MEASUREMENT OF ZOOPLANKTON BIOMASS BY CARBON ANALYSIS FOR APPLICATION IN SOUND SCATTERING MODELS James Carlton Radney LIBRARY GRADUATE SCHOOO ,'. CALIFORNIA 9394Q if PI1 0 1\ n U U n I L n onterey, California 35 t-sast w a eLwEsm 3* MEASUREMENT OF ZOOPLANKTON BIOMASS BY CARBON ANALYSIS FOR APPLICATION IN SOUND SCATTERING MODELS by Jatnei ; Ci art ton Radney Thesis Advisor: E. D. Traganza Approved for public release; distribution luilimitcd. September 1974 T164087 Unclassified SECURITY CLASSIFICATION OF THIS PAGE (Whan Detm Entered) REPORT DOCUMENTATION PAGE READ INSTRUCTIONS BEFORE COMPLETING FORM I. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER 4. TITLE (end Subtitle) Measurement of Zooplankton Biomass by Carbon Analysis for Application in Sound Scattering Models S. TYPE OF REPORT & PERIOD COVERED Master's Thesis; September 1974 6. PERFORMING ORG. REPORT NUMBER 7. AUTHORfa,) 8. CONTRACT OR GRANT NUMBERft; James Carlton Radney 9. PERFORMING ORGANIZATION NAME AND ADDRESS Naval Postgraduate School Monterey, California 93940 10. PROGRAM ELEMENT. PROJECT, TASK AREA 4 WORK UNIT NUMBERS It. CONTROLLING OFFICE NAME AND ADDRESS Naval Postgraduate School Monterey, California 93940 12. REPORT DATE September 1974 13. NUMBER OF PAGES 14. MONITORING AGENCY NAME 4 ADDRESS!-/' dllterent from Controlling Office) Naval Postgraduate School Monterey, California 93940 15. SECURITY CLASS, (of thtt report) Unclassified 15«. DECLASSIFI CATION/ DOWN GRADING SCHEDULE 16. DISTRIBUTION STATEMENT (of thla Report) Approved for Dublic release; distribution unlimited 17. DISTRIBUTION STATEMENT (of the ebctrnct entered In Block 20, If different from Report) 16. SUPPLEMENTARY NOTES Supported by the Office of Naval Research 19- KEY WORDS (Continue on reverse aide i/noc«*t«ry end identify by block number) Carbon, biomass, zooplankton 20. ABSTRACT (Continue on revorho aide If noceeemry end Identity by block number) Estimates of zooplankton biomass were made by use of a LEC0 Carbon Analyzer. The methodology developed in this study is a rapid (70 seconds per sample), precise (±3%) and accurate (±3%) measurement of total carbon. Casein and benzoic acid were used interchangeably as standards. The tech- nique was further tested on rigriopus californicus which yielded a value of 38.6% C by weight. Estimates of total, living, and dead zooplankton DD | JANc' biomass were made in a joint experiment by carbon analysis and ATP-C mea- surements. Field studies in Monterey Bay demonstrated a definite seasonal trend over the period of three cruises. DDlJan'7:< H73 (BACK) Unclassified S/N 0l6li-Ol-l-()G01 SECURITY CLASSIFICATION OF THIS P A G Cf »*.n :>•'• Bnffd) Measurement of Zooplankton Biomass by Carbon Analysis for Application in Sound Scattering Models by James Carlton Radney Ensign, United States Navy B.S., United States Naval Academy, 1973 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN OCEANOGRAPHY from the NAVAL POSTGRADUATE SCHOOL September 1974 DUDUV KNOX . «AVAL POSTGRADL'Vc * ' WL/FOaNia 93U4U ABSTRACT Estimates of zooplankton biomass were made by use of a LECO Carbon Analyzer. The methodology developed in this study is a rapid (70 seconds per sample), precise (±3%) and accurate (±3%) measurement of total carbon. Casein and benzoic acid were used interchangeably as standards. The tech- nique was further tested on Tigriopus caiifornicus which yielded a value of 38.6% C by weight. Estimates of total, living, and dead zooplankton biomass were made in a joint experiment by carbon analysis and ATP-C mea- surements. Field studies in Monterey Bay demonstrated a definite seasonal trend over the period of three cruises. TABLE OF CONTENTS I. INTRODUCTION 10 A. BACKGROUND — — 10 B. OBJECTIVE 11 C. CARBON AS A MEASUREMENT OF ZOOPLANKTON BIOMASS 12 II. METHODS 14 A. BACKGROUND OF CARBON ANALYSES 14 1. Dissolved Organic Carbon 15 2. Total Organic Carbon 15 3. Particulate Carbon 17 a. In Sediments 17 b. In Seawater 18 c. In Zooplankton 18 B. APPARATUS DESCRIPTION — 19 1. Operation 19 2. Maintenance 21 3. Accuracy and Precision 21 4. Designed Uses and Applications 27 C. ATP-CARBON ANALYSIS — 27 D. FREEZE-DRYING OF FIELD SAMPLES 27 E. REGRESSION ANALYSIS 29 III. EXPERIMENT DESCRIPTIONS 31 A. STANDARDIZATION 31 1. Benzoic Acid 31 2. Casein 32 3. Infrared Analysis 32 B. CARBON IN Tigriopus californicus 34 C. ATP-CARBON TO TOTAL CARBON RATIO IN Tigriopus californicus— 35 D. ATP-C AND CARBON ANALYSES IN ASSOCIATION WITH FIELD STUDIES- 41 IV. RESULTS - 57 A. STANDARDIZATION 57 1. Benzoic Acid 57 2. Casein 63 3. Infrared Analysis 63 B. CARBON CONTENT IN Tigriopus californicus 74 C. ATP-CARBON TO TOTAL CARBON RATIO IN Tigriopus californicus— 83 D. ATP-C AND CARBON ANALYSES IN ASSOCIATION WITH FIELD STUDIES- 90 V. DISCUSSION AND CONCLUSIONS 96 A. LABORATORY WORK — 96 B. FIELD STUDIES -- — 99 VI. RECOMMENDATIONS 101 APPENDIX A Volume Reverberation Theory 102 APPENDIX B Thermal Conductivity 104 APPENDIX C Calibration Instructions 105 APPENDIX D Cruise Data 108 LIST OF REFERENCES 117 INITIAL DISTRIBUTION LIST 120 FORM DD 1473 122 LIST OF TABLES I. ACCURACY AND PRECISION USING STEEL CALIBRATION RINGS II. DATA FOR BENZOIC ACID STANDARD CURVES III. DATA FOR CASEIN STANDARD CURVES IV. PERCENT CARBON DETERMINATION FOR tigriopus californicus V. ATP CARBON TO CARBON RATIO IN tigriopus californicus LIST OF DRAWINGS 1. LECO High Frequency Induction Furnace and LECO 70 second Determinator. 2. Freeze drying unit. 3. Infrared 10 cm NaCl cell. 4. Sieve columns 5a. Tigriopus caiifornicus sieved onto nylon screen. • b. Three Size fractions Of Tigriopus caiifornicus. 6. Oven and filtration unit 7a. R/V ACANIA. b. Laboratory aboard the R/V ACANIA 8. Cruise Area for all stations 9a. Drogue track and geographic station plot for May cruise (7403). b. Water mass station plot relative to a drogue for May cruise (7403). 10a. Drogue track and geographic station plot for July cruise (7404). b. Water mass station plot relative to a drogue for July cruise (7404). 11a. Drogue track and geographic station plot for August cruise (7405). b. Water mass station plot relative to a drogue for August cruise (7405). 12a. First generation net system on deck. b. First generation net system on Clarke-bumpus sampler attached to the hydrowire. 13a. Second generation net system on deck. b. Second generation net system attached to the hydrowire. 14a. Composite graph of benzoic acid; DVM vs. mg benzoic acid. b. Standard curve for benzoic acid; DVM vs. mg C. 15a. Composite graph of casein; DVM vs. mg casein. b. Standard curve for casein; DVM vs. mg C. 16a. Infrared spectrum for casein samples. b. Near infrared spectrum for casein samples. 17a. Infrared spectrum for plankton sample. b. Near infrared spectrum for plankton sample. 18a. First experiment, DVM vs. sample weight, Tigriopus caiifornicus. b. Second experiment, DVM vs. sample weight, Tigriopus caiifornicus. 19a. First experiment, mg C vs. sample weight, Tigriopus caiifornicus. b. Second experiment, mg C vs. sample weight, Tigriopus caiifornicus. 20. Combined data plot, mg C vs. sample weight, Tigriopus caiifornicus. 21a. First experiment, ATP-C VS. C in Tigriopus caiifornicus. b. Second experiment, ATP-C VS. C in Tigriopus caiifornicus. C Combined data plot, ATP-C in Tigriopus caiifornicus. 22a. Time series plot of total carbon for all cruises. b. Time series plot of living carbon for all cruises. c. Time series plot of dead carbon for all cruises. ACKNOWLEDGEMENTS The author wishes to express his gratitude for the help of many people in the conduct of this research and the production of this manuscript: To Dr. Eugene D. Traganza, the advisor, for his willingness to allow the author to be a part of his project (Inorganic Chemical Nutrients and Volume Reverberation Limitations in the Ocean, Part I, Biocnemical Relationships of Secondary Biomass and Dissolved Nutrients) sponsored by the Office of Naval Research; his assistance in joint experiments and during all cruises; his perceptive guidance and sug- gestions; his encouragement; and his experienced insight in the final corrections to the manuscript. To Mr. Kenneth J. Graham, chemist, for his unselfish efforts to solve technical problems; for his generosity in allowing the author to use various equipment and materials; for his. and Betty his wife's assistance on the final cruise; and for his varied contributions in the final draft of this thesis. To Mr. Scott Anderson, oceanography technician, for his invaluable assistance on all cruises; for his technical contributions to shipboard and oceanographic equipment; and for his help in the reduction of raw cruise data. To Miss Georgia P. Lyke, the reference librarian at NPS, for her help in the procurement of many relevant papers, journals, and books. To the Captain, Woodrow W. Reynolds, and the crew of the R/V ACANIA for their kind assistance during the cruises. To LT Tom Pearson, oceanography student, for his helpful suggestions and comments during the course of the author's research and for his assistance on the cruises. To Dr. Charles F. Rowell , for his experienced analysis of various phases of the research. To Mr. Pete Wisler and his staff at the NPS Machine Facility for their cooperation and help in fabricating the net systems. To the personnel of the NPS Photo Division for their work involving many of the figures and photographs in this work. To Mr. Pat Collelo, for his extra efforts in finding relevant material using the Defense Documentation Center computer search. To the secretaries, Mrs. W.L. Estes, Mrs I. Eid, and Mrs. H. H. Hale, in the Oceanography Department for their help and encouragement. To the Department of Physics and Chemistry for the use of the Wang 700 Computer System used in much of the data reduction. And, finally to my wife Cindy, for her assistance, understanding, patience, and faith throughout the period of research and writing. I. INTRODUCTION The sea contains within its volume inhomogeneities of many different kinds, ranging in size from microscopic particles to schools of fish. "These inhomogeneities form discontinuities in the physical properties of the medium and thereby intercept and reradiate a portion of the acoustic energy incident upon them. This reradiation of sound is called scattering and the sum of the scattering contributions from all scat- ters is reverberation. It is heard as a long, slowly decaying, quivering tonal blast following the ping of an active sonar system and is often the primary limitation on system performance" (Urick, 1967). In recent years it has become clear that volume reverberation or sound scattering strength (see Appendix A for explanation of these terms) is associated with variations in populations of marine organisms whose density and distributions are closely dependent on the zooplankton and other links in the food chain. A. BACKGROUND Traganza and Stewart (1973) accumulated a data base of volume re- verberation measurements at 3.5 kHz for the development of a prototype model for forecasting operationally useful information to the fleet. Tnrough the use of a regression equation, volume reverberation coverage was extended to most of the Northern Hemisphere on the basis of histor- ical zooplankton data. "There are a number of obvious reasons why there may be disagreement between current predicted volume reverberation from the zooplankton model and observed volume reverberation. Some can be attributed to 10 questionable uncalibrated scattering strength observations, poor zoo- plankton estimates, the lack of a sufficient number of observations to make significant regression analyses for all oceans, a need for a better understanding of food chain dynamics, and a better delineation of hydrographic, acoustic, and biological provinces" (Traganza and Stewart, 1973). It is the intent of this study to more accurately define zoo- plankton estimates of biomass for possible effective acoustic modeling. B. OBJECTIVE The objective of this thesis has been to evaluate carbon analysis for the chemical measurement of zooplankton biomass. The three phases of this research which were accomplished to make this evaluation in- cluded the following: (1) An investigation of the capabilities of the LECO*Carbon Analyzer, which employs high-temperature dry combustion and a. thermal conductivity sensor, to measure carbon in benzoic acid and casein was conducted. The biomass measurement using carbon analysis was tested by the determination of carbon content in a single marine copepod species, Tigriopus calif 01- nicus. (2) An estimate of the ATP (adenosine triphosphate) - carbon to total carbon ratio in this species was also determined in a joint ATP and carbon experiment with Dr. Traganza. This ratio was later employed to determine the living biomass of field net samples. (3) Estimates of total, living, and dead biomass of oceanic popula- tions were made from three cruises in the Monterey upwelling area. In short, a rapid (70 seconds on dry samples) and accurate (±3%) total particulate carbon analysis was developed and combined with ATP measure- ments for the determination of total, living, and dead zooplankton biomass *Laboratory Equipment Corporation C. CARBON AS A MEASUREMENT OF ZOOPLANKTON BIOMASS Many researchers, e.g. Cushing (1959), Beers and Stewart (1970), and Mullin (1969), have been concerned with the distribution of the standing crop of zooplankton over characteristic regions of the ocean. The preferred measure of zooplankton biomass is the total amount of zooplankton under a unit area of sea surface expressed in terms of 2 organic content or weight of dry organic substances {e.g. mg carbon/m ). The standing stock of plankton samples is also reported as the amount of zooplankton in a unit volume of water. "For various reasons, zooplankton sampling is presently inadequate to obtain a meaningful measure of zooplankton biomass as defined above. Various measures of amount are currently used, including displacement volume, wet weight, ash-free dry weight, dry weight and weight of dry organic matter."(National Academy of Sciences, 1969). There is some prospect that it may be possible to estimate zoo- plankton biomass and production using chemical means. Sutcliffe (1965) has used deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) deter- minations for this purpose with some success. Also, possibilities do exist for the measurement of zooplankton biomass using the phytoplankton adenosine triphosphate (ATP) method of Holm-Hansen and Booth (1966). Some enzyme determinations have already been utilized for the deter- mination of biomass (Aleem, 1955). The measures of zooplankton most commonly used in the past have been displacement volume and wet weight (Nakai and Honjo, 1962). The displacement volume is determined on a drained plankton sample. It is not a measure of zooplankton alone but includes the interstitial liquid held between and by the bodies of planktonic organisms. Interstitial 12 liquid commonly accounts for 30 to 40 percent of volume of the sample. Wet weight determinations are made on plankton after blotting. Most of the interstitial liquid will have been removed before weighing, but "over-drying" of the plankton should be avoided. Zooplankters can differ markedly in the amount of organic constituents in their bodies per unit volume (water plankters such as salps compared with crustacean plankters such as copepods). The only consistent measure of biomass appears to be the amount of dry organic matter per unit volume of water sampled (National Academy of Sciences, 1969). Such analyses have been published by Krey (1958) and Curl (1962). Previous approximations of biomass have been calculated in terms of organic content by Beers and Stewart (1970) for various size groups of microzooplankton (20-200 ym) . From the volume estimates assuming a specific gravity of one, a water content of 80 percent of the wet weight, and organic carbon as 40 percent of the dry weight, carbon is computed to be 0.08 times the volume. This common type of approximation leads to highly inaccurate biomass carbon estimates and points out the need for a rapid and accurate determination of biomass the result of which are 2 values directly expressable in mg-C/m . For this reason direct carbon analysis of freeze-dried plankton was pursued. Measurement of carbon by high temperature dry combustion represents a rapid and accurate estimation of the zooplankton biomass. 13 II. METHODS Carbon comprises close to 50% by weight of the organic matter in living organisms (Curl, 1962). Piatt et al. (1969) found a high degree of correlation (r = 0.94) between the carbon content and calorific equivalent of marine zooplankton. Thus, organic carbon analysis may provide the most sensitive and reliable test of the energy content of the biogenous material in marine organisms. Its measurement should be equatable to the biomass of zooplankton. Despite the abundance of carbon, it is one of the most difficult elements to measure. For ex- ample, the presence of carbonate (considered inorganic carbon) in many marine organisms provides an added complication in measuring organic carbon. Curl (1962) points out that washing plankton with distilled water to remove chloride in interstitial seawater which can also inter- fere with carbon analyses can result in an appreciable loss for water- soluble carbon containing compounds from the organisms. A. BACKGROUND OF CARBON ANALYSES If seawater is filtered through a membrane filter on the order of 0.45 ym pore size, the organic carbon retained on the filter is referred to as "particulate carbon." The organic carbon passing through the filter is termed "dissolved" . The particulate fraction is small and in oceanic waters rarely exceeds a few percent of the total organic carbon (Sharp, May, 1973) and at times the carbon content is determined on unfiltered samples. Thus, for evaluating carbon analysis of zooplankton it is useful to discuss literature on dissolved organic carbon and that of total organic carbon in seawater in the same context. 14 1. Dissolved Organic Carbon Several methods for the determination of dissolved organic carbon (0.1 to 20 mg/1) have been used. Analysis of organic carbon in seawater is hindered by the existence of large quantities of inorganic salts which make the organic constituents minute in comparison. Accepted methods are generally based on wet oxidation of carbon by acid di chroma te. Menzel and Vaccaro (1964) developed an analysis of dissolved carbon as an adaptation of a method by Wilson (1961) which consists of the wet oxidation of 1 - 5 ml of filtered seawater by potassium persulfate in a sealed glass ampoule after inorganic forms of carbon have been removed. The samples were subsequently flushed through a nondispersive infrared COp analyzer. Approximately 100 samples can be analyzed in a single day with a precision of ±0.1 mg/1 using a sample volume of 5 ml. Menzel (1967), and Williams and Gordon (1970) have also used this method for determination of dissolved carbon in the deep sea while Williams (1967) modified the procedure somewhat for sea surface chemistry. The Menzel and Vaccaro analysis was slightly modified and is now con- sidered a standard analysis in seawater (Strickland and Parson, 1968). Morris and Foster (1971) utilized ultraviolet photoxidation of one liter samples followed by gravimetric estimation of the liberated carbon dioxide. The estimated precision by this method is ±5%. 2. Total Organic Carbon Van Hall et ai (1963) proposed a method for analysis of organic carbon in aqueous solutions by high temperature combustion. In their method, liquid samples were injected into a 950 C furnace and carbon was oxidized to C0^ in an oxygen atmosphere. The resulting gas, after removal of water, was measured in an infrared analyzer. Their method 15 was not useable at carbon concentrations below 2 mg/1 nor was it tested extensively with solutions of high salt concentrations. It has not been used for oceanographic work because of these limitations. Wangersky (1965) began development of an analyzer for use with seawater, similar to that of Van Hall et ai (1963). Sharp (March, 1973) has developed a prototype for a high temperature combustion method for liquid samples. Oxygen is purified and used as a carrier gas. The combustion products pass through a condenser and Mg(C10J2 drying column into a Beckman IR-215 nondispersive infrared analyzer with C0?-filled detectors and cells 34.3 cm in length. Sharp and Wangersky (Sharp, March, 1973) have suggested that the most accurate method for measurement of organic carbon in seawater should involve high temperature combustion rather than wet chemical oxidation. The average precisions of the two methods are comparable and averages of errors are 5.0 - 5.5%; however the standard method of persulfate oxidation measures an average of 78% as much carbon as the high temperature method (Sharp, March, 1973). The works of P.J. Williams (1969) and P.M. Williams (1969) also seem to give evidence for incom- plete analysis by the persulfate oxidation method. Gordon and Sutcliffe (1973) have studied the feasibility of combusting freeze-dried seasalt at 730 C in a commercial elemental analyzer (Perkin-Elmer Model 240 Elemental Analyzer). This instrument was selected because it has a large sample volume capacity and combusts samples in pure oxygen. The combustion products (N~, COp and H^O) are subsequently measured by a series of thermal conductivity detectors separated by appropriate scrubbers. Each complete analysis takes twelve minutes. The error of this carbon method was estimated to be 16 about 8%, somewhat greater than that reported by Sharp (March, 1973) for both his high temperature and wet oxidation methods. Hewlett-Packard has developed a system for the quantitative analysis of particulate samples using a commercial carbon, hydrogen, and nitrogen analyzer and calculator system. This model (185B) provides a reproducible dry combustion method for the rapid oxidation of carbon in organic matter to gaseous products which are then determined by gas chromatography. The system was operated at sea and handles five samples per hour. Furnace temperatures reach 1100 °C and helium is used as the carrier gas (Atlantic Oceanographic Laboratory Report No. BI-R-73-14). 3. Particulate Carbon a. In Sediments 'The most satisfactory method for analysis of the total carbon content of sediments of sedimentary rocks appears to be com- bustion of a sample at temperatures exceeding 1500 C in an atmosphere of dry, C0?-free oxygen. Such temperatures can be reached in high- frequency induction furnaces in which the sample is mixed with iron and heated". Similar to seawater analyses, depending on the amount of C02 evolved, combustion gases are then passed through a gasometric analyzer, through absorption trains for gravimetric analysis, or detected by nondispersive infrared C0? analyzers or thermal conductivity sensors (Carver, 1971). The LECO Carbon Analyzer, used for carbon analysis of plankton in this report, uses the high frequency induction furnace and a thermal conductivity sensor and has been used in sediment studies (Andrews , personal communication) . 17 b. In Seawater The method of Menzel and Vaccaro (1964) for determination of particulate carbon consists of concentration of the particulate matter from a one to four liter sample on a glass fiber or membrane filter, combustion in an automated furnace at 800 °C in the presence of CuO, using oxygen as a carrier, and the detection of the resulting CO,, by infrared absorption. The precision of this method is 10 yg in a range of 0 - 500 ygC. Approximately six samples can be analyzed in an hour (Menzel and Vaccaro, 1964). c. In Zooplankton The total carbon content of the major taxonomic groups in net zooplankton from the upper 500 m of the Sargasso Sea off Bermuda was determined by Beers (1966). Samples of approximately 0.3-2.5 mg were combusted in a carbon analyzer furnace, the CO- liberated was collected, and the carbon determined by infrared analysis as in the Menzel and Vaccaro method. Curl (1962) also analyzed total carbon in 19 species of marine organisms and mixed collections. Weighed, oven-dried samples or standards were placed in zirconium-ceramic crucibles together with one gram each of low-carbon iron chips and fine granular tin. Analyses consisted of combustion within the crucibles in the induction furnace (temperatures up to 2000 °C) and subsequent measurement of evolved C0~ in a gas burette. The average sample was 20 mg of dry weight, and the carbon concentrations were from 6.6 to 46.8% of dry weight. According to Curl (1962), the advantages of rapidity and ease of operation of the induction combustion method for carbon are offset to some extent by decreased accuracy. However, the LEC0 instrument used in the author's 18 study which has an induction furnace and thermal conductivity sensor, was both rapid and accurate. B. APPARATUS DESCRIPTION 1. Operation (after LECO instruction manual, 1974) The instrument used for particulate carbon analyses in this study was the LECO (Laboratory Equipment Corporation) High Frequency Induction Furnace and the LECO 70 Second Carbon Determinator (Figure 1). In the method, a sample of known weight (0 - 100 mg) is placed in a ceramic crucible to which is added roughly 1 g of iron accelerator and 0.7 g of copper accelerator. The crucible containing the sample is then placed in the high frequency induction furnace within a com- bustion tube through which oxygen is passed. Since nearly all organic substances have high dielectric properties and are poor conductors of magnetic flux (Curl, 1962), samples must be heated indirectly through the use of the accelerators mentioned previously. The carbon in the sample is converted to C0~ at temperatures in excess of 1600 C. Metal oxides either remain in the crucible or are filtered out in a dust trap, while sulfur gases are absorbed in a trap containing manganese dioxide. Any carbon monoxide formed is converted to C0? in a heated catalyst tube. The dust trap, sulfur trap, catalyst tube and heater are mounted on the side of the induction furnace, as seen in Figure 1 . Moisture is removed in a Dehydrite trap which is mounted on the determinator. Both CO and moisture, is allowed to pass into the determinator, will cause erroneous results. The carbon dioxide formed and the carrier oxygen are collected in a cylinder. The thermal conductivity of the gas mixture contained in the cylinder is measured by a thermistor type conductivity cell 19 20 (See Appendix B). The output of the thermal conductivity cell is read on a special DC digital voltmeter. With the source oxygen in the cylinder, the thermal conductivity cell is balanced to yield 0.000 output as indicated on the digital voltmeter (DVM). The determinator utilizes the difference in thermal conductivity between oxygen and carbon dioxide. With the instrument thus balanced, the output of the thermal conductivity cell is indicated on the DVM and is proportional to the amount of C0? in the cylinder, assuming there are no other gases present in significant concentration which would affect the DVM reading. The preamp, bridge circuit, and cylinder are housed in a temperature controlled oven where the temperature (45 C) is set above ambient to eliminate variations which would cause an unbalance of the thermal conductivity cell or preamp. All the timing functions are controlled by solid state timers and reliable relays. 2. Maintenance The only routine maintenance items performed after each run (20 - 30 samples)were the replacement of Dehydrite, glass wool, and metal screens in the moisture trap mounted on the side of the determinator and the cleaning of the dust trap and combustion tube found on the induction furnace. Another moisture trap for the oxygen supply is mounted on the side of the determinator. Its Ascarite, Dehydrite, and glass wool were replaced about every fifth run. 3. Accuracy and Precision To achieve reliable and consistent results with the LECO Carbon Analyzer several conditions must be met. The instrument must be turned on three to four hours prior to the start of the analysis to allow metal and glass components to reach thermal equilibrium. An electronic 21 checkout and apparatus blanking process are required before analyzing any samples in each run to ensure that the instrument is calibrated correctly (See Appendix C). The samples to be analyzed must be dry since moisture will affect the thermal conductivity sensor. Finally, in the analysis of a sample the induction furnace plate current should go to between 400 and 500 mi Hi amperes, which indicates that enough current has been attained to induce sufficient heating for good com- bustion. Steel rings of 1 g each (less than 1% carbon) are recommended standards for the LECO Carbon Analyzer. During the course of the author's research, these steel rings were used to determine the accuracy and precision of this high-temperature dry combustion method for measuring carbon. On almost all runs in each type of experiment, one to several of the steel rings were analyzed for their carbon content. Since each steel ring represents a 1 g sample, the DVM readout is a direct expression of percent carbon. The estimates of accuracy and precision were found to be different for each type of steel ring analyzed. Three steel rings of different carbon content (0.074, 0.383, 0.854% C) were used. The compiled data for the tests and computations are given in Table I. The lowest carbon steel ring (0.074% C) demonstrated the largest inaccuracy. The percent average error, a measure of accuracy, was com- puted to be ±10.4%, and the percent average deviation was found to be ±11.4% (Masterton and Slowinski, 1970). This seems, at first glance, to be very disappointing. However, a more realistic test for accuracy and precision on organic substances from 10 to 50 mg was the 0.383% C steel ring. Computed estimates of percent average error and deviation were ±3.1% and ±2.6% respectively. The 0.854% C steel ring yielded 22 TABLE I ACCURACY AND PRECISION IN STEEL CALIBRATION RINGS 0.383 %C DVM Error Deviation .380 .003 .005 .383 .000 .008 .386 .003 .011 .396 .013 .021 .381 .002 .006 .363 .020 .012 .365 .018 .010 .350 .033 .025 .406 .023 .036 .372 .011 .003 .365 .018 .010 .374 .009 .001 .370 .013 .005 .378 .005 .003 .363 .020 .012 .373 .010 .002 .368 .015 .007 .386 .003 .011 .378 .005 .003 .376 .007 .001 .368 .015 .007 23 0.383 %C (con.) DVM Error Deviation .368 .015 .007 .362 .021 .013 .387 8.630 .004 .271 .012 Sum .224 Mean .3752 .0118 0.854 %C .0097 .861 .007 .001 .856 .002 .006 OCR .001 .007 .871 .017 .009 .856 .002 .006 .853 .001 .009 .840 .014 . .022 .877 .023 .015 .871 .017 .009 .841 .013 .021 .857 .003 .005 Sum 9.8411 .100 .110 Mean .862 .0091 .01 24 0.074 %C DVM Error Deviation .073 .001 .003 .074 .000 .004 .069 .005 .001 .077 .003 .003 .054 .020 .016 .056 .018 .014 .080 .006 .010 .076 .002 .006 .059 .015 .011 .081 .699 .007 .077 .011 Sum .079 Mean .0699 .0077 .0079 Error = observed - true value Deviation = observed - average value 25 .383% C Steel Ring % Average deviation = ^||||| x 100 = 2.595% (from mean) Precision - ± 2.6% % Average error = -™™3 x 100 =.3.0765% Accuracy = ± 3.1% .854% C Steel Ring % Average deviation = -'§§2 x ^00 = ^*^6^ Precision - ±1.2% Bl « - .0090909091 v inn - l nfi4? % Average error = gp x 100 - l.Obff* Accuracy = ±1.1% .074% C Steel Ring % Average deviation = ]Q699 x 100 = 11.302 Precision - ±11.3% % Average error = ^p x 100 = 10.405% Accuracy * ±10.4% 26 even better results of ±1.1% average error and ±1.2% average deviation. For the purpose of estimation of zooplankton biomass by the use of the carbon analyzer, the author considers the reported estimates of accuracy and precision for the mid- range (0.383% C) steel ring to be the most applicable. That is, the average percent error (accuracy) and average deviation (precision) were both determined to be approximately ±3%. 4. Designed Uses and Applications It must be remembered that the LECO Carbon Analyzer was designed for sensitive, low carbon measurements (<10%) as evidenced by the recom- mended use of low carbon steel rings as standards. This carbon analyzer has been adapted to measuring high (-50%) carbon in organic substances. The use of small samples, on the order of 10 mg of dry zooplankton for the estimation of biomass, allowed the resultant measurement of carbon in these samples since the DVM readouts were in the same order of mag- nitude as the steel rings. C. ATP - CARBON ANALYSIS ATP was measured by the Holm-Hansen and Booth (1966) method as mod- ified by Dr. Traganza for zooplankton analyses using a JRB Model 1 ATP Photometer. ATP was converted to ATP-C by the equation: 0.2382 ATP = ATP-C (Traganza, personal communication). D. FREEZE-DRYING OF FIELD SAMPLES Net plankton and laboratory samples (described in section III) were filtered onto Whatman GF/C glass fiber filters (0.45 ym pore size mesh) which had been precombusted at 450 - 500 °C for 2-3 hours. These samples were held in a freezer at -3 C and later "freeze-dried" in the laboratory at -196 °C with liquid nitrogen. The glassware unit 27 Q- 03 c/) 03 3 U Q- ro S_ 4-> S_ OJ en o o c 0) Dl o s- 4-> ■r- c "O •r— 3 CT •r~ i — +-> ■1 — .C C -f-> rs ■i — 3 0) -C ■»-> -M •i — C o =3 4-J cn+-> c o • ^ OJ >> c S- c X3 o o CD N _e a> u ai • r- s_ .c Ll_ 3 0) s_ en 28 shown in Figure 2 employed the use of a vacuum pump (29.2" Hg); a liquid nitrogen cooled protective trap; a liquid nitrogen cold finger trap; a condenser (merely the outside surface of the cold finger); and a set of sample flasks which connect to the unit. A set of 10 to 15 glass fiber filters and samples were freeze-dried for approximately 20 hours to ensure complete dryness of the sample before carbon analysis. Replen- ishment of the liquid nitrogen was required approximately every three hours. The resultant total use for a twenty hour freeze-drying process was up to 40 liters of nitrogen at a approximate cost of twelve dollars. Unfortunately, this procedure is quite costly and time consuming. It appears to be the single most important limiting factor in the presented carbon analysis scheme. E. REGRESSION ANALYSIS A tape program was used for all fitting of curves to experimental data by the method of "Best Fit" (Wang Laboratories, Program S. 107-7. 3) which, according to the program description, calculates the equation of the line by minimizing the squares of the perpendicular deviations of the points from the line. This line takes into account deviations due to the variability in both X and Y value, in contrast to the method of "Least Squares," which minimizes the squared deviations from the line in the Y-direction or X-direction. This is especially helpful with the experimental data in section III since the X (i.e. an approx- imated mass) and Y (i.e. DVM readout) values are not considered absol- ute and are subject to deviations. Figures 18-21 show the output of this program and the "best fit" line which fits the presented data. Neither X nor Y can equal zero when entering data points to be plotted. Very small values of X and Y were used with apparatus blanks made both prior to and following each combustion series. 29 In addition to plotting the axes, data points, and line of "best fit" this program types out the number of data points plotted (n) and the correlation coefficient (r) of the line of best fit. This coef- ficient is based on the following formulas: Y - Y = - 2^[2x2 - Zy2 - /(EX2 - Zy2)2 + 4(zxy)2}(X - X) Where: zx2 = zx2-M)2 s zy2 = EY2_M)2 zxy = m . (M?n zxy /z? /zy7 The equation used to calculate the line of best fit was Y = a + bX where a and b are the following relationships: 1 [zx2 - zy2 - /(zx2 - zy2)2 + 4(zxy)2] 2zxy = ^Y _ biX n n 30 III. EXPERIMENT DESCRIPTIONS According to Chester and Riley (1971), the quantitative estimation of zooplankton biomass (and production) is a difficult task. "The estimation of biomass is perhaps the field calling most for standard- ization but at the present time no one method seems to offer the poten- tial for widespread adoption" (Tranter, D.J. and Fraser, J.H., eds, 1968), In order to evaluate the measurement of zooplankton biomass using the LECO Carbon Analyzer, the following experiments were undertaken. A. STANDARDIZATION Since the LECO Carbon Analyzer was specifically designed for low carbon measurement in steel (e.g. less than 10% carbon in steel rings of one gram each), there was a question to its applicability to the measurement of carbon in relatively high carbon content compounds. An average of 50% carbon in dry weight is commonly assumed for pelagic marine invertebrates (Curl, 1962). Two organic compounds with nearly this carbon content were used in small amounts (0 to 75 mg) to stay within the detection limits of the analyzer. 1. Benzoic Acid Benzoic acid is a fairly common chemical standard of known carbon content (68.8487%), as computed from the total molecular weight (Hodgman, 1957). Sharp (March, 1973) used benzoic acid solutions for standardization of his combustion analyzer. Combustion of low mass samples occurs nicely with copper and iron accelerators. Greater than 50 mg samples of benzoic acid, however, did undergo rapid combustion and small deflagrations did occur. 31 A series of three runs of various masses of benzoic acid was performed and the readouts of the digital voltmeter (DVM) were recorded. Each of the three runs occurred on a different day. The weight of each sample was determined on a Mettler microbalance ( 0.1 mg). The normal apparatus blanking process was accomplished prior to each run (See Appendix C). 2. Casein Casein is a common protein with a carbon content of 53.13% (Heilbron, 1946), roughly paralleling the carbon content of marine invertebrates. The use of casein as a second standard was employed for two reasons. First, it was used to determine a second standard curve to compare with that of benzoic acid. Secondly, casein allowed the evaluation of possible contamination of the thermal conductivity de- tector by oxides of nitrogen. The same procedure used with benzoic acid was applied to casein. A series of three runs of various masses of casein was performed, and the DVM values and weights were recorded. Each of the runs again was done on a different day. Combustion was complete on all but a few samples, as indicated by the furnace plate current meter. These samples were discarded. 3. Infrared Analysis To further test the possible interference of gaseous products, specifically nitrogen oxides, an analysis of the combustion products collected in the cylinder of the carbon determinator was performed. A modification was made in the rear of the determinator to allow passage of the combustion products from the cylinder to a small 10 cm NaCl infrared cell (Figure 3 ). Analysis of each sample containing these 32 o CD +-> 0J o c o o o u CD o p". o o CD +-> X3 o c_> CD S- nD CD s_ ZS cn 33 combustion products was performed on a Perkin-Elmer 337 Grating Infrared Spectrophotometer. Samples from one of the casein standardization runs and from field sample analyses were taken as specimens to be analyzed. The carbon evaluations for these samples are in Table III. B. CARBON IN Tigriopus californicus In order to adequately determine the applicability of the LECO Carbon Analyzer to zooplankton studies, a preliminary test to evaluate the carbon content of a single copepod species was made. "Tigriopus californicus is a harpacti coi d, supra- littoral benthic copepod that is related to the pelagic, planktonic calanoid copepods. Convenience in choosing a test organism was considered, but at the same time it was important to use one that was similar to typical zooplanktonic species" (Baugh, 1974). The Copepoda comprise over 60% of the pelagic animal families and are, as such, the most common of the zooplankton in number. Populations of t. californicus were easy to obtain since they live in splash pools and occur exclusive of any other species along the west coast of California (Egloff, 1966). Field collections of natural popu- lations of t. californicus were taken from splash pools above the mean high-water mark along the rocks that line the beach around Lover's Point at the southern end of Monterey Bay, California. Identification of the species was simplified due to the distinctive reddish-orange color and essentially homospecific nature of the catches. The animals were scooped from the pools with a number 10 plankton bucket (160 m mesh size). Once collected, the copepods were kept in a plastic container at room temperature near a source of sunlight. The only source of nourish- ment for the copepods was the natural food in the seawater from the 34 splash pools. After allowing the collection of r. caiifornicus to stand for approximately one week, the animals were fractionated into size groups with the use of a sieve column (Figure 4.) and small "Nitex" nylon screens of various size meshes. The sieve column was back filled with pre-filtered seawater prior to fractionation to maximize the effectiveness of the process. This experiment was run twice. In the first experiment, three mesh sizes were employed: 297 ym, 177 pm, and 125 ym. The largest mesh size was placed at the top of the sieve column and so on, down to the smallest mesh size at the bottom. In the second experiment, as a result of lessons learned in size fractionation of this species, larger mesh size screens were used: 420 ym, 320 ym, 297 ym, and 250 ym. It was hoped that this approach might give answers as to the variation of carbon content in the different life stages of t. caiifornicus as delineated by the dif- ferent size fractions (see Figures 5(a) and (b). After fractionation, the nylon screens, with hundreds of copepods of each appropriate size range, were placed in the freeze-drying unit (Figure 2.) for approximately 20 hours (minimum drying time may be less). Carbon analysis was done on the freeze-dried organisms which were care- fully scraped into the ceramic crucibles. The crucibles were pre- weighed in order to determine the mass of each sample being analyzed. The mass and DVM reading of each sample were recorded in both experiments. The data from these experiments is given in Table IV. C. ATP-CARBON TO TOTAL CARBON RATIO IN Tigriopus caiifornicus Two joint experiments were conducted with Dr. Traganza to examine the ATP-C to total C (carbon) ratio in the test organisn and to deter- mine a method of finding the carbon present in living cells of organisms. 35 it if Figure 4. Sieve columns used for screening size fractions 36 Figure 5(a). Tigrii 7 < ali fom 37 \ M .<*** o o CD I/) If. 3 (/I a cd o <— T-, Q. ^ E Cr E- O 4- +-> O C •i — o cu ■I- (J +-> 03 U i — n3 a. s_ 4- T3 QJ ro N •i- cr. l/l c ai c cu ai s- cu X: s. I— u • s- .— - CD -O +-> — '4- LT) rC 333 urn; 333-243 pm; 243-202 pm; 202- 160 pm; and 160-102 pm. The net sample from the bucket was then concen- trated on Nitex nylon screens with the sieve column. In some cases, the sample was split with a "plankton splitter" before being poured into the sieve column in order to lessen the bulk of organic material. The nor- mal extraction and grinding routine followed for ATP analysis, and the particulate material collected on glass filters was frozen in the ship's freezer (<-3 C). These samples were later freeze-dried and analyzed for carbon. ATP samples were also frozen until analysis ashore. In addi- tion to the accumulation of total (i.e., the sum of living plus dead) and living carbon data, other relevant oceanographic observations were 45 4Q 0425 1,3,7 O6 02 -36-501 N -36"45'N .%*40'N 122* 10' W 05 122-00' W CRUISE /// Figure 9(a). Drogue track and geographic station plot for May cruise (7403). 46 CRUISE 111 STA^7 Figure 9(b). Water mass station plot relative to a drogue for May cruise (7403) 47 8-B K40 O A(450 Q 7-A 1025 9-0 2310 (g)23O0,003O ; A 1900,2050 3-B OS 15 Q A 1300 A 1100 O 2-A 0610 4-0 0950 (TOO! >40 O 6"° '330 \07IO ^ 5C 1145 O I 1 ®0200 1-0 0215 1 DR06UE TRACK JULY CRUIS£ 5 Mifes M Z2S Hours iii 02' 122*05' °^ OS 36*45' CRUISE IV Figure 10(a), Drogue track and geographic station plot for July cruise (7404). 43 — 0 - 1 -2 -3 -4 L5 Nautical Miles CRUISE IV Figure 10(b). Watermass station plot relative to a drogue, for July cruise (7404) 49 4-B 1030 O 0200 Ik 0*00%- 6-B Q20IS 3-A 0830 O 1143 , * 0955 0600 0750 o 8-C 2315 Q 3-A 1835 |^^^ o 9-0 0020 -36*SO'« -36'45'N -36'40'^ 122 05' w 122" 00 W CW/S£ 7 El'SS'W Figure 11(a). Drogue track and geographic station plot for August cruise (7405). 50 w / 3A Q ^6-B W Ds-a 4-b( ^vNfc *^5?<. // ^Sl ^v^^. \2-0 \l-0 v 'V V 7 \ f ■ - 0 - 1 - 2 -A L5 Nautical Miles CRUISE Y Figure 11(b). Watermass station plot relative to a drogue for August cruise (7405) 51 K^^ \ \ s ¥% \ \ \\ \ \ \ \ A \ \ \ \ \ \ \ N \\\\\\\\ \ '///// !\\U\\vV / / / / 4 A \ \ \ \ \ V V :\ \ \ \ ■■■ \ \ \ V A / / / / / / / / / \ \ \ \ \ \ //// fH\\\\\\\\ \ \ v m \\\v\Ai • v\\ \ \ \ \W\\\\ \ \ \'\\\U Uv\ A\\\\\\\\\\\i \\\\\\\\\W k\\\\\\\\\\\y\\ A \ \.\ \ \ \ \ \ \ \ \ \ \ \\\\\\\\\\\\Y u m\\\\\\\ \ \ \ \ \ \ \ v Figure 12(a). First qeneration net system on deck. S2 Fiqure 12(h). First generation net system on the Cldrke-bumpus sampler attached to the hydrowire. S3 Figure 13(a). Second generation net system on deck with nets and buckets exposed 54 SfeL Figure 13(h) Second generation net system attached to the hydrowire. 55 taken while at sea. Nansen bottle casts were used to determine T-S dia- grams and to collect nutrient samples for analyses with a "Technicon Autoanalyzer". Weather, sounding, and sea surface temperature data were also observed. A mechanical BT was attached to the second generation net system to determine the exact depth of the tow and to examine the temperature profile. The results of this data will be presented at a later date (Traganza, 1974). 56 IV. RESULTS The experiments described above were conducted to determine the quantitative capability of estimation of zooplankton biomass by carbon analysis with the LECO Carbon Analyzer. A. STANDARDIZATION 1. Benzoic Acid The various weights (o - 50 mg) and corresponding DVM readings from each of the runs of benzoic acid are given in Table II. The ap- paratus blank values, which are included despite the small ness of their numbers, were considered to be part of the analysis procedure. Thus, by approximation of negative and near zero readings and of small weights, the analysis of copper and iron accelerators alone (called "blanks") contributed to the data taken and the subsequent results. Since benzoic acid has a known carbon content (68.8487%C), each of the weights of the standard were converted to carbon (mg). The composite graph of benzoic acid data points for the DVM readings and weights was plotted by the method of "Best Fit" and is shown in Figure 14(a) . The linear relationship and high correlation coefficient (r = 0.9830) of this graph indicates that the DVM reading was directly proportional to the weight of the benzoic acid sample. After conversion of all weights to carbon, the composite graph of DVM readings and corresponding carbon values were plotted using the method of "Best Fit," also (Figure 14(b). This plot represents a "standard curve" since the carbon content of any benzoic acid sample of unknown weight could be determined from it. This can be done by solving for 57 TABLE II PLOTTED DATA FOR BENZOIC ACID CALIBRATION RUNS Run #1 Mass (mg) 24.5 20.5 13.9 9.2 45.6 27.4 12.7 22.7 8.2 22.3 15.9 Blank 0.1 Blank 0.1 Blank 0.1 DVM ■ Carbon (mg) (0.688487 x mass) 1.111 16.87 .946 14.11 .660 9.570 .464 6.334 2.025 31.40 1.330 18.86 .687 8.744 1.045 15.63 .434 5.646 .929 15.35 .753 10.95 .012 .06885 .004 .06885 .003 .06885 Total number of data points on Run #1 : n = 14 Run #2 Mass (mg) DVM Carbon (mg) (0.688487 x mass) .722 13.29 1.380 16.87 1.210 19.97 58 19. 3 24. ,5 29. ,0 Mass (mg) 19.3 24.5 29.0 25.1 13.2 18.1 19.0 8.4 10.0 8.7 4.3 24.3 32.7 Blank 0.1 Blank 0.1 Blank 0.1 DVM Carbon (mg) (0.688487 x mass) .722 13.29 1.380 16.87 1.210 19.97 1.153 17.28 .601 9.088 .773 12.46 1.069 13.08 .431 5.783 .595 6.885 .489 5.990 .248 2.960 1.377 16.73 1.343 22.51 .008 .06885 .003 .06885 .001 .06885 Total number of data points on Run #2: n = 16 Mass (mg) DVM Carbon (mg) . (0.688487 x mass) 4.6 .250 3.167 6.1 .310 4.200 7.9 .409 5.439 10.6 .488 7.298 15.2 .723 10.47 59 Run #3 (con. ) Mass (mg) DVM 1.143 Carbon (mg) (0.688487 x mass) 22.3 15.35 30.0 1.219 20.65 18.0 .814 12.39 7.5 .390 5.164 8.7 .449 5.990 25.5 1.263 17.56 12.8 .443 8.813 13.6 .691 9.363 15.0 .747 10.33 15.0 .704 10.33 10.7 .543 7.367 Blank 0.1 .001* .06885 Blank 0.1 .001* .06885 Blank 0.1 .001* .06885 Total number of data points on Run #3: n = 19 *DVM of these blanks represent averaged small positive values <+.002 Note: Small mass of 0.1 mg was used with all blanks. 60 + + + + + + 2.400-+ % n = 50 r = 0.9830 + I y = 0.0100 + 0.0^655 x + + 2.100-+ + + + + + + + 1.800-+ + X t > X 1.500-+ 0.6 OO-+ + + 0.3 0 0-+ + + + + +. t 1.200-1 * ++ t + + t + t + + 0.9 0 0-+ t .+ i + + ++ + + + r + i +..+ t / +t t • i + v ?+++++++++++++++++++++++++++++++++++++++++++++++++++H + +++-I 0 10 20 30 40 50 mg BENZOIC ACID Figure 14(a). Composite graph of the three runs of benzoic acid. 61 2.70 0 ; 2.400 : 2.100-: 2 > 1.800-; 1.500-; 1.200-; 0.900 : n = 50 r = 0.9832 Y = 0.0335 + 0.0652 X 0.600-; BENZOIC ACID 0.300-; *n rH 1 y + t v + ++ + + + + k + + +++ ¥ + t--* ^^■+f^ + +^+■^^• H++ + + +++ + + + + + +* + + + 4 t-t ► + * *■ + + + + +* ;** 0 5 10 15 20 25 30 35 mg C Figure 14(b). Standard curve for benzoic acid. 62 the carbon content, X, by using the equation of the standard curve and substituting the DVM reading for Y or by using the graph itself. The linear plot and high correlation coefficient (r = 0.9832) of the benzoic acid standard curve suggest a direct relationship between DVM and carbon. 2. Casein The various weights (0-71 mg) and corresponding DVM values and weights, including apparatus blanks are given in Table III. As was the case in benzoic acid, the linear relationship and high correlation coef- ficient (r = 0.9802) of this graph demonstrate that the DVM reading was directly proportional to the weight of casein sample (Figure 15(a)). Note that the equation of this line is very different from that of the DVM versus sample weight for benzoic acid (Figure 14(a)). The "standard curve" for casein (53.13% C) was also plotted after conversion of each weight sampled to carbon. Again, a linear plot and high correlation coefficient (r = 0.9802) of the casein standard curve propose a direct relationship between DVM and carbon (Figure 15(b)). 3. Infrared Analysis Interference with the DVM reading was evaluated by infrared analy- sis of the gaseous combustion products which entered the thermal conduc- tivity cell in consideration of possible nitrogenous oxides. Two casein samples denoted in Table III, and one freeze-dried plankton sample (see Appendix D, Cruise 7403) were selected for evaluations. All three were potential sources of oxides of nitrogen in the combustion products since they contained protein, a source of nitrogen. Because temperature reached 1600 C in the induction furnace during the combustion process, this reac- tion was a source of NO (nitric oxide). At the high temperatures, NO,, was not formed rapidly enough to appear in the exhaust gases (Stoker and 63 TABLE III PLOTTED DATA FOR CASEIN CALIBRATION RUNS Run #1 Mass (mg) 5.9 ** 16.9 35.8 71.0 2.9 6.0 4.8 1.4 14.9 15.8 10.1 ** 0.6 13.1 8.0 . 24.2 8.5 2.7 Blank 0.1 Total number of data points on Run #1 : n = 18 * DVM of these blanks represent small negative values. **These samples were also used for Infrared analysis. 64 DVM Carbon (mg) (0.5313 x mass ) .290 3.135 .490 8.979 1.058 19.02 2.532 37.72 .146 1.541 .296 3.188 .209 2.550 .069 .7438 .590 7.916 .578 8.395 .377 5.366 .027 • .3188 .429 6.960 .258 4.250 .702 12.86 .335 4.516 .115 1.435 .0001 * .05313 Mass (mg) Blank 1. 3 8. 3 10. 9 27. 8 22. 6 12. 1 9. 7 16. 5 9. 5 6. ;8 5, ,7 17 .7 25 .0 18 .7 30 .4 20 .1 42 .6 0 .1 Run #2 DVM Carbon (mg) (0.5313 x mass) .062 .6907 .385 4.410 .496 5.791 1.125 14.77 .779 12.01 .496 6.429 .456 5.154 .507 8.766 .299 5.047 .215 3.613 .246 3.028 .584 9.404 .714 13.28 .580 9.935 .975 16.15 .633 10.68 1.501 22.63 .0001 * .05313 Total number of data points on Run #2: n - 18 * DVM of these blanks represent small negative values 65 Run #3 Mass (mg) 0. 9 2. 6 8. 4 24. 6 36. 9 8. 3 17. 7 11. 1 13. 4 33. ,7 28. ,0 20, ,7 13. ,9 7, .7 4 .4 10 .2 12 .5 15 .5 0 .1 DVM Carbon (mg) (0.5313 x mass) .048 .4782 .126 1.381 .398 4.463 1.173 13.07 1.285 19.60 .339 4.410 .754 9.404 .334 5.897 .496 7.119 1.287 17.90 1.187 14.88 .752 11.00 .420 7.385 .340 4.091 .167 2.338 .476 5.419 .559 6.641 .413 8.235 .0001* .05313 Blank Total number of data points on Run #3: n = 19 * DVM of these blanks represent small negative values. 66 - + + + + + + + + 2.700-* n = 55 r = 0.9802 + I X = 0.0073 + O.03592 X + 2.400-J + + + + + 2.100-j + + + + + + + 1.800-| a t 1.500-J + + + 1.200-j , .. .• + + + + + + 0.900-: 0.600-: + + + • + + + + + + . + ++- .+ + + + : * * + ♦*. + 0.300-: ♦ x ♦*•♦ + * is + ♦+ + + + + + + + + + + T + + + ++ V + + •■ + + + >+ + + H- + + + + + + + + + J + + + + + + + + T + + + + + + + + + + + + + + + + + + + * + + 0 10 20 30 40 50 60 70 mg CASEIN Figure 15(a). Composite graph of the three runs of casein. 67 2.700-; 2.400-: 2.1 00-; 2 > 1.800-: 1.500- 1.20 0-: 0.900-; n = 55 r = 0.9302 r = 0.0276 + 0.06^9 x + V 0.600-; 0.300; + ►♦•>• % r • "+ t CASEI 0 5 10 15 20 25 30 35 nig C Figure 15(b). Standard curve for casein 68 Seager, 1972). The presence of NO can be seen in all samples in the infrared spectra (Figures 16(a) and 17(a)) at a frequency of about 800 cm (Pouchert, 1970). Even the blank and carrier gas (0?) samples show a peak in the spectrum at a frequency near 800 cm" , indicating that some nitro- gen entered the system at some time, e.g., when the furnace was unloaded and loaded. By visual inspection the relative magnitudes of the NO peaks appear to be roughly equal and low compared to C0?, which implies a con- stant amount of NO formation. Thus, due to the constancy of the NO peak, no biased results occurred from NO interference. No significant peaks occurred in the spectrum due to the presence of N02 as suspected (Pouchert, 1970). The near-infrared spectra of Figures 16(b) and 17(b) show a strong C0p peak at a frequency of about 2349 cm (Pouchert, 1970) in both casein samples and plankton sample. The carrier gas (0?) and "blank" samples (accelerators only) do not have this peak as one might expect (Figure 23). A possible harmonic peak of C02 exists in the infrared spectrum near 1270 cm" (Dr. C. F. Rowell , personal communication) of the samples shown in Figures 16(a) and 17(a). A final possibility for nitrogen was the presence of N~0 (nitrous oxide). This gas has a thermal conductivity at 27 C of 4.13 cal sec cm" deg" compared to 3.96 cal sec" cm" deg" for C02 (Ewing, 1972). Not only might this gas affect the DVM reading like other oxides of nitrogen, but its presence would affect the thermal conductivity sensor in very much the same way as C0?. Only by infrared analysis or some other means could this be tested. All of the samples indicate that there is no noticeable N?0 peak (Pouchert, 1970) in the infrared spectrum and that there was an insignificant amount of interference from this gas. 69 »o Z3 S- -t-J CJ o C — - 3 m to ro -r- CD > E S- i— CD t- o •i- 4- o o S- QJ CN QL ro c U. for c ystyre o E •— Z3 O o S- CL o (J QJ ed spe s. Th o S- OJ ro ■ — o s- a. M- E a) . In sa o o vo CN QJ S_ CD •r— o o CO (%)3DNVU.IWSNVyJL 70 o o C E •i- rs oj s- 1/7 4-> ro o O O) Q- "O CO c IN( (T3 (T3 r— S_ -O JD r— -— « O CNJ o o CO TD ra OJ D"; CO o X" 3 o t S- to OJ co c^ s_ * — ' CO -Q CO r— OJ i- rs CD •r- o o o O V (%)3DNVlllWSNVai 71 cd CD c: CD 5- O a. CD CD O- 03 +-> <_> CD CD CL-M (/) 13 s_ T3 -Q CD ■«- S- r— 03 rO S- <_> 4- E O r-«. CD S- en (%)3DNVlllWSNVyi 72 c a> s- CO >> o CL CD ai CL E to o • +-> E c s- res +j i— o O- u; Q. s- to O <+- ZS S- +J U rO CD S- CL_a to •!— -a v cd o s- f0 o S- +J E "O 1—1 CL) to S- Z3 f0 (1) 00 cd O) (%)3DNViJLIWSNVai 73 B. • CARBON CONTENT IN Tigriopus calif ornicus Two different experiments were attempted to quantitatively determine the total carbon content of the test organism. In both sets of carbon analyses, the DVM readings were converted to carbon by using a combined standard curve based on benzoic acid and casein. Its particular use is further explained in the Discussion and Conclusions section. Weights (0-20 mg) of freeze-dried t. californicus and corresponding DVM readings from both experiments are given in Table IV. In the first experiment, all samples were of one size fraction, (greater than 297 m) while in the second experiment several size fractions were used. The DVM readings and weights were plotted by the method of "Best Fit" and are shown in Figures 18(a) and (b). Blanks were included in this and all other graphs associated with the carbon content experiments. The linear relationships and high correlation coefficients (r = 0.9903 and r - 0.9928 respectively) of the first and second experiments indicate that the DVM reading was directly proportional to the weight of the sample. After con- version of all DVM readings to carbon (Table IV) two graphs of carbon and sample weight were plotted in the same manner for each experiment (Figures 19(a) and (b)). The linear relationships and high correlation coeffi- cients (r = 0.9893 and r = 0.9917, respectively) of these graphs indi- cate a direct proportionality of the carbon to weight in t. californicus. The final "Best Fit" plot (Figure 20) is a combination of the carbon and sample weights from both experiments. Table IV also shows the percentage of carbon to freeze-dried weight for each sample analyzed. A mean carbon percentage was computed to be 40.27, for the first experiment (excluding the very small sample of 0.3 mg which gave an abnormally high DVM reading) and 37.87 for the second 74 TABLE IV PERCENT CARBON DETERMINATION FOR tigriopus californicus Experiment 1 Fraction Size Mass (mg) DVM Carbon (mg) %C (C/mass x 100) >297 vm 0.3 .019 .2808 93.60 >297 1.2 .042 .6208 51.73 >297 0.6 .011 .1626 27.10 >297 0.3 .001 .0148 49.33 >297 1.5 .042 .6208 41.39 >297 1.1 .026 .3843 34.94 >297 3.0 .068 1.005 33.50 >297 2.4 .040 .5912 24.63 >297 5.6 .167 2.468 44.07 >297 5.2 .164 2.424 46.61 >297 14.3 .389 5.750 40.21 >297 10.3 .340 5.026 48.90 blank 0.1* .001 Blank 0.1 .0001 n = 14 for DVM vs. mass N = 12 for C(mg) vs. mass (mg) * Blanks given represent a blank of 0.000 75 Experiment 2 Fraction size Mass (mg) DVM Carbon (mg) %C (C/mass x 100) >420 yfll 12.9 .292 4.318 . 33.47 >420 6.6 .143 2.114 32.03 420 - 320 3.5 .089 1.316 37.60 420 - 320 8.1 .192 2.840 35.06 420 - 320 11.3 .298 4.407 39.00 320 - 297 2.0 .055 .8135 40.68 320 - 297 11.9 .298 4.407 37.03 320 - 297 3.8 .098 1.449 38.13 297 - 250 0.6 .027 .3993 66.55 297 - 250 0.2 .007 .1035 51.75 >420 28.7 .708 10.47 36.48 >420 8.9 .222 3.283 36.89 >420 10.1 .252 3.727 36.90 >420 9.7 .242 3.579 36.90 >420 12.2 .332 4.910 40.25 420 - 320 11.3 .332 4.910 43.45 420 - 320 14.0 .354 5.236 37.40 420 - 320 16.0 .375 5.546 34.66 320 - 297 9.4 .215 3.179 33.82 320 - 297 11.5 .252 3.727 32.41 320 - 297 21.0 .496 7.336 34.93 297 - 250 5.0 .101 1.494 29.88 76 Fraction size Mass (mg) 297 - 250 m 3.8 297 - 250 8.6 Blank 0.1 Blank 0.1 DVM ,084 ,176 ,0001 ,012 Carbon (mg) %C (C/mass x 100) 1.242 2.603 32.68 30.27 n = 26 DVM vs. mass N = 24 C vs. mass 77 + + + + + 0.4 0 0-+ + + + + +■ + + 0.3 50-+ + + + .+ + + + 0.300-+ + + + + + + + 0.250-+ + + + + + + + 0.200-+ -t + + + + + 0.150-| + + + + + + + 0.1 00-+ '+ + + + + + + 0.050-+ + + + > n « 1> r = 0.9903 Y = -0.0030 + 0.029*+ X T. c a I Hornicus + + • + &►++++++++++++++++++++++++++++ »■++++++++++* + + + + + + + + + + ++++ + + 5 7.5 10 12.5 2.5 SAMPLE WEIGHT , mg Figure 18(a). First experiment. 78 > + + + + + + + 0.9 00-+ + + + + + + + 0.8 00-+ + + + + + + + 0.700-+ + + + + + + + 0.5 00-+ + + + + + + + 0.400-+ + + + + + + - + 0.300-+ + + + + + + + 0.200-+ + + + + + + + 0.100-+ + + + + + + + n s 26 r = 0.9928 Y = 0.0004 + 0.02^3 X • + + • + + + +. +• + +• T. cal i iornicu 4. + + +++♦ + + + + + + + + + + + +++-♦«• + + ++ + + + ++ + + + + + + +++ + + + + + + + + +■»• + + + + + + + + + 5 10 15 2*0 25 SAMPLE WEIGHT , mg Figure 18(b). Second experiment. 79 + + + + + + 5.4-t + + + + + + 4.8-: + + + + + + • + 4.2-! + + n = 12 r = 0.9893 Y = -0.05^3 + 0.^366 X 3.6-: u en E 3.0-: 2.4-1 + + + + + + + 1.8-: + + + + + + T. caEifornicus 0.6-: + + . + : * + . 6 2.5 ■ + + + [ + + + + + + + + + + + •»4 + + + + + + + + + + + ++ + + -^■^•«■ + + + + + + + + + + ^ 5 15 10 125 15 SAMPLE WEIGHT , mg Figure 19(a). First experiment, .% carbon determination 80 T + + + + + + + 9.6-+ + + + + + + 8.4 + + + + + + + + 7.2 | U E + + + + + 6.0 + + + + + + + + 4.8-+ + + + + + + + 3.6-+ + + ■ + + + + + 2.4-+ + + + + + + + 1.2-J + + + + + + + n = 2k r = 0.9917 Y = -0.0124 + 0.3622 X + + + + + + 4. + T. caliiornicus •F+-t+f f + + + +- "H 0 5 10 15 20 25 SAMPLE WEIGHT , mg Figure 19(b). Second experiment, % carbon determination. 81 u en + + + + + + + 9.6-+ + + + + + + + 8.4-: + + + + + + + 7.2-+ + + + + + + + 6.0-+ + + + + + + + 4.8-+ + + + + + + + 3.6-+ + + + + + + + 2.4-+ + + + + + + + 1.2 + n = 36 r = 0.9882 Y = 0.0^06 + O.3658 X + + + ■? *■ + . + + ++ • +• + T. cal ifornic u : % * + -H- + 4-*f+ + + + + + + + -t++f !• + + + + •>• + + +(• + h + + + + 4+ + + + + ++ + -(• *- + + + + + + + + + + + + + ► f+ + + 0 5 10 15 20 25 SAMPLE WEIGHT , mg Figure 20. Combined data plot from both experiments 82 experiment. No significant deviations were found in the carbon percent- ages of different size fractions in the second experiment. Finally, a combined mean carbon percentage of both experiments was computed to be 38.6%. Curl (1962) determined the total and organic carbon in terms of percent dry weight in "mixed copepods" to be 35.6% and in Caianus finmar- chicus, a well-known pelagic copepod , to be 39.8%. The inorganic carbon in all but one of the 19 species and mixed collections of marine organisms that Curl analyzed was found to be negligible. Assuming that no signifi- cant inorganic carbon was present in Tigriopus caiifomicus , the value of 38.6% carbon agrees well with Curl's data of total and organic carbon in copepods. C. ATP-CARBON TO TOTAL CARBON RATIO IN Tigriopus caiifomicus The ATP-C to total C ratio was examined in order to determine the ratio of ATP to cell carbon and its constancy by different size groups in the test organism. The organisms analyzed for ATP and carbon were assumed to be alive (dead and detrital particulate matter passed through the screens). Table V shows the ATP-C and carbon data from the first and second experiments, respectively. Also noted are the corrections applied and the size fractions of t. caiifomicus used in the experiments. The "Best Fit" plots of ATP-C against total C are given in Figures 21(a) and (b) for the first and second experiments, respectively. A third graph was done of the combined data from both experiments (Figure 21(c)). The 297-250 ;im fraction was excluded since it was not common to both experiments and may have contained detritus. No signifi- cant differences were observed in the ATP-C total C ratio (Table V) between different size fractions. The linear relationship and correlation coefficient (r= 0.9505) of the composite plot suggests a constant ATP-C 83 TABLE V ATP CARBON TO CARBON RATIO IN tigriopus californicus Experiment 1 Fraction size Mass (mg) DVM Carbon (mg) (ym) Approximate Uncorrected > 420 6.2 .292 4.319 > 420 8.0 .230 3.400 420 - 297 7.3 .275 4.067 420 - 297 12.0 .487* 7.203 420 - 297 8.0 .218 3.222 Control .5 .011 .163 Before Filtration CL (mg) ATP-C (mg) ATP-C/C, xlOO L 4.392 .0095 .2163 3.437 .0107 .3114 4.127 .0118 .2859 7.432 .0177 .2382 3.239 .0083 After Filtration .2562 CL (mg) ATP-C (mg) ATP-C/C, xlOO 4.393 .0108 .2458 3.436 .0100 .2910 4.125 .0097 .2352 7.431 .0172 .2315 3.239 .0083 84 .2562 Extraction volume Correction (mg) .2273 .1889 .2113 .3742 .1718 Three corrections applied to DVM derived carbon: 1. Volume extraction correction (added) 2. ATP-Carbon in ATP analysis (added) 3. Control contributed carbon (subtracted) * Question in data - DVM is either .287 or .487. 85 Experiment 2 Run 1 Fraction Size( m) >420 Mass (mg) Approximate 10.9 DVM Carbon (mg) Uncorrected CL (mg) ATP-C (mg) Ratio .384 5.679 5.636 .0158 .2803 >420 19.5 .387 5.724 5.679 .0144 .2536 420-297 23.6 .388 5.739 5.695 .0154 .2704 420-297 23.5 .357 5.280 5.235 .0143 .2732 297-250 12.4 .119 1.760 1.707 .0060 .3515 Control 0.1 .004 .059 Run 2 >420 15.3 .311 4.660 4.556 .0138 .3029 >420 10.5 .141 2.086 1.973 .0060 .3041 420-297 17.8 .325 4.807 4.701 .0115 .2445 420-297 18.1 .468 6.922 6.821 .0166 .2434 297-250 15.8 .073 1.080 .9643 .0023 .2385 Control 1.5 .008 .118 Two corrections applied to DVM derived carbon: (1) Control contributed carbon (subtracted) (2) ATP-Carbon in ATP analysis (added). There was no volume extraction correction. 86 + + + + + + + + 0.016-+ + + + 0.010-+ + o i i I *- 0.008 + < + t t + + 0.006-+ + + + + + + + 0.004-+ + + + + + + • 0.002-+ + + + + t' + .+ n * 10 r = 0.9569 % 1 = 0.0022 + 0.0020 X + + + + 0.014-+ + + + + + + + . 0.012-+ + / + + + + + + . + + + t' T. californicus t,. + + + f + 4+ + + + + + + + + + + + + + + + + + + + + + + + + -f + + + + + + + + + + + + + + + + + * + + + + + 0 1.5 3 4.5 6 7.5 mg C Figure 21(a). First experiment, ATP-C to total C ratio. 87 u en E + + + + + + + + 0.016-+ + + + + + + + 0.01 4-+ + + + + + + + 0.01 2-1 + + + + + + + 0.0 10-; + + ■ + + + + + 0.GG8-: + + + + + + + 0.0 06-+ + + + + + + + 0.004-+ + + + + + + + 0.002-+ + + • n = 10 r = 0.9813 Y = 0.0010 + 0.002^ X . + + + alif amicus + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 1 + + + + + + + + + + + + + + + + + + + + + + + + + + + + 1.5 1 45 m g C 7.5 Figure 21(b). Second experiment, ATP-C to total C ratio, 88 + . + + • + + + • + * + % + •• + 0.016-+ n = 18 r = 0.9505 + + * + + . + Y = 0.0019 + 0.0021 X + • + + + + * 0.014 + + + + • + + * + • + + • + 0.012 + + + + • + + • + + + + * + + • + ■ 0.010-: 1- + • + + u * • w ♦ 1 ♦ • ♦ CL ♦ * . + *- 0.008* • < * • 4 + • » : • E • 4 • 0.006-+ . + • + • i • 4 + • + + • 0.004-+ • + t + + + + + • %' T. californicus 0.002-} + + + + + * + + + ■» + + + + + + + + + + + + + + + + + + + J- + + + + + + + + ++ + + + + + + + + + + + + + + + + + + + +• + + + + + 0 1.5 3 4.5 6 75 mg C Figure 21(c). Combined data plot of both experiments, ATP-C to total C ratio. 89 to total C ratio in both size fractions. A mean percentage of ATP-C to total C from the data of the graph yielded 0.26%. Balch (1973) has re- ported a mean of 0.89% for the ATP-C to total C ratio, 0.78% for total carbon, and 1.89% for non-1 ipid body carbon, in Caianus finmarchicus . Holm-Hansen (1970) determined that "during exponential growth in batch culture, cellular contents of ATP remained at fairly uniform levels in unicellular algae and averaged 0.35% of the cellular organic carbon con- tent." From the assumption that analyses were made on live organisms, the total carbon determinations were considered to be a measure of the carbon present in living cells, i.e., "living carbon" or C. . Thus, the ratio can be expressed such that ATP-C/C, = R, where R is the ratio and C, is "living carbon". Since this mean ratio was determined for r. californicus as 0.2534 x 10 , ATP measurements converted to ATP-C can be combined with this ratio to solve for living carbon: CL = 1/R(ATP-C) which is approximately 380(ATP-C). All determinations of living carbon stemmed from this equation. D. ATP-C AND CARBON ANALYSES IN ASSOCIATION WITH FIELD STUDIES ATP and carbon analyses were made on "net zooplankton" samples from three cruises taken in the Monterey upwelling area (Appendix D). These analyses were used to compute total (C), living (C. ), and dead carbon (Cn = C - C. ) , in each of the five net fractions for each station. (H3 = > 333 urn; //5 = 333-243 imr, #8 - 243-202 urn; //10 = 202-160 nm; #14 = 160-102 nm). The values were computed in terms of surface area (per square meter). The nose cone used in both net systems sampled a 0.041 m column (0-200 m) of ocean and the resulting determinations were 90 converted to a square meter. Total carbon determinations on the "net zooplankton" samples were made using the LECO Carbon Analyzer. The use of ATP measurements and the ATP-C to carbon ratio in t. califomicus was employed to compute living carbon estimates. The dead carbon estimates were computed by taking the difference between total and living carbon. Again, based on Curl (1962) the inorganic carbon in the marine organisms was presumed to be negligible. Thus, total particulate analysis of the "net zooplankton" samples represented total organic carbon. Time series plots of total, living, and dead carbon for all cruises are given in Figures 22(a), (b), and (c), respectively. Night and day stations are indicated. Trend lines are also shown by connecting night and day mean values for each net fraction for each cruise. No adjustments were made to allow for minor errors in the data indicated in Appendix D. The trend lines shown in the summary plots indicate that there was an increase in biomass and in total, living and dead carbon mean estimates in each of the net fractions collected on the cruises from May to July, followed by a slight decrease in early August. Two net fractions pro- vided exceptions to this trend, since #8 (243-202 um) and #10 (202-160 ym) net fractions remained relatively constant. Living (C, ) and dead (Cn) carbon in fractions #6, #8, #10 and #14 reached a maximum of about 200 mg/m m whereas the highest value for any station for total carbon (i.e., 2 the sum of living plus dead (C)) was about 300 mg/m . The unbounded upper 2 fraction (#3) contained up to 2000 mg/m of total carbon and up to 800 2 2 mg/m for living and 1600 mg/m for dead carbon. Significant differences in night and day stations of each cruise occurred, e.g., the difference was greater than 50% of the mean C. (day) in the #6 fraction in July (see Discussion and Conclusions). Since the size fractions were only approxi- mately 50 vjiti, a skewed distribution of biomass occurred in the net 91 mg C m 1500- 1000 200- 100 0 300- 200- 100 o-t- 300- 200- 100- 0 300 200H 100 0 7403 MAY 7404 7405 JULY AUG. — i ' ^=3 =#= 8 -^10 # 14 L Figure 22(a). Total carbon (C) per surface area. + = night station, o = day station; — = night, — - day 92 1500 10 0 0- 500 0 300 200 100 0 300 rrig CL X 20 0 m 100 o 300^ 200 100 0 300 200-I 100 7403 MAY 7404 7405 JULY AUG. ^3 -W=Q o o ^8 ^=10 ^-14 — — "^ ^1-. Figure 22(b). Living carbon (CL) per surface area. + = night station, o = day station; = night, — = day. 93 7403 MAY 1500- 1000- mgCD m 200- 100 0 -100 200H 100 0 -1001 7404 7405 JULY AUG. 1 ' ^3 o ♦ =#=8 *M0 ^=14 Figure 22(c). Dead carbon (Cn) per surface area. + = night station, o :: day station. night, day. 94 fractions. For example, in the most obvious case, the #3 net fraction (> 333 ym) which had no upper limit contained greater biomass than the lower fractions. The #6 net fraction (333-243 ym) also had a greater size range than #8, #10, and #14, and more biomass than did these frac- tions. 95 V. DISCUSSION AND CONCLUSIONS A. LABORATORY WORK It has been established that organic carbon analysis is a sensitive and reliable test of the biomass of marine organisms, i.e., zooplankton. The use of the LECO Carbon Analyzer, adapted and used for all total car- bon analyses, demonstrated a rapid (70 seconds), consistent, and accurate measurement of carbon in benzoic acid (a non-protein) and casein (a pro- tein), the two chemical standards, in samples of less than 71 mg. Further- more, the comparison of the two standard curves in Figures 14(b) and 15(b) of these organic substances reveals a \/ery close relationship of DVM to carbon in the two different standards. If a benzoic acid or casein sample were combusted, therefore, the use of either standard curve to convert the DVM reading would yield the same measurement of carbon. There seems to be no interference of gaseous combustion products, i.e., oxides of nitro- gen, with the thermal conductivity cell of the LECO Carbon Analyzer as demonstrated by the similarity of the two standard curves and the infra- red analyses. Thus, it was assumed for further carbon analyses that the measurement of carbon in zooplankton could be based on the "standard curve" for benzoic acid and casein, i.e., direct conversions of DVM read- ings to zooplankton carbon could be made. All DVM conversions to carbon in all but the standardization experi- ments were actually based on combined standard curve for benzoic acid and casein. The equation of the composite graph of DVM against weight of benzoic acid for three runs was found to be Y = 0.0101 + 0.0465507 X (Figure 14(a)). For the range of plankton samples in this study, (^5-20 mg) 96 the above equation was approximated by Y = 0.04656 X for computational purposes. The resulting standard curve, using this equation, a 10 mg sample of benzoic acid, and its carbon content, was Y = 0.06763 X. This equation was manipulated to yield X = 14.79 Y, where Y is the DVM reading and X is the carbon in mg. The equation of the composite graph of DVM against weight of casein for three runs was determined to be Y = 0.0073 + 0.03592. As before, for the sample range of interest, this equation was approximated by the following form: Y = 0.03592 X. The conversion of this equation to that of the standard curve, using a 10 mg casein sam- ple and its known carbon content, resulted in the equation: Y = 0.06761 X. Manipulating this casein standard curve equation lead to the same result determined by the benzoic acid standard curve equation: X = 14.79 Y, where Y is the DVM reading and X is carbon in mg. Thus, measurements of carbon (mg) were obtained on the basis of multiplying the DVM value times 14.79. These approximations of slope to obtain a zero intercept could have lead to an error in carbon calculations of up to ±3% at high or low values by the standard curves. This error is still less than that found in other total carbon analyses which are usually greater than 5%. The determination of the total or organic carbon content in t. cali- fornicus demonstrated a good measurement of carbon in a single copepod species. This mean carbon content of 38.6% of the freeze-dried weight agrees well with Curl's data on the carbon content in copepods (1962). Thus, it was shown that carbon analysis using the LEC0 Carbon Analyzer could be extended to freeze-dried plankton in determining total carbon (i.e., the sum of living plus dead). The examination of the ATP-C to total C ratio in r. califomicus resulted in a mean ratio of 0.2634 x 102 or 0.2634%, which could be used 97 to convert ATP to living carbon. The constancy of the determined ratio from the data in these experiments suggests that no nonliving particulate matter was present and that there was no difference in two different size groups. The two experiments were each conducted in a slightly different manner. The effect of filtration, tested in the first experiment, caused no significant changes in the ATP measurements of before and after filtra- tion. Control analyses consisted of filtering the TRIZMA buffer solution by itself and were done to investigate the effect of the ATP extraction on ATP measurements of the solution passing through the filter, as well as carbon measurements of the freeze-dried filter. The result in both experiments was a small control correction, usually subtracted from the original measurements of ATP and carbon. Suspended carbon carried over in the ATP aliquot was corrected for in the first experiment, i.e., 2/extraction volume x total C was added to total C. No corrections were made for dissolved non-ATP-C or possible losses on experimental apparatus. Corrections inherent in the ATP and carbon methods of measurement might have been necessary, but the constancy of the ATP-C to cell carbon ratio in the test organism suggests that consistent analysis procedures were used, and the ratio is accurate, assuming these corrections were insigni- ficant. Separate sievings of the test organisms produced no significant differences in the ATP-C to total C ratio. No significant differences occurred between different size fractions. Two replicated size fractions (297-250 vjm) were excluded from the combined data graph (their ratios were 0.24 and 0.35%) of ATP-C against total C in the determination of the mean ratio, since these fractions were not common to both experiments and may have contained detritus. Based on the assumption of live organisms in this experiment, the ATP-C to total C ratio examination provided a 98 ATP-C C L measurement of the amount of living carbon (C. ) expressed as R This relationship was later used in the form C. = -jHATP-C) % 380(ATP-C) to compute living carbon estimates in "net zooplankton" samples. These estimates of living biomass were based on the constancy of the ATP-C to total C ratio in t. caiifomicus and its further application to all mar- ine organisms caught by the nets. B. FIELD STUDIES No existing sampling net takes a representative sample of all types of zooplankton in any given area (Mullin, 1969). However, the in-line mul- tiple net system utilized in this study was designed by Dr. Traganza in an attempt to separate trophic size groups by approximate size fractions which would include organisms which were predominantly herbivorous. The importance of herbivores in sound scattering models is a subject of another study (Traganza, 1974). The sampling technique employed several hypotheses on the distribu- tion of zooplankton in the ocean. For example, as illustrated by Black- burn, et al (1970) the amount of chlorophyll a_ below 150 m in the eastern tropical Pacific is negligible. Thus, the herbivores and phytoplankton were assumed to be in this upper layer of the ocean. The herbivorous plankton as grazers of the phytoplankton were generally assumed to be indigenous to this region or they migrate vertically to it from depth. A sampling depth of 200 m, which would include the euphotic zone, was thus chosen as the limiting depth of the vertical net tow. No correction was made for unintentional sampling at greater or lesser depths. Appendix 0 gives the depths of the tows. The zooplankton populations were assumed not to be uniformly or ran- domly distributed, but rather in "patches" (Margelef, 1967) which are 99 strongly dependent on space and time manifestations of an organization related to hydrographic distributions such as eddies, areas of strong vertical mixing, upwelling domes, internal waves, Langmuir circulation, convergences and divergences, etc. (Traganza and Stewart, 1973). These "patches" were assumed to be on the order of miles or tens of miles (Cush- ing and Tungate, 1963). The practical approach used to consider these "patches" was to study a volume or water mass defined by coordinates relative to a drogue. Each cruise began from nearly the same geographic location. The use of the drogue to mark the water mass eliminated hori- zontal advectional effects and resulted in the station positions being taken relative to the drogue. The X-shaped station pattern was based on an optimum search technique which was used to minimize the zooplankton patchiness problem and locate maximum concentration of zooplankton in a search area (Traganza, 1974). Estimates of total, living, and dead zooplankton biomass by carbon analysis and ATP measurements demonstrated a definite seasonal trend over the period of the three cruises. Explanation of these results is beyond the scope and intent of this thesis. However, it is clear that the meth- odology developed in this study is a rapid (70 seconds), precise (±3%) and accurate (±3%) measurement of zooplankton biomass carbon. If such estimates of zooplankton biomass by carbon analysis are com- bined with biomolecular characteristics of zooplankton, e.g., ATP, and related to chemical properties of the environment, predictive sound scat- tering models of the ocean may be feasible. 100 VI. RECOMMENDATIONS After reviewing carbon research and reflecting on the experiences and knowledge gained in this thesis, the author makes the following rec- ommendations: (1) Further verification of the derived standard curves of the LECO Carbon Analyzer should be done with other standards in future studies which use them directly for carbon calculations in zooplankton. (2) A sieve column for carbon samples, which would use stainless steel bolting cloth screens for concentration of the sample (these steel bolting cloth screens would fit into a crucible for direct carbon analysis after freeze-drying) should be tried. (3) A more efficient commercial freeze-drier (2-3 hour drying time) is needed. (4) Carbon analyses should be attempted at sea with the LECO Carbon Analyzer combined with one-week or longer cruises to make dirunal and several day studies to allow duplicate sampling. (5) Acoustic measurements should be made in the same region to test the application of zooplankton biomass estimates to sound scattering. 101 APPENDIX A - VOLUME REVERBERATION THEORY "The scattering of sound by biological populations in the upper layers of the ocean can place practical limits on the operation of low frequency (2 - 20 kHz) sonar" (Traganza and Stewart, 1973). The argu- ment is outlined quantitatively from Batzler, Vent, and Davis (1968) as follows. The echo level, EL, depends on the target strength, TS; the source level, SL; and a logarithmic function of the depth of the water column, H, such that EL = SL + TS - 2H. The volume reverberation level, RL , the most variable and unpredict- able of the reverberation sources, depends on the source level, SL; the area insonified, A; the integrated water column scattering strength, S ; and a logarithmic function of the depth, H, such that RL = SL + 10 log A + S - 2H v 3 v where A = r x ct/2 x sec 0 c = speed of sound in seawater, x = pulse duration, = beam width, 0 = transducer tilt, and r = horizontal range to the target. The integrated water column scattering strength, S , is defined as z2 Sv = 10 log J sv(z)dz Zl where s (z) is the scattering strength of any segment of the layer lying between depths z, and z?. 10? S is dependent on the type and density of scatterers that give rise to reverberation and represents the amount of reverberation by all scatterers in a water column one square meter in area to a given depth. 10? The a6i'» t„ detect made the «, ' atV organic nr ■ T"e™> co„duct,. / etect- re,,*,. a„d nf,. nSa transport nhp. rec°gmsed. °f k^c energy Pneno^enon fn that ft T. ^ US t0 a ^Perature craw transfe^ Mie aeterminfno th* „ Terence fn fh 9 ' Nation 0f the T/r „ 7n the™a7 conductivity of dStector * a SamP?e mixture r^ a car^er gas *nft Ure- ™e heated f11amo , ~ ' and a carrier qas ;-- — a_ ^r coo,ed to — - - *■ — ,*. ,eCause 0 tta e :™ * *»«** are commonly u-oo ca] sec rm~'^ -7 cm de9 ' at 27°c. ion APPENDIX C - CALIBRATION INSTRUCTIONS (After LECO Instruction Manual, 1974) Listed below are the basic steps which one would take prior to analysis of samples using the LECO Carbon Analyzer after completion of the electronic checkout: 1. Turn on the FILAMENT switch of the induction furnace; after one minute, turn on the HIGH VOLTAGE switch of the induction furnace. 2. Open and close the leading tray of the induction furnace. 3. Turn the BLANK switch on the determinator OFF if it is on. 4. If the DVM HOLD switch is not glowing, depress it to turn it on. 5. Turn the FUNCTION SELECT switch to the OPERATE position. 6. Depress the ANALYZE switch to initiate the timing sequence. The switch will glow green before starting and will glow white at the time it is depressed and throughout the timing sequence. The oxygen should start flowing at this time also. 7. Set the oxygen flow through the purifying train to 1.5 liters per minute during the first 20 seconds after depressing the ANALYZE switch. 8. When the DVM HOLD and ANALYZE switches stop glowing, signifying an end to the 70 second timing sequence, open and close the furnace, but do not load a crucible. 9. Depress the ANALYZE switch to initiate a bridge balance determin- ation. The determinator will go through the timing sequence again. 10. When the ANALYZE light stops glowing, the reading on the DVM should be 0.000 + 0.002. If not, adjust the BRIDGE BALANCE FINE and/or COARSE controls located behind the swing-out access panel to bring the reading within specifications. Id1 11. Repeat steps 8 through 10 with the DVM HOLD glowing for verifi- cation of bridge balance zero. Repeat if necessary. 12. Place a scoop of LECO iron chip accelerator and a scoop of LECO copper metal accelerator in a crucible. 13. Open the loading tray and place the loaded crucible on the cera- mic pedestal. Close the loading tray and swing the counterweight arm against the POWER button of the induction furnace. 14. Depress the ANALYZE and DVM HOLD switches to start the timing sequence. The oxygen flow will not go to zero during pre-burn as before. The induction furnace plate current should go to between 400 to 500 milli- amps before the red light of the induction furnace stops glowing. 15. The reading on the DVM will give an indication of the percentage content of argon in the oxygen tank. Depress the BLANK switch to turn it on and depress the DVM HOLD to turn it off. Adjust the BLANK control to bring the DVM reading to approximately 0.000% C. This is a rough setting of the BLANK. 16. Repeat steps 12 through 16 with the BLANK switch on for a verifi- cation of the blank adjustment. 17. Repeat steps 8 through 10 with the' DVM HOLD glowing for verifica- tion. 18. Repeat steps 12 through 16 with the BLANK switch on for a final "blank" adjustment. The blanking process can be further amplified by finer calibration with LECO steel rings. These instructions are given in the LECO Instruc- tion Manual. This amplifying procedure was not used due to the time involved with further calibration and the number of samples required to complete the steel ring calibration. The ^ery slight increase in accu- racy due to this calibration was not worthwhile for the scope of the 106 "rbon analyses reported in tM K ' Leu 'n this work tv •»■'" "«e ManHng process„ cn„ , • St6pS ' th™gl> 18 COm- S- COmP,eted b^re each analysis. 107 t) _i t > V + « l L 0 2 cc O LL o z Q O o Im c/) CL CL < UJ z IS la o 1*) 0 la « 0 < i- < 0 uj Cl Z J o < o s ? N ^ .. 4 *> W,1 i — [.. -1 - - : • r- ^ : - 1 - * _ — o 3 - - 1 <\J S | (< H rr - - •-I - =-i *- I Ml1 jo" -1 >i -i »C ■ •Jl : - <5^ : - qfn j Qi j C5 3 : - •51 1.1 <3l - — "1 - ^i - : o « si > « it" if ^ : 3JS - o hi - - - too - - 4 S <1 "I 0<9" - 8 - _ - 1:^ _ h * N - - - - Q bo ij U H - jSa Is ^ - U 3s 3d : s - \S1 • j - — 8 i - h - _ - - ft 1 \ *1 ra ji ?! ; ^ u h-. : i - i r _ - - - - 0 R ■ < *3 - ^ ■ - t — i *, , V\. . , sS. -d o ! Qi 1 Q, i 0 , O l-Ckl 0 ■1 108 i s 3 0 -a I « DC O U- O z: IQ O o j'CO o CL [tr < 111 z UJ o S > 8 1° \»* V t •*- ■S < o I H < 0 M r l«J J «J J 1*; I fci O <*,, Q 1 O 0 J »q jH a Ult in o in1 ' uV S? Q 1 0 5* PS f rS -- rj • fxl <

H \n o i q! In! 1 a O o o - 109 n 0 i n •v* — - - - 1 — 0" Q K O m l •u £ *^> Q-< r- c*> "V • £ • 0r >a rn ^ ^1 n\ ~i>r 5 r 3 "=* J! u V V I 0 u < o £ » j o~- ■^ <3- "1 Is- s \M-: 1^. it- _ 0 sa V * jj Vi H - I1 vr - » 1 ^ > ^ Cr Do 4 o m m m m <• * o! H ^ - 1± r 1- < ri si H n: ~c N; gj "<" "< i j « ™ in 1 i _ 1 v< H vi cr J-4 «i *<; - ^1 " «i ^ £ 0 «ij : ^ »T N" VI 4 try «T vf VT - WT a- ^r t* uy - - o u. te • ^ ^ . » • O to - ■ NT - or ^ - - - c s D in fr oj a - - o CO 0" t1 : o llll < X UJ Q. R 6 - • ■_ H ^b K H 1 ! . - -1 ^ - :4: ^ <- _4i i " O 6 j . q. i uv i vnj ; q, ; o, 1 o; o a n < ^ • cvj. H o; i - - - ; 0 1 p ^ t > W. Im! \h iVy |v&. '."- 1 "I 110 « 1 *2 = o • 0 o >3~ X £ 3 5 C^ 1 Q> - — ^5 ^ ^ 3i ^ -5 *l - o 0 M 1 ■ j «oj q eg : Cc * i s « < •i i 1 .' * J "J ** . .• . 1 j >S! j Q~ — 3 " v3 cr ■ — - f i C 5 V) • $ ■•.17 ^ j^j _ ^» r- T*T fn >b $ CV, •+• K ^ -i bi -i $ ~ _ V£ 1 ns " § ^S - « » 2 DC O LL t - *sfl j s= 1 w\i i o^ 1H *v - O Q <" O" -' 1 cy In 1 vs: " " s= •* f*sj ui _j»r p n5 "■ *<" -j •a : tjj i - s3 « tX3 ^ - • i o: : 3 Jl ll - - > 1? in t \-^ O-J 1 -< >nj is 1°° v> r^. : 1 O: Q. 10 ! 0j 1 Q; 1 J ] 0, ■ _ sac ^0 . O. '*n |0 j Vn | Q| hn| £- ^ i^j ja ja ui jsy ^ 5j ^ - - ■^1 ■ K -»i 1 ^i 1 13 ^ _ * 01 < - if - BH j nil i : i - 0 ft - • - - - o ■ Q ■ " i s ° _ 3 ■ !8 - •a • O q • - 4 Q d G 1 - - o. i J: q - - - «02 H ut. 4 0 0 »n *l. , in, J j Oo -5^ 00 - - - *••; 1 »1| 1 iO| 1 *• i An: 1 sSi M (/S. <5^ 0 I oi I oi i n Ci <) O CJ O 112 <-? H w- 31 < a « 5 J4 ico |Ol o " Q. „ .. cc • - M Q- i-1 ; cc uu I IS ,0 Is 0 % '4 I c 8!! v4, to 0 1 |N| -j *b -1 t-i is 0, J UU l^j n j*ti ^ 1 jvJ N" i 1 «J] -i lot : r4] ^ «>^> in. 6of 106* 21 -a * .1 ^ H H j «i| s 1 cr *5 K ^ c<\ 4^ cxy prv t^. •" 1 3 3^ 1 ff j vS1 14 -1 -1 o- 1 -S" ] 0 ■ <4 1 N K7 01 1 j kj o; N , 1 1 r4 -1 ssr 1 I -2X1 ^r : *: _01 TTT ' I-' 1 'JSL :V !n$. -S: -I -\'- Si v Wl ' £ = J_i i .^o: 1 & 1 ^ a J^ h55 ^ ^ h | 01 rs •si : j lo, . "So -1 - -i i i - "gt"Tnh: to i (0)i Oo - ^ - ^ 0 ? * * ci 0 ol 1 «s1 " i\ ' - _ t * i- < D 3 d 1 cr - d ^oi i^ ics 3 - 1 cr - 01 Ql _ oL _ c^ 1 rtj' <> 1 <£ ~ cr a] - »0j - a - - ■i * IDC o 4 crl i s js : - cii o 1 o - — — 6 «>3 0 3 z : : 01 ^ «ti V) J <3> \ 0_j J n3h j i^ bo 0 s o ^ 1) «> j oj j o_ 1 S ^ <4J _ >t 0. ; 111. D o o! M ml" i i ;£; i 2 J oJ .. ■■ or - " 1* * fl .1 1 * >^ N «c 1 Hffl : cv 1 QJ - iliil: cy - 3 a ■5) : 6\ - _ J Q 8; ^ In 1 "< ~ 1 H - s 1 "^ " % # K ^ 1 j. - 1 1 : -i ■ -i 1 ; - 0 E ■ j If) vS is ^«t : <-3L J Q, _ ^ 1 1 1 >*4 ^^A- • «r 1 ^r d si > q o s \i ci 1 cri j oJ jo; : o: 1 : § :s ra :$ 7f - -i 1 - i 3 : •v : : - 8 • ; If q o; •d i+1 ' - t 0^ Q~- -a «*n . t~ j —j . ^ -a , O % fc* <** J V\ ^ i s ; ^: 1 n; ; - ; h ; -: ; i : « ia r CO jj -~- -<: 1} S Oo 0-» Oo Q ' C\( fT) ' O' • 0 I 1 Vi_ .. Q O O O >» ' »sj W N ' O K 1 ^ v\. . >a. r*j .^, <5- 0 • Q . Q. . o o. lol id .o. 1 1'. 116 LIST OF REFERENCES Aleem, A. A., Measurement of Plankton Populations by Tri phenyl -tetrazoli urn Chloride, Kieler Meeresforsch. , v. 11, p. 160-173, 1955. Andrews, R. S., Personal communication. Atlantic Oceanographic Laboratory Report No. BI-R-73-14, A Seagoing Sys- tem for the Measurement of Particulate Carbon, Hydrogen, and Nitrogen, by G. T. Hagell and R. Pocklington, p. 1-7, November 1973. Balch, N. , ATP Content of caianus finmarchicus, Limnology and Oceanography, p. 906-908, 1973. Baugh, D. E., RNA/DNA Ratios in the Estimation of Growth Stages of Oceanic Zooplankton Populations, M.S. Thesis, Naval Postgraduate School, 1974. Beers, J. R. , Studies on the Chemical Composition of the Major Zooplankton Groups in the Sargasso Sea Off Bermuda, Limnology and Oceanography, v. 11, p 520-528, October, 1966. Beers, J. R. and Stewart, G. L., Numerical Abundance and Estimated Bio- mass of Microzooplankton (Part VI of "The Ecology of the Plankton Off La Jolla, California, in the Period April through September, 1967"), Bulletin of the Scripps Institution of Oceanography, U. of California, San Diego, La Jolla, California, v. 17, p. 67-87, 1970. Blackburn, M., Laurs, R. M. , Owen, R. W. , and Zeitzschel , "Seasonal and Area! Changes in Standing Stocks of Phytoplankton, Zooplankton and Micro- nekton in the Eastern Tropical Pacific", Marine Biology, v. 7, p. 31, 1970. Carver, R. E. , Procedures In Sedimentary Petrology, p. 580-587, Wiley- Interscience, 1971. Curl, H., Jr., Analyses of Carbon in Marine Organisms, Journal of Marine Research, v. 20, p. 181-188, 1962. 1 C u s h i n g , D . H . , The Seasonal Variation in Ocean Production as a Problem in Population Dynamics, Journal de Conseil, v. 24, p. 455-464, 1959. Cushing, D. H. and Tungate, D. S. , Studies on a caianus Patch - The Iden- tification of a caianus Patch, Journal of Marine Biology, v. 43, p. 327- 337, 1963. Eg! off, D. A., Ecological Aspects of Sex Ratios and Reproduction in Experi- mental and Field Popul ations of the Marine Copepod Tigriopus caiifomicus, Ph.D. Thesis, Stanford University, August, 1966. Ewing, G. W. , Instrumental Methods of Chemical Analysis, McGraw-Hill, p. 459, 1969. 117 Gordon, D. C, Jr., and Sutcliffe, W. H., A New Dry Combustion Method for the Simultaneous Determination of Total Organic Carbon and Nitrogen in Seawater, Marine Chemistry, v. 1 , p. 231-244, 1973. Heibron, Sir Ian (Editor), Dictionary of Organic Compounds, Oxford Univ- ersity Press, v. I, p. 408, 1946. Hodgman, C. D. , Handbook of Chemistry and Physics, 39th ed., Chemical Rubber Publishing Co., 1957. Holm-Hansen, 0. , ATP Levels in Algae Cells as Influenced by Environmental Conditions, Plant and Cell Physiology, v. 11, p. 689-700, 1970. Holm-Hansen, 0. and Booth, C. R. , The Measurement of Adenosine Triphos- phate in the Ocean and Its Ecological Significance, Limnology and Ocean- ography, v. 11, p. 510-519, 1966. Krey, J. , Chemical Determinations of Net Plankton, with Special Reference to Equivalent Albumin Content, Journal of Marine Research, v. 17, p. 312- 324, 1958. Margelef, R. , Some Concepts Relative to the Organization of Plankton, Oceanography and Marine Biology Annual Review, v. 5, p. 257-271, 1967. Masterton, W. L. and Slowinski, E. J., Mathematical Preparation for Gen- eral Chemistry, W. B. Saunders Company, p. 163-170, 1970. Menzel , D. W. , Particulate Organic Carbon in the Deep Sea, Deep Sea Research, v. 14, p. 229-238. Menzel, D. W. and Vaccaro, R. F. , The Measurement of Dissolved and Parti- culate Carbon in Seawater, Limnology and Oceanography, v. 9, p. 138-142, 1964. Morris, A. W. and Foster, P., The Seasonal Variation of Dissolved Organic Carbon in the Inshore Waters of the Menai Strait in Relation to Primary Production, Limnology and Oceanography, v. 16, p. 987-989, 1971. Mull in, M. M. , The Production of Zooplankton in the Ocean: The Present Status and Problems, Oceanography and Marine Biology Annual Review, v. 1 , p. 293-315, 1969. Nakai, Z. and Hon jo, K. , Comparative Studies on Measurements of the Weight and the Volume of Plankton Samples, Proc. 9th Session Indo-Pacific Fisheries Council, Section II, p. 9-16, 1962. National Academy of Sciences, Recommended Procedures for Measuring the Productivity of Pla nkton _S tanding S tock and Related Oceanic Properties , prepared by the Biological" Methods "Panel Committee on Oceanography Divi- sion of Earth Sciences National Research Council, 1969. Piatt, T. L., Dickie, L.M., and Trites, R. W. , Caloric, and Carbon Equi- valents of Zooplankton Biomass, Journal of I isheries Research Board of Canada, v. 26, p. 1453-1473, 1970. 118 Pouchert, C. J., The Aldrich Library of Infrared Spectra, Aldrich Chemical Co., Inc., p. 179, 1970. Rowel 1, C. F., Personal communication. Sharp, J. H., Size Classes of Organic Carbon, Limnology and Oceanography, v. 18, p. 441-446, May 1973. Sharp, J. H., Total Organic Carbon in Seawater - Comparison of Measure- ments Using Persulfate Oxidation and High Temperature Combustion, Marine Chemistry, v. 1, p. 211-229, March 1973. Stoker, H. S. and Seager, S. L. , Environmental Chemistry: Air and Water Pollution, Scott, Foresman and Company, p. 25-27, 1972. Strickland, J. D. and Parsons, T. R. , A Practical Handbook of Seawater Analysis, Fisheries Research Board of Canada Bulletin, v. 167, 331 pp. Sutcliffe, W. H., Jr., Growth Estimates from Ribonucleic Acid Content in Some Small Organisms, Limnology and Oceanography, v. 10 (Suppl.) R 253, 1965. Traganza, Office of Naval Research Project: Inorganic Chemical Nutrients and Volume Reverberation Limitations in the Ocean, Part I, Biochemical Relationships of Secondary Biomass and Dissolved Nutrients, 1974 Deleted Urick, R. J., Principles of Underwater Sound, McGraw-Hill, 1967. Van Hall, C. E., Safranko, J. and Stenger, V. A., Rapid Combustion Method for the Determination of Organic Substances in Aqueous Solutions, Analytical Chemistry, v. 35, p. 315-319, 1963. Wangersky, P. J., The Organic Chemistry of Seawater, American Science, v. 53, p. 385-374, 1965. Willi ams , P . J . , The Wet Oxidation of Organic Matter in Seawater, Limnology and Oceanography, v. 14, p. 292-297, 1969. Williams, P. M. , Sea-Surface Chemistry: Organic Carbon and Organic and Inorganic Nitrogen and^hojghcjn^nJSj^ Fi1ms and Subsurface Waters, Deep Sea Research, v. 14, p. 791-800, 1967. Williams, R. M. The Determination of Dissolved Organic Carbon in Seawater: A_Comp_arison of the Two Methods, Limnology and Oceanography, v. 14, p. 297-298, 1969. Williams, P. M. and Gordon, L. I., Carbon-13: Carbon-12 Ratios in DIssoJK and Particulate Organic Matter in the Sea, Deep-Sea Research, v. 1, p iy-'t>, 1970. Wilson, R. F., Measurement of Organic Carbon in Seawater, Limnology and Oceanography, v. 6, p. 259-261, 1961. 119 INITIAL DISTRIBUTION LIST 1. Defense Documentation Center Cameron Station Alexandria, Virginia 22314 2. Library, Code 0212 Naval Postgraduate School Monterey, California 93940 3. Dr. Eugene D. Traganza, Code 58Tg Naval Postgraduate School Monterey, California 93940 4. ENS J. C. Radney, USN 493 West "J" Street, Apt L Benicia, California 94510 5. Department of Oceanography, Code 58 Naval Postgraduate School Monterey, California 93940 6. Oceanographer of the Navy Hoffman Building No. 2 200 Stovall Alexandria, Virginia 22332 7. Office of Naval Research Code 480 Arlington, Virginia 22217 8. Dr. Robert E. Stevenson Scientific Liaison Office, 0NR Scripps Institution of Oceanography La Jolla, California 92037 9. Library, Code 3330 Naval Oceanographic Office Washington, D.C. 20373 10. SIO Library University of California, San Diego P. 0. Box 2367 La Jolla, California 11. Department of Oceanography Library University of Washington Seattle, Washington 98105 120 12. Department of Oceanography Library 1 Oregon State University Corvallis, Oregon 97331 13. Fleet Numerical Weather Central 1 Monterey, California 93940 14. Environmental Prediction Research Facility 1 Monterey, California 93940 15. Department of the Navy 1 Commander Oceanographic System Pacific Box 1390 FPO San Francisco 96610 121 - 2 2 *» 9 5 'Thesis fil37 Radney c.l Measurement r.* to» ^-omass by0cfJb00°P,anfc n3 yc!. , y carbon a- eFs. atter'"9 mod- 2 2 4 9 5 Thesis R137 c.l 1 rn Radney Measurement of zooplank- ton biomass by carbon a- nalysis for application in sound scattering mod- els. U 9 thesR137 Measurement of zooplankton biomass by ca 3 2768 002 05248 2 DUDLEY KNOX LIBRARY