THE :M.A.RFI·~ SC1ENCE LIBRARY Tho 7J:;is·1;;.;il:1' of Texas at Austia P. 0. i-':~l{ 1267Port AruriSaS, Texas 78383-1267 A Report To E.I. DuPont de Nemours & Co. June 30, 1978 The Ecological Significance and Fate of Carbon Tetrachloride and FREON* 113 in the Estuarine Environment by William Brogden Carl H. Oppenheimer The University of Texas Marine Science Institute Port Aransas Marine Laboratory Port Aransas, Texas 78373 *Trademark E.I. DuPont de Nemours & Co. TABLE OF CONTENTS Page 1 List of Tables • 4 List of Figures . . . . . . . . . . . . . . . . . . . . 6 I Abstract •• . . . . . . . . . . . . . . . . . . 8 II Introduction • 12 III Chromatographic and Extraction Techniques 12 A. Gas Chromatography B. Recovery cc14 and Freon 113 From Solid Samples 16 IV Physical and Biological Processes Effecting Carbon Tetrachloride and Freon 113 in Estuarine Environments 19 19 A. Loss to Atmosphere 27 B. Interaction with Sediments 33 v Methodolog~ Results and Discussion . A. Short Term Experiments with Phytoplankton 33 1 Light/dark bottle experimental methods 33 34 2 Stock culture preparation •.•.. 36 3 Growth rate experimental methods 4 Experiments with natural populations 37 and CC1 4 . . . . . . . . . . . . . . . . . 5 Experiments with cultured phytoplankton 38 and CC1 4 . . . . . . . . . . . . . . . 43 6 Growth rate experiments with CC14 . . . . . . 7 Light/dark bottle experiments with natural 50 populations and Freon 113 . . . . . . . . . 52 8 Growth rate experiments with Freon 113 . . . 9 Discussion _of short-term experiments with phytoplankton . • . . • • • • • • . • • 56 Page B. Short-term Effects on Larger Animals ·· !' • .-. .• • • 5 9 . . . . . . . 59 1 Introduction •· 60 2 Methodology • a Collection and maintenance of organisms 60 b Respiration measurement apparatus 60 c Dose control and aeration 63 64 d Recording oxygen levels and activity 69 3 Results ••.••• a Respiration experiments 69 b Short-term exposure of shrimp to • • • • • • • • • • 75 Freon 113 c Short-term toxicity experiment with trout and cc14 • • • • • . • •. • • . • • • • • 76 d Short-term toxicity experiment with trout and Freon 113 • • • • • • • 83 c. Long-term Experiments in Ponds 84 84 1 Introduction and summary 2 Pond ecosystem description 86 91 3 Description of ponds 4 Pond stocking and sampling history 98 5 Discussion of organism survival in ponds 109 6 Halogen content of animals exposed in ponds . 113 120 7 Treatment of pond oxygen data • 158 8 Nutrient analysis 9 Fluoride analysis • • 161 164 VI Summary and Recommendations 170 VII Bibliography • . . • • • • • . 173 VIII Appendix A Nutrient Analysis Data 1 LIST OF TABLES IV-1 Physical properties of cc14 , Freon 113 and oxygen. IV-2 Results of least squares fits to the equation Y = AeBt for rate of halocarbon loss and oxygen deficit decrease. IV-3 Concentration of cc14 in sea water exposed to various amounts of sediment. IV-4 Concentration of cc14 and Freon 113 in sterile and nonsterile sea water. IV-5 Halocarbon concentration from test bottles left in ponds. V-1 ASP-2 medium for PR-6 culture. V-2 Respiration measurement results obtained with Penaeus setiferus at 34 ppt salinity and 26° c. V-3 Respiration measurement results obrained with Penaeus setiferus under several different salinity and temperature regimes. V-4 Summary of 48-hour tests of penaeid shrimp exposed to Freon 113. V-5 Summary of cc14 toxicity in large outdoor pond, starting December 12, 1973. V-6 Summary of Freon 113 short-term toxicity experiment in large outdoor pond with Cynoscion nebulosus (spotted sea trout) starting March 18, 1974. V-7 Estimates of the accuracy, repeatability and lower limit of detection of the various analyses. V-8 Diurnal oxygen changes before and after the first introduction of halocarbons into the ponds, in percent saturation. 2 V-9 Organism observations after the first test interval, 5/9/73 to 6/19/73. v-10 Organism observations after the second test interval, 6/20/73 to 8/29/73. V-11 Organism observations after the third test interval, 8/30/73 to 10/26/73. V-12 Organism observations after the fourth test interval, 10/27/73 to 1/9/74. V-13 Organism observations after the fifth test interval, 1/28/74 to 3/8/74. V-14 Organism observations after the sixth text interval, 3/12/74 to 4/24/74. V-15 Organism observations after the seventh test interval, 4/13/74 to 6/15/74. V-16 Organisms sent to the pathologist. V-17 Halocarbon content of organisms exposed in long-term pond experiments. Samples taken June 15, 1974 after 15 months of exposure. V-18 Short-term average oxygen concentrations in ponds. V-19 Short-term average oxygen "delta" values. V-20 Analysis of some representative temperature data from the ponds. V-21 Oxygen concentrations at 100% saturation for some representative pond salinity and temperature values. V-22 Correlation coefficients indicating the degree of correlation in average oxygen values between the ponds for the period 5/10/73 to 4/8/74. 3 V-23 Sign test applied to average oxygen values. V-24 Sign test applied to "delta" oxygen values. V-25 Sign test comparison of pond 2 (high CC14 ) with the control ponds (3 and 9) with respect to "delta" oxygen values for two different periods. V-26 Nutrient concentration and turbidity means and standard deviations for the pre-treatment period, 1/15/73 to 5/9/73. V-27 · Nutrient concentration and turbidity means and standard deviations for the period 5/14/73 to 8/27/73. V-28 Nutrient concentration and turbidity means and standard deviations for the period 9/10/73 to 5/26/74. V-29 Fluoride analysis results. 4 LIST OF FIGURES III-1 Typical gas chromatographic curves. III-2 Digestion and distillation apparatus for recovery of halocarbons from solid samples. V-1 Light/dark bottle experiments with natural sample from Aransas Pass. V-2 Light/dark bottle experiments with plankton from Aransas Pass. V-3 Light/dark bottle experiments with pond populations. V-4 Light/dark bottle experiments with pond populations. V-5 Light/dark bottle experiments with Agmenellum quadruplicatum strain PR-6 in enriched sea water. Experiment one, cc14 • V-6 Light/dark bottle experiments with Agmenellum quadruplicatum strain PR-6. Experiment two, cc14 . V-7 · Light/dark bottle experiments with Agmenellum quadruplicatum strain PR-6. Experiment three, cc14• V-8 Growth curve of Agmenellum quadruplicatum at nominal concentrations of 1.0, 3.0 and 10 ppm cc14 • V-9 Various parameters of the growth curves shown in V-8 plotted vs. cc14 concentration at the end of the experiment. V-10 Light/dark bottle experiments with pond phytoplankton-Freon. V-11 Growth rates obtained in first experiment with Freon 113 Agmenellum quadruplicatum. V-12 Results of second growth rate experiment with Freon 113. V-13 Results of third growth rate experiment with Freon 113. V-14 Experimental set up for respiration and activity monitoring. 5 V-15 The systems used to provide a constant concentration of halocarbons in the air supply to the test chamber. V-16 Experimental pond layout. v-17 · Typical diurnal oxygen curve from pond ecosystems, with solar energy input, from Odum et al., (1963). V-18 Pathologists report on samples of 1/9/74. V-19 Pathologists report on samples of 6/15/74. V-20 Average oxygen concentration in ponds 1 and 8, including all periods of 2 or more days of records. V-21 Average oxygen concentrations in ponds 2, 3, 6 and 9. V-22 Short-term average oxygen "delta" values for ponds 1, the "low" Freon pond and pond ·8, the "high" Freon pond. V-23 Short-term average oxygen "delta" values for ponds 2, 3, 6 and 9. V-24 Graphs of the averages for all ponds of oxygen, oxygen "delta", salinity and temperature. V-25 Ammonia nitrogen over a four-month period for all ponds. V-26 Nitrate nitrogen measurements over a four-month period. V-27 Phosphate measurements during the period from 5/1/73 to 8/27/73. V-28 Silica measurements during the period from 5/1/73 to 8/27/73. V-29 Turbidity measurements over a four-month period. 6 CHAPTER I ABSTRACT This report describes research on the fate and effects of carbon tetrachloride and FREON 113 (1,1,2-Trichloro, 1,2,2 Trifluorethane) released into the estuarine environment. Both the air at rates compounds are rapidly released from the water to controlled by turbulent diffusion in the surface water layers. Degradation processes are very slow compared to loss into the atmosphere. As the halocarbons were rapidly released to the air, average values were used for the experiments based on the high of addition and the loss between additions. Growth and photosynthesis experiments with estuarine algae showed slight stimulation at an average of 1.0 ppm water level for both CCl4 and FREON 113 with suppression at higher levels. An attempt was made to detect short term sublethal effects using shrimp respiration measurements. However, sublethal effects would not be detected due to high individual variability of the shrimp responses. Short term acute toxicity experiments indicated lethal concentrations for penaeid shrimp close to 3.0 ppm for both compounds under laboratory conditions, and 6 to 8 ppm cc14 for seatrout (Cynoscion sp.). Seatrout survived FREON 113 levels up to 7 ppm in short term experiments. Long term exposure experiments were conducted in pond ecosystems containing common estuarine fish and shrimp. All organisms survived 5 ppm FREON 113. Seven ppm killed shrimp and 10 ppm killed fish and shrimp. All organisms survived levels of CCl4 to 4 ppm. Nutrient cycling was unaffected by the halocarbons. Concentration factors for halocarbons was generally less than one for grasses and less than 10 for animals. 8 CHAPTER II INTRODUCTION The halocarbons, Carbon Tetrachloride and FREON 113 (1,1,2-· Trichloro, 1,2,2-Trifluoroethane), are commonly used chemicals in our daily life. Under certain ·concentrations the materials may become toxic to living creatures. The following report describes experiments designed to show the ecological impact of the halocarbons in sea water to various living forms under short and long term experimental conditions. The methods for determining the amounts of the materials in water, the solubilities, loss to the air, and degradation will be described in detail as a setting to the experiment.al conditions with living organisms (Chapter III). All sample aliquots were collected by a method designed to hold the water samples out of contact with air to avoid evaporation loss of the halocarbons. Samples of laboratory air were analyzed occasionally to ensure that levels of carbon tetrachloride and FREON 113 were within OSHA recommended limits. Air samples were collected in 100 microliter syringes lubricated with water and silicone rubber septa were used to plug the needle. As solid samples were to be analyzed a methodology for halocarbon recovery had to be developed. This was accomplished by a digestion and distillation technique. The experiments on physical-chemical behavior of the halocarbons are described in Chapter IV. Evaporation rates into 9 the air were shown to be rapid and controlled by turbulent diffusion in the water layers. Adsorption by sediments was measured using typical estuarine sediments. Research on biological effects was divided into experiments involving short term effects that could be studied in the laboratory cultures and long term effects which were studied with model ecosystems in large outside ponds. Respiration and activity measurements were used as a method to detect sublethal effects of the halogenated hydrocarbons on shrimp, fish and oysters. Unfortunately, difficulties with the equipment prevented us from completing the respiration work and only the shrimp studies were completed. Many of the deleterious effects of man-made organic chemical cannot be predicted from short term tests on isolated organisms. A classic example is the accumulation and magnification of DDT related degradation products through the food chain, eventually resulting in damage to higher organisms. Thus more realistic long term exposure tests were conducted in small pond ecosystems containing as many of the natural estuarine ecosystem components as possible. To some extent this approach was experimental since not very much research has been reported on the creation and maintenance of model ecosystems large enough to contain higher organisms. Response of the organisms to the various concentrations of halocarbons was determined by time of death, respiration changes, and for the phytoplankton growth rate changes. Standard techniques such as light-dark bottle experiments were used to determine effects 10 of the halocarbons on the system tested. For long term experiments nutrients were monitored and correlated with oxygen to determine community metabolism changes in the outside experimental ponds. For the long term exposure tests, organisms were selected to represent the phytoplankton, marine grasses, crustaceans, invertebrates and fishes. Grasses and organisms were examined for halocarbon concentration effects. This was determined by extracting various tissues and determining halocarbon concentration. Some animals were examined for pathological tissue change. It was difficult, because of the low solubility and rapid loss of the materials to air, to maintain a given concentration of the halocarbons in the experimental ponds. Thus a "target" concentration approach was used in which periodic amounts of the materials were added to the water and the loss determined during the interim period. From these data the average concentrations were determined. 11 Acknowledgements are given to the DuPont de Nemours and Co. who supplied the basic funding for the research. The University of Texas Marine Science Institute Port Aransas Laboratory, and colleagues of the Institute who were generous of time and helpful during the course of the complicated experiments. The use of the word Freon III through the report must be associated with the DuPont Trademark FREON. Special acknowledgements are given to Julie E. Collins for work with the respirometer, James H. Collins for monitoring the outdoor ponds, Deborah A. White for the phytoplankton work and Dr. J. S. Holland for assistance with the pond experiments. 12 CHAPTER. III CHROMATOGRAPHIC AND EXTRACTION TECHNIQUES A. Gas Chromatography Analysis of halocarbons was performed on a Varian 1700 series gas chromatograph (GC) equipped with an electron capture detector using tritium as an ionization source. The chromatographic column was 1/8 inch outside diameter stainless steel, 23 inches long, packed with Poropak QS 100-120 mesh, an~ was operated at 160 degrees C for analysis. Detector temperature was approximately 190 degrees C and the on-column injector was operated at 180 degrees. The column inlet pressure was adjusted to give a retention time for CC14 of 6.0 + 0.5 minutes. Electron capture detectors are extremely sensitive to trace contaminants, including oxygen, in the carrier gas so "ultrahigh purity" nitrogen was used at first. Later we found that addition of an "Oxy-Trap" column in the nitrogen supply line helped to improve the standing current, presumably by removing oxygen from the carrier gas. The following procedure was found suitable for analysis of halocarbons in water at concentrations up to about 1.0 ppm cc14 and 5.0 ppm Freon 113. Samples were stored in completely filled glass stoppered bottles because even a small air bubble could cause significant loss from the water. Plastic bottles rapidly absorbed and/or desorbed the compounds. A 5 microliter syringe, used to inject samples into the GC, was rinsed repeatedly with distilled water and filled partly with distilled water. Just before the sample was taken, the distilled water was expelled, 13 thus insuring that the syringe needle was filled with distilled water. A 0.5 microliter sample was drawn into the syringe, the tip of the needle touched to tissue paper to remove any clinging sample water, and the sample was immediately injected. The syringe needle was held in the injection port for 5 seconds after injection to allow a small fraction of the distilled water in the needle to vaporize. A standard deviation of 5% was typically obtained with this technique, with extremes within ~ 10% of the mean. The lower limit of detection was on the order of 0.02 ppm for cc1and 0.1 ppm for Freon 113 using this sample volume. 4 More concentrated samples were found to give a non-linear calibration curve, so they were analyzed by rapidly diluting the sample in small glass volumetric flasks just before analysis. Air samples were typically analyzed by injecting 50 microliters from a 100 microliter syringe lubricated with distilled water. Because of the extreme volatility of the halocarbons and the small quantities required, calibration standards were made up by a two-step dilution procedure. A 300 ppm solution of cc14 or Freon 113 was prepared by using a 5 microliter syringe to draw up 4.7 microliters of halocarbon and inject it into 25 ml of acetone in a volumetric flask. Due to the high solubility of the halocarbons in acetone, this solution was stable for several hours if kept stoppered. This solution was further diluted into tap water in "BOD" bottles of 300 ml capacity. The small acetone peak which resulted did not interfere with the analysis. A few typical chromatographic curves are shown in Figure III-1. The large initial peak is of course water which is not retained b b A B e a a FIGURE III-1. Typical Gas Chromatographic curves from the injection of: A) 0.5 ppm CC14in water, B) 2.7 ppm Freon 113 in water. Note the following features: a-injection, b-water peak offscale, c-unknown peaks always seen in water samples, d-Freon 113 peak, e-cc14• CONDENSER~ DELIVERY ~TUBE / DIGESTION ~ FLASK -..... ICE BATH MAGNm'IC STIRRING HOT PLATE FIGURE III~2Digestion and distillation apparatus for recovery of ha.locarbons from solid samples. 16 by the Poropak QS, the Freon 113 peak occurs on the tail of the peak, and the cc14 peak occurs a few minutes later. It water was found that reproducible and accurate results were obtained by simply drawing a tangent to the baseline, as shown and measuring the height of the peak. The sensitivity of the gas chromatograph detector declined with time due to "bleed" from substances accumulated on the column. Sensitivity could be restored by baking the column was overnight at 210 degrees C. During bake out, the column detached from the detector inlet to avoid accumulation of organic matter in the detector. Some "memory" effects were noticed, these were reduced by injecting several distilled water samples before starting an analysis series. B. Recovery of ,cc14 and Freon 113 From Solid Samples At several points in the research it was necessary to determine the concentration of carbon tetrachloride or Freon 113 in a solid phase such as sediment or fish. A digestion and distillation procedure appeared to be the best way to accomplish this. The procedure used by Bionomics, Inc. (1972), was adapted to smaller glassware, as shown in Figure III-2. This apparatus consisted of a Kontesware 50 ml Erlenmeyer flask, and distillation head with a received chilled in an ice bath. For the more bulky samples, 125 ml Erlenmeyer flasks were used. The sample was weighed and inserted in the glask with a and the Teflon-coated stir-bar. The apparatus was assembled digestion reagent, 9 N surfuric acid, added by pipette. was 17 The receiver was chilled with ice, and the digestion mixture was heated and stirred. The heating rate was adjusted so that distillate was not collected .until the sample had disintegrated. The receiver was a Kontesware volumetric flask of 5, 10 or 25 ml capacity, and distillate of water plus dissolved halocarbon was collected until the receiver was about 75 to 90 percent full. Tap water was ciruclated through the condenser. After distillation, the volume in the receiving flask was adjusted with distilled water and inverted once for mixing. Gas chromatographic analysis was started immediately, using the standard procedure. If the resulting peak was out of the calibration range, the sample was diluted with distilled water. Preliminary experiments with carbon tetrachloride in water solutions with added sulfuric acid digestion mixture indicated essentially 100 percent recovery in the condensed water, so this procedure was used for a number of measurements of recoverable carbon tetrachlorida and Freon 113. However, doubt was cast on the efficiency of recovery in a series of experiments which measured recovery from an actual shrimp digestion mixture. In addition several unknown peaks appeared in the gas chromatograms of water distillation samples. These peaks were shown to be due to polar compounds as resolved by partitioning the distillate with heptane. Thus, it is presumed that these peaks were not due to halogenated hydrocarbons but were probably due to low molecuiar weight oxygen containing compounds. Apparently, recovery using water as a condensing medium was not entirely reliable. We, therefore, explored the use of THE lvIAKiNE SCIENCE LIBRARY Marine Scienci;; Institute The University of Texas at Austio P. 0. Box 1267 Port Aransas, Texas 78383-1267 18 hydrophobic solvents in which the halocarbons would be more soluble, and which would not interfere with gas chromatography. Normal heptane was found to elute after the carbon tetrachloride peak and also to have a high enough boiling point although some batches of heptane were found to have interfering peaks. The modified procedure for the use of heptane was as follows: the weighed sample was placed in the previously described flask with a stir-bar, and the apparatus was assembled. The room temperature sulfuric acid digestion mixture was added by pipette, then 10 to 25 ml of heptane was added to the flask. A small amount of heptane was added to the receiving flask, so that the end of delivery tube was covered. The receiver was iced down, and the digestion mixture was heated with stirring. Due to the high boiling point of heptane (98 C), the digestion process could proceed as before. After the sample disintegrated, the temperature was raised slightly to allow the heptane to distill over, carrying the more volatile halocarbons. Heptane was used to adjust the volume in the receiving flask and dilution if necessary. Small droplets of water were often distilled over into the receiver but since the halocarbons are much more soluble in heptane, these droplets were expected to have no effect. Gas chromatography, using a 0.5 microliter sample volume was carried out as described previously. The response factors for the halocarbons were the same as with water samples, however, an additional wait of about five minutes was necessary after the carbon tetrachloride peak for the elution of the heptane. The electron capture detector did not respond to the heptane with a definite peak but the baseline was disturbed. 19 CHAPTER IV PHYSICAL AND BIOLOGICAL PROCESSES AFFECTING CARBON TETRACHLORIDE AND FREON 113 IN ESTUARINE ENVIRONMENTS Introduction and Summary A number of experiments were conducted to determine the importance of physical processes such as evaporation and adsorption, and the importance of biological processes such as microbial degradation to the persistence of halocarbons in the estuarine environment. It was shown that evaporation is the dominant process of halocarbon loss, controlled by turbulent diffusion rates in the surface water olayers. Adsorption from water to sediments is insignificant. Carbon tetrachloride apparently undergoes a slow reaction with sediments and there is some evidence for an even slower reaction by Freon 113. No microbial degradation occurs. Physical Processes -Loss to the Atmosphere Some of the physical properties of cc14 , Freon 113, and oxygen are shown in Table IV-1. Solubilities in fresh water and salt water for Freon 113 and in salt water for cc14 were determined by shaking bottles filled with water and at least 50 ml of halocarbon for twodays at approximately 25 degrees c. There are two ways of looking at the water to air transfer of these halocarbons; (1) in terms of the air-water interface. An equilibrium situation is not likely to come about in the estuarine environment, except in the unusual case of air bubbles surrounded by water, possibly as a result of active breaking waves TABLE IV-1 Physical properties of CCl4, Freon 113, and Oxygen. Entries marked * were experimentally determined. Property Units CCl4 CCl2F-CClF2 02 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-~~~ Molecular weight Density of liquid at 25 C Boiling point at 1.0 Atmosphere Solubility at 25 C -fresh water -seawater 32 ppt Vapor pressure at 25 C Partial pressure -std. atmosphere Calculated quantities vapor pressure of a 1 mg/l solution -fresh water -seawater 32 ppt (molecular weight)-~ ratio of (mw)-~ to oxygen gm/cc -c mg/l mg/l mm Hg mm Hg mm Hg mm Hg 153.B 1.595 76.B 800 820* 110 0.14 0.14 0.0806 0.455 187.4 1.565 47.6 190* 136* 336 1.77 2.47 0.0731 0.413 32.0 8.6 7.1 158 18.4 22.3 0.177 1.0 "" 0 21 or mechanical aeration. In this case the lower solubility and higher vapor pressure of Freon 113 will cause the ratio of Freon to cc1in the air phase to be about 12.6:1, so the 4 Freon will be lost 12.6 times faster than the CC14 • However, in the natural ·estuarine environment, there are few air bubbles and the loss rate will generally be controlled by diffusion. Both experimental results and theoretical consideration (Liss, 1973; Liss and Slater, 1974) show that for chemically unreactive species, the rate controlling factor is the diffusion rate through the liquid boundary lay·er. This is in contrast to the diffusion of water vapor which is controlled by the air boundary layer. Furthermore, the resistance should be the same for movement in or out of the water. In extremely calm waters, molecular diffusion through the liquid boundary layer will account for most of the transport. The molecular diffusion coefficient is inversely proportional to the square root of the molecular weight, so the lower molecular weight of cc1would allow it to diffuse faster. 4 In highly turbulent waters, turbulent diffusion will account for most of the transport, so we would expect both halocarbons and oxygen to be transported at close to the same rate. Because extensive studies have been made on the diffusion of oxygen through natural air-water interfaces, and theoretical considerations indicate that there should be a relation between halocarbon and oxygen diffusion rates, several experiments were conducted to simultaneously measure the rate of oxygen diffusion into water and the rate of cc14 and Freon diffusion out. 22 To determine the oxygen reaeration coefficient and ccl4 and Freon 113 loss rate coefficients, the oxygen percent saturation values were converted to oxygen deficit values so that an equation of the general form: dN/dt = k; N where N is oxygen deficit or halocarbon concentration, could be used. This integrates to Nt = ln N+ k t, and a least-squares approach can 0 be used to determine k for oxygen and for the halocarbon compounds. The first experiment was conducted with 2 beakers of 3 liter capacity, filled to 3/4 of their 20.5 cm depth with seawater. One was deaerated with nitrogen and the other was saturated with cc14 • Two containers were necessary because the oxygen was measured with a membrane type probe, (YSI type 51 probe and model 54 meter) and it was considered possiblethat'high concentrations of cc14 might damage the membrane. The beakers were placed on the ground outside the laboratory in the shade, the wind speed was about 10 knots. Oxygen concentration was monitored with a YSI probe (V-B) and cc14 was determined by GC after dilution after O, 45, 260, 487 · and 1145 minutes of exposure. During this period the oxygen concentration rose from 7~2 _to 67 percent of saturation and the cc1 4 concentration fell from 680 to 310 ppm. Rainfall overnight caused stratification in both beakers so these results were considered crude. Nevertheless, a least squares fit to an exponential equation gave a first order constant for cc1of 0.65 times that 4 of oxygen, indicating that molecular diffusion was controlling 23 the loss rate. The next experiment was conducted with shallow form glass culture dishes, 18 cm in diameter and 6 cm deep. One dish was filled with sea-water (25 ·ppt) desaturated using nitrogen, the other with similar water which had been saturated with a mixture of cc14 and Freon 113. These dishes were exposed outside in the "pond yard", with winds on the order of 15 to 20 knots. Oxygen was measured as before and cc14 and Freon were measured by GC after dilution. A least squares fit to an exponential equation gave a constant for ccl4 of 0.83 times that of oxygen; for Freon 113, 0.86 times that of oxygen. In this case, there was obviously more turbulent diffusion than in the first experiment. A final experiment was conducted to compare cc14 and Freon 113 diffusion rates. A single shallow form glass culture dish was used, filled to within 0.5 cm of the top with distilled water and placed on a magnetic stirrer in the laboratory. A slow stirring rate was used, with no bubbles being produced. Water temperature was 19.5 c. A mixture of CC1 4 and Freon 113 in 0. 37 ' ml of acetone was added to give a starting concentration of about 0.7 · ppm CC1 4 , 3.9 ppm Freon 113, and 230 ppm acetone. Sarrples were taken directly into a syringe and injected into the GC, thus each data point represented one GC run rather than the average of three as in the previous experiments. Sixteen samples were taken at times out to 326 minutes, at which point the concentrations of the halocarbons had fallen to about 1/3 of the initial value. A least squares fit to an exponential equation 24 gave a constant for cc14 of 1.129 times that of Freon 113. The expected ratio fonnolecular diffusion is 1.104. The results of the least squares fits are shown in Table IV-2 along with the 95% confidence interval for the loss rate constant calculated according to Hoel (1971). The least squares fit was done by transforming the equation to the form: lnY = lnA + Bt, and fitting this linear equation by the standard method. The "Coefficient of Determination", r 2 , indicates how closely the equation fits the data. It appears that the loss rate coefficients of cc14 and Freon 113 are not statistically different. The significance of these results is that the many extensive studies of reaeration coefficients in ponds, streams, etc. can be utilized to predict the halocarbon loss rate coefficients for these environments. Additional information on the rate of loss of cc14 and Freon to the atmosphere is provided by the pond experiments. The biological aspects of these experiments will be detailed in a later section, the physical description of these ponds is as , follows. I, , Nine concrete ponds are arranged in a square pattern in the "pond yard" at the MSI. Each pond is in the form of a square with the corners cut off and rounded, the bottom of each pond is flat concrete at the level of the ground, and the walls are built up to a height of about 24 inches above ground level. An electric pump with an intake reaching to the bottom of the pond, and a submerged outflow pipe maintain a minimum circulation and prevent stratification. The circulation rate induced by the pump TABLE IV-2 Results of least squares fits to the equation Y = AeBt for rate of halocarbon loss and oxygen deficit decrease. 95% conf. EXPERIMENT y A B interval r2 April 5, tall beakers 02 deficit 81.4 -.001045 + .000219 outside, 10 knot wind 0.977 5 points, max t = 1145 min. CCl4 702.9 -.000683 + .000160 0.971 April 26, shallow dish 02 deficit 82.7 -.005420 outside, 15-20 knot wind 0.9983 CCl4 1951.0 -.004535 5 points, Freon 341. 5 +++ .000695.000302.000300 0.9976 max t -.004681 0.9882 = 403-min. February 21, 1975shallow dish inside, CCl4 7.80 -.003544 + .000267 0.9695 slow stirring16 points, Freon 8.95 -.003140 + .000267 0.9683 max t = 326 min. Note. A is in percent saturation for oxygen deficit, arbitrary units for halocarbons. N U1 26 is strongest near the wall and is slower towards the center, flow rates at the surface appear to range from 1 to 5 cm/sec. The surface circulation is also influenced by strong winds. The concentration of halocarbon in the experimental ponds is controlled by daily analysis of samples, followed by addition of the amount of cc14 or Freon 113 required to bring the concentration back up to a target level. The compounds are added to the pond water by injecting the pure liquid into the pump intake hose with a hypodermic syringe, the impeller breaks up the fluid into a fog of small droplets which then dissolve. The change in concentration over a 24-hour period can be converted into a loss rate coefficient, unfortunately the oxygen reaeration coefficients can not be measured at the same time due to the active photosynthesis and respiration. The variation of loss rate with weather conditions is quite pronounced, losses during windy weather were much higher than during calm winds, and rain caused very rapid loss. The loss rate coefficients of cc14 and Freon 113 appeared to be identical under all conditions, as long as air bubbles were not introduced by the pumps, considering all of the possible sources of error, this probably means identical to within + 15%. The loss rates observed during June, July and August, 1973 can be summarized as follows: calm weather, wind 5 -10 knots, average daily loss is about 45% corresponding to k = 0.025 hr-l (base e), winds 10 -15 knots, daily loss is about 55%, corresponding to k = 0.033 hr-1 , winds 15 -20 knots average loss is about 70% corresponding to k -0.050 hr-1. 27 To extrapolate the rates observed in the ponds to large ponds, channels or an open bay would require the consideration for the following factors: loss rate will be decreased by increasing depth or by any tendency to stratification, and loss rate will be increased by turbulence caused by waves, high current velocity, and rough bottom topography. Liss and Slater (1974) have recently made estimates of the flux rates of various gases through the air-sea interface.· Their estimate for cc14 under average sea conditions is about 4 to 5 times faster than the rate observed in the ponds during high winds. ~hey also predict that the rate constant for oxygen transfer should be about twice that for cc1 4 , and that the rate of transfer should increase with the square of the wind velocity. These results are compatible with our measurements, if we assume that the turbulence in the small ponds is much less than that in the open sea, so that the thickness of the diffusion limiting layer is greater. Interaction with Sediments Two possible forms of interaction with sediments were considered: degradation by sediment microorganisms, and adsorption by sediment particles. A survey experiment was conducted to determine if either type of interaction could be detected. A grab sample of sediment was collected from the edge of La Quinta channel across from the DuPont loading dock site. The sample was a largely anaerobic mud with sand and shell fragments and numerous benthic organisms. After collection, the sample 28 was placed in a covered plastic bucket held at laboratory temperature, the disturbed sediment quickly settled, and a thin tan oxidized layer formed over the anaerobic grey mud. Subsamples were taken for the various experiments by "coring" the sediment in the bucket with a 10 ml plastic syringe with the end cut off. In the first experiment, BOD bottles were made up in duplicate with 0, 10, 20, and 40 ml of sediment and seawater with an initial cc14 concentration of approximately 1.0 ppm. The bottles were shaken until all particles were suspended twice a day, and allowed to stand in the dark at approximately 23 C. Samples were analyzed by gas chromatography at various times as shown in the table. The cc14 concentration decreased at a rate proportional to the volume of sediment added. The data from the first experiment are shown in Figure IV-3, along with the least squares fit to a linear equation. The loss rates, k, show a linear dependence on volume of sediment added. Based on further experiments, the loss of cc14 from the bottles without sediment is attributed to unavoidable contact with air during sampling. All of the sediment from one of the bottles containing 40 ml of sediment was transferred to a distillation flask, and an attempt was made to recover the adsorbed cc14 by adding acid to the sediment, steam distilling off the cc14 , and trapping it in a cooled receiver. No cc14 could be detected in the 25 ml of distillate. The detection limit is estimated at 0.04 ppm cc1 4 in the distillate. TABLE IV-3 Concentration of cc14 in seawater exposed to various amounts of sediment (average of two bottles). Also shown are the coefficients obtained by a least squares fit to a linear equation: Ct = c0 kt, and r2, the coefficient if determination. Sediment Least Squares Fit Volume 0.5 64 88 110 136 Co k r2 0.0 1.05 0.94 0.91 0.98 0.85 1.04 0.0012-3 0.68 10.0 ml 1.02 0.76 0.69 0.68 0.58 1.00 0.00316 0.966 20.0 0.98 0.76 0.54 0.41 0.26 1.02 0.00543 0.970 40.0 1. 07 0.59 0.25 0.06 lt .01 1.10 0.00939 0.984 tv\.0 30 The second experiment was designed to determine if the adsorption was due to microbial activity or a chemical process. Four bottles were made up containing 40 ml of sediment and sea water, and 2 with seawater alone. Two of the sediment bottles and one seawater bottle were sterilized in an autoclave at 121 C for 30 minutes. Filtered seawater saturated with a 1:1 mixture of cc14 and Freon 113 was added to the bottle to give a final concentration of approximately 1 ppm CC1 4 and 1 ppm Freon 113. Table IV-4 gives the results of analysis after various lengths of time. Apparently degradation of cc14 is not due to microbial activity. There is also an indication of loss of Freon 113 in the bottles with sediment, but much more slowly than CC1 4 • It was reasoned that if it was possible for bacteria or other microorganisms to degrade cc14 or Freon 113, such organisms should be present in the experimental ponds exposed to continuous doses of the halocarbons. On Sept. 24, 1973, four glassstoppered BOD bottles were filled with water from pond # 2, and four were filled from pond # 8. One bottle from each group was analyzed at once, and the remaining bottles were placed upright on the bottom of the ponds they were filled from. Great care was exercised to exclude air bubbles during fi-ling of the bottles. Placing the bottles in the ponds ensured that they would be exposed to a natural regime of temperature and light. The remaining bottles were analyzed after 25 hours, 73 hours, and 552 hours as shown in Table IV-5. Since the standard deviation of determinations done during a single day is typically 31 TABLE IV-4 Concentration of CCl4 and Freon 113 in sterile and nonsterile seawater, with and without sediment. Concentra tions are given in ppm, as CCl4/Freon 113. Conditions Time in hours 0.5 45 77 117 sediment non-sterile 1.18/1.02 0.88/0.92 0.47/0.91 0.24/0.83 sediment sterile 1.21/1.04 0.84/1.01 0.41/0.92 0.13/0.78 seawater non-sterile 1. 00/0. 81 1.10/0.91 1.02/0.83 1.03/0.76 seawater sterile 1. 05/0. 86 1.21/1.06 1.09/0.91 1.08/0.91 * * * * * * * * TABLE IV-5 Halocarbon concentrations from test bottles left inponds, in ppm. Time given in hours. Elapsed time 0 25 73 552 1.33* 1.38 1.12 1.24 Freon 113 2.50* 2.40 2.23 2.72* (Each point represents the average of 2 gas chromatograph determinations except * which are 3) 32 :t 5%, with somewhat larger variation from day to day due to instrument drift, these variations are not significant. We must conclude that any aerobic degradation of halocarbons by natural plankton and bacteria, if present, is extremely slow. The most interesting point about the sediment experiments is that the loss rate is clearly a linear function of time and sediment volume. This implies that the reaction rate depends on the availability of reaction sites within the sediment. Since cc14 could not be recovered from the sediment, it must be assumed that it either reacted with the sediment or was hydrolyzed in a reaction catalyzed by the sediment. In view of the rather large amount of sediment for a given volume of water in these experiments, compared with actual estuarine conditions, it is expected that losses to the sediments will be much smaller than losses to the air in the natural environment. CHAPTER V METHODOLOGY, RESULTS AND DISCUSSION A. Short-term Experiments with Phytoplankton Experiments on the effect of carbon tetrachloride and Freon 113 on photosynthesis, respiration and growth of phytoplankton were conducted with both natural populations and with Agmenellum quadruplicatum (PR-6), a marine coccoid blue-green alga. Photosynthesis and respiration were measured suing light and dark bottle techniques with both natural populations and cultured organisms. The growth rate experiments required the development of some special techniques due to the volatility effects of the halocarbons. 1. Light/dark bottle experimental methods. The general theory of light and dark bottle experiments for the determination of photosynthesis and respiration is well established. A well-mixed and oxygenated plankton sample is portioned out into a number of identical bottles, some of the bottles are incubated in the light and some are incubated at the same temperature but with coverings to prevent the admission of any light and thus, photosynthesis. The initial bxygen concentration is also determined. Both photosynthesis and respiration can proceed in the light bottles, but only respiration occurs in the dark bottles, thus the two rates can be measured. Three hundred ml capacity glass "BOD' bottles were used in these experiments. The starting plankton inoculum was mixed and aerated in a 20 liter glass carboy, then siphoned into the individual bottles. Oxygen concentrations were determined by 34 Winkler titration using standard methods. Duplicate bottles were analyzed for each treatment and duplicate titrations were made of each bottle. The standard error of duplicate titrations was typically 1 percent. Initial experiments indicated that halocarbons injected into the test bottles with a microsyringe would not dissolve rapidly enough. However, it was found to be possible to create a dense dispersion of very small droplets of halocarbon in water using the following procedure: Ten grams of carbon tetrachloride or Freon 113 with distilled water to make 100 ml and 50 microliters of "Tween 80" (a nonionic detergent) were subjected to a short treatment with an ultrasonic probe. This produced a milky dispersion, stable for several hours. Experimental concentrations of halocarbons were established by injecting this dispersion with a microsyringe into the BOD bottlesojust before stoppering each bottle. The cloudy dispersion rapidly dissolved. 2. Stock culture preparation. A stock culture of PR-6 {Agrnenellum quadruplicaturn) was maintained in ASP-2 medium {Table V-1) in a 2.5 L flask. The culture was grown at optimal types growth temperature (39 degrees) and light saturation (approx. 600 ft. c.). After reaching a density of about 10% T, the culture was kept at room temperature under normla light conditions. As a result, some clumping occurred. Before use, however, free cells were evenly distributed through the medium by bubbling with 1% C02 + air, and inoculum clumps are allowed to settle to the bottom. Before each experiment, a mass culture is grown up according TABLE V-1 ASP-2 medium for PR-6 culture. Component Concentration Comments NaCl 18.0 gm/l add as solid MgS04•7H20 5.0 gm/l from stock solution KCl 0.60 gm/l " " " CaCl2·2H20 0.370 gm/l " " " NaN03 1.0 gm/l add as solid KH2P04 0.050 gm/l from stock solution Na2 EDTA {complexing agent) 0.030 gm/l " TRIS (buffer) 1.0 gm/l (pH = 8. ~ Fe2{S04)3·7H20 0.0039 gm/l " II" II "" " " Trace metals mix 10 ml/liter see below for compositior Trace Metals Mixture H3B03 3.426 gm/lCuS04 0.3 mg/lC0Cl2·6H20 1.215 mg/lMnC1 2 ·4H20 432.0 mg/lZnCl2 31.5 mg/lMo03 {85%) 3.0 mg/lVitamin B-12 0.8 mg/l 36 to the following procedure: A 12 L carboy was acid-cleaned and autoclaved with a 90 ml millipore system attached. 10 L of 75% enriched seawater media (.2 g/l NaN03 and .02 g/l NaP04 ) are then drawn through the MF and into the carboy. The media was then inoculated with about 50 ml stock culture and stock Bl2 (1 ml/L 8 ppm) and immersed in a water bath at 39 degrees. A mixture of 1% co2 and air was bubbled through the culture. When the mass culture reached a density of 85% T at 560 nm in a 1/2 inch test tube, it was siphoned into 300 ml BOD bottles in a darkened room. These bottles were subsequently injected with a cc14 solution and incubated in the constant temperature room at 39 degrees for 8 hours. Light intensity in the CTR was 600 ft. c. 3. Growth rate experimental methods. Use of algal growth rate measurements as an index of water toxicity has been described by Ban Baalen, Pulich and O'Donnell (1973). This technique, however, required that the growth tubes.be continuously bubbled with 1% carbon dioxide in air to ensure adequate carbon supply and agitate the cells. In order to conduct these tests with the highly volatile halocarbons, it was necessary to use a closed system. Erlenmeyer flasks with 500 ml screw cap were used. They were fitted with a side arm which could be inserted into spectrophotometer to measure the culture density. Aluminum foil was used under the screw caps to improve the seal. Flasks were prepared by adding 20 ml of ASP-2 medium (Table V-1) and autoclaved. Just before inoculation, each flask was flushed with 1.0% carbon dioxide in air for three minutes using sterile procedures. Each flask was inoculated with 0.20 ml of a 37 PR-6 stock culture just starting the log growth phase. Flasks were sealed and placed on a shaker table in a constant temperature/ light room for two hours. At this point, halocarbons were directly injecting into the side arm of the flask through the aluminum foil seal so that the carbon dioxide mixture was not diluted. The liquid halocarbon evaporated quickly and reached an equilibrium in which most of the added compound was in the vapor state. The correct volume to add was determined experimentally, using uninoculated culture medium. All flasks were incubated on the shaker table at constant light and temperature. The culture density was determined at intervals by tilting the flask so that the liquid ran into the side arm, then reading the percent transmission with a "Spectronic 20". The liquid was drained from the side arm back into the flask and returned to the shaker table. Growth rate data was presented by converting the transmission data to absorbance, multiplying by 1000 and taking the log base 10. The resulting "growth units", when plotted versus time, show a linear increase during the log phase of algal growth, and the slope of this line can be used to compare growth rates under different conditions. 4. Experiments with natural populations and CCl4 Two sources for the mixed phytoplankton populations were used, Aransas Pass surface water and one of the control ponds used in the long term experiments. In both cases, the water sample was well mixed and aerated and did not contain any macroscopic zooplankton or detritus. Source and incubation 38 conditions are given in Figures V-1 to 4 which present the initial oxygen concentration, and the oxygen concentrations in the light and dark bottles for control and treated samples. In all of these figures, the oxygen concentration for each bottle is plotted as a dot, and the mean for each treatment is shown on the connecting line. In experiments with plankton from Aransas Pass, the total oxygen change was relatively small, relative to the variation between duplicates (Figures V-1 and 2). The results at 1.0 ppm carbon tetrachloride can not be distinguished from the control, while at 10 ppm there is indication of increased respiration. One hundred ppm completely suppresses both photosynthesis and respiration. Two experiments were also conducted with the plankton of pond #3, a control pond in the long term exposure experiments (Chapt. V-) , these are shown in Figures V-3 and 4. In the first experiment, little net photosynthesis occurred, the 1.0 ppm bottles showed stimulation of photosynthesis (p = .005, t test, df = 2), but the 10 and 30 ppm were indistinguishable from controls, while 100 ppm showed suppression of respiration (p =.OS). In the second experiment, 1 ppm again shows slight stimulation of photosynthesis (p = 0.1), 10 and 30 ppm are not significantly different, and 100 ppm greatly reduces photosynthesis while respiration is unaffected by even the highest level. 5. Experiments with cultured phytoplankton and cc1 4The stock cultures of A. quadruplicatum used were much denser than natural phytoplankton and were also near optimum 39 cc14 ppm INIT 0 1 10 • 8.5 I • I I I s.o • s~ K ' ' ' • c:l. r ' FIGURE V-1 Light/dark bottle experiments with natural sample from Aransas Pass, incubated under natural light in pond 3 from ·1000 to 1400, partly cloudy conditions. Predominant plankton genera were Chaetocerus, Biddulphia and Rhizoselenia. The zooplankton population was sparse. 40 INIT 0 1 10 100 • a.o _, t.• \ \ FIGURE V-2 Light/dark bottle experiments with plankton from AransasPass, incubated under artificial light, 8 hours. cc14 ppm 41 mIT 0 1 10 30 100 •-· • ...\ ''\ \ \ ' ' '\ B.o l--: • FIGURE V-3 Light/dark bottle experiments with pond population,incubated under artificial light for 8 hours. CC1ppm 4 IllIT 0 1 10 30 100 I 6.o ' \ t---+-.___·---t'~ . t • FIGURE V--4 Light/dark bottle experiments with pond population,incubated under artificial light for 8 hours. 43 growth conditions. Therefore, the oxygen production rates were high while respiration was relatively low. In experiment one, similar in technique to V A-3, 1 ppm carbon tetrachloride stimulated photosynthesis ( p = 0.025, t test, n = 2) while 10 ppm was undistinguishable from the control and 100 ppm greatly suppressed photosynthesis. Respiration was not significantly affected (Figure V-5). For experiments two and three, (Figures V-6 and 7), three bottles were used for each treatment. In experiment two, respiration was increased in all cc14 treated bottles (p = .025). The 1 ppm light bottle had significantly lower oxygen than the control, but this was apparently due to increased respiration. Photosynthesis was greatly reduced at 100 ppm, but not eliminated. In experiment three, extremely high oxygen production was observed for all but the 100 ppm light bottle. Photosynthesis was distinctly stimulated at 1 ppm (p = .005) and reduced at 10 ppm (p'= .05). 6. Growth rate exper~ments with CCl4 Several preliminary experiments were run using the growth flasks, but the growth was scored visually rather than spectrophotometrically. The 72 hour observation after inoculation indicated that essentially no growth occurred in the 10 and 100 ppm flasks, with growth similar to or slightly less than controls at 1 ppm. One experiment at 3 ppm also showed growth similar to controls. Gas chromatographic analysis of the cultures at the end of 72 hours indicated that the CCl4 concentration fell during 44 CC14 ppm !NIT 0 1 10 100 • light bottles 1 5.0 I I I I I I I [ I Q. 1o.o I e~ I 0 ' ' \ \. ~ dark bottles 5.0 • FIGURE V-5 Light/dark bottle experiments with Agmenellum quadruplicatumstrain PR6 grown in enriched seawater, experiment number one. 45 INIT 0 10 100 l -·-.·~l 20.0 ...,,,. •I light bottles • 1 o.o • - •' ' dark bottles ' • -. '~-----4-i.-----TT! ~ FIGUREV-6 Light/dark bottle experiments with Agmenellum qua.d.ruplicatumstrain PR6 grown in enriched seawater, experiment number two. 46 CC14 ppm INIT 0 1 1 0 100 30.0 e. '1. 20.0 ~ ~ 0 1 5.0 T•• • 1 o.o ~... ... -t-• • ~ .••------+ 5.0 FIGURE V-7 Light/dark bottle experiments with Agmenellum quadruplicatumstrain PR6 grown in enriched seawater, experiment number three. 47 the experiment to as little as 12% of the starting concentration. This problem may have been due to losses of CCl4 vapor during the course of the experiment through imperfectly sealed caps or through degradation of the cc14 . The results of the main growth rate experiment are shown in Figures V-8 and V-9. Initial concentrations in this experiment were 1.0, 3.0 and 10.0 ppm cc14 with two flasks at each concentration; however, the concentrations at 72 hours are used in the figures. Peak growth rates were estimated from the maximum slope of each line by eye, and lag time was estimated by extrapolating this line to 2.0 growth units. Effects can be seen for this data, in terms of all three parameters, peak growth rate, lag time, and final growth level. However, the results are not consistent between flasks at the same nominal concentration. The most consistent result is a lower final growth level, however, this may be partially due to C02 escaping from the flasks through bad seals as suggested earlier. The algae would then be limited in their growth by availability of C02. The lag time is quite variable and has no clear relation to dose. Peak growth rate is also variable, with both nominal 1.0 ppm flasks and one nominal 10 ppm flask within the range of the controls. An experiment was conducted to determine the loss rate from the growth flasks in the absence of algae. Six flasks were prepared with sterile growth medium. Sufficient carbon tetrachloride was injected to produce 1.0 ppm in the liquid phase, half of the flasks were covered with aluminum foil, and 48 TIME hours 20 30 40 603.0------..a----------'----------+--------------------- -2.5 CD COHTRO~ 0 - a'f 0 m ~ 2.0 M 0 00 0.13(1) ...._,,.-0 •-• tiE"' • 0 2.5 r-4 ...._,,. ~ !3 = 2.0 ~ 0 ~ .26(1 ) • EXPERIMmTAL 1 • 5 1 .o FIGURE V-8 Growth curves of Agmenellum quadruplicatum at nominal concentrations of 1.o, 3.0 and 10 ppm carbon tetrachloride. Actual final concentrations are shown next to each curve. with the starting concentration in parentheses. Below about 1 .4 the growth units are very imprecise. 49 .15 • PEAK GROWTH RATE § • • •10 i ~ • ==e • .05 0 0.5 1 .o 1.5 2.02--ppm cc1450 • LAG TD!E 40 =~ 30 • • 20 I • 10 0 0.5 1.0 1.5 2.0 ppm cc14 3.0 FINAL GROWTH LEVEL 2.9 •••• ~ 2.8 H • ~ == 2.1 ~ • • t!:J ~ 2.6 • • 0 1.0 2.0 ppm cc14 FIGURE V_:·9 Various parameters of the growth curves shown in Figure v-8plotted versus carbon tetrachloride concentration at the end of theexperiment. 50 all flasks were held on the shaker table in the constant temperature and light room. One light and one dark flask were taken for analysis after 3, 24 and 48 hours. The concentration in both types of flask after 48 hours was 0.8 ppm indicating about 20% loss. 7. Light/dark bottle experiment with natural phytoplankton and Freon 113 One experiment was conducted using the phytoplankton population from pond number 3, a control pond in the long term exposure experiments with levels of 1, 10, 30 and 100 ppm Freon. 113. The samples were incubated under artificial light for eight hours. Figure V-10 shows the results obtained. The oxygen level in the 1 ppm bottles was significantly higher than the controls (t test, df = 2, p = .01). It was felt that the t test might have been unduely affected by the coincidence that both light control bottles had oxygen titration results identical to the 0.01 mg/liter level, in contrast to the typical range for these titrations of 0.10 mg/l. So a second t-test was run with the same mean value but a range of 0.10 for the light-control oxygen, this indicated significance only at the 0.10 level. The oxygen level in the 10 ppm light bottles were not significantly different from the light control; however, the 10 ppm dark bottles were significantly lower (p = .05), indicating increased respiration. When a concentration of 100 ppm Freon was reached, photo synthesis was about 65 percent of the value for the controls, similar to results obtained with carbon tetrachloride at 100 ppm with the pond number 3 population. This indicates that some 51 FREON 11 3 ppm INITIAL 0 1 10 30 100 6.0 I I 3.0 FIGURE V-10 Light/dark bottle experiments with pond phytoplankton incubated under artificial light for 8 hours. 52 phytoplankton are extremely resistant to the effects of these compounds. 8. Growth rate experiments with Freon 113 Three growth rate experiments with the blue-green alga Agmenellum quadruplicatum strain PR-6 were conducted using Freon 113 doses between 1 and 10 ppm. All three experiments showed the unexplainable result that the 1 and 3 ppm cultures grew more slowly or not at all while the 10 ppm flasks grew almost as well as the controls. The results of the first experiment are shown in Figure V-11; the units are the same as used in the carbon tetrachloride experiments. The 1 and 3 ppm flasks showed no measurable growth while the 10 ppm flasks grew about 1/3 to 1/2 as fast and reached somewhat lower maximum levels than the controls. Measurement of the Freon in one of the 10 ppm flasks at the end of the experiment gave 3. 9 ppm. The second experiment shown in Figure V-12, again the 1 and 3 ppm flasks showed no measurable growth. The maximum growth rates of the 10 ppm flasks were slightly less than the controls, and the peak density was not quite as high. In the third experiment (Figure V-13), both 1 ppm flasks and one of the 3 ppm flasks showed measurable growth, but there was a distinct lag time on the order of 7 hours, the maximum growth rate was slower and the peak density was not as high as either the 10 ppm flasks or the controls. In this experiment, the 10 ppm flasks grew just as fast as the controls and had final density levels only slightly less than the controls. 53 TIME hours 10 20 30 40 •/-·-- CONTROIB .--. ""'"' G) J "4 0 2.0 0 m ,c < -H 00 1.5 '-' 0 0 ~ L./ ......• ' ....• 0 ,.... (10) L.-J ... · 2.5 ·-·~ ~ ..... = ......../..9 (10) • ~ .~/ ~ = / FR!DN113 CJ 2.0 FIGURE V-11 Growth rates obtained in first experiment with FREON 11 3. The alga used was Agmenellum guadruplicatum, just as inthe experiments with carbon tetrachloride. Although flasks wereprepared at nominal concentrations of 1, 3 and 10 ppm, only the10 ppm flasks showed measurable growth. At the end of theexperiment, a concentration of 3.9 ppm was measured in one flaskwhich had an initial concentration of 10 ppm. 10 20 30 40 50 /:===· :~-· r-. ......... m 2.5 ./~OW G> 'f . 0 2.0 ,c < 0 a0 l;I 0 0 ..._., -H ~10 1.5 0 ....~ ~ ~ H (10) ~ •/' = 2.5 / · (1 o) ~ ~ I I FRmB113 2.0 ,• 1.5 FIGURE V-12 Results of the second growth rate experiment withFREON 11 3. The flasks with initial concentrations of 1 and 3 ppm showed no measurable growth. 55 TIME hours 1 5 20 25 30 35 - 2.0l'""'"'!1 G> 0 .Da M 0 m ~ 1 • 5 H 0 0 0 .._.. -2.0 (10)0 • ~·------------·----------· ~ ,...0 2.5 ~:/·=-·--6-.1-(1_0_)·----· ~ ~ 1./I ~:.54{1) ~ 2.0 /J/ ~·~· = ·/ ·/·,. .61 (1) ~ 0~ I 1 .o FIGURE V-13 Results of the third growth rate experiment with FREON 113.The nominal starting concentrations are shown in parentheses next tothe measured final concentrations. 56 Analysis of individual flasks at the end of the experiment gave the concentrations of Freon shown on the figure, with the starting concentration in parentheses. Final concentrations ranged from 20 to 70 percent of initial dose. The following experiment was conducted to determine the loss rate of Freon 113 from the growth flasks in the absence of algae. Six flasks were prepared with sterilized medium, but not inoculated. Sufficient Freon was injected into each flask to produce 2.6 ppm in the liquid phase. Three of the flasks were wrapped in aluminum foil, and all were held on the shaker table in the constant temperature and light room. Flasks were removed for analysis at approximately 1, 24 and 96 hours. The loss rate was essentially identical from both light and dark flasks, approximately 21 percent in 96 hours. A 50 percent loss in 24 hours in one flask was traced to an incomplete seal under one cap. The possibility that the algae were simply absorbing the halocarbon was tested by taking the dense algal culture resulting in each of four control flasks from the second experiment and adding enough Freon to give 10 ppm in the liquid phase. Two of the flasks were covered with aluminum foil and all were held on the shaker table in the constant temperature and light room for 2.5 hours. Analysis of the liquid phase gave .an average of 9.50 ppm for the "light" flasks and 9.45 ppm for the dark flasks. Considering the accuracy of the method, this value is not significantly different from the starting value. 9. Discussion of short term experiments with phytoplankton The light/dark bottle experiments with both natural 57 phytoplnakton and the pure culture of blue-green algae generally showed stimulation of photosynthesis at the 1 ppm level by both carbon tetrachloride and Freon 113, while 100 ppm always showed considerable suppression of photosynthesis. Results at intermediate concentrations were variable from one experiment to the next. The effect on respiration is also variable. The variation between duplicate bottles seems to be higher at the higher concentrations in some experiments. The stimulation of photosynthesis at 1.0 ppm carbon tetrachloride was typically in the 6 to 15 percent range. This would probably be hard to detect under natural conditions due to other sources of variability. Even at the highest concentration tested, 100 ppm, there was some photosynthesis in 5 out of 6 experiments with carbon tetrachloride. In a single experiment with Freon 113, the photosynthesis at 100 ppm was 65 percent of the control. One hundred ppm is about 12 percent of saturation for carbon tetrachloride and 74 percent of saturation for Freon 113. The greatest difficulty encountered in the light/dark bottle experiments was the low activity and high variability of the natural samples. Furthermore, the necessity of using a dispersant to stabilize the halocarbon suspension added an additional degree of variability, even though the final concentration was extremely low. One approach which could be used to solve both problems would be to concentrate the natural phytoplankton using a centrifuge, then the experimental concentration of halocarbons could be established in larger volumes of seawater and the phytoplankton concentrate added to form the final mixture. Growth rate experiments with A. quadruplicatum, strain PR-6 58 were more sensitive than the light/dark bottle experiments, but were more variable between replicates. In the carbon tetrachloride experiments, all of the measurable parameters, peak growth rate, lag time and final growth level were affected but without a clear relation to dose. At the lowest dose level used, 1 ppm initial and .13 to .26 ppm final, peak growth rate was ~imilar to controls. The loss of large fractions of the starting concentrations during the course of the experiment may be due to either imperfect seals in the caps or to degradation of the halocarbon by the algae. Degradation seems to be unlikely since very large amounts of halocarbon were present in the flasks in the vapor phase, and other experiments with natural plankton and bacteria populations indicated that any degradation must be slow. The results obtained with Freon 113 in growth rate experiments are hard to interpret, since the 10 ppm flasks grew almost as well as the controls while the 1 and 3 ppm experiments gave little or no growth. In the light/dark bottle experiments with natural phytoplankton, 1 to 10 ppm gave photosynthesis similar to or greater than the controls, while in the long term pond experiments the community photosynthesis/respiration at 1 ppm was similar to the controls. Since the natural plankton were not damaged at this level, there seems to be little call for alarm, but this effect should be investigated further. Unfortunately, not enough time was available to further investigate this phenomenon. As in the other experiments, maintaining a constant level of halocarbon is the greatest problem in the algal growth experiments. In these experiments we tried to satisfy the 59 requirement for a constant supply of carbon dioxide by providing a large volume of gas in the closed flasks. A better approach which would require more elaborate equipment would be to bubble a mixture of halocarbon vapor, carbon dioxide and air through the culture in test tubes held under constant temperature and light conditions. B. Short Term Effects on Larger Animals: I. Introduction The initial plan of the research called for respiration and activity measurements on shrimp, trout, and oysters because it was believed that these measurements would be more sensitive than approaches which rely on the death of the organism as a toxicity indicator. The detection of sublethal stress by means of monitoring fish movements has been reported by Waller and Cairns (1972) and Sparks, Cairns and Waller (1972). Respiration response to sublethal stress has been reported for fish (Steed and Copeland, 1967) and (Cech 1970). An instrument for measuring respiration and movement simultaneously was devised by Gordon and Oppenheimer (1972), and this instrument was used to measure the response of shrimp to Galveston Bay water as an indicator of toxicity (Oppenheimer, et al., 1973). A full evaluation of the use of the instrument in the Galveston Bay project was not possible until after the present work was started. Results from the Galveston Bay research were discouraging, mainly because the variation in respiration between individuals was so large that it was hard to ascribe changes in respiration to toxicity. 60 It was also planned to use several extremes of salinity and temperature while testing the organisms, however, great mortalities occurred when we attempted to acclimate shrimp to the extremes of temperature and salinity. Therefore, it was found necessary to test the shrimp at temperatures and salinities close to those at which they were caught. 2. Methodology a. Collection and maintenance of organisms The majority of the organisms were collected aboard the R/V Lorene, research boat of the University of Texas Marine Science Institute, in a flat net (1 1/8 -2" mesh size) with a down time of 5-10 min. and placed in onboard wet boxes equipped for continual aeration. Local sampling areas included the areas adjacent to the Corpus Christi Ship Channel, Redfish and Aransas Bays and the Gulf waters approximately 2-7 fathoms off shore. Upon returning to the Institute, the organisms were placed in an outdoor holding pond equipped with recirculating filters, constant inflow of salt water from a clarified holding tank and a heated water source for maintaining seasonal averages during the winter months. Prior to testing, the organisms were randomly selected from the main holding tank and placed in laboratory aquaria for acclimation to temperature and salinity. While in both holding stages, they were fed a protein enriched pellet developed by Oppenheimer and Subrahmanyam (1970). b. Respiration measurement apparatus The continuous flow respiration measurement apparatus is shown diagramatically in Figure V-14. Each test animal was ~ AIR SUPPLY __.. o=41 'm>T CELL _I' STYROFOAM ICE CHrnT PROBE ELECTRONICS AND RECORDER 0 0 0 0 0 0 CONSTANT TEMPERATURE WATER CIRCULATION FIGURE V-14 Experimental set-up for respiration and activity monitoring. Up to three test cells could be run simultaneously in one water bath. °' I-' 62 confined to a one liter clear plastic cylinder closed with rubber stoppers and equipped with inlet and outlet tubes. Adjustable perforated plastic baffles confined the organism to the central area of the cylinder. The hot-wire 1hermister activity probe was placed slightly posterior to the animal at the top of the chamber. YSI oxygen probes placed in the inlet and outlet tubes were used to measure influent and effluent oxygen concentration. A 20 liter capacity dosing reservoir was continuously aerated to maintain oxygen concentration near saturation. A Cole Parmer Masterflex multichannel variable speed pump was used to supply each of three test cells with a constant flow of water at rates between 3 and 7 liters per hour. Flow rates were adjusted so that the animal's respiration produced a 10 to 50 percent depletion of oxygen in the water during passage through the test chamber. Pumping rate was determined by catching effluent water in a graduated cylinder at the start of a run, and was maintained by monitoring the rotation rate of the pump with a tachometer and correcting the rotation rate as needed. To reduce build up of wastes during a run, effluent from the chambers was run through a filter tube containing glass wool and activated charcoal as it returned to the reservoir. Each reservoir and up to three test cells were placed in a large water bath to maintain constant temperature during a run. In our initial attempts to make respiration runs at extremes of salinity and temperature, temperature was controlled by a circulating water bath which balanced a continuous cooling unit (Forma Scientific Model 2535) against an adjustable heating unit (Poly Science 63 Model 73 Immersion circulator). Due to-difficulties in acclimating shrimp to these extremes, we had to abandon this approach and subsequently used laboratory ambient temperatures of 22 to 26 degrees c. Seawater from the Aransas Pass channel was filtered through Whatman #4 paper before use. Salinity was adjusted with distilled water or with hypersaline water from a controlled evaporation pond. The same water was used in both the reservoirs and in the water bath to facilitate transfer of animals into the test cells. c. Dose control and aeration It was quickly determined that the rate of aeration necessary to maintain dissolved oxygen levels near saturation rapidly removed cc13 or Freon 113 from the water. A once-through constant flow system could not be adopted because of the large quantities of seawater which would be required. It was found that satisfactory control of the level of dissolved halocarbon could be obtained by maintaining a constant vapor pressure of the halocarbon in the air used for reaeration. The quantities of halocarbon which had to be metered into the air stream are small, so the use of metering pumps was not attempted. It was found possible to maintain a very constant level of CCl4 by the use of a mermeation tube arrangement. The air stream was passed through a short length of silicone rubber tubing, bathed on the outside in liquid CCl4. The cc14 diffuses rapidly through the wall and into the air stream, the rate of diffusion remained constant over several months as long as the tubing was not disturbed. The desired level of CCl4 in the air 64 stream could be obtained by adjusting the air flow in inverse proportion to the desired change in CCl4 concentration. The resulting equilibrium concentration was monitored by gas chromatographic analysis of water samples from a small container of water which was heavily aerated, since the small volume of water responded more rapidly to changes in concentration than the large volume in the reservoirs. The experimental set-up is shown in Figure V-15. The silicone tubing used was 4 mm I.D. with 3 mm wall thickness. An exposure length of from 3 to 10 mm with an air flow of 1 to 2 liters/min gave a range of concentration in water of from 1 to 4 ppm CCl4. The silicone tubing swells on contact with cc14 , but equilibrium is established within 24 hours. In swollen condition, the tubing is fragile and should not be disturbed. Once equilibrated, this system provided constant carbon tetrachloride levels for days without adjustment. Control of Freon 113 level was found to be much more difficult, permeation tubes of silicone rubber, tygon, and latex were tried, but a stabilized concentration could never be obtained. The reason for this is unknown. The method which was finally adopted was to pass a slow stream of air through liquid Freon 113 and then mix the vapor laden air with a more rapid flow of pure air, as shown in Figure V-15. d. Recording oxygen levels and activity Each oxygen probe (Clark electrode, Fig. V-14) was biased at 0.80 volts, and the current passed was measured using an operational amplifier in the current measuring mode, followed A) AERATION SYST:E>t FUR cc14 3-WAY VALVE _I' EXHAUST FLOW 4 TEl>T CHAMBER AERATION MEI'ER ~ ~CC14 RF.SERVOIR AIR SUPPLY ---t SILICONE RUBBER TUBING NEEDLE REGULATOR ./ VALVE B) AERATION SYSTEM FUR FREX>N 113 ~TEST CHAMBER AERATION • FLOW ~ME1'ERS ~1. AIR FREDN 113 RESERVOIR SUPPLY ~ 0 0 0 REGULATOR J '-._ NEEDLE VALVE5 J FIGURE V-15 The systems used to provide a constant concentration of halocarbon vapor in the air supplyto the test chamber reservoirs. The system for carbon tetrachloride used a permeation °'U1tube while the system for Freon 113 used a controlled flow of air saturated with Freonvapor mixed with a controlled flow of pure air. 66 by an amplification stage with variable gain and voltage offset. The resulting signal was recorded with a Hewlett Packard 4 channel recorder on thermosensitive paper. Since inflowing water for each of the three chambers was drawn from the same well mixed reservoir, only one channel was used for influent water and the other three were used to measure oxygen content of effluent water. In spite of rigorous attention to grounding, electrical interference problems were encountered from time to time. When the recording system was not operational, oxygen measurements could still be made using a "switch box" with a YSI model 54 oxygen meter. The switch box was used to maintain all probes under 0.80 volts bias and switch them into the measuring circuit of the meter when desired. The activity probe (Figure V-14) consisted of a pair of thermistors mounted in a plastic tube with silicone rubber so that one thermistor was exposed to water movements within the chamber, and the other was relatively shielded. The thermistors were wired in a wheatstone bridge configuration with balance resistors, and a stable, reg~lated power supply provided enough voltage so that the thermistors underwent self-heating. The out-of-balance signal produced by the bridge was fed through an operational amplifier for signal gain control and zero adjustment and recorded on a Hewlett Packard recorder similar to that used for the oxygen channels. The signal produced by this circuit is roughly proportional to the rate of water movement (cooling) past the exposed thermistor, this is the principle of the "hot wire anemometer". Calibration of the oxygen recording system. The following steps were necessary to establish the calibration of each oxygen channel: 1) zeroing the recorder stylus while the probe was immersed in water deaerated by bubbling with nitrogen, 2) adjusting the gain to give a scale reading of about 90 when freshly aerated water from the experimental reservoir is pumped through the probe. The gain was adjusted to give 90 rather than 100 to allow for drift. Periodically during each experiment, the amount of drift occurring in the calibration of the oxygen probes was determined by reversing the flow of water through the chambers until a new equilibrium was established. The new scale readings corresponding to 100% saturation were recorded, and the gain was reset to give a reading of 90. Calibration of the activity probe. No way has been found to produce an absolute calibration of this probe in terms of energy expenditure by the organism, because different types of activity produce different responses. Therefore, the gain was simply adjusted to give a moderate response to moderate movements, under these conditions, extreme movement sent the stylus off scale. Interpretation of oxygen and activity recordings. The continuous recordings of oxygen concentration and activity were converted into numerical form as follows: For each 15 minute period, the average oxygen concentration was determined for each channel. The activity for that period was scored on a scale from 0 to 5, with 0 for no appreciable activity and 5 for 20 or more 68 significant deflections during the period. There is a delay between respiration of oxygen by the organism and measurement of the loss of oxygen from the water, this delay can be calculated on the basis of the volume of the chamber and the pumping rate. If the complete replacement time as calculated by volume divided by pumping rate was greater than 3 minutes, the time frame for reading activity was offset from that for reading oxygen by the replacement time. The numerical values for oxygen and activity were punched in paper tape for computer processing. Computer processing was accomplished in three steps, first, the data was read from the punched paper tape into the computer, second, the data was run through a program which corrected for drift in the oxygen channels, third, the corrected data was used to calculate respiration rates and the correlation between respiration and activity. The following calculations are made: respiration mean, maximum and minimum standard deviation, activity mean, maximum and minimum and standard deviation, product moment correlation coefficient between activity and oxygen uptake, and the slope and intercept of the least squares fit linear equation relating respiration to activity. Respirometer test sequence. Shrimp which were acclimated to the test salinity and temperature were placed in individual test cells and allowed to adapt for at least one hour. Next, oxygen probe electronics were adjusted to give predetermined scale readings when receiving saturated water, and activity probe sensitivity was adjusted to give a roughly mid-scale deflection for moderate swimming activity. 69 A twenty-four hour monitoring period was then used to obtain an initial respiration rate for each animal. At this time the aeration of the reservoir was switched from "clean" air to air with added halocarbon, and the animals were exposed for 48 hours while still in the respiration chambers. Next, the oxygen probes were readjusted and another 24 hour record was obtained. Finally, each animal was weighed and measured and the recorded results were interpreted. The condition of the animals was observed at roughly three hour intervals and dead animals were removed from the system. 3. Results a. Respiration experiments We had initially assumed that up to 1,500 hours of experimental time with the respirometer would be available which would allow a number of 24 hour experimental runs to be made with shrimp, oysters and speckled trout. Two factors prevented us from accomplishing this many experiments: 1) we had initially planned to acclimate the shrimp to extreme temperature and salinity conditions so that they would show maximum sensitivity to stresses from the halocarbons. In repeated efforts we were unable to acclimate sufficient numbers of shrimp to justify experimental runs; 2) problems with instability of the instrumentation prevented the completion of many runs and eventually forced us to go to manual recording methods. When it became obvious that it would not be possible to complete all of the planned respirometer testing, short term exposure experiments in aquaria and in outdoor ponds were planned. 70 These tests included penaeid shrimp treated with Freon 113 and spotted seatrout tested with both Freon 113 and carbon tetrachloride. The results of these experiments are described later in V-D. The results of all respirometer runs resulting in usable data sequences are shown in Tables V-2 and V-3 with runs at 34 ppt salinity and 26 degrees C in the former and at various salinities and temperatures in the latter. In all cases in Table V-2 the first line for a given individual gives the results of the initial 24 hour monitoring sequence while the second line gives the results after 48 hours of exposure to test conditions. In 5 out of 9 cases with exposure to 3.0 ppm CCl4 the shrimp died during the final monitoring period. This result indicates a TLM close to 3 ppm. However, the response in terms of respiration was extremely variable, since 5 shrimp gave higher respiration rates and 3 gave lower rates during exposure. In tests without exposure to CCl4, 4 shrimp gave higher respirations and 2 gave lower respirations after confinement in the test chambers. Exposure to 1.0 ppm CCl4 (Table V-3) gave one shrimp with higher respiration, two with lower, and one with essentially the same respiration. In general, the shrimp exposed to 3.0 ppm cc1 4 showed increased activity, possibly as avoidance behavior. However, due to equipment failures, only 3 control runs are available for comparison. Correlation between activity and respiration was significant in only about 50 percent of all runs, and the sign of correlation was variable. Individual variability between shrimp was quite large as shown by the variation in initial respiration rate. Jackson (1975) TABLE V-2 Respiration measurement results obtained with Penaeus· setiferus at 34 parts perthousand salinity and 26 degrees C. Within each group, each individual is listedfirst for the control measurement period and then for the experimental measurement. Treatments are given in mg/l of CCl4. Batch Individual Wt. Sex Treatment Events Res;eiration ActivityCCl4 N Mean SD Max N Mean SD HK716 A 5.45 m 0 -72 0.210 .13 .484 72 2.4 2.4 3.0 d 93 .307 .18 .724 93 4.6 1. 0 B 7.11 m 0 -72 .143 .14 .421 72 3.4 2.3 3.0 d 93 .461 .14 .789 93 4.6 1. 0 c 5.65 f 0 -72 .307 • 31 1.129 72 3.6 2.2 3.0 d 83 .614 . 46 1. 411 83 4.5 1. 0 LP717 A 15.83 f 0 -94 .165 .06 .337 94 1.1 1.9 3.0 d 64 .043 .08 .361 31 4.6 1. 2 B 13.77 f 0 -94 .180 .06 .312 94 2.1 2.2 3.0 -106 .015 .03 .191 100 3.3 2.0 c 8.43 f 0 -94 .270 .11 .482 94 2.1 2.3 3.0 -106 .019 .06 .504 100 2.8 2.0 HK724 A 8.25 f 0 -78 .024 .05 .176 78 1.6 2.2 3.0 -81 .500 .26 .878 81 4.4 1. 3 B 6.73 f 0 -78 .198 .17 .489 78 0.5 1. 3 3.0 c 8.08 f 0 -78 .018 .06 .393 78 1.3 1. 9 3.0 -45 .228 .13 .489 45 4.5 1. 3 Note: Events d = death, m = moult '1 1--' TABLE V-2 continued Batch Individual Wt. Sex Treatment Events Respiration ActivityCCl4 N Mean SD Max N Mean SD JM726 A 4.56 f 0 -86 .294 .29 1.33 84 2.54 2.40 -80 .433 .47 1.28 80 2.21 2.4 B 8.08 f 0 -86 .358 .18 0.66 84 0.26 0.90 d 58 .059 .25 1.56 17 1. 20 1. 8 c 8.25 f 0 -86 .403 .18 0.75 84 4.89 0.50 -80 .286 .48 1.66 79 3.21 1. 9 LP725 A 8.25 f 0 -91 .154 .15 0.43 91 0.88 1. 7 B 6.73 f 0 -91 1.148 .24 1.63 91 1.51 2.2 c 8.08 f 0 -91 1.187 .22 1. 52 91 1.65 2.2 GJ821 A 3.00 f 0 -90 .432 .20 -90 1.86 1. 8 0 -94 1.547 B 3.88 f 0 -90 .456 .23 -90 2.45 2.40 m 94 1. 66 c 4.58 f 0 -90 .389 .08 -90 2.140 m 94 1.898 2.2 JM718 A 4.83 f 0 -99 .169 .15 0.63 99 0.78 1. 4 B 5.40 m 0 -99 .073 .13 0. 64 -99 1. 37 1. 4 c 5.42 f 0 -89 .179 .45 0.94 89 3.60 1. 7 GJ827 A 4.84 f 0 -116 .285 .26 1.00 90 4.32 1. 4 B 5.75 f 0 -116 .126 .23 1.16 89 1. 81 2.0 c 6.44 f 0 -116 .255 .21 0.67 89 2.15 1. 8 TABLE V-3 Respiration measurement results obtained with Pena·eus ·setifer·us under several different salinity and temperature regimes. within each group, data is given for each individual first for the control period and then for the experimental period if any. Batch Individual Wt. Sex Treatment Events ResEiration mg 02/gm/hr Activity Sal. T CCl4 N Mean SD Max N Mean SD GJ323 A 8.96 f 26 22 0 -91 1.087 0.32 2.461 d 69 1.057 0.31 1. 82 B 9.52 f 26 22 0 -91 0.683 0.39 1. 301 d 90 1.504 0.27 1.66 c 6.25 m 26 22 0 -91 0.627 0.33 1.221 -91 0.463 0.20 1.45 GJ423 A 5.03 f 32 30 0 -93 0.677 0.50 1. 96 91 0.3 0.9 B 6.40 f 32 30 0 -79 0.094 0.29 1.60 77 1.12 1. 5 c 8.51 f 32 30 0 -93 0.737 0.21 1.17 91 4.33 1.41 -93 0.524 0.17 1.09 88 0.32 0.2 GJ426 A 15.78 -35 30 0 -95 0.436 0.13 0.81 95 0.87 1.6 B 9.94 -35 30 0 -95 1.137 0.21 1.44 95 0.21 0.8 HK520 A 15.8 f 35 30 0 -72 0.324 0.23 1.21 52 1.90 2.0 B 13.8 m 35 30 0 -73 0.419 0.17 0.92 66 2.74 2.3 c 12.4 m 35 30 0 -72 0.086 0.16 0.72 56 1. 00 1. 5 HK613 A 7.92 f 32 26 0 -36 0.153 0.20 0.59 36 3.13 1.9 B 11.8 m 32 26 0 ~ -36 0.302 0.25 0.83 36 1.13 1.0 wc 13.4 f 32 26 0 -35 0.170 0.16 0.52 35 0.11 0.5 TABLE V-3 continued Batch Individual Wt. Sex Treatment Events Reseiration m9 02/gm/hr Sal. Activit:t T CCl4 N Mean SD Max N Mean SD HK628 A 3.96 f 35 23 0 -80 2.400 1.58 5.68 80 0.46 1.1B 12.1 f 35 23 0 -80 0.467 0.15 0.81 80 2.50 1.5c 12.1 m 35 23 0 80 -0.058 0.09 0.42 80 2.24 2.4 ""' .i::i 75 also observed large variations in respiration rate, however, his procedure measured large numbers of individuals so that better average respiration data could be obtained. In general it appears that the respiration and activity measurement approach as implemented in this study was not satisfactory as a detector of low level stress in shrimp due to carbon tetrachloride. In order for this approach to be usable, a much more rapid .measurement technique suitable for large numbers of individuals would be needed. b. Short term exposure of shrimp to Freon 113 Static tests were conducted in ten gallon glass aquaria equipped with undergravel filters and external activated charcoal and spun glass filters. The concentration of Freon 113 was controlled by aeration with air containing a controlled amount of Freon vapor. This aeration ,kept dissolved oxygen on the order of 80 percent of saturation. Temperature was the laboratory ambient of 23.5 to 24.5 degrees. Salt water for the tests was drawn from the Aransas Pass channel, and was diluted if necessary with a well aerated tap water. When salinity in the channel water dropped due to rains, hypersaline water from a recirculating evaporation tank was used to adjust final salinity. Shrimp were trawled from local channels and kept in a large covered pond until transferred to the laboratory. Due to fluctuations in availability, both Penaeus aztecus and P. setiferus were used for these tests. Initially ten shrimp, selected at random from acclimating stock were placed in each tank. Since shrimp are cannibalistic under crosded conditions, in some cases 76 exceptionally aggressive or active individuals had to be removed. After at least 12 hours acclimation to the test tank, the standard aeration "air stone" was removed and aeration with halocarbon vapor was started. At one to three hour intervals (except midnight to eight AM), observations of dead, moribund or lethargic animals were made, dead animals were removed and weighed. Gas chromatographic analysis samples were taken in small glass stoppered bottles at frequent intervals. Dissolved oxygen and pH values were checked at roughly three hour intervals, but little variability was noted within a tank. The first set of data were obtained using Penaeus setiferus (white shrimp) at 35 parts per thousand salinity. Before the second set of tests could be started, local rains dropped the salinity so these tests were run at 25 parts per thousand. Both P. setiferus and £· azetecus (brown shrimp) had to be utilized to obtain sufficient numbers. As shown in Table V-4 both sets-of data indicate a TLM between 3.0 and 3.5. Chi square tests were run on the total of both sets of data with the results as indicated. In the long term pond tests (discussed in Chapter V-D) it was found that during some test intervals, penaeid shrimp could survive Freon 113 concentrations up to 5.0 mg/l. The organisms in the ponds, however, were not under the stress due to crowding which the static test organisms were subject to. The first symptoms exhibited in the static tests were usually lethargy and disorientation. c. Short term toxicity tests with trout and carbon tetrachloride. Due to instrumental difficulties, it was not possible 77 TABLE V-4 Summary of 48 hour tests of Penaeid shrimp exposed to Freon 113. Average Freon 113Concentration 35 ppt Salinity 25 ppt Salinity Significance vsmg/l Initial Dead Initial Dead Controls 5.4 9 8 ** 5.0 7 7 ** 3.5 8 8 ** 3.2 9 8 ** 3.0 10 3 20 1 * 2,9 9 4 ** 2.2 10 1 0.0 15 0 19 0 Significance by Chi square, ** p lt .001, * P lt 0.1, -notsignificant, calculated for totals at any given concentration. 78 to conduct the planned respiration measurements on trout. When this problem became clear, short term exposure experiments were designed to obtain some data on the sensitivity of these fish. · Since these animals are rather large, tests had to be conducted in one of the outdoor ponds described in detail in Chapter V-D. The pond used for these experiments had been set up primarily for test organism storage, it was covered with a wooden louver system to reduce temperature extremes and had an external sand bed filter for the removal of particulates. The fish for these experiments, Cynoscion nebulosus (spotted seatrout) and C. arenarius (sand seatrout) could not be caught in large enough numbers on demand. It was necessary to gradually accumulate enough individuals for a test as they turned up in the research trawls of other projects. The. cooperation of Capt. Elgie Wingfield of the Lorene in setting aside these animals is gratefully acknowledged. The animals used were typically in the 6 to 10 inch size range and were presumably less than one year old. The spotted seatrout is one of the major sport and commercial fish of this area. They spawn in grassflats and other parts of the bay system throughout much of the year. The sand seatrout is also an important sport fish. The experiment with carbon tetrachloride was started December 12, 1973 with 35 trout in water with 25 ppt salinity. Monitoring of oxygen and temperature during the previous week indicated that the test pond was typically about 1 degree cooler than the uncovered ponds, but exhibited an appreciable oxygen concentration fluctuation due presumably to filamentous algae attached to the 79 tank sides and bottom. It was found that dispersion of the carbon tetrachloride within the tank could be achieved by a combination of injecting the liquid into the filter pump intake and dispersing droplets of liquid over the sand filter bed. The slow solution of these droplets tended to counteract the rapid loss to the air. The concentration in the pond was determined two or more times per day and additional doses calculated on the basis of these results. Results are summarized in Table V-5. Initial symptoms exhibited by all fish were a "logy" appearance, with all swimming close to the surface. The four which died within the first 24 hours were all gaping, a sign of respiratory distress, however, the four which died the next day were not gaping. Raising the target concentration from 4.7 to 7 ppm did not produce any more deaths, however, some fish seemed to have equilibrium problems. Raising the target concentration to 8 ppm killed all but five of the remaining fish within 48 hours. At this point, aeration of the pond was started to remove remaining carbon tetrachloride. Two more fish died within the next two days. Our interpretation of these results is that the 9 trout killed at levels of 3.2 to 4.7 ppm carbon tetrachloride were more susceptable due to fin rot or other stresses since no other deaths resulted at levels between 4.0 and 7.0 ppm. The initial symptom of carbon tetrachloride poisoning was swimming at the surface as if oxygen starved. The 96 hour LC-50 for these fish probably lies in the range 6.0 to 8.4 ppm. Live fish were removed for analysis of muscle tissue on the 14th at the peak exposure, on the 17th after the concentration TABLE V-5 Summary of carbon tetrachloride toxicity experiment in large outdoor pondstarting December 12, 1973. Day Time Temp. Oxygen CCl4 Remarks oc ppm 7 AM PM 12.6 14.2 77 107 4.7 Initial dose at 1100, 35 trout 8 PM 3.8 4 trout dead -all gaping, others appear logy 9 PM 3.4 all fish close to surface 10 AM PM 12.5 14.0 77 107 3.2 5.0 4 dead, 1 moribund (show symptoms of tail rot) 11 AM PM 14.2 14.5 76 121 4.7 7.0 target concentration raised all trout simming at surface, some equilibrium problems 12 AM PM 18.0 21.0 84 122 4.0 6.7 all swimming normally most swimming at surface 13 AM PM 19.3 20.8 75 124 6.0 8.4 most swimming at surface, act irritable 14 AM 19.3 77 7.9 one dead PM 22.4 134 8.4 trout acting logy and irritable 15 AM PM 4.1 all but 5 dead, strong norther started start aeration to remove CCl4 16 PM 1.1 one dead 17 AM 12.6 100 PM 16.1 82 0.7 one dead, one live trout removed for analysis 00 0 TABLE V-6 Summary of Freon 113 short term toxicity experiment in large outdoor pond with ~noscion nebulosus (spotted seatrout), starting March 18, 1974. Day Time Temp. Oxygen Freon Remarks oc ppm ppm 18 AM 22.9 78 approx. 20 trout present, salinity 20 ppt PM 25.8 106 19 AM -start dose PM 3.0 20 AM 23.8 67 2.2 PM 25.8 110 3.4 no symptoms visible 21 AM 16.2 66 1. 0 strong wind is increasing exchange rate PM 16.3 136 4.3 22 AM 14.3 57 3.0 no symptoms visible PM 16.l 90 4.0 rain 23 AM 2.5PM 5.9 24 AM 0.8 strong wind 25 AM strong wind and rain 27 PM start dose 28 AM 17.7 62 4.5PM 21.1 124 29 AM 19.5 50 3.3PM 23.3 126 7.1 CX> ....... 30 AM 3.6 one dead TABLE V-6 continued Day Time Temp.Oc 4/1 AM PM 25.5 2 AM PM 23.0 24.9 3 AM PM 21. 7 25.3 4 AM PM 19.1 5 AM Oxygen ppm Freon ppm Remarks 114 32 110 53 140 72 4.1 8.9 6.9 2.6 10.4 2.8 9.3 start dose high wind two dead three dead, remainder missing from tank "" 83 had declined to 0.7 ppm and on the 18th. Two samples of dorsal muscle tissue were taken from each fish and analyzed by digestion and distillation. The samples taken on the 14th, when the pond was at least 8 ppm carbon tetrachloride gave concentrations of 17.5 and 17.8 ppm using the water distillation method. The samples taken on the 17th gave 12.0 and approximately 10 ppm (problems with distillation caused lack of accuracy). One of the samples taken on the 18th was analyzed using the water distillation method, and wasone analyzed using heptane distillation, the water method gave 13 ppm while the heptane method gave 72 ppm. This difference is taken to indicate that the recovery obtained from standard solutions in water could not be obtained with actual digestion mixtures. Heptane apparently gives a much higher degree of recovery. Recovery with the water distillation method, however, appears to be consistent, and indicates that the carbon tetrachloride was not lost rapidly from the animals. d. Short term toxicity tests with trout -Freon 113 Short term tests with Freon 113 and spotted seatrout were conducted in March and April 1974 in the outdoor pond same used for the carbon tetrachloride tests. These experiments were hampered by unfavorable weather and had to be terminated before a clearly toxic level had been defined, nevertheless, some usable results were obtained as summarized in Table V-6. The experiment started with approximately 20 animals on March 19, the initial target concentration was 4 ppm. The halocarbon was dispersed into the tank using an ordinary laboratory aspirator attached to the exit line of the filter pump. The inlet line of 84 the aspirator was kept underwater so that no bubbles were introduced into the system, and the Freon was introduced into this line slowly with a syringe, resulting in a finely d~spersed fog of droplets which rapidly dissolved and mixed into the tank. The fish exhibited no detectable symptoms during the four days of exposure. Considerable difficulty in maintaining a constant level was caused by high winds which increased turbulent mixing in the tank. . The experiment was restarted on March 27 and levels of up to 7.1 ppm were obtained for a short period. The one fish which died in this period appeared to have been weakened by fin rot. In the final phase of the experiment, levels of up to 10.4 ppm were obtained, but the overnight loss of Freon was high. Three of the five fish which died in this period exhibited gaping symptoms. The experiment was terminated abruptly when the remaining fish vanished from the tank, presumably stolen by someone who did not realize that they were experimental animals. We were unable to collect enough trout within the available time to repeat the experiment. We can only conclude that the LCSO lies somewhere above the roughly 6 to 7 ppm achieved in this experiment. c. Long term exposure experiments in ponds 1. Introduction and Summary Long term exposure experiments were conducted in pond ecosystems designed to test as many as possible of the direct and indirect effects of carbon tetrachloride and Freon 113 in 85 the estuarine environment. Factors monitored in the ponds included: the survival of common estuarine fish and shrimp, photosynthesis/ respiration, and nutrient cycling. Two ponds were maintained as controls, two were exposed to carbon tetrachloride and two were exposed to Freon 113 ("Freon" and ~Cl4" will be used for the balance of the chapter). Because of the rapid loss of the Freon and CCl4 from the ponds to the air, it was not possible to maintain constant levels of these compounds. Therefore we used a daily analysis and correcting dose to produce a peak or "target" concentration. The concentration fell over the course of 24 hours to from 60 to 20% or less of the peak value, depending on weather conditions. Exposure periods for organisms ranged from 33 to 69 days. All organisms survived Freon target levels as high as 5 ppm, however, 7 ppm killed penaeid shrimp but not fish, and 10 ppm quickly killed shrimp and fish. In ponds with targets of up to 4 ppm CCl4 all test organisms survived as well as in control ponds for most test intervals. Considerable variability in organism survival was found in both test and control ponds, presumably due to the variability of natural stresses. Photosynthesis and respiration rates calculated from oxygen measurements were not affected by Freon. A statistically significant reduction was observed for the pond exposed to a target concentration of 0.24 ppm CCl4. In this pond, photosynthesis/ respiration was about 76% of that in control ponds. Interpretation of the significance of this effect is unclear because the pond exposed to higher concentrations showed less of an effect. 86 Clearly the effect of low levels of CCl4 on photosynthesis and respiration needs more research. Nutrient cycling results showed no effect attributable to CCl4 or Freon. 2. Pond ecosystem description The experimental ponds were designed to simulate a shallow nearshore estuarine environment, with primary production by benthic algae, planktonic algae, and benthic higher plants (grasses). Herbivores and omnivores added to the system were oysters, shrimp, fish, zooplankton, and a variety of benthic invertebrates. The population density of higher organisms which be maintained in can a pond without external food supply is probably limited by the primary production and degradation or turnover rates of nutrients. Experimental work in local grassflats has yielded biomass measurements from 18 to 337 pounds wet weight of fish and shrimp per acre depending on the season (Hellier, 1961, 1962). The average for a year was 89.0 pounds per acre, mostly mullet. For our ponds of 0.00282 acres area, this corresponds to a range of 0.05 to 0.95 pounds with an average of 0.25 pounds (23 to 431 grams, average 113 grams). The ponds, of course, might be expected to support more biomass due to the lack of predators. In Hellier's study, the most abundant top carnivore, the spotted seatrout (Cynoscion nebulosus), averaged 3.8 pounds per acre which corresponds to about 0.01 pound for the area of the ponds. Obviously, it would take much larger ponds to support the higher carnivores. Short discussion of the ecological significance of the organisms used in these experiments: Mugil cephalus -Striped Mullet, are found throughout Texas estuarine systems and the nearshore Gulf and are frequently the most abundant fish found in terms of biomass. They tolerate a wide range of environmental conditions and are hardy laboratory animals. Mullet feed mainly on organic detritus and algae, and they are major food items for important sport fish such as trout and for fisheating birds. They tolerate a wide range of environmental conditions and have been found in waters with salinities from 0.5 to 70 parts per thousand {ppt) and temperatures from 6.0 to 34.5 c. They are hardy laboratory animals and have been used for numerous studies by MSI scientists. Micropogon undulatus -Atlantic Croaker, are found throughout local estuarine systems and the nearshore Gulf and are typically fairly abundant. The juveniles feed on organic detritus, micro crustacea and benthic organisms, and have been found at salinities from 10 to 45 ppt and temperatures from 4.0 to 39.0 C. Croaker are an important food item for trout, flounder, and other sport fish. Cyprinodon variegatus -Sheepshead Minnow, are small fish frequently found along the shallow edges of Texas estuaries feeding on algae and detritus. They are adapted to withstanding extremes of temperature {4 . 8 to 34 C) and salinity {2.0 to 75 ppt), and and are important prey for many wading birds and sport fish. Crassostrea virginica -American Eastern Oyster, a commercially valuable oyster found in Texas estuaries at a wide range of salinities, from 1.4 to 42 ppt, however, the most valuable reefs are generally found in areas with relatively high fresh water 88 input. Oysters feed on suspended particulate organic matter and thus can concentrate pesticides by a large factor. Fundulus grandis -Gulf Killifish, is a small fish common in shallow bay areas of Texas bays and also found in open bay areas. It has a wide tolerance for salinity, 0.4 to 45 ppt and temperature, 5.0 to 34.o0 c. Major food items include benthic organisms, insects and fish, and the Gulf killifish is eaten by· sport fish and fish-eating birds. Fundulus similis -Longnose Killifish is a small fish common in shallow bays, marshes and grassflats and is also found in open bays. The longnose killifish eats benthic organisms and insects, and is in turn eaten by sport fish and birds. It tolerates temperatures from 4.8 to 34.9°c and salinities from 2.0 to 75 ppt. Ruppia maritima -Widgeongrass, is a seagrass fairly common in local bay systems although not as common as Halodule wrightii (shoal grass) or Thalassia testudinum (turtlegrass). Widgeongrass tolerates salinities up to \ 45 ppt but prefers about 25 ppt. As the name implies, this is an important food for ducks. In grassflats, the leaves serve as the substrate for algae and small invertebrates which are fed on by a variety of organisms. When the leaves die and break off, the detritus formed is an important food source to many estuarine organisms. Penaeid shrimp. Three species of penaeid shrimp are found frequently in Texas estuaries and nearshore Gulf of Mexico; Penaeus aztecus, the Brown Shrimp, Penaeus duorarum, the Pink Shrimp, and Penaeus setiferus, the White Shrimp. Although they differ in details of life history and preferred habitat, they 89 all eat similar food items including benthic organisms, micro crustacea, algae and detritus. In captivity they are frequently cannibalistic. They are, of course, the most important commercial organisms on the Gulf coast and are also important as food for most major sport fish. Thalassia testudinum -Turtlegrass. This wide bladed seagrass is an important component of local grassflats. It grows at salinities up to 60 ppt but prefers the range 33 to 38 ppt. The ecological significance of this plant is similar to that of widgeongrass. Sources and treatment of organisms stocked in ponds. Organisms used for stocking the ponds came from the local bay systems, Corpus Christi Bay, Redfish Bay, or Aransas Bay, except that some shrimp were obtained offshore in the Port Aransas area. Fish were typically obtained by seining in shallow flats on the west side of Mustang Island. Oysters were obtained from Mud Isiand in Aransas Bay. Shrimp were obtained by trawling in the bays or Gulf. The ani~als were quickly transported to the Marine Science Laboratory and placed in covered holding ponds similar in design to the experimental ponds. The animals were typically held for 2 days to 2 weeks before they were introduced into the experimental ponds so injuries suffered in catching and handling could become apparent. The benthic community of the ponds was established as follows. A thin layer of sand ("plaster sand" from a local building supply company) about 1.5 inches thick was spread in each pond and washed with tap water. Next, a few inches of sea water was allowed to 90 stand in the ponds for several days. This layer was drained, and the ponds filled with a mixture of seawater and tap water to produce a salinity of 14 ppt. Next, on November 29, 1972 approximately one cubic foot of mud from Corpus Christi Bay was placed in each pond, presumably carrying with it a variety of benthic organisms. Seagrasses were established by introducing roughly one square foot per pond of mixed Thalassia testudinum and Ruppia maritima from Redfish Bay grassflats (at first we thought that we had Halodule wrightii instead of Ruppia, however, when the plants grew, the correct identity was established.) The grasses were planted roughly in the center of each pond where they were least likely to be disturbed by currents. Procedure for inventorying organisms Preliminary work established the fact that it was impractical to harvest the fish and shrimp from the ponds, measure them and return them, because most fish died from this treatment, and because the nets disturbed the bottom too much. We evolved a technique which enabled us to get fairly good counts of organisms with a very high survival rate. In this procedure, a pond was drained to a depth of 2 to 4 inches of water using both the circulating pumps and large siphons with screens over the intakes. The circulation in each pond tended to pile up sediment in the center, so when the water was drained, the fish and shrimp were confined to the deep areas near the edges of the pond. Three or four of the project personnel stood around the periphery of the pond and counted the organisms within their sector, while one person acted as recorder. 91 The pond was refilled as rapidly as possible and the procedure moved to the next pond. By properly timing the draining cycles all ponds could be inventoried within 3 hours. 3. Description of ponds The outdoor ponds used for simulated ecosystem experiments were originally constructed a number of years ago for ecological research (Odum et al., 1963). Six of the set of nine were used for this project. These ponds are constructed of concrete, with a flat floor at about ground level, and approximately 2 ft high walls in the form of a square with rounded corners, giving a total surface area of about 123 square feet. Inside each pond there is a six foot length of concrete wall about 24 inches from one outer wall for the purpose of channelizing the flow from the pumps. The numbering scheme used for these ponds is shown in Figure V-16. The experimental ponds were essentially identical with the following exceptions: pond 6 was painted inside with epoxy paint about a year before the start of the project, and ponds 8 and 9 have the circulation pumps mounted on the side away from the interior wall. Circulation is induced in every pond by a 1/2 horsepower centrifugal pump with a plastic impeller. Intake of water is through a PVC pipe extending to the bottom of the pond and having slots from the usual waterline to the bottom. Near the end of the project, the intake pipes were surrounded with plastic mesh bags to prevent damage to organisms which might be accidentally drawn close to the intake. 92 FIGURE V-16 Experimental pond layout. N 0 1 -low Freon 2 -high cc14 3 -control 4 -organism storage 5 -organism storage 3°0°0 6 -low cc147-·-not used 00 0 8 high Freon 9 -control 0 93 To prevent the local bird population from disturbing the organisms, heavy nets with about 2 inch openings were suspended over the ponds. To avoid temperatures in the ponds reaching unrealistically high values during the summer, canvas sunshades were suspended over the center of each pond. These sunshades were 64 square feet in area and constructed from plain, untreated canvas and rope. The pumps produced a circular circulation pattern with flows of at least 1 cm/sec in every part of the pond except the center. Currents were strong enough to scour away the sand in a small area near the pump outlet. Additional circulation was provided by the frequent strong winds. Seawater for the ponds was provided from a large settling tank which was in turn filled from the Aransas Pass channel during incoming tides. During periods of low rainfall, tap water was added to maintain water levels in the ponds. This water was from the Nueces County Water District #4 system which draws ultimately from the Nueces River. Generally tap water was sprayed into the ponds to allow the chlorine to escape and provide aeration. During periods of high rainfall the ponds were allowed to overflow. Pond monitoring methods One of the main purposes of the pond experiments was to determine if the functioning of an estuarine ecosystem would be disturbed by additions of carbon tetrachloride or Freon 113. Therefore it was necessary to monitor as many ecosystem parameters as possible for lond periods. Photosynthesis and respiration are 94 very basic ecosystem parameters which can be measured indirectly by monitoring the dissolved oxygen in the water (see Odum and Hoskin1 1958 for a discussion of this technique applied to Texas estuaries}. Oxygen and temperature were monitored in the morning and evening for at least five days each week. In addition, notes were made on the weather, appearance of the ponds, dead organisms found, etc. Ideally, the oxygen in the ponds should have been monitored continuously to locate the minimum and maximum con centrations, however, this would have required a large· amount of equipment and manpower. In the work on pond ecosystems by Odum et al. (1963), a number of diurnal oxygen curves were constructed by measurements at frequent intervals. One of the so curves obtained is shown in Figure V-17. Other curves have shown the maximum oxygen occurring from 1500 to 1800, with the minimum always in the early morning around 0500 to 0600. Measurements were typically made at 0700 and 1700 ~ith slight variations due to the schedule of the student working on the project. Another aspect of the functioning of ecosystems is the cycling of nutrients. Nutrient concentrations were monitored in the ponds on a roughly weekly basis for the first year of research then switched to a biweekly basis for the remaining six months. The nutrients measured were nitrate, nitrite, phosphate, silica and ammonia. Turbidity was also measured at the same time as an indicator of phytoplankton density. Nutrients and turbidity were measured with a Hach Chemical Company model DC-DR meter and Hach reagents. Salinities were measured with an American Optical refractometer. Temperatures FIGURE V-17 Typical diurnal oxygen curve from pond ecosystems (lower curve), with solar energy input (upper curve), from Odum et. al. (1963). 1 00 1 0 I I \ \ I \ \ \ ~ I \~ 5 ,. I I I ~ I \ \ \ \ i .... I 0 6 1 2 18 24 HOURS 96 were measured with a mercury in glass thermometer after 8/9/73 and with the oxygen probe thermister before. Oxygen was measured with a Yellow Springs Instrument model 54 meter and probe using the calibration in air procedure recommended by the manufacturer. Estimates of the lower limits of detection accuracy and repeatability are given in Table V-7. Samples were filtered through a 0.45 micron Millipore filter before chemical tests were run. As discussed in the section on "evaporation" experiments CCl4 and Freon 113 are rapidly lost to the air from shallow ponds. Therefore to maintain a relatively constant level, it was necessary to measure the existing level, compute the addition needed, and add the necessary halocarbon on a daily basis. Typically, a water sample was taken from each pond being dosed about .1500 and analysis made by gas chromatography. The amount required to bring the concentration up to the "target" level was calculated on the basis of the current water level in the ponds. The calculated dose was administered to each pond about 1630 by slowly injecting the liquid halocarbon into the intake pipe of the circulating pump. The impeller broke the liquid up into very small droplets which rapidly dispersed into the circulation of the pond. A small amount was vaporized and apparently lost as gas, but the extra amount needed to compensate for this was determined experimentally. Occasionally samples were taken about 30 minutes after administering the halocarbon. Analysis of these samples usually indicated that the desired level + 1520 % had been obtained. TABLE V-7 Estimates of the accuracy, repeatability and lower limit of detection of the various analyses. Repeatability estimates for values near the middle of the range. Parameter Method Units Estimated Repeatability Lower Limit Range Accuracy Oxygen YSI probe % sat. + 10% + 5% 5% 0-200 N03+N02 HACH NitraVer IV ppm-N + 5% + 5% 0.01 0-1. 5 (no dilution) N02 HACH NitriVer ppm-N + 20% + 10% 0.005 0-0.2 - NH4 HACH Nessler ppm-N + 10% + 5% 0.05 0-3.0 - P04 HACH StannaVer ppm-P04 + 10% + 5% 0.05 0-2.0 - Turbidity HACH Turbidity JTU + 10% + 10% 10 0-500 Silica HACH Heteropoly ppm-Si02 + 5 to 10%* + 5% 0.15 0-15 - blue {Molybdate) dilution 1:5 Salinity Refractometer ppt + 1 + 0.5 o.o 0-70 Temperature YSI probe oc + 1 + 0.5 0.0 0-70 - - Mercury in Glass Thermometer oc + 0.05 + 0.05 0-50 \0 *Silica scale found to give values 5 to 10% too high. -...J 98 Under this procedure, the pond ecosystems were subjected to concentrations of halocarbon ranging from 20 to 115% of the "target" concentration. To obtain a finer control over the concentration it would have been necessary to devise some method of continuous dosing. Most of the continuous dosing systems which were suggested, however, ran the risk of catastrophic failure which could subject the ponds to much higher doses, possibly ruining an extended experiment. 4. Pond stocking and sampling history Analysis of nutrients started on 1/15/73, and the first stocking of animals started shortly thereafter. Oysters were first stocked on 1/25/73, and some experimentation was necessary to find a way of supporting the oysters off the bottom. Finally we settled on a system of small plastic "strawberry baskets" suspended in the water by nylon monofilament line, holding a total of 48 oysters per pond. This system enabled us to withdraw the oysters for examination without disturbing the substrate. Introduction of fish began on 2/7/73 with 24 mullet and 32 Fundulus similis added per pond. The mullet were juveniles between 70 and 110 mm in standard length, the Fundulus were presumably adult, about 1/2 the size of the mullet. Nets were placed over the ponds on 3/27/73 to discourage birds. Samples of typical fouling organisms were obtained by hanging approximately 1 foot square plates of asbestos cement sheet in the university marina for about 3 months. On 3/29/73 two plates were placed in each pond, suspended by rope so that they could be pulled out and examined easily. At this time 6 hermit crabs were also introduced into each pond. 99 An experimental attempt was made on 3/30/73 t.o drain pond 3 and count all animals. This treatment proved so rough that all 15 mullet still present died. This pond was refilled and restocked with 24 juvenile mullet. At this time 50 penaeid shrimp and 12 juvenile croaker (Micropogon undulatus) were stocked in each pond. Daily oxygen, salinity and temperature monitoring was started at 0700 on 5/7/73. The day/night oxygen changes for the first 3 days are shown in Table V-8. All of the ponds are seen to be undergoing substantial oxygen changes typical of high production and respiration, however, ·the range is rather large. The first doses of halocarbon were given about 1600, on 5/9/73 with the "target" concentrations as follows: Low Freon, 1.0 ppm in pond l; high Freon, 10.0 ppm, in pond 8; low CCl4, 0.2 ppm in pond 6; high cc14 , 2.0 ppm, in pond 2. Ponds 3 and 9 were used as controls. Samples taken about 30 minutes after injection of the liquid halocarbons into the pump intakes indicated that the concentrations achieved were about 80% of the target concentrations. Subsequent monitoring indicated that the concentration dropped about 40% over night. On the basis of these observations, routine daily analysis and dosing was started as discussed in the methods section. The only obvious effect of the halocarbons was the appearance of dead mullet and Fundulus, floating on the surface of pond 8, the 10 ppm Freon pond, within one day of the first dose. The oxygen readings seemed to simply continue previous trends. In spite of the dead fish, oxygen readings were not unusually low 100 TABLE V-8 Diurnal oxygen changes before and after the introduction of halocarbons into the ponds, first in percent saturation. TIME SPAN POND NUMBER Day Time Day Time 1 2 3 6 8 9 5/7 0700 5/7 1700 51 66 48 39 26 37.5 5/7 1700 5/8 0700 52 66 51 36 25 40 5/8 0700 5/8 1700 65 70 . 59 43 36 47.5 5/8 1700 5/9 0700 63 70.5 65 46 39 58.5 5/9 0700 5/9 1700 39 77.5 56 41 51 42 average delta 02 54.0 70.0 55.8 41.0 35.4 45.1 std dev delta 02 10.5 4.7 6.7 3.8 10.6 8.4 average 02 cone. 94.8 106.4 95.2 92.5 87.5 82.3 5/10 0700 5/10 1700 49 72 68.5 54.5 79.5 46 5/10 1700 5/11 0700 52 81 63.5 51 53 40.5 5/11 0700 5/11 1700 57 94 70.5 49 53 45.5 average delta 02 52.7 82.3 67.8 51. 5 51. 8 44.0 std dev delta 02 4.0 11.1 3.6 2.8 2.0 3.0 average 02 80.0 103 94.6 94.1 88.4 79.6 101 in pond 8. Results obtained on examination of the fish for halocarbon residues are given later. Actual measurements on all ponds are given in the Appendix. Dosing with these target concentrations continued until 6/19/73 when the ponds were drained and organisms counted. The results are shown in Table V-9. Some obvious points from this data are that: 1) only the hermit crabs and seagrasses survived in pond 8, the 10 ppm Freon 113 pond; 2) the oysters generally did poorly; 3) there was considerable variation in organism survival between the controls. In addition to those organisms which had been deliberately stocked, some accidentals were found. The possible juvenile redfish probably were mistaken for juvenile croaker on the original stocking, and the blue crabs probably came in as very small individuals on the seagrasses. The flounder in pond 1 was probably introduced into this pond accidentally before the nets were placed over the ponds by someone who did not realize that an experiment was in progress. Pond 1 is next to a pond used for storage of a variety of specimens. The flounder was seen once, and then vanished, probably by burying itself in the mud. The presence of this flounder probably accounts for the relatively small number of fish and shrimp left in pond 1. Before refilling the ponds, small clumps of the seagrass, Ruppia maritima were transplanted from pond 3 to pond 2 and from pond 8 to ponds 6 and 9. Halocarbon addition was continued for ponds 1, 2 and 6 starting on 6/21/73 with slightly increased target concentrations TABLE V-9 Organism observations after first test interval, 5/9/73 to 6/19/73. POND NUMBER 1 8 2 6 3 9 Target-ppm 1 Freon 10,Freon 2.0 CCl4 0.2 CCl4 CONTROL CONTROL Penaeid shrimp 0 0 6 (large) 5 0 2 mullet 6-8 0 8-10 12 18 4 Fundulus 0 0 10-12 4-5 0 0 croaker 0 0 2-3 8-9 0 0 hermit crabs 4 4 4 4 2 4 Thalassia condition fair poor fair fair fair fair Ruppia condition 2 patches good none none good none oysters 2 l? 0 6 0 2-3 fouling plates fair poor fair fair fair fair Additional organisms not itentionally stocked -small blue crab in 2 and 3, possible juvenile redfish in 2, 3 and 6, a small flounder in 1. ...... ""' 103 for 2 and 6 to 2.4 and 0.25 ppm cc14 respectively, and the same target of 1.0 ppm Freon for pond 1. On 7/5/73 mullet and Fundulus similis were added to bring each pond up to 20 of each. Thirty penaeid shrimp were added to each pond on 7/10/73. On 7/18/73 dosing of pond 8 was started with a target concentration of 5 ppm Freon. The ponds were drained on 8/29/73 and the observations shown in Table V-10 were made. The flounder was again seen in pond 1 and this time we were able to capture it. A redfish approximately 8" in length was found in pond 9, presumably this was introduced by someone not connected with the project. This opportunity for a taste test was not neglected; the flounder from pond 1 (low Freon), the redfish from pond 9 (control) and_ 2 five inch mullet from ponds 2 and 6 (high and low CCl4) were cleaned and broiled (without spices) and were found to taste delicious. The ftounder and redfish probably accounted for the lack of shrimp in ponds 1 and 9. For the third test interval dosage began on 9/1/73 with target concentrations the same as for the second interval. Approximately 60 shrimp, 20 mullet and 20 Fundulus were stocked into each pond early in this period, unfortunately the exact number and date have been lost. This experimental period was notable for the near passage of a hurricane during September 4 and 5, and considerable rain during the entire period. The ponds were drained and refilled on 10/26; the observations are shown in Table V-11. It was apparent at this time that the Thalassia was not surviving, and that the Ruppia (Widegeon grass) TABLE V-10 Observations of organisms after the second test interval 6/20/73 to 8/29/73. POND NUMBER 1 8 2 6 3 9 Target -ppm 1.0 Freon 5.0 Freon 2.4 cc14 0.24 CCl4 Control Control after 7/18 t. Penaeid shrimp 0 8 2 2 3 0 mullet 9 5 12 12 22 6 Fundulus --0 1 2 6 1 0croaker 0 1 0 0 0 0 hermit crabs 5 4 5 6 6 3Thalassia condition small 1 leaf none 1 good 3 small 1 smallpatch clump clumps clumpRuppia condition 3 small 10 patches 2 small 2 patches none 1 floatingpatches much floating patches plantoysters none none none none 4 1fouling plates good poor good poor fair fair Additional organisms flounder in 1, gulf crab (Callinectes danea? sp.) in 3sheepshead minnow (Cyprinodon variegatus) in 6, redfish in 9. ~ 0 ~ TABLE V-11 Organism observations after third test interval 8/30 to 10/26/73. POND NUMBER 1 8 2 6 Target -ppm 1.0 Freon 5.0 Freon 2.4 CC1 4 3 9 0.24 CCl4 Control Control Penaeid shrimp 55 35 19 40 50 67 mullet 8 5 5 11 11 9 Fundulus 3 1 2 1 0 1 croaker 0 0 0 1 0 0hermit crabs 3 5 1 2 3 4 Thalassia condition none none none none none noneRuppia condition none good good none none none large patches patch oysters 0 0 0 0 0 0 fouling plates -no notes Additional organisms "spot croaker" in 3? sp.sheepshead minnows (Cyprinodon variegatus) in 2,3 and 6; ...... 0U1 106 was doing well only in ponds 2 and 8. Small patches of grass were transplanted into the other ponds from 2 and 8. Dosage for the fourth test interval began on 10/27/73 and ran to 12/17/73. Since the shrimp and mullet had survived well, no organisms were added. Very low temperatures occurred during this period, on January 4th a temperature of 3.9°c was measured in pond 2. These low temperatures were probably the cause of the low numbers of shrimp found when the ponds were drained on 1/9/74, since a number of dead shrimp were found at that time. As shown in Table V-12, however, some mullet and Fundulus survived in all ponds. Samples of fish from each pond and shrimp from ponds 1 and 9 were preserved in buffered 10% formaldehyde. These preserved organisms were sent to the Department of Veterinary Medicine at Texas A&M University for pathological examination. For the fifth test interval, the high cc14 and high Freon dose levels were increased since both fish and shrimp had survived these levels during the previous test intervals. The target concentrations were thus increased from 5.0 to 7.0 ppm Freon and .from 2.4 to 4.0 ppm CCl4, while the low concentrations remained at 1.0 ppm Freon and 0.24 ppm cc14 • The following organisms were added to each pond on 1/22/74: 60 shrimp, 8 mullet, 11 Fundulus, 1 croaker and 1 hermit crab. Dosage started on 1/28/74 and continued to March 5, the ponds were drained and organisms counted on March 8 (Table V-13). Good survival was obtained for all species except the shrimp in pond 8, the 7.0 ppm Freon pond. Once again, the seagrass, Ruppia was growing well only in ponds 2 and 8, so small patches were transplanted into the other ponds. TABLE V-12 Organism observations after fourth test interval 10/27/73 to 1/9/74. POND NUMBER 1 8 2 6 3 9 Target -ppm 0.0 Freon 5.0 Freon 2.4 CCl4 0.24 CCl4 Control Control Penaeid shrimp 1 0 0 0 0 2 mullet 6 5 15 10 6 8 Fundulus 2 1 3 2 1 2 croaker 0 0 0 1 0 1 hermit crabs 1 1 0 3 2 2 Thalassia none none none none. none none Ruppia condition none fair good poor none none patch fouling plates fair no live poor no live fair no live barnacles barnacles barnacles sediment organism few many few many many very few holes Additional observations, recently dead shrimp were found in 1, 3, 8 and 9 pond 6 contained one sheepshead minnow (Cyprinodon variegatus). I-' .....J TABLE V-13 Organism observations after the fifth test interval 1/28 to 3/8/74. POND NUMBER 1 8 2 6 3 9 Target -ppm Penaeid shrimp 1.0 Freon 7.0 Freon 4.0 CC14 22 0 14 0.24 11 CCl4 Control 18 Control 18 mullet 12 10 11 18 10 14 Fundulus 8 12 7 8 10 5 hermit crabs 5 2 1 1 3 2 Ruppia condition dormant? good patch good patch dormant? dormant? dormant? fouling plates - all are covered with algae, no barnacles - Additional observations --one Fundulus in pond 9 killed by pump during draining. I-' 109 For the sixth test interval, the only organisms added were 40 shrimp to pond 8, and 15 oysters were added to each pond. The oysters were simply placed on the sediment of each pond, resting on the most highly curved valve. Dosage ran from March 12 to April 1. The ponds were drained and organisms counted on April 24, results are shown in Table V-14. Shrimp from 1, 3 and 9 were preserved in buffered 10% formaldehyde. Small patches of Ruppia were transplanted into ponds 1, 3, and 9, and 20 shrimp were added to each pond. Each pump inlet was covered by a small mesh nylon bag to reduce the chance of small organisms being caught in the inlet hoses. Target dose levels remained the same, 1.0 and 7.0 ppm Freon in ponds 1 and 8, and 0.24 and 4.0 CCl4 in ponds 6 and 2. Treatment started on May 13 and continued to June 14. During this period, the seagrass in pond 8 grew rapidly and bloomed extensively. Samples were taken for fluoride analysis on June 11. This period was also notable for a windstorm which tore the canvas sunshades so that they had to be removed. The results of draining the ponds on June 15 are shown in Table V-15. The ponds were refilled but no further experiments were undertaken. 5. Discussion of organism survival in ponds First, a number of general points should be mentioned. The survival of organisms in test and control ponds was quite variable from one period to the next, except for those conditions which resulted in clear toxicity. One has only to compare the two control ponds to realize that there are significant random factors affecting the survival of organisms. It would have TABLE V-14 Organism observations after the sixth test interval 3/12 to 4/24/74. POND NUMBER 1 8 2 6 3 9 Target -ppm 1.0 Freon 7.0 Freon 4.0 CCl4 0.24 CCl4 Control Control Penaeid shrimp 2 0 0 0 8 14 mullet 10 8 7 13 8 13 Fundulus 4 8 8 4 5 7 hermit crabs 5 1 1 1 3 0 Ruppia condition none very large good large good none none patch patch patch oysters 13 11 10 10 8 9 fouling plates --all remain covered with algae Additional observations --sediment in 2 is darker than that in other ponds. I-' I-' TABLE V-15 Organism observations after seventh test interval 4/13 to 6/15/74. POND NUMBER Target -ppm Penaeid shrimp mullet Fundultis hermit crabs Ruppia condition Oysters 1 1.0 Freon 9 10 3 2 small patch 0 8 7.0 Freon 0 3 (see note) 3 very extensive 1 2 4.0 CC14 1 5 5 1 large patch 2 6 0.24 CC14 1 11 5 0 medium patch 0 3 Control 3 7 3 2 single sprig 1 9 Control 12 10 6 2 small patch 1 Additional observations -the grass in 8 is so dense that the Fundulus cannot be counted, several are present; many zooplankton and small fish are present in 8. 1--' 1--' 1--' 112 been better to have collected more data on the survival of organisms, and the variability of other parameters for several control periods. Because of the variability in the survival of organisms, no statistical treatment of this data has been undertaken. However, some clearcut distinctions can be seen. The Freon results clearly show good survival of all organisms at 5.0 ppm "target" concentration (it should be remembered that this is the daily peak concentration and the average was less). A target concentration of 10.0 ppm Freon quickly killed everything except Ruppia (Widgeongrass) and hermit crabs. Seven ppm Freon killed penaeid shrimp consistently while mullet, Fundulus spp. and oysters survived. The pond with high Freon dose (pond 8) produced a lush growth of the seagrass Ruppia maritima, but it is not clear whether this was due to stimulation of the grass by Freon, the lack of shrimp in pond 8 over most test intervals, or some other factor. In the ponds treated with CCl4, organisms survived as well as in the control ponds during most test intervals. Apparently, levels even higher than 4.0 ppm cc14 would have been necessary to show a definite organism survival effect. The successful growth of the seagrass in pond 2 became apparent as early as the third test interval (Table V-11) under a dose "target" concentration of 2.4 ppm, and continued when the dose was increased to 4.0 ppm. Pathological examination results Samples for histopathological examination were taken on two occasions. The first sample, taken 1/9/74, contained fish 113 from all ponds and shrimp from ponds 1 and 9. The report of the pathologist is shown in Figure V-18. Since no effect had been shown in the fish, the next sample consisted entirely of shrimp. The results, shown in Figure V-19, indicate that the "fatty change" originally suspected to be related to Freon exposure also appeared in control shrimp and thus was probably related to conditions common to all ponds. A list of all organisms examined is shown in Table V-16. The largest gap in the pathological results is the lack of shrimp exposed to high Freon and high cc14 doses. Furthermore, only one shrimp exposed to low CCl4 levels was available for examination. 6. Halocarbon content of animals exposed in ponds. At several points during the long-term exposure experiments organisms from the ponds were analyzed for halocarbons using the digestion and distillation procedure described in section I. The first set of analyses was made on fish from pond 8 (high Freon) which were found dead on the third day after exposure began (May 11, 1973). One mullet and one Fundulus was similis were analyzed at once, and one of each species frozen in a plastic bag filled with pond water. The whole fish were digested without further treatment. The water distillation procedure was followed, and analysis indicated Freon content of 19 ppm for the mullet and 0.9 ppm for the Fundulus. A striking difference in concentration was also found when the frozen samples were processed the same way 14 days later, the mullet gave 17 ppm and the Fundulus gave 2.0 ppm. Since the pond 8 concentration varied from 4.4 to 8.1 during the exposure period, these Figure V-18 Pathologist report on samples of 1/9/74. 115 TEXAS A&M UNIVERSITY COLLEGE OF VETERINARY MEDICINE COLLEGE STATION, TEXAS 7784'3 Deparlme11t of Ap+il 19, 1974 VETERINARY PATHOLOGY Mr. William B. Brogden The University of Texas Marine Science Institute Port Aransas, Texas 78373 Dear Mr. Brogden: This is the final report concerning the histiopathologic examination of the specimens I received in February 1974. The tissues examined in each case included the gills, liver, pancreas, spleen, intestine, kidney, gonad, and skeletal muscle. Additional organs including endocrine glands were examined in some specimens. No lesions compatible with those produced by· halogenated hydro~ carbon toxicity were found in any fish. In fact, the control mullet, 9-A, had moFe hepatic lipidosis than any of the exposed fish. The shrimp were more fruitful in that shrimp 1-A had severe fatty change of the digestive gland. This lesion, absent in the control shrimp, is compatible with that produced by certain halogenated hydrocarbons. All of the fish had some degree of gill epithelial hyperplasia. I suspect that this is related to your management and could be either dietary or toxic in origin. The only other microscopic findings included renal myxosporidiosis in mullet 9-A, microsporidial myozooiasis in mullet 8-A, and anomalous gill filaments in croaker 6-B. I hope this report will be of value to you. Please contact me at 713-845-2651 if you have any questions. Sincerely, !(«-L-ttt_ ·y. R. A. Bendele, Jr., D.V.M: Assistant Professor 'Ril/cs 116 TABLE V-16 Organisms sent to pathologist. The first group was exposed from 9/1/73 to 12/17/73 and removed from the ponds on 1/9/74. The second group was exposed from 5/13 to 6/15 and sampled on 6/15. Pond Group 1 1 (low Freon) 2 (high CCl4) 3 (control) 6 (low CCl4) 8 (high Freon) 9 (control) Group 2 1 (low Freon) 3 (control) 6 (low CCl4) 9 (control) Number Organism 1 Mugil cephalus (mullet) 1 Fundulus grandis (killifish) 1 Penaeus aztecus (brown shrimp) 2 Fundulus grandis 1 Mugil cephalus 1 Mugil cephalus 1 Fundulus grandis 1 Mugil cephalus 1 Fundulus 9:randis 1 Micropo9:on undulatus (croaker) 1 Cyprinodon varie9:atus (sheepshead 1 Mugil cephalus 1 Mu9:il cephalus 1 Fundulus 9:randis minnow) 1 Penaeus 2 Penaeus 3 Penaeus 1 Penaeus 1 Penaeus 1 Penaeus 3 Penaeus setiferus (white shrimp) setiferus (white shrimp) aztecus (brown shrimp) setiferus aztecus aztecus setiferus Figure V-19 Pathologists report on samples of 6/19/74. / 118 TEXAS A&M UNIVERSITY COLLEGE OF VETERINARY MEDICINE COLLEGE STATION, TEXAS 77843 July 25; 1974 Deparlment of VETERINARY PATHOLOGY Mr. William B. Brogden Research Associate University of Texas Marine Science Institute Port Aransas, Texas 783 73 Dear Sir: The results of the histopathologic examination of the shrimp you senton 7-3-74 are not very exciting. Two of the three control animals, M-9and J-3, had fatty change in the cells of the hepatopancreas. Likewiseonly two of the six exposed shrimp had the same lesion. The exposed shrimp,J-1-B, had the most severe lesions of the four animals with fatty change.However, none was as severe as those seen in the affected shrimp from thefirst test. In view of these results I believe that the fatty change isprobably associated with some dietary deficiency rather than the freon.Specific amino acid deficiencies are probably the most likely cause. One major problem encountered in this sample was that the specimenswere not well fixed. Therefore, many of the internal organs, especiallythe hepatopancreas, were distorted due to post-mortem autolysis. Thisdictated that I make my interpretations based on only the periphery ofthe glands that were not autolytic. Thus, some bias was introduced intothe results. The animals should be collected immediately after death and shouldbe left in neutral, buffered 10% formalin for at least 24 hours. Enclosed you will find a bill for $5.00/specimen. Please contact me if you have any questions. R. A. Bendele, Jr., D.V.M.Assistant Professor . RAB:cs Encl. 119 concentration factors were about 3 for the mullet and 0.2 for the Fundulus. There is no readily apparent reason for this difference. The next analysis was performed on a mullet which had been exposed for 35 days in pond 6 to low levels of carbon tetrachloride. The viscera and dorsal muscle were analyzed separately, yielding a carbon tetrachloride content of 0.82 ppm in the muscle and 0.10 ppm in the viscera. Large unknown peaks were found in the gas chromatogram, both before and after the carbon tetrachloride peak. These peaks were shown to be due to polar compounds by adding a small amount of heptane to the water distillate. Almost all of the carbon tetrachloride went into the heptane, while the other compounds remained in the water. This was taken to mean that the unknowns were polar. When the ponds were partially drained on October 26, 1973, one shrimp was removed from each pond for analysis using the digestion and distillation procedure. Shrimp were kept on ice in whirl-pak bags for 1 to 4 hours before digestion of the whole animals. No Freon was recovered from the shrimp, and the recovery of carbon tetrachloride was 0.55 ppm for the shrimp from the high level pond (1.4 -2.4 ppm) and 0.017 ppm for the shrimp exposed to .24 to .16 ppm. These results suggested that either substantial amounts of halocarbon were lost before analysis took place or that recovery was poor or both. Two major unknown peaks appeared in control and experimental samples. One of the digestion mixtures was saved, and recovery of halocarbons was checked by adding 5 ml of standard solution containing 1.0 ppm carbon tetrachloride and 5.0 ppm Freon 113. Five ml of 120 water was distilled and collected. The recovery of carbon tetrachloride was only 24 percent and the recovery of Freon was 22 percent or less. The Freon peak was partially obscured by an unknown with a similar retention time. After these results the technique for distillation with heptane was devised. The final analysis was made on June 15, 1974 on organisms captured when the ponds were drained. Samples were kept on ice in whirl-pak bags until digestion and distillation using the heptane technique. The results are summarized in Table 17. All of the results with grass show a concentration factor less than one, however, there was undoubtedly some loss from the plants before digestion since the blades have a very high surface area. The analyses of fish dorsal muscle show generally a concentration factor in the range 3.7 to 5.1 with the exception of the one killifish sample which was too small to skin and was theref9re analyzed skin and all. It is possible that this is due to a high concentration of Freon in the subcutaneous fat which was not sampled in the other fish. 7. Treatment of pond oxygen data As discussed in the section on pond monitoring methods, oxygen measurements were taken at about 0700 and 1700 each day for five days per week. Due to problems with the oxygen probe and meter, rain, etc., few continuous five day records were obtained. However, a number of satisfactory 2 to 4 days sequences were obtained. We have chosen to analyze the oxygen data in two ways, first with respect to the average oxygen concentration and TABLE V-17 Halocarbon content of organisms exposed in long-term pond experiments. Samples taken June 15, 1974 after 15 months of exposure. Pond Concentration Organism Tissue Target Final 6 CC14 0.24 0.16 mullet (Mugil cephalus) dorsal muscle grass (Ruppia maritima) whole 2 CCl4 4.0 2.0 killifish (Fundulus dorsal muscle A similis) " B " grass (Ruppia martima) whole 1 Freon 1.0 0.45 shrimp (Penaeus sp. ) tail muscle " " head mullet (Mugil cephalus) dorsal muscle 8 Freon 7.0 2.7 killifish (Fundulus tail with skin similis) grass (Ruppia martima) whole Tissue Concentration ppm 0.86 0.07 11.1 12.8 2.78 3.9 5.5 3.6 185 1.34 Concentration Factor 4.3 0.35 3.7 4.3 .92 5.5 7.9 5.1 41. 0.3 t--J l\J t--J 122 second with respect to the diurnal oxygen change. The pond ecosystems imported and exported very little external organic matter and we assume that there was little accumulation of organic material. Therefore, over the long term, oxygen production and consumption should balance. If the reaeration coefficient and temperature were constant, this implies that the average oxygen concentration should be 100% of saturation over the long term. Phytoplankton blooms in which photosynthesis was in excess would be balanced by periods in which respiration was in excess, and this should be true for ponds in which little productivity occurred as well as those for which productivity was high. The long term oxygen average for all ponds, using the averaging procedure discussed below, was found to be 85.16 percent of saturation. This is lower than the 100 percent expected and can probably be explained by the following factors. 1) The calibration procedure for the oxygen meter, although repeatable from day to day, may not be accurate. 2) Readings at 0700 and 1700 may not reflect the true minimum and maximum oxygen concentrations. 3) Reaeration coefficients are undoubtedly higher during the day due to higher wind speed, thus depressing the size of the oxygen maximum. 4) The oxygen surplus and deficit parts of the diurnal curve are not necessarily symetrical, therefore the maximum and minimum do not accurately reflect total production and respiration. In spite of these deficiencies, the oxygen readings are a consistent set of data and can be used to compare the functioning of the control and experimental ecosystems. 123 All oxygen measurement sequences of two days or longer were used for analysis, starting with a morning reading and ending with an afternoon reading. Thus a two day sequence would have four readings to be averaged and would contain three diurnal change measurements, a three day sequence would have six readings to be averaged and 5 diurnal change intervals. Data analysis was performed with a Hewlett-Packard model 25 programmable calculator which took a sequence of oxygen percent saturation values and computed the average oxygen concentration and the average diurnal change in percent saturation. Average oxygen concentrations are tabulated for each pond in Table V-18 and presented graphically in Figures V-20 and V-21. The average diurnal change or "delta" values are tabulated in Table V-19 and graphed in Figures V-22 and V-23. Furthermore, the average of all six experimental ponds for both oxygen and delta oxygen is given in Figure V-24, along with the average temperature and salinity. Interpretation of pond oxygen data To interpret the oxygen data we must first consider all of the factors which can contribute to differences between ponds and to changes in time for all ponds. These differences can be conveniently divided into physical and biological factors. By physical factors we mean wind, temperature, light, salinity and circulation. Under biological factors we include nutrients, organisms, and the experimental treatments. Differences in physical factors have been reduced as much as possible by using physically identical ponds, however, the 124 TABLE V-18 Short term average oxygen concentrations in ponds. Start Days Ponds Date Averaged 1 2 3 6 8 9 5/07/73 3 94.8 106.4 95.2 92.5 87.5 82.3 5/10/73 2 80.0 103.0 94.6 94.1 88.4 79.6 5/22/73 4 93.9 101.1 85.8 96.1 75.9 96.3 5/29/73 4 80.8 86.7 93.4 83.l 86.7 93.5 6/26/73 2 82.9 80.1 73.9 81.5 93.3 86.5 7/11/73 3 89.8 89.8 86.0 75.0 87.6 93.7 7/16/73 5 84.3 88.6 82.8 87.0 101.1 90.9 7/25/73 2 83.9 82.8 77.8 83.0 94.1 83.l 7/31/73 4 84.4 84.4 83.9 85.4 98.3 78.5 8/06/73 4 87.6 78.9 82.9 91. 3 98.4 72.8 8/13/73 4 78.9 74.3 79.2 84.6 92.3 80.3 8/22/73 3 94.2 69.9 86.7 81. 2 86.5 80.2 8/27/73 2 72.3 70.6 74.0 73.9 45.1 69.3 9/12/73 2 106.3 87.8 97.4 99.8 92.6 133.8* 9/17/73 4 89.6 70.5 83.7 83.4 84.2 93.5 9/24/73 5 88.1 79~7 82.9 76.6 83.9 83.4 10/07/73 3 87.9 84.3 92.6* 79.6 88.3 79.6 10/22/73 2 79.5 73.0 72.8 73.3 75.6 86.6 11/05/73 3 83.2 67.0 83.3 75.0 81.3 79.8 11/14/73 2 98.1 78.4 97.1 78.9 86.5 111. 3 11/28/73 3 82.6 81. 2 83.8 77.8 79.3 89.0 12/05/73 3 80.9 76.0 78.5 67.8 74.8 80.8 12/10/73 5 75.0 76.8 80.6 70.2 76.0 75.1 12/17/73 2 83.1 86.3 86.0 76.0 80.9 61.5 125 TABLE V-18 continued Start Days Ponds Date Averaged 1 2 3 6 8 9 1/28/74 4 94.9 88.2 94.0 83.3 93.2 91.4 2/04/74 2 87.3 87.6 85.5 83.4 88.9 89.4 2/11/74 2 81.9 84.3 90.3 80.0 92.8 86.6 2/18/74 2 87.3 76.8 95.8 78.5 86.8 95.0 2/26/74 2 87.9 86.0 98.9 81. 4 88. o· 90.0 3/04/74 4 92.8 93.2 103.8 86.5 87.4 89.7 3/21/74 2 72.4 84.3 82.3 77.0 84.5 82.0 3/28/74 2 85.3 99.3 89.3 80.1 96.5 93.l 4/02/74 2 93.5 100.9 89.0 80.1 86.6 98.0 4/08/74 2 84.3 81.1 86.5 76.3 87.8 88.0 *circulation pump problems 126 Figure V-20 Average oxygen concentrations in ponds 1 and 8, including all periods of 2 or more days of records. Pond 1 was the "low" Freon dose pond and pond 8 was the "high" Freon dose pond. The time scale shown along the bottom is days from \ 1/1/73, while the months are shown along the top. 1973 1974 M J J A S 0 N D J F M 100 ~ 90 \ I 0 H V\ ./\_,./y . '·-·"'//\,_v \.1·-· I./\. 8 'f"" ~ e so 8 C5 . < 0... Cf) 70 8 ~ 100 /\l\0 P::: • f:::l 0... I 90 \/\._._. ·""'· '\ !\~. • ~ 80 \_/\ /'\_ i "\/ ~ / ''\.... • 0 CX) ~ A 70 0 z P1 ~ 60 ~ ')0 1 20 160 200 240 280 320 360 400 440 460 DAYS FROO 1 /1 /73 t--' N -...J Figure V-21 Average oxygen concentrations in ponds 2, 3, 6 and 9. Ponds 3 and 9 were control ponds, pond 6 was "low" CCl4 and pond 2 was "high" cc14 . 1973 1974 M J J A S 0 N D J F M A 1 00 (\J 90 A :z; ~ 80 70 100 ~ H 8 ~ \() 90 8 B t t > calculated value - 9 control 3 control 32 17 0.18 0.429 3 control 2 high CCl4 33 20 1. 04 0.149 3 control 6 low cc14 33 24 2.44 0.007* 3 control 8 high Freon 32 17 0.18 0.429 1 low Freon 3 control 33 17 0.00 0.500 9 control 1 low Freon 32 19 0.88 0.311 8 high Freon 9 control 32 17 0.18 0.429 9 control 6 low CCl4 31 24 2.87 0.003* 9 control 2 high CCl4 32 20 1.24 0.108 2 high CCl4 6 low CCl4 33 20 1.04 0.149 1 low Freon 8 high Freon 33 17 0.00 0.500 *significant at better than .99 level 1--1 ~ U1 TABLE V-24 Sign test applied to "delta" oxygen values. Pond Count of Probability of A B total n A > B t t > calculated value 9 control 3 control 33 18 0.35 0.363 3 control 2 high CCl4 32 22 1.95 0.026* 3 control 6 low CCl4 33 30 4.53 0.001** 1 low Freon 3 control 33 19 0.70 0.242 1 low Freon 9 control 32 18 0.530 0.298 8 high Freon 9 control 32 18 0.53 0.298 9 control 6 low CCl4 32 25 3.00 0.001** 9 control 2 high CCl4 32 19 0.88 0.188 2 high CCl4 6 low CCl4 32 24 2.65 0.004** 1 low Freon 8 high Freon 33 17 0.00 0.500 * significant at better than .95 level** significant at better than .99 level t-J °'~ Pond 2 (high CCl4) was found to average 94% of pond 3. The sign test clearly allows us to say that the Freon treated ponds were not different from the control ponds with respect to average oxygen levels or changes in oxygen level due to photosynthesis and respiration. However, the CCl4 data is more difficult to interpret. The low dose pond is different from the controls but the high dose pond is not significantly different with respect to average oxygen level, and, in one out of two comparisons, with respect to diurnal oxygen changes. Since the high CCl4 dose -pond had a target concentration of 2.4 ppm from 5/9/73 to 12/17/73, and a 4.0 ppm target concentration from 1/28/74 to 6/15/74, it is interesting to compare the oxygen data for these two periods separately. As the number of comparisons decreases, of course, the power of the statistical test decreases. The results of comparing pond 2 with the control ponds for the two different dose periods are shown in Table V-25. It seems there is possible significant difference · for the period of 2.4 ppm target concentration and no significant difference for the period of 4.0 ppm CCl4 target concentration. It is hard to imagine what effect could show more prominance at a low dose than at a high dose. There is obviously some effect which has to be explained. The only physical difference in construction between 6 and the control ponds was the epoxy paint in 6, which should have been inert since it was more than one year old at the start of the experiment. On many occasions pond 6 was the warmest of the ponds, as discussed previously, however, the difference was always small and does not seem adequate to explain the differences observed. 148 TABLE V-25 Sign test comparison of Pond 2 (high CCl4) with the control ponds (3 and 9) with respect to "delta" oxygen values for two different periods.Target concentration during the 5/9/73 to 12/17/73 was 2.4 ppm CCl4, while during the 1/28/74 to 4/8/74 it was 4.0 ppm CCl4. Count of Probability of t > Pond A Pond B total n A > B t calculated value Period 5/9/73 to 12/17/73 3 2 23 16 1.67 0.047 9 2 22 15 1.49 0.068 Period 1/28/74 to 4/8/74 3 2 9 6 0.67 0.251 9 2 10 5 o.o 0.500 149 The biological differences between the ponds which might influence the magnitude of the diurnal oxygen change include the fairly dense growth of seagrass which occurred in pond 2 after October 1973 but never in the control ponds and not in pond 6 until about April 1974. The possibl~ interactions between seagrass and phytoplankton are potentially quite complex. In ponds with dense grass, there appeared to be less phytoplankton but a huge population of epiphytic algae on the grass blades. In this section we will discuss the implications which can be drawn from the nutrient analysis data, given in full in the appendix. Figures V-25 through V-29 show a 4 month section of the 16 months of data, starting 2 weeks before the initiation of halocarbon dosing. Table V-26 contains the means and standard deviations for the period preceding dosing, while Tables V-27 and V-28 give the figures for the periods 5/14/73 to 8/27/73 and 9/10/73 to 5/26/74 respectively. Measurements of components in the nitrogen cycle included ammonia, nitrate, and nitrite, with ammonia found to be the dominant form in all cases. Ammonia analysis did not start until 3/5/73 due to problems in adapting Hach reagents and procedures to saline waters. As shown in Table V-26, ammonia mean levels were quite variable from pond to pond, during the first period, ranging from 0.14 ppm-N in pond 9 to 0.65 in pond 1. These differences are due mainly to spectacularly high values which occurred on 3/19 and 4/6 in certain ponds. The cause of this is believed to be a problem with the ammonia test which· gave a precipitate interfering with color development. If these values are throw out, the values for all ponds become quite similar. 4/73 ~/73 6/73 7/73 8/73 25 1 14 21 28 5 11 18 25 2 9 16 24 30 6 13 24 27 1 • 5 POND 1 1 • 0 0.5 .·-. !'""· . / "'·-·"--./'""'/. \./~~.--· o.o 1 4 --· 1 .o POND 21 • r, o. C) POND 3 1 .o _!_.-.1:=·"·-·-·~·"'·~.li~~-·-:-·~.--· ~ tt: P.. 0.5 1 • ') 3 :z; I ·----.-·""'·/·, /.II. .""·./ ""'·"·-· o.o --·-1\ POND 6 o.o 3 . "· 1 .o ~ :z; ·---·-· f I ·--·--· 0.5 ~ 1.1) 6 6 ./ ""·/'"--·-'"-.•...-. y' '\ . ~.-· POND 81 .o --·~ /'-... o.o • ~. 8 0.5 /.-. . ! "'·--· o.o 8 •---· ""'·_.... --·-./ -------·--. 1.0 1--' • /·-·~. /·~ ././"-....,_,_, 9 POND 9 0U1 o. c:; "" .--· ~-- 9.---· ---~ o.oFIGUHE V-25 Ammonia nitrogen over a four month period for all ponds. Note that the vertical scales are overlapped. 5/73 6/73 7/73 8/73 1 7 14 21 28 5 11 18 25 2 9 1 6 24 30 6 1 3 20 27 .1 5 •1 0 POND 1 .05 • 1 • /"" ___.1· '\ 1 • • 1 c:> .oo . .--·, ."".~·---.--·-·---.-----· I'\ . POND 2 / . . .05 ~2."' !'""."'" /·~·---. \2/. .1 0 .1 0 .oo p... .-· 3 • ·-· ·""'· """"'· ·-·-·-·/./\: POND 3 •0'1 • z as .oo :--·----...............___ / ~--·-·---·-----· . I .1 0 ("('") z0 • OC) POND 6 .1 0 POND 8 •05 \ . . / .oo .1 5 8/·, .oo . ""·---.I"".-...............~~·-·-·---·--·---. //\ . / .1 0 I-' U1 9 . ;·"" •05 POND 9 I-' ;· __,/""-·-· ·""_,.,.......-·~./·-----·-. .oo FIGUHE V-26 Nitrate nitrogen measurements over a four month period. Note that the vertical scales are overlapped. f)/73 6/73 7/73 8/737 14 21 28 ') 11 18 25 2 I I I 0.3 9 16 24 30 6 • I t I I I I I f 1 3 20 27 I I I I I I . 0.2 POND 1 • 1/."'·-·/·"-.. /\ __._,/._\ 1;·to.4 0.1 ---·-. -.............~·-·-· \ I-0.3 o.o -POND 2 /.-.............. ! \ ·-2/·-·~ . .---·-· / .-·""'2. 1-0.2 ·-· ' ~-·-·1-0.1 0.2 • .-.-·-............ . r o.o aPOND 3 0.1 . ·-·-3/'-......../ '"-.,_,_,...,..-·............. / ............... _~--~ . . 0. o.o -I 0.... • ~· ~ 0.2 I • ·-6 o:::j 0 6 /"' .-· "" 0.... .. ·-....._._./ "'·/ /""-/ ·----.----· ·---·-·---· "\.. POND 6 t-0.1 0.2 . POND 8 0.1 • /" /""' r-o.o /·---...._ / ""' . s/ ~--· .............._ ,.,./" •.......__a /"', . -.........--· . ·-·... 0.0 ~ . L . r 0.29 / "' • •-t .---·---./ "'·---·,.,./"----.----~·-·/ •-----·~./ •-........... • I-' ~~---·~0.1 POND9 U1 --~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~---10.0 N FIGUHE V-27 Phosphate measurements during the period from i::;/1/73 to 8/27/73. Note that the vertical scales are overlapped. 1 0 C) POND 1 0 1 s (\J 0 ·rl 1 0 rn I POND 3 s B: 0.... 0 ~ 0 H ...:i H rn 1 0 POND 8 5 0 FIGURE V-28 5/73 6/73 7/73 8/73 7 14 21 28 !) 11 18 2S 2 16 24 30 6 9 1 3 20 27 I I I I I l I I I I ' ' I I I I I I / __ .-·--· /' .--·~·"""./·-· --.-·-. f . I ,1 -.-·-· -... / ~· 2 -·/ "--.. --·-· .............. --· t-10 " /" .... / " .----.----.,---. 5 2 ~·--. .-·~ / ' -· __,. .... 0 3 /. ..........--· "·--"·--· __,,,,,,.. -• ,,.,.---. ·--. "". /.- .---· 3 . . ~ /.~.-·-·--6 6 ~· ~·-·\ ............. ·-· . / --· ~·~·--· .-·,,.,.. . ~ 8 ·-·-/.--·-·-........_ ay·-·, 8 ....--· ---•::::::>-·.\_....--· •---·-;:::/·-·-· .---·-·-·-· ............. _. 9 / .~·/ .-·---./· 9 9 0 Silica measurements during the period from s/1/73 to 8/27/73. Note that the verticalscales are overlapped. POND 2 5/73 6/73 7/73 8/73 1 00 7 14 21 28 5 11 18 2S 2 9 16 24 30 6 1 3 20 27 • POND 1 50 -1/I / .............._""· _,,,,. ~~·---•--·-·-·---·----·--• ..........._ .- --.-· ·-· -·-·/ 1 [ 100 ._,.---·-·-·--·--·, 1 50 -2 / -· " • .-·- .o -·-·/ /' \. ·-·-.-·------.-·---: 50 POND 2 1 00 /. ............... 0 POND 3 ---· \. ----·-·-· ./.-2 / "".-·-· ~./ 3 p 50 -3 ~ .---·../ 0 -. _,.---·-............ 100 H~ . ~/""' .-·-·---. A .----·, / 6 • 8~ -I--· '~ "'·-·-·~·/ "/. 50 POND 6 1 00 -8 ... 0 soi ·--.8---......._.-·-·-·--.~_/·-·/ / ............. ../'·-·~,/ .....--· -[ 1 ~o POND 8 . . .'• ./ ... 0 -·---... 9 .-·/~·---.---·-·-.-· I . ~· ,_.Y -·--.---·/ "" 100 POND 9 -........ 9 • ... 50 ....... lJ1 ~ 0 FIGURE V-29 Turbidity measurements over a four month period. Note that the vertical scalesare overlapped. 155 TABLE V-26 Nutrient concentration and turbidity means and standard deviations for the pre-treatment period,1/15/73 to 5/9/73. 1 2 3 6 8 9 Analysis mean 0.020 0.026 0.020 0.041 0.028 0.030 N03-N ppm std dev 0.014 0.021 0.013 0.032 0.023 0.061 mean* 0.65 0.15 0.39 0.21 0.34 0.14 NH4-N ppm std dev 0.86 0.15 0.66 0.17 0.54 0.14 mean .003 .003 .003 .004 .003 .003 N02-N ppm std dev .003 .003 .003 .004 .002 .002 mean 2.93 Sio2 ppm 5.38 5.85 2.87 3.09 4.18 std dev 2.10 3.18 3.42 2.05 2.50 2.62 mean 0.077 0.160 0.054 0.059 0.048 0.053 P04 ppm std dev 0.114 0.340 0.075 0.031 0.030 0.033 mean 18.5 24.9 17.5 27.5 24.6 18.9 Turbidity jtu std dev 8.5 17.3 11.3 17.0 14.3 9.7 *period 3//5/73 to 5/7/73 only TABLE V-27 Nutrient concentration and turbidity means and standard deviations for the period 5/14 to 8/27/73. Analysis Pond 1 2 3 6 8 9 mean 0.033 0.032 0.032 0.042 0.035 0.038N03-N ppm std dev 0.035 0.032 0.035 0.033 0.034 0.035 mean 0.61 0.53 0.60 0.56 0.72 0.60NH4-N ppm std dev 0.40 0.23 0.39 0.29 0.60 0.25 mean .004 .006 .005 .007 .006 .007N02-N ppm std dev .002 .005 .004 .006 .005 .005 mean 7.85 6.94 8.87 8.73 3.96 9.79Sio2 ppm std dev 3.19 3.00 3.36 3.89 2.57 3.92 mean 0.139 0.144 0.108 0.119 0.107 0.113P04-P ppm std dev 0.057 0.093 0.043 0.039 0.042 0.037 mean 67.8 67.4 102.8 72.1 47.9 85.6 Turbidityjtu std dev 16.1 12.6 28.3 21.0 28.4 27.2 157 TABLE V-28 Nutrient concentration and turbidity means and standard deviations for the period 9/10/73 to 5/26/74. Analyses were weekly until 12/14/73 and every two weeks thereafter. Analysis Pond 1 2 3 6 8 9 mean 0.036 0.036 0.031 0.034 0.037 0.038 N03-N* ppm std dev 0.025 0.029 0.032 0.022 0.030 0.030 mean 0.46 0.45 0.46 0.38 0.37 0.42 NH 4-N ppm std dev 0.23 0.24 0.24 0.21 0.17 0.20 mean 5.23 4.56 4.98 6.04 5.38 5.68 Si02 ppm std dev 2.78 3.01 2.61 2.24 2.92 2.64 mean 0.13 0.11 0.11 0.10 0.10 0.10 P04-P ppm std dev 0.04 0.05 0.04 0.03 0.02 0.02 mean 83.6 56.9 95.9 73.4 63.2 73.6 Turbidity std dev 26.6 18.3 64.0 22.2 14.1 23.l *9/10 to 11/20/73 only, 10 measurements. Interpretation of nutrient analysis results Ammonia values for the period after the starting of dosing are similar for all ponds, with pond 8 having a somewhat higher mean value due to a very high value of 2.6 ppm-N on 7/16/73. Inspection of Figure V-25 shows that all ponds had some sort of peak on or near this day, possibly related to the addition of organisms the previous week. Pond 8 was not being treated with Freon in the period 6/19 to 7/18. During the period 9/10/7~ to 5/26/73 (Table V-28), the ammonia concentration was quite similar in all ponds. The highest concentration during this period was 1.10 ppm-N found in ponds 1, 2 and 3 on 1/28/74. Nitrate values for the pre-treatment period (Table V-26) are similar for all ponds except number 6 which is somewhat high, apparently due to unusually high values on 4/6, 4/16, and 5/1/73. Values for all ponds seem to be similar during the 5/14 to 8/27 period and the 9/10 to 11/20/73 period. It can be seen from Figures V-26 and V-26 that high nitrate values tend to follow high ammonia values by a few weeks, as expected from the known properties of the nitrogen cycle. In almost all cases, nitrite values were at the lower limit of detectability, therefore little significance is attached to these values. Corpus Christi Bay during the period from December 1972 to June 1973 had values of ammonia plus nitrate nitrogen ranging from 0.13 to 0.23 mg-N/l, averaging 0.18. Therefore, the ponds were considerably richer in nitrogen than the bay system. 159 Phosphate mean concentration values for the 1/15 to 5/9 period were similar for ponds 3, 6, 8 and 9; slightly higher for pond 1 and much higher for pond 2. These differences in mean concentration were due to abnormally high values on 2/26 and 3/12 in ponds 1 and 2, during the rest of the period, all ponds had similar values. The cause of these unusual peaks is not known, it may be due to contamination of the sample since the values observed are much larger than those observed at any other time. During the period of dosing, ponds 3, 6, 8 and 9 were again very similar, with somewhat larger mean phosphate concentrations in 1 and 2. Again there is one occurrence of unusually high phosphate in pond 2 on 7/2/73. Figure V-27 shows that the phosphate values during the 5/14/73 8/27/73 period go up and down coherently. Phosphate mean concentrations for the 9/10/73 to 5/26/74 period were very similar for all ponds (Table V-28). For Corpus Christi Bay as a whole, phosphate values were found to average about 0.015 mg P/l by Holland et al., 1974. The ponds, therefore, were considerably richer in phosphate than the bay system. Dissolved silica concentration was quite variable from pond to pond during the initial period, with ponds 2 and 3 relatively high and 6 and 8 low (Figure V-28). During the next period, Table V-27, pond 8 remained low while the others were nearly uniformly high. One possible explanation for this is that during this period, pond 8 had few or no organisms, therefore the sediment was less disturbed and less silica was released. However, the level of silica was not low enough to inhibit diatoms. 160 During the 9/10/73 -5/26/74 period, silica values were similar in all ponds. Dissolved silica values in Corpus Christi Bay were determined on 11/13/73 to range from 7.3 to 24.3 ppm Si02 with high values at the lower salinities as expected. Turbidity was measured at the same time as the nutrients as an indication of suspended matter, including phytoplankton, detritus and inorganic clays, etc. During the initial, no dose, period of monitoring, the ponds seemed to fall into two groups with respect to turbidity. Ponds 1, 3 and 9 were low, around 18 JTU, while the other ponds averaged around 25 JTU (Figure V-29). During the first experimental period turbidity increased in all ponds, with the control ponds, 3 and 9 having highest values, 1, 2 and 6 intermediate values and pond 8, the high Freon pond, the lowest value. The low value in pond 8 was presumably due to a lack of organisms to keep things stirred up. During the period from 9/10/73 to 5/26/74, control pond 3 remained high in turbidity while pond 9 dropped back to the range of the experimental ponds. The-relatively greater growth of seagrass in ponds 2 and 8 would tend to reduce the turbidity by filtration. Holland (1974) found turbidities in Corpus Christi Bay from 0 to 200 JTU with an average around 14. In Nueces Bay the range was 20 to 260 with an average around 102. Thus the ponds were more turbid than Corpus Christi Bay but less turbid than the shallower Nueces Bay. General discussion of nutrient results In teneral the greatest differences between ponds were observed in the pre-experimental period. Presumably during this period the model ecosystems were still "settling down" with 161 respect to the nutrient cycles between water and sediment. Nutrient concentrations during the experimental period were quite similar for all ponds with the exception of silica in pond 8, which was probably low due to absence of fish and shrimp. This implies that the micro-organisms responsible for nutrient regeneration were not affected by the experimental treatments, and that the relatively low photosynthesis and respiration observed in pond 6 could not be due to a lack of nutrient supply for phytoplankton and bacteria. 9. Fluoride analysis It was felt that if Freon was being degraded to any appreciable extent by microbial or other processes in the ponds, it might be possible to detect an excess of inorganic fluoride ion in the treated ponds. Fluoride analyses were carried out on 6/11/74 using an Orion pH meter, a Corning fluoride electrode, and Orion TISAB buffer and standards. Salinity was measured on the same samples with a refractometer. Precision of the electrode measurement is estimated at + 2%, and the salinity measurement is estimated to be within 1.0 ppt. For comparison, fluoride in seawater is about 1.3 ppm. The results are shown in Table V-29. The essentially identical level of fluoride for all ponds with similar salinity is taken to indicate that little fluoride was released during the 47 days since the ponds were last refilled. Pond 8 was dosed from 5/13/74 to 6/11/74 with a target of 7 ppm Freon, so the average must have been in the vicinity of 4 to 5 ppm. Assuming that the fluoride test could have detected an excess on TABLE V-29 Fluoride analysis results. Pond Dose Fluoride Salinity 1 (1.0 ppm Freon target) 0.72 ppm 21 ppt 8 (7.0 ppm Freon target) 0.69 20 3 (Control) 0.72 20 9 (Control) 0.72 20 6 (0.24 ppm CCl4 target) 0.66 15 2 (4.0 ppm CCl4 target) 0.72 20 164 CHAPTER VI SUMMARY AND RECOMMENDATIONS This project was designed to look at these aspects of the carbon tetrachloride and Freon 113 in the estuarine environment: the chemical and physical behavior, the effect of short exposures on important organisms, and the effect of long exposures on organisms. Physical-chemical behavior The most significant pathway for these compounds when discharged to a shallow estuary will be loss to the air. It was found that under field conditions loss to the air can be described as limited by diffusion within the water column (in contrast to evaporation), similar to oxygen. The loss rate is increased by wind induced turbulence, bubbles, and waves. An an example, under most weather conditions our 2 ft. deep experimental ponds usually lost 1/2 of their Freon 113 or carbon tetrachloride in less than 24 hours. Loss to sediments occurs slowly for carbon tetrachloride, this is not a biological process since it occurs in sterile samples at the same rate as non-sterile. Some chemical transformation is probably involved since the cc14 could not be recovered from the sediment. The sediment samples used in these experiments came from La Quinta channel. Freon 113 does not appear to be lost to sediments at an appreciable rate. Short-term exposure biological effects We had initially planned to use respiration measurements to show effects of short term exposure with shrimp, trout and oysters. However, due to difficulties with the equipment and with acclimation of organisms to the test conditions, we were only able to test shrimp. Most runs were made with carbon tetrachloride and only a few were made with Freon 113, which was more difficult to control in concentration. Respiration results were highly variable between organisms and we were not able to detect any sublethal effects. Short term exposure of speckled and sand seatrout (Cynoscion sp. -1 to 2 lb. adults) indicated that concentrations around 3.0 ppm of Freon 113 or carbon tetrachloride could be withstood for several days. A concentration of around 8 ppm carbon tetrachloride or -round 10 ppm Freon 113 killed most trout within two days. One of the first symptoms exhibited was loss of equilibrium. More exact relation of concentration to death rate could not be obtained due to the difficulty in controlling dosage in the outdoor ponds needed to hold these relatively large animals. Short term photosynthesis/respiration experiments using the light and dark bottle technique with natural phytoplankton and with a blue-green alga culture showed that concentrations around 100 ppm carbon tetrachloride were required to depress photosynthesis. There were some indications of stimulation of photosynthesis and reduction of respiration at 1.0 ppm carbon tetrachloride. Similar experiments with Freon 113 showed a possible stimulation of photosynthesis and reduction of respiration at 1.0 ppm, but rates 166 similar to the control at 10 ppm. Concentrations of 30 ppm Freon 113 definitely increased respiration and 100 ppm depressed photosynthesis. Growth rate experiments with the blue-green alga indicated no effect at 1.0 ppm carbon tetrachloride, slight inhibition at 3.0 ppm and definite inhibition at 10 and 100 ppm. Growth rate experiments with Freon 113 gave the unexplainable results of growth similar to the control at 10 ppm with inhibition at 1.0 and 3.0 ppm. Sufficient time was not available to pursue these experiments at lower concentrations. These results might be cause for alarm if the pond experiments had not indicated normal photosynthesis at levels around 1 ppm Freon 113 with natural populations. The effects of Freon 113 and carbon tetrachloride on phytoplankton deserve more detailed study, with special attention paid to exact control of concentration while maintaining suitable conditions for growth. Long term exposure biological effects In the long term pond experiments various estuarine organisms including shrimp, small fish, oysters and sea grasses were exposed to various levels of carbon tetrachloride and Freon 113 for periods of several months. Due to high rates of loss of the volatile compounds from the outdoor ponds, even daily analysis and dosing could not control the concentration to better than + 20% to -70% of the "target concentration" however, continuous exposure for several months under close to natural conditions was obtained for several batches of animals. No external food was supplied 167 so that these animals were living on the natural food chain. "Low" and "high" target concentrations were initially 1. 0 and 10.0 ppm Freon 113; 0.24 and 2.4 ppm carbon tetrachloride. The 10.0 ppm Freon 113 dose was rapidly fatal to the fish and shrimp so it was reduced to 5.0 ppm; after good survival was noted at this level, it was raised to 7.0 ppm. The 2.4 ppm dose of carbon tetrachloride was raised to 4.0 ppm after good survival was obtained. Pathological examination of a limited number of fish and shrimp which had been exposed to these levels was conducted by Dr. R. A. Bendele of Texas A&M. No abnormalities which could be attributed to the halogenated hydrocarbons were found. Because there was only one pond for each concentration, with two control ponds, and because large numbers of organisms died even in the control ponds, the data on survival is hard to interpret. However, it appears that the survival in the low dose ponds {0.24 ppm carbon tetrachloride, 1.0 ppm Freon 113) was similar to controls for all organisms tested. One obvious difference between the ponds at the end of the experiment was the lush growth of the sea grass, Ruppia maritima in the pond receiving the high dose of Freon ·113 compared with almost no growth in the controls and moderate growth in the other ponds. Daily measurements, morning and evening, were made of oxygen and temperature, and weekly measurements were made of nutrients in an attempt to detect differences in overall ecosystem functions such as photosynthesis, respiration, and nutrient cycling. The ponds did not behave entirely synchronously 168 with respect to plankton blooms, etc., but averages over fairly long periods indicate no differences in photosynthesis, respiration, and nutrient cycling due to carbon tetrachloride or Freon 113 with the exception of the "low" dose carbon tetrachloride pond. This pond showed significantly lower average oxygen concentration and diurnal changes in oxygen. This effect was not seen at higher concentrations or in laboratory experiments, and may be due to uncontrolled physical differences between ponds. This observation also suggests further work with carbon tetrachloride effects on phytoplankton is needed. The possibility that a microbial flora in the ponds might become adapted to these compounds and become able to actively degrade them was checked and no degradation was found. Fluoride measurements were made as an alternate way to detect Freon 113 degradation but no excess fluoride was found. Fish and shrimp from the ponds were found to have concentrated carbon tetrachloride and Freon 113 in their tissues by a factor generally from 3 to 8 times the water concentration. A single fish was found to have concentrated Freon 113 by a factor of 41. To check the possibility that a degradation byproduct might spoil the flavor of exposed fish, several were cooked and found to be perfectly normal. Considerations for future toxicity experiments Our greatest difficulties in these experiments have been in obtaining , stable exposure levels over periods long enough to get useful data, and still maintain a high enough oxygen level for the organisms. Due to the highly volatile nature of these compounds indoor experiments must have special provision for ventilation around experimental tanks, and outdoor experiments are strongly affected by weather conditions. Using the aeration air to control concentration appears to be reliable and could be adapted to flowing tests as well as static tests. The major problems with the gas chromatographic analysis procedure were lack of detector linearity at high levels, sensitivity drift and ghost peaks following injections of concentrated samples. Electron capture detectors with linear response over a wide range are now available. The problems with ghost peaks might be solved by analyzing only the gas phase in equilibrium with a sample, thus reducing the effects of large quantities of water vapor. Recovery of the halocarbons from solid samples with the distillation technique were quite variable. A better method of collecting volatiles from the digestion mixture is needed. Absorption in cooled hydrophobic solvent or onto a polymeric gas chromatographic column packing would probably work. BIBLIOGRAPHY Bionomics, Inc. 1972. Accumulation and persistence of carbon tetrachloride residues in killifish (Fundulus heteroclitus) continuously exposed to the chemical in water for 28 days. Research Report, Bionomics, Inc., Wareham, Mass. Cairns, T. Jr., R. E. Sparks and w. T. Waller. 1972. The design of a continuous flow biological early warning system for industrial use. Paper presented at 27th Pardee Industrial Waste Conference May 2-4, 1972. Cech, J. J.Jr. 1970. Respiratory responses of the striped mullet Mugil cephalus to three environmental stresse-s. M.A. Thesis. Univ. of Texas at Austin. 115pp. Hoel, P. G. 1971. Introduction to Mathematical Statistics. Fourth edition. John Wiley & Sons, Inc. N.Y. 409pp. Holland, J.S., N.J. Maciolek, R.D. Kalke and C.H. Oppenheimer. 1974. A benthos and plankton study of the Corpus Christi, Capano, and Aransas Bay Systems. A Report on Data Collected During the Period July, 1973 -April, 1974. Second Annual Report to the Texas Water Development Board. University of Texas Marine Science Institute at Port Aransas. 12lpp. Jackson, R. G. 1975. Effects of zinc and mercury on the white shrimp, Penaeus setiferus. Dissertation presented to the Faculty of the Graduate School of The University of Texas at Austin in partial fulfillment of the requirements for the degree of Doctor of Philosophy. Liss, P. S. 1971. Exchange of S02 between the atmosphere and natural waters. Nature 233:327-329. Liss, P. s. 1973. Processes of gas exchange across air water interfaces. Deep Sea Research 20:231-238. Liss, P. s. and P. G. Slater. 1974. Flux of gases across the air-sea interface. Nature 247:184-188. Mackay, D. and P. J. Leinonen. 1975. Rate of evaporation of low solubility contaminants from water bodies to atmosphere. Environ. Sci. and Tech. 9:1178-1180. Odum, H. T., et al. 1963. Experiments with engineering of marine ecosystems. Publ. Inst. mar. Sci. 9:373-403. Oppenheimer, C. H. and K. G. Gordon. 1972. Texas Coastal Zone Biotopes: An Ecography Interim Report for the Bay and Estuary Management Program (CRMP), University of Texas Marine Science Institute, Port Aransas, Tex. 118pp. Oppenheimer, c. H., T. Isensee, w. B. Brogden and D. Bowman. 1974. Biological uses criteria, final report. Establishment of operational guidelines for Texas Coastal Zone Management. The University of Texas Marine Science Institute at Port Aransas. Division of Natural Resources and Environment, University of Texas at Austin. Steed, D. L. and Copeland, B. J. 1967. Metabolic responses of some estuarine organisms to an industrial effluent. Contr. Mar. Sci. 12:143-159. Subrahmanyan, c. B. and c. H. Oppenheimer 1970. Food preferences and growth of penaeid shrimp. In Food-Drugs from the Sea. Proceediings 1969. H.W. Youngken, Jr. (ed). Sympoxium XIV. Marine Technology Society, Washington, D.C. pp. 65-75. 172 Van Baalen., C., W. Pulich and R. O'Donnell. 1973. A blue-green algal assay of water quality. In Toxicity Studies of Galveston Bay Project, Sept. 1, 1971 to Dec. 1, 1972. Final report to the Texas Water Quality Board, Galveston Bay Study Program for Contract IAC(72-73) 183. Oppenheimer, C.H., et al. (eds.). 32lpp. Waller, W. T. and J. Cairns, Jr. 1972. The use of fish movement patterns to monitor zinc in water. Water Research 6:257-269. 173 APPENDIX A Nutrient Analysis Data N0ppm-N 3 Date Time :1 2 J. 6 8 :~2 01 /1 5/73 0950 0.01 5 0.025 0.025 0.021) 0.005 0.01 5 01 /22/73 101 5 0.027 0.030 0.017 0.028 0.006 0.042 01/29/73 1000 0.041 0.030 0.017 0.045 0.035 0.038 02/05/73 1045 0.018 0.027 0,036 0.013 0.034 0.031 02/12/73 11 00 0.020 0.013 0.003 0.049 0.013 0.026 02/20/73 091 ') 0.007 0.070 0.034 0,028 0.027 0.029 02/26/73 1000 0.010 0.001 0.001 0.001 0.002 0.001 03/0~=/73 091 5 0.023 0.069 0.031) 0.013 0.036 0.034 03/12/73 0930 0.01 2 0.029 0.031 0.014 0.024 0.023 03/19/73 0900 0.009 0.01 2 0.01 3 0.053 0.029 0.033 04/06/73 091 5 0.017 o.01 5 0.001 0.080 0.035 0.01 5 04/16/73 1030 0.017 0.037 0.033 0.097 0.033 0.067 04/25/73 0900 0.01 5 -0 -0 o.04c:; 0.025 0.051) 05/01 /73 1400 o.oc:;8 0.010 0.030 0.1 0') 0.100 0.023 05/07/73 0730 0.008 0.018 0.020 0.018 0.01 2 0.025 05/14/73 0900 o.04c:; 0.042 0.033 0. 01 0 0.042 0.052 05/21 /73 0900 0.01 2 0.01 3 0.01 5 0.008 0.010 0.01 0 05/28/73 0900 o.ooc:; 0.010 -0 0.028 0.003 0.008 06/05/73 0800 0.059 0.073 0.080 0.074 0.068 0.080 06/11 /73 0. 030 0.033 0.037 0.013 0.018 0.043 06/18/73 0730 0.005 0.010 0.004 0.033 0.003 0.004 06/25/73 0900 0.020 0.036 0.011 o.oso 0.022 o.02c:; 07/02/73 0.01 0 0.020 0.01 0 0.030 0.030 0.01 0 07/09/73 0730 0.018 0. 01 5 0.025 0.017 0.028 0.040 07/16/73 0700 0.020 0.010 0.01 0 0.040 0.01 5 0.025 07/24/73 0730 0. 01 0 0.01 0 0.008 0.022 0.020 0.01 5 N0pprn-N (cont.) 3 Date Time 1 2 J. 6 8 2 07/30/73 1 200 0.02') 0.008 0.022 0.040 0. 01 3 0.010 08/06/73 0800 0. 045 0.037 0.052 0.030 O.Ot)8 0. 067 08/13/73 0730 0.1 38 0.1 25 0.1 28 0.11 9 0.1 33 0.1 35 08/20/73 071 5 0.01 2 0.01 2 0.008 0.040 0.028 o.01 Cj 08/27/73 071 5 0.075 0.062 0.073 0.11 3 0.070 o.o6c:; 09/1 0/73 0730 0.020 0.01 3 0.008 0.040 0.013 0.022 09/17/73 0800 0.1 00 0.11 0 0.100 0.090 0.11 0 0.11 0 09/24/73 0730 0.013 0.008 0.003 0.028 0.008 0.003 10/01 /73 0900 0.050 0.045 0.073 0.047 0.062 0.062 10/08/73 0830 o.oy; 0.027 0.025 0.019 0.023 0.030 1 0/1 5/73 0800 0.045 0.01)0 0.038 0.022 0.033 0.038 10/22/73 0900 0.030 0.025 0.020 0.020 0.030 0.030 11 /05/73 0830 0.020 0.028 0.022 0.030 0.040 0.048 11 /1 2/73 0900 0.020 0.023 0.010 0.022 0.030 0.02') 11 /20/73 0930 0.025 0.027 0.008 0.018 0.020 0.012 NH + ppm-N 4 Date Time 1 2 l 6 8 2 0.01 0.02 0.01 0.23 0.01 0.04 03/05/73 091 5 03/1 2/73 0930 O. S7 0.20 0.1 8 0.22 o.oc:; 0,08 03/19/73 0900 2.oi' 0.20 1.88¥ 0.30 1•')3. 0.1 5 04/06/73 091 5 1.68~ 0.43 0.32 0.!'.)3 0.43 0.43 04/16/73 1030 0.18 0.1 0 0.07 0.1 2 0.14 o. 1 c; 0900 0.02 0 O. P) 0.03 0.01 0.03 04/25/73 05/01 /73 1400 05/07/73 0730 0.04 0.10 0.1 2 0.04 0.23? 0.08 05/14/73 0900 0.1 s 0.14 0.17 0.25 0.30 0.22 05/21 /73 0900 0.78 o.83 o.88 0.84 0.47 o.87 0900 o.83 0.82 0.73 0.80 o.83 05/28/73 o. 75 06/05/73 0800 0.48 o. 51 o.68 o. 7"5 0.77 o.68 06/11 /73 0.48 0.46 o. 56 0,25 0.27 0.38 06/18/73 0730 0.23 0.47 o. 21 0.60 o.4c; 0.24 06/25/73 0900 o.68 0.64 0.44 0.40 0.30 o.86 07/02/73 0.20 0.30 0.1 2 0.40 0.1 c; 0.45 07/09/73 0730 0.62 0.1 3 0.')2 0.20 0.47 O. S1 07/16/73 0700 1 .66 o.83 1.64 o. 57 2.60 0.73 07300 1.1 0 0.73 1.1 0 1 .28 1.43 0.98 07/24/73 07/30/73 1200 o. 51 o.67 0.62 0.67 0.78 0.72 08/06/73 0800 1.05 o.63 0.82 0.73 1 • oc; 0.72 08/13/73 0730 0.42 0.56 0.47 o.67 0.73 o.68 08/20/73 071 5 0.27 0.29 o.28 0.27 0.44 0.30 0.36 0.42 0.30 0.34 0.52 0.35 08/27/73 071 5 09/10/73 0730 09/17/73 0800 o.66 0.72 0.78 0.72 o.7c; o.66 09/24/73 0730 0.34 o.5c; 0.53 0.37 0.42 o.63 10/01 /73 0900 0.29 0.23 o. c:;8 0.55 0.28 0.28 1 0/08/73 0830 0.29 0.1 9 0.18 0.19 0.23 0.25 10/1 5/73 0800 0.27 0.24 0.1 3 0.05 0.22 0.13 10/22/73 0900 0.29 0.32 0.33 0.33 0.28 0.28 11 /05/73 0830 0.20 0.23 0.26 0.1 2 0.32 0.22 11 /12/73 0900 0.30 0.28 0.25 0.18 0.22 0.23 11 /20/73 0930 0.20 0.23 0.18 0.1 5 0.23 0.17 * possible error due to precipitate formation during test NH+ ppm-N (cont.) 4 Date Time 1 2 l 6 8 .2. 0.22 0.20 0.20 0.22 12/03/73 0930 0.26 0.1 9 0. S21 2/17/73 0900 o.67 0.48 0.42 0.24 0.40 1.1 0 0.60 0.'12 0.32 01/28/74 0830 1.1 0 1.1 0 o. 56 0.33 0.03 o.83 02/17/74 0900 0.56 0.78 0.48 o.66 03/04/74 0900 o.67 0.54 0.69 0.74 0.50 03/1 8/7.Li 0730 0.50 0.41 0.45 0.42 0.49 ,.. --===------=·.. . . -- 04/01 /74 0830 0.34 0.28 0.33 0.39 0.42 0.38 04/15/74 0830 0.58 o.63 0.58 o. 56 o. 54 0. 53 04/30/74 0830 0.50 0.48 0.53 0.52 o. 51 0.55 05/13/74 1300 o. 51 o. 52 0.52 0.38 0.C)3 0.52 05/26/74 0830 o,63 0.58 0.49 o. C)3 0.42 0.52 N0ppm-N 2 Date Time 1 2 l 6 8 .2 ~5/73 0950 <-0.00C) <0. 005