1440 1330 1600 310 1970 1790 1970 1630 1670 370 300 1940 1860 1500 1730 2080 1870 2110 2170 2280 2200 1100 1470 1360 1450 1480 1690 1990 1500 1610 2090 2380 2270 C1 87132 98628 105000 19992 124680 120969 122681 119023 117055 24392 19795 138964 90000 88150 109099 144655 134610 143799 140605 150157 128861 65632 98048 96937 89730 100130 120586 128074 109091 108602 131197 151006 142094 SO 11684 11470 11254 13830 16407 21560 21361 137 154 317 280 177 175 200 172 228 205 186 261 182 18 16 l8 17 19 15 17 l4 14 18 16 24 28 13 13 11 11 ll 17 21 19 19 18 19 22 20 22 19 13 11 321 326 362 65.9 447 437 452 417 410 48.0 62.1 486 421 313 382 526 455 535 532 555 463 242 355 362 303 375 442 452 410 408 468 555 529 Well 148 158 168 173 183 9D 8D AMlD 7D 6D 5D 4D 3D 2D 13D 12D 11D 14D 15D 16D 17D 18D Temp 21.5 21.5 21.5 21.0 21.0 23.0 23.0 23.0 23.2 24.0 23.5 23.0 23.0* 23.0 22.5 21.0 21.8 21.8 21.5 21.5 20.8 21.0 pH 6.70 7.00 6.70 6.90 6.65 6.45 7.25 6.52 7.39 6.75 6.55 6.60 6.45 6.48 6.50 6.95 6.88 6.35 6.90 6.60 6.80 7.10 density Ca 1.1772 1.1830 1.1892 1.1616 1.0824 1.1115 1.1190 1.1311 1.1312 1.1311 1.1465 1.1312 1.0763 1.0871 1.1497 1.1800 1.1739 1.1716 1.1845 1.1900 1.1723 1.0951 510 500 450 560 840 940 1210 780 1010 1040 860 1070 720 1210 860 490 510 490 500 480 560 810 -106- March Analysis (continued) Mg 11830 12470 12040 9130 5200 6800 7840 8690 7790 8080 8830 7990 4760 5290 8670 13660 11860 11630 11740 11800 10560 5240 Na 81000 83900 88800 80800 42200 53800 54800 61700 61900 6300 70300 61100 33900 39200 71100 84600 81700 80600 88900 91000 83500 45000 2470 2620 2730 2510 1280 1410 1380 1760 1760 1580 1970 1520 870 1030 2000 2440 2500 2290 2640 2720 2600 1330 C1 147606 153990 160396 144205 76545 96366 101801 111836 112490 114322 126965 111427 61250 71390 127962 157703 148519 146318 160102 163093 149494 80719 SO 20172 20172 20172 13830 8479 12486 11906 15577 12100 12872 14032 12486 *Field temperature and therefore field density estimated 8 555 581 8 593 12 44 17 15 16 22 21 21 23 12 16 18 11 10 516 275 329 359 380 411 403 474 351 221 270 458 560 532 9 528 9 574 8 631 9 543 45 294 14T August analysis performed on non-acidified formaldehyde treated sample. Appendix E Hydrology ~107- APPENDIX E - height or well uepth from top - land elevation " lullb’th 2):- ".4811 - level of water (‘1 (measured) Determining the piezometric potential pipe above the ground (measured) of well pipe to water (measured) with sea level as reference (measured) pipe (measured) in well pipe with land elevation as reference Pictumetric potential = awry hp = ;r"‘sunrACE W B fif~‘SEA LEVEL 0 h IOOcm P -z ¥L.DATUM P , .;¥§§RE§_} Well 6 shallow for August 1979 A = 15.24 cm D - B = 190.0 - 55.88 = 134.12 = hp B = 55.88 cm D - (A + C + 100 cm) = C - 19.20 cm 190.0 - (15.24 + 48.79 + 100) = 25.97 = -Zp D = 74.80 cm Density = DEN = 1.1353 ¢ = uuxgnp + 2p) = 1.1353(134.12 - 25.97) = 122.78 Seawater 100 cm Zp = 0.0 DEN = 1.023 Piezometric potential = ¢ = 1.023(100 + 0.0) = 102.3