1. CRUISE NARRATIVE: A16S (Last update: 2004.09.24) 1.1 HIGHLIGHTS WHP CRUISE SUMMARY INFORMATION WOCE section designation A16S Expedition designation (EXPOCODE) 318MSAVE5 Chief Scientists/affiliations DR. WILLIAM A. SMETHIE / LDEO DR. MICHAEL MCCARTNEY/WHOI Cruise dates 1989.JAN.23 - 1989.MAR.08 Ship R/V MELVILLE Ports of call Cape Town, South Africa to Montevideo, Uruguay Number of stations 73 CTD/Rosette (13 Large Volume) 32°16.70' S Stations' geographic boundaries 50°15.60' W 00°59.10' E 53°58.10' S Floats and drifters deployed 0 Moorings deployed or recovered 0 CHIEF SCIENTISTS' CONTACT INFO. Dr. William A. Smethie • Lamont-Doherty Earth Observatory Columbia University • Palisades NY 10964 TEL: 914-359-2900 x566 • FAX: 914-365-8157 EMAIL: bsmeth@ldeo.columbia.edu Dr. Michael McCartney • Dept. of Physical Oceanography, Clark 344 Woods Hole Oceanographic Institution • Woods Hole MA 02543 TEL: 508-289-2797 • FAX: 508-457-2181 EMAIL: mike@gaff.whoi.edu SOUTH ATLANTIC VENTILATION EXPERIMENT (SAVE) Leg 5 R/V MELVILLE 23 January 1989 - 8 March 1989 Cape Town, South Africa to Montevideo, Uruguay Data Report Prepared by: Oceanographic Data Facility Scripps Institution of Oceanography University of California, San Diego April 1992 Sponsored by National Science Foundation ODF Publication No. 232 Grand OCE-86 13330 SIO Reference 92-10 CHIEF SCIENTIST'S OVERVIEW Leg 5 of SAVE (denoted Hydros 3 by Scripps, and A16C for the WOCE project) was carried out on the R/V Melville and was actually a combination of a meridional section across the Argentine Basin at 41°W and a short section perpendicular to the western boundary at about 35°S planned as part of SAVE, and the southern end of a meridional. long line in the Atlantic Ocean planned by M. McCartney, L. Talley and M. Tsuchiya that extended from 32.5°S, 25°W to South Georgia Island The cruise departed from Cape Town on January 23,1989 at 1400. The first station, a reoccupation of AJAX station 44 at 36°S, 1°E, was taken on January 26. Work began on the 25°W line on February I with station 237 taken at 32°30' S, 25°00' W. The southernmost station (278) just north of South Georgia Island was taken on February 17. The station spacing was about 30 na. mi. along this line. The section along 41°W (stations 279-304) was occupied between February 18 and March 2 with a station spacing of about 65 na. mi. The portion of the western boundary section outside Uruguay's 200 mile zone (stations 305-308) was taken between March 4 and March 6. Stations inside Uruguay's 200 mile zone were taken on the following leg when an Uruguayan observer was on board. The cruise ended in Montevideo on March 8, one day later than scheduled. The one day delay was caused by a breakdown of the main engine, which was repaired at sea. The total number of miles steamed was 5989 and the total number of CTD- rosette stations occupied was 73. These stations were taken using a Neil Brown CTD interfaced to a Scripps Ocean Data Facility rosette equipped with thirty-six 10 liter bottles. At 13 of the stations large volume samples were collected using Gerard barrels. Generally 18 large volume samples were collected between the surface and the bottom. The total steaming time for the cruise was 25 days, 15 hours and the total station time was 17 days, 5 hours. Originally, 81 stations had been planned, but 8 stations were dropped because of time lost during rough weather. Approximately 1.25 days were lost because of slow steaming and greater time required to complete casts in rough weather and an equal amount of time us lost because of equipment failures during rough weather. Most of the equipment failures were Gerard barrels not tripping properly and breaks in the CTD termination.1. LIST OF PARTICIPANTS Ship's Captain | Curtis D. Johnson | Scripps Institution of | | Oceanography -------------------|-------------------------|-------------------------- Chief Scientist | William M. Smethie, Jr. | Lamont-Doherty Geological | | Observatory -------------------|-------------------------|-------------------------- Co-chief Scientist | Michael S. McCartney | Woods Hole Oceanographic | | Institution IFREMER, Brest, France Kevin G. Speer Lamont-Doherty Geological Observatory Michael T. Benjamin Richard P. Cember Mieczyslawa Klas Kathy A. Tedesco Massachusetts Institute of Technology Stephen G. Zemba Physics Institute of Bern, Switzerland Jose M. Rodriguez Princeton University Richard J. Rotter Scripps Institution of Oceanography/ODF David L. Bos Carol Conway Frank M. Delahoyde Arthur W. Hester John K. Jain Leonard T. Lopez Douglas Masten David A. Muus Scripps Institution of Oceanography George H. Bouchard Ronald L. Comer Frederick A. Van Woy Texas A&M University Bret B. Bergland Woods Hole Oceanographic Institution Danuta Kaminski Jan C. Zemba Leg | STATIONS | DEPARTURE | ARRIVAL ---------|-----------|-------------------------|------------------------ SAVE 1 | 1 - 43 | Recife, Brazil | Abidjan, Ivory Coast | | 23 November 1987 | 13 December 1987 ---------|-----------|-------------------------|------------------------ SAVE 2 | 44 - 105 | Abidjan, Ivory Coast | Rio de Janeiro, Brazil | | 18 December 1987 | 23 January 1988 ---------|-----------|-------------------------|------------------------ SAVE 3 | 106 - 170 | Rio de Janeiro, Brazil | Abidjan, Ivory Coast | | 29 January 1988 | 07 March 1988 ---------|-----------|-------------------------|------------------------ SAVE 4 | 171 - 235 | Punta Arenas, Chile | Cape Town, South Africa | | 07 December 1988 | 15 January 1989 ---------|-----------|-------------------------|------------------------ SAVE 5 | 236 - 308 | Cape Town, South Africa | Montevideo, Uruguay | | 23 January 1989 | 08 March 1989 ---------|-----------|-------------------------|------------------------ HYDROS-4 | 309 - 379 | Montevideo, Uruguay | Bridgetown, Barbados | | 13 March 1989 | 19 April 1989 INTRODUCTION The broad objectives of the South Atlantic Ventilation Experiment (SAVE) are to investigate the rates of ocean circulation, mixing, ventilation, inter-ocean exchange, and carbon, oxygen and nutrient cycling on an ocean basin scale. Transient and radioactive tracers are well suited for pursuing these objectives because they integrate over various spatial and temporal scales, providing strong constraints on time-averaged rates of circulation and mixing, and revealing pathways between the air/sea interface and the abyssal ocean that are not always discernible from traditional hydrographic and nutrient observations. The results should lead to a better understanding of the ocean's major role in the earth's climate, via its control of the global heat and carbon dioxide budgets. The specific problems that will be addressed include: 1) Exchange of South Atlantic waters with the North Atlantic, Pacific, Antarctic and Indian Oceans, 2) Formation and spreading rates of intermediate waters, 3) Thermocline ventilation rates and locations, 4) Rates of biologically-mediated metabolic processes, 5) Diapycnic processes, 6) Origins of deep and bottom waters and their spreading routes and rates. Underway/XBT Program XBT casts were usually taken between CTD stations. Where the station spacing was 30nm, one XBT was taken and for greater spacing 2 XBTs were taken. The XBT casts were accompanied by water samples for dissolved oxygen, pCO2, salinity and nutrients taken from the Melville's clean seawater line. Bathymetry Depth was measured continuously along the entire cruise track using a 12KHz precision depth recorder. The raw data was digitized by hand at 5- minute intervals and a tape of the reduced data will be produced by the Geological Data Center at Scripps. A16S (SAVE 5) • Smethie/McCartney • R/V Melville • 1989 __________________________________________________________________________ __________________________________________________________________________ 2. BOTTLE DATA COLLECTION, ANALYSES, AND PROCESSING ODF CTD/rosette casts were carried out with a 36-bottle rosette sampler of ODF manufacture using a General Oceanics pylon. An ODF-modified NBIS Mark 3 CTD and a SeaTech transmissometer provided by Texas A&M University (TAMU) were mounted on the rosette frame. Seawater samples were collected in 10-liter PVC Niskin bottles mounted on the rosette frame. These bottles were numbered 1 through 36, if one of these 36 bottles needed servicing and repairs could not be accomplished by the next cast, the replacement bottle was numbered 71-78. Added CTD levels, no water samples, were assigned bottle numbers 95-99. Subsets of CTD data taken at the time of water sample collection (a 10 second average) were transmitted to the bottle data files immediately after each cast to provide pressure and temperature at the sampling depth, and to facilitate the examination and quality control of the bottle data as the laboratory analyses were completed. After each rosette cast was brought on board, water samples were drawn in the following order: Freon, Helium-3, Oxygen, Oxygen-18, pCO2. …CO2, Tritium, Nutrients, Salinity and Suspended Particulate Matter. Other ancillary program samples were drawn after the core samples. The samples and the Niskin sampler they were drawn from were recorded on the Sample Log sheet. Comments regarding validity of the water sample (valve open, lanyard caught in lid, etc.) were also noted on the Sample Log sheets. Table 1 is a tabulation of samples collected during all six legs [SAVE Legs 1 through 5 (STS/ODF, 1992) and HYDROS Leg 4 (STS/ODF, 1992)] unless otherwise noted and includes the Principal Investigators and their institutions. Samples collected on each station are tabulated in the Station and Cast Description. Gerard casts were carried out with -270 liter stainless steel Gerard barrels on which were mounted 2-liter Niskin bottles with reversing thermometers. Samples for salinity, 14C, 228Ra, 39Ar, and 85Kr Were obtained from the Gerard barrels. The Gerard barrels were numbered 81 through 91 and the piggy-back Niskin were numbered 41 through 70. Surface samples taken from the ship's underway pump line were assigned a bottle number of 98 through 99. Salinity check samples were always drawn from the Niskin bottles for comparison with the Gerard barrel salinities to try to verify the integrity of the Gerard sample. Occasionally, barium and some of the samples normally taken from the rosette were also drawn from the Gerard-mounted Niskin bottle. These were also recorded on a Sample Log sheet. The discrete hydrographic data were entered into the shipboard data system and processed as the analyses were completed. The bottle data were brought to a useable, though perhaps not final, state at sea. ODF data checking procedures included verification that the sample was assigned to the correct level. This was accomplished by checking the raw data sheets, which included the raw data value and the water sample bottle, versus the sample log sheets. Any comments regarding the water samples were investigated. The raw data computer files were also checked for entry errors. Investigation of data included comparison of bottle salinity and oxygen with CTD data, and review of data plots of the station profile alone and compared to nearby stations. If a data value did not either agree satisfactorily with the CTD or with other nearby data, then analyst and sampling notes, plots, and nearby data were reviewed. If any problem was indicated the data value was flagged or deleted. (However, ODF preserves in its archives all bottle data values). The Bottle Data Processing Notes section includes comments regarding deletion of samples. If it was determined that an entire 10 liter water sample was contaminated by leakage or other bottle or rosette malfunction, the level is reported with just the CTD data (pressure, temperature and salinity). This has been done to preserve the profile and accommodate investigators who prefer using bottle data files exclusively. # Of || # Of Parameter Samples || Parameter Samples ----------------- ------- || --------------------- ------- discrete salinity 3179 || pCO2 845 discrete temp. 546 || suspended part. 369 oxygen 2530 || Keeling C02 Neodynium 20 nutrients 2517 || Barium 280 CFMs 1592 || 14C 187 Helium-3 346 || Ra-228 237 Tritium 346 || Kr-85 87 TCO2 1040 || Ar-39 13 2.1. PRESSURE AND TEMPERATURES All pressures and temperatures for the Niskin bottle data tabulations on the rosette casts were extracted from the processed CTD data, usually those from the corrected 10-second average bottle trip files collected during the up cast (see CTD DATA COLLECTION, ANALYSES, AND PROCESSING). Gerard pressures and temperatures were calculated from Deep-Sea Reversing Thermometer (DSRT) readings. Each DSRT rack normally held 2 protected (temperature) thermometers and I unprotected (pressure) thermometer. Thermometers were read by two people, each attempting to read a precision equal to one tenth of the thermometer etching interval. Thus, a thermometer etched at 0.05 degree intervals would be read to the nearest 0.005 degrees. Each temperature value is therefore calculated from the average of four readings. IT SHOULD BE CLEARLY NOTED THAT THE TEMPERATURES PRODUCED AND PUBLISHED BY ODF IN THIS REPORT ARE BASED ON THE INTERNATIONAL PRACTICAL TEMPERATURE SCALE OF 1968, RATHER THAN THE CURRENTLY USED INTERNATIONAL TEMPERATURE SCALE OF 1990. (The expedition took place before I January 1990, the starting date for ITS-90). 2.2. SALINITY Salinity samples were drawn into ODF citrate salinity bottles which were rinsed three times before filling. Salinity was determined after sample equilibration to laboratory temperature, usually within about 8-36 hours of collection. Salinity has been calculated according to the equations of the Practical Salinity Scale of 1978 (UNESCO, 198 1) from the conductivity ratio determined from bottle samples analyzed (minimum of two recorded analyses per sample bottle after flushing) with a Guildline Autosal Model 8400A salinometer standardized against Wormley P- 108 standard seawater, with at least one fresh vial opened per cast, or from the corrected CTD conductivity, temperature, and pressure. Accuracy estimates of bottle salinities run at sea are usually better than 0.002 psu relative to the specified batch of standard. Although laboratory precision of the Autosal can be as small as 0.0002 psu when running replicate samples under ideal conditions, at sea the expected precision is about 0.001 psu under normal conditions, with a stable lab temperature. Still, because a small droplet of fresh water, or the residue from a small evaporated droplet of seawater, can affect a bottle salinity in the third decimal place, and because the Autosal salinometer is sensitive to environmental fluctuations, salinities from bottle samples have a lower true precision under field conditions than in the laboratory. ODF typically deleted the Niskin bottle salinity from this report, and substituted the corrected CTD salinity, whenever there was any question regarding its validity (see BOTTLE DATA PROCESSING NOTES). 2.3. OXYGEN Samples were collected for dissolved oxygen analyses soon after the rosette sampler was brought on board and after CFC and Helium were drawn. Nominal 100 ml volume iodine flasks were rinsed carefully with minimal agitation, then filled via a drawing tube, and allowed to overflow for at least 2 flask volumes. Reagents were added to fix the oxygen before stoppering. The flasks were shaken twice; immediately, and after 20 minutes, to assure thorough dispersion of the Mn(OH)2 precipitate. The samples were analyzed within 4-36 hours. Dissolved oxygen samples were titrated in the volume-calibrated iodine flasks with a I ml microburet, using the whole-bottle Winkler titration following the technique of Carpenter (1965). Standardizations were performed with 0.01N potassium iodate solutions prepared from pre- weighed potassium iodate crystals. Standards were run at the beginning of each session of analyses, which typically included from I to 3 stations. Several standards were made up and compared to assure that the results were reproducible, and to preclude basing the entire cruise on one standard, with the possibility of a weighing error. A correction (- 0.014 ml/l) was made for the amount of oxygen added with the reagents. Combined reagent/seawater blanks were determined to account for oxidizing or reducing materials in the reagents, and for a nominal level of natural iodate (Brewer and Wong, 1974) or other oxidizers/reducers in the seawater. The assay of the finest quality K103 available to ODF is 100%, ±0.05%, but the true limit in the quality of the bottle oxygen data lies in the practical limitations of the present sampling and analytical methodology, from the time the rosette bottle is closed through the calculation of oxygen concentration from titration data. Overall precision within a group of samples has been determined from replicates on numerous occasions, and for the system as employed on this expedition, one may expect ±0.1 to 0.2%. The overall accuracy of the data is estimated to be ±0.5%. TABLE 1: SCIENTIFIC PROGRAMS Participating | Principal | Scientific | Institutions | Investigators | Programs | Leg --------------|-----------------------------|---------------------|---- STS/ODF | Mr. David Wirth | Salinity | ALL | | Oxygen | ALL | | Nitrate, Nitrite, | ALL | | Phosphate, Silicate | ALL | | CTD Profiles | ALL --------------|-----------------------------|---------------------|---- LDGO | Dr. Wallace S. Broecker | Carbon-14 | ALL MIAMI | Dr. H. Göte Ostlund | | MIAMI | Dr. Zafer Top | | --------------|-----------------------------|---------------------|---- WHOI | Dr. William J. Jenkins | Helium-3 | ALL | | Tritium | ALL --------------|-----------------------------|---------------------|---- PRINCE | Dr. Jorge L. Sarmiento | Radium-228 | ALL PRINCE | Dr. Robert M. Key | Radium-226 | ALL SIO | Dr. Ray F. Weiss | Freon-11 | ALL LDGO | Dr. William M. Smethie, Jr. | Freon-12 | ALL LDGO | Dr. William M. Smethie, Jr. | Krypton-85 | ALL --------------|-----------------------------|---------------------|---- BERN | Dr. Heinz Loosli | Argon-39 | ALL LDGO | Dr. William M. Smethie, Jr. | | BERN | Dr. J. H. Oeschger | | --------------|-----------------------------|---------------------|---- LDGO | Dr. Taro Takahashi | Total CO2 | ALL | | pCO2 | --------------|-----------------------------|---------------------|--- LDGO | Dr. Arnold Gordon | XBT Profiles | ALL LDGO | Mr. Stanley Jacobs | | ANCILLARY PROGRAMS Participating | Principal | Scientific Institutions | Investigators | Programs | Leg --------------|------------------------|----------------------------------|----- TAMU | Dr. Wifford Gardner | Suspended Particulate Matter | ALL | | Transmissometer | ALL --------------|------------------------|----------------------------------|----- SIO | Dr. Charles D. Keeling | Total CO2 | ALL SIO/GDC | Mr. Stuart M. Smith | Bathymetry | ALL LDGO | Dr. James K. B. Bishop | Barium | ALL --------------|------------------------|----------------------------------|----- SIO | Dr. Ray F. Weiss | Underway pN20, pCO2, pCH4 | ALL | | Underway Surface Measurements | ALL --------------|------------------------|----------------------------------|----- LDGO | Dr. Andrew McIntyre | Coccolith | 1 URI | Dr. Michael Bender | Oxygen-18 | 2 LDGO | Dr. K. K. Liu | Nitrogen-15 | 2 --------------|------------------------|----------------------------------|----- UH | Dr. Chris Measures | Beryllium Isotopes | 3 | | Aluminum | 3 --------------|------------------------|----------------------------------|----- CNES/GRGS | Dr. Catherine Jeandel | Neodymium |4 & 5 SIO | Dr. Martin Whelen | 13C, 14C in methane (air samples)| 5 INSTITUTION CODES: ----------------------------------------------------------------------- BERN Physics Institute of Bern, Switzerland CNES/GRGS Observatorie of Midi-pyrenees, Toulouse, France LDGO Lamont-Doherty Geological Observatory of Columbia University MIAMI University of Miami PRINCE Princeton University SIO Scripps Institution of Oceanography SIO/GDC Scripps Institution of Oceanography/Geological Data Center STS/ODF Shipboard Technical Support/Oceanographic Data Facility TAMU Texas A & M UH University of Hawaii URI University of Rhode Island WHOI Woods Hole Oceanographic Institution Oxygens were converted from milliliters per liter to micromoles per kilogram using the equation: O2[µm/kg]=O2[ml/l]/(.022392*(1.0+sigma theta/1000.0)) The potential density anomaly, sigma theta, is the potential density in kg/m3 referenced to pressure=0, from which 1000 has been subtracted. 2.4. NUTRIENTS Nutrients (phosphate, silicate, nitrate and nitrite) analyses, reported in micromoles/kilogram, were performed on a Technicon AutoAnalyzer(r). The procedures used are described in Hager et al. (1972) and Atlas et al. (1971). Standardizations were performed with solutions prepared aboard ship from pre-weighed standards; these solutions were used as working standards before and after each cast (approximately 36 samples) to correct for instrumental drift during analyses. Sets of 4-6 different concentrations of shipboard standards were analyzed periodically to determine the linearity of colorimeter response and the resulting correction factors. Phosphate was analyzed using hydrazine reduction of phosphomolybdic acid as described by Bernhardt & Wilhelms (1967). Silicate was analyzed using stannous chloride reduction of silicomolybdic acid. Nitrite was analyzed using diazotization and coupling to form dye; nitrate was reduced by copperized cadmium and then analyzed as nitrite. These three analyses use the methods of Armstrong et al. (1967). Sampling for nutrients followed that for the tracer gases, CFCs, He, Tritium, and dissolved oxygen. Samples were drawn into -45 cc high density polyethylene, narrow mouth, screw-capped bottles which were rinsed twice before filling. The samples may have been refrigerated at 2 to 6°C for a maximum of 15 hours. Nutrients were converted from micromoles per liter to micromoles per kilogram by dividing by sample density calculated at an assumed laboratory temperature of 25°C. A16S (SAVE 5) • Smethie/McCartney • R/V Melville • 1989 __________________________________________________________________________ __________________________________________________________________________ 3. CTD DATA COLLECTION, ANALYSES, AND PROCESSING Year-2 of SAVE (Legs 4 and 5) were processed with Hydros-4 therefore there may be references made to Hydros-4. 209 CTD casts were completed using a 36-bottle rosette sampling system; 64 casts were completed on Save Leg 4, 74 on Save Leg 5 one of which was a test and not reported in this report, and 71 casts on Hydros-4. STS/ODF CTDs #1 and #2, modified NBIS Mark IIl-B instruments, were both used during SAVE-4; CTD-1 was used exclusively on SAVE-5 and Hydros-4. The CTD data were initially processed into a filtered, half-second average time-series during the data acquisition. The pressure and PRT temperature channels were corrected using laboratory calibrations. The conductivity/salinity channels were calibrated to salinity check samples acquired on each cast. The CTD time-series data were then pressure-sequenced into 2- decibar pressure intervals, and the pressure series oxygen channel was corrected to match oxygen check samples acquired on each upcast. 3.1. CTD LABORATORY CALIBRATIONS 3.1.1. PRESSURE TRANSDUCER CALIBRATION Each CTD pressure transducer was calibrated in a temperature-controlled bath by comparison with pressures generated by a Ruska Model 2400 piston gage. The mechanical hysteresis loading and unloading curves were measured both pre- and post-cruise at cold temperature (-I to 0.5°C bath) to a maximum of 8830 psi, and at warm temperature (28-30°C bath) to a maximum of 2030 psi. 3.1.2. PRT TEMPERATURE CALIBRATION The CTD-1 and all CTD-2 PRT temperature sensors were calibrated in a temperature-controlled bath by comparison with temperatures calculated from the resistance of a Rosemount Model 162CE standard platinum thermometer, measured by a NBIS model ATB 1250 resistance bridge. The Rosemount standard PRT was checked periodically in water and diphenyl ether triple-point cells. Seven or more calibration temperatures, spaced across the range of 0 to 30°C, were measured both pre- and post-cruise. The CTD-2 primary PRT temperature transducer was replaced after SAVE-4, station 185, so the pre- and post-cruise calibrations for that temperature channel are not comparable. 3.2. CTD DATA PROCESSING 3.2.1. CTD DATA ACQUISITION Eight data channels (pressure, temperature, second temperature, conductivity, dissolved oxygen, transmissometer, altimeter and elapsed time) were acquired by CTD-2 at a data rate of 25 Hz. CTD-1 acquired seven cbannels, having only one temperature transducer. The FSK CTD signal was demodulated by an STS/ODF-designed deck unit and output to an RS-232 bus interface. An Integrated Solutions, Inc. (ISI) Optimum V computer served as the real-time data acquisition processor. Data acquisition consisted of storing all raw binary data on hard disk, then on magnetic cartridge tape, and generating a corrected and filtered half-second average time-series. Data calculated from this time series were reported and plotted during the cast. A 10-second average of the time-series data was calculated for each water sample collected during the data acquisition. Generating the half-second time-series data set involved applying single-frame absolute value and gradient filters, then performing a two- pass 4,2 standard-deviation data rejection to all channels. During the acquisition, the pre-cruise laboratory calibration data were applied to pressure and temperature. Pressure, conductivity and oxygen were matched to the thermal response of the PRT temperature transducer. This lag time was determined using raw CTD data from the cruise. The conductivity and oxygen channels were corrected for thermal and pressure effects. 3.2.2. PRESSURE, TEMPERATURE AND CONDUCTIVITY/SALINITY CORRECTIONS A maximum of 36 salinity and oxygen check samples, plus 4 thermometric pressure and temperature measurements, were collected during each CTD cast. A 10-second average of the CTD time-series data was calculated for each trip time sample. The resulting data were then used to verify the pre- and post-cruise pressure and temperature calibrations, and to derive CTD conductivity/salinity and oxygen corrections. 3.2.2.1. CTD PRESSURE CORRECTIONS 3.2.2.1.1. CTD #2 The pre- and post-cruise pressure calibrations were compared. The warm/shallow calibration curve shifted by almost 3 decibars from pre- to post-cruise, and the cold/deep calibration curve shifted by 3.5 decibars. There were no notable slope differences between the two sets of calibrations. Offsets are adjusted automatically for each station as the CTD enters the water, so there were no significant differences between the two calibration results for CTD-2. Since all CTD-2 casts were done during the first of the three Hydros legs, closer to the pre-cruise calibrations, no change was made to the shipboard CTD-2 pressure data, to which the pre-cruise pressure calibrations had already been applied. Thermometric pressures collected during the CTD-2 casts were minimal because of numerous lanyard hangups or other problems. The few acceptable check-pressure differences gave no indication of any problems with the CTD-2 calibrated pressures. 3.2.2.1.2. CTD #1 The laboratory calibration for CTD-1 at the end of Hydros-4 disclosed a pressure problem affecting all three Hydros (SAVE/Year-2) legs: a. Although pre- and post-cruise pressure calibrations matched (to within 0.2 decibars) below 1190 decibars, the 0-decibar calibration point differed by 2.5 decibars (loading) and 3 decibars (unloading) between pre- and post-cruise. b. There was a sudden 6-decibar jump in pressure response between the 1050- and 1190-decibar calibration points on the loading curve during the post-cruise calibration. c. The jump does not appear in the unloading curve, but shallower than the 1400-decibar point, the post-cruise begins to deviate from the pre-cruise calibration to a maximum of 3 decibars at the 0-decibar calibration point. d. The post-cruise pressure calibration was repeated, giving similar results. The pre-cruise calibrations were re-checked, and a smaller/shallower indication of the same problem could be seen on the cold loading curve at 365 decibars. CTD-1 downcasts from all three legs of the cruise were checked for sudden pressure shifts between 800 and 1600 decibars. The CTD-1 half-second time-series downcasts were checked for a ratio of pressure difference to time difference ("r") greater than 2.5 (equivalent to a winch speed of 150 m/min). For each station with a sudden pressure shift visible on Pressure vs. Time plots, there was exactly one area where r was between 4 and 13. All but six of the jumps occurred between 1000 and 1550 decibars, where the temperature ranged from about 2.5 to 4°C. The other six jumps, between 900 and 1000 decibars, occurred during casts at the beginning of SAVE-4 and at the end of Hydros-4 All detectable rapid changes in pressure occurred in 0.6 to 1.5 seconds and were 1.5 to 10.2 decibars in magnitude. 95 out of 192 CTD-1 casts were affected by a pressure jump as described above. There were problems on many sequential casts, then no detectable problem for many more casts in a row. The pressure at which the jump occurred, and the magnitude of the jump, seemed to be random: there was no apparent correlation to water temperature or elapsed time from the cruise start. A "baseline" pressure correction curve was generated by averaging the pre- and post-cruise calibration curves for each of the shallow/warm and deep/cold data sets, then removing the 500- to 1500-decibar distortion of the post-cruise calibrations by using the shape of the more typical pre-cruise curve. This averaged out the surface differences and maintained the deep consistency in the calibrations. This average correction curve was used for all casts in which no pressure problem was detected. Customized pressure corrections were generated for problem casts, as identified by areas of r greater than 3.5. The "baseline" curve described in the previous paragraph was used as the starting point for each cast. The "average" curve was used from the surface to 365 decibars and from the post-"jump" pressure to the bottom. The correction for the beginning pressure of the jump was the post-jump correction value plus the jump size, minus the real pressure change occurring in that time. A straight-line correction was used between 365 decibars down and the beginning pressure of the jump on the cold curve. The average upcast calibration curves were retained for each cast unless the last corrected in-water surface pressure of an upcast was less than -0.6 decibars. In these cases, the upcast correction curve was adjusted by 0 at 365 decibars to a maximum of +1.5 decibars at the surface, which allowed the curve to resemble the shape of the pre-cruise curve in cases where the end-pressure warranted it. The corrections applied were intended to correct the distorted pressures to within 2 decibars to maintain the WOCE CTD salinity standards. It should be noted that other problems may still exist, but are not detectable at this time. CTD-1 pressure data above 1600 decibars should not be considered accurate to better than 2 decibars, while deeper data should be within more typical error ranges. Thermometric pressures collected during the casts were compared with CTD- 1 corrected pressure data. Any differences between the two sets of pressures appeared to be rack-dependent. 3.2.2.2. CTD TEMPERATURE CORRECTIONS 3.2.2.2.1. CTD #2 CTD-2 had two PRTs. PRT-1 was the main temperature sensor and began drifting during station 177, seven stations into SAVE-4. The drifting increased by stations 183-185, where it was finally isolated as a temperature problem rather than a conductivity problem. PRT-2 was a secondary temperature sensor installed to provide a check for the primary PRT; it was used as the primary PRT for stations 177-185 and 210. PRT-2 was not located as near to the conductivity sensor as PRT-1 and therefore generated noisier CTD salinity data, as it was measuring slightly different water and could not be matched properly to the conductivity sensor response. CTD-2 was used for all stations from 171- 185, plus 210 and 218: CTD-2 was re-installed at these last two stations to test the new PRT-1 sensor, to ensure that CTD-2 could be used as a backup if CTD-1 failed later in the expedition. The pre- and post-cruise laboratory temperature transducer calibrations for PRT-1 were not comparable because a new sensor was installed during SAVE-4 PRT-1a and PRT-2 data from the pre-cruise laboratory calibration and from shipboard casts were compared to see if the relative difference between them changed. There was good agreement between these two CTD-2 PRTs, the differences being consistent at calibration, within a given station, and from station to station. No shift in either sensor occurred until station 176, when the difference increased by .002. Since PRT- la began shifting noticeably in the same direction during station 177, an additional +.002 temperature offset was applied to station 176 data, assuming some change may have begun at that time. A new sensor (PRT-1b) was installed in CTD-2 after station 185. The CTD was used/tested during station 210, where PRT-1b also displayed a drifting problem. CTD-2 was worked on again, then tested at station 218: PRT-1b functioned normally and was used as the CTD-1 primary PRT for that cast. PRT-1b and PRT-2 differences from station 218 and the post- cruise calibration were compared. The post-cruise difference was an average -.022 compared to the station 218 difference. PRT-2 shifted by -.013 from pre-cruise to post-cruise laboratory calibrations, which points to a -.035 shift in PRT-1b between station 218 and the post- cruise calibration. Two deep thermometric temperatures from station 218 also support this shift. The post-cruise PRT-1b laboratory calibration curve, with an additional -.035 offset, was applied to station 218 CTD temperature data. A comparison of the pre- and post-cruise laboratory CTD-2/PRT-2 temperature transducer calibrations showed an average -.013°C shift for PRT-2. There was also a small slope difference between the two PRT-2 calibrations. Since CTD-2 was used mostly at the beginning of the expedition and there is no evidence to show that PRT-2 shifted by the last cast it was used for, the pre-cruise PRT-2 laboratory temperature calibration was applied to stations 177-185 and 210 CTD data. The few thermometric temperature differences available for comparison with CTD-2 data support the temperature corrections used. Additional verification of the temperature corrections was obtained by comparing bottle salinity differences vs. CTD values generated from corrected pressure and temperature, plus raw conductivity. There were no sudden shifts in preliminary salinity differences on casts where a different primary PRT was used or when temperature corrections were offset from the laboratory calibrations. 3.2.2.2.2. CTD #1 CTD-1 had a single temperature sensor and was used for stations 186 through 379, except stations 210 and 218. A comparison of the pre- and post-cruise laboratory CTD-1 PRT temperature transducer calibrations showed a shift varying from -.001°C at cold temperatures to -.0017 at warm temperatures. An average of the two laboratory calibrations was applied to the CTD-1 temperature data. Thermometric temperature data from the cruise were compared to the calibrated CTD-1 temperature data. Any offsets in the differences tended to correlate with thermometer rack number, no shifts in CTD-1 temperature data could be detected over the course of the cruise. 3.2.2.3. CTD CONDUCTIVITY CORRECTIONS 3.2.2.3.1. CTD #2 Check-sample conductivities were calculated from the bottle salinities using CTD-2 pressures and temperatures. The differences between sample and CTD conductivities at all pressures were fit to CTD conductivity using a linear least-squares fit. A conductivity correction slope was not warranted at that time. Conductivity differences were then calculated: offsets were determined for each cast and fit to station number. Preliminary offsets were applied to stations 171 through 175 individually, based on conductivity differences at pressures below 100 decibars. Stations 176-185 conductivity differences below 1000 decibars were re-fit, and smoothed offsets were applied to those casts. Stations 210 and 218 conductivity offsets were determined individually based on differences below 1000 decibars. The offsets were manually adjusted to account for discontinuous shifts in the conductivity transducer response, or to insure a consistent deep theta-S relationship from station to station. After applying the conductivity offsets, the need for a conductivity slope as a function of conductivity was reevaluated: no slope was required. 3.2.2.3.2. CTD #1 Check-sample conductivities were calculated from the bottle salinities using CTD pressures and temperatures. The differences between sample and CTD-1 conductivities at all pressures were fit to CTD conductivity using a linear least-squares fit. Values greater than 2 standard deviations from the fit were rejected. The resulting conductivity correction slope was applied to each CTD-1 cast. Conductivity differences were calculated for each cast after applying the preliminary conductivity slope correction. Residual conductivity offsets were then computed for each cast and fit to station number. Smoothed offsets were determined in three groups: stations 186-189 (first-order fit), stations 190-202 (0-order/same offset for all casts) and stations 203-379 (0-order/same offset for all casts). The resulting smoothed offsets were then applied to the data, then conductivity slope as a function of conductivity was re-checked: no change was warranted. Some offsets were manually adjusted to account for discontinuous shifts in the conductivity transducer response, or to insure a consistent deep theta-S relationship from station to station. It was noted that stations 190-202 had numerous casts of inconsistent bottle data values caused by a scum buildup in the Autosal cell. Adding to this problem were CTD-1 conductivity offsetting problems caused by sensor fouling on stations 192 and 202. The Autosal was thoroughly cleaned and checked before station 200 deep salts were run, and the CTD conductivity sensor was thoroughly cleaned both before and after station 202. No effect on conductivity values was noted after opening up CTD-1 two separate times to repair the non-functioning multiplexer channel (after stations 192 and 255). 3.2.2.3.3. BOTTLE VS. CTD CONDUCTIVITY STATISTICAL SUMMARY The Hydros (SAVE/Year-2) calibrated bottle-minus-CTD conductivity differences yield the following statistical results: Mean conductivity Cruise Pressure difference Standard #values leg range(dbars) (bottle-CTD mmho/cm) deviation in mean ------ ------------- -------------------- --------- ------- SAVE-4 all pressures -.00076 .00578 1944 allp (4,2rej) -.00039 .00227 1805 press < 1500 -.00113 .00769 991 press > 1500 -.00034 .00238 952 SAVE-5 all pressures -.00011 .00400 2311 allp (4,2rej) -.00012 .00137 2176 press < 1500 -.00002 .00552 1174 press > 1500 -.00021 .00104 1137 -------------------------------------------------------------- "4,2rei" means a 4,2 standard-deviation rejection filter was applied to the differences before generating the results. 3.2.3. CTD DISSOLVED OXYGEN DATA 3.2.3.1. CTD OXYGEN CORRECTIONS Dissolved oxygen data were acquired using a Sensormedics dissolved oxygen sensor. During the three legs of the cruise, three oxygen sensors were used. The original sensor on CTD #2 was used for stations 171 through 183, then failed during station 184. A replacement sensor did not work on stations 185 and 210, then functioned normally on station 218 following various repairs. The CTD #1 oxygen sensor was used for all other CTD casts except station 279, where a new sensor faded; the original sensor was re-installed. CTD raw oxygen currents were extracted from the downcast pressure-series data (upcast for station 202) at isopycnals corresponding to the upcast check-samples. The differences between CTD and check-sample dissolved oxygens were used to generate coefficients for the sensor model on a station-by-station basis. Bottle oxygen values were weighted as needed to optimize the fitting of CTD oxygen to discrete bottle samples. 3.2.3.2. BOTTLE VS. CTD OXYGEN STATISTICAL SUMMARY After the CTD oxygen fitting was completed, the upcast bottle values were compared to the corrected CTD oxygen values. The bottle-minus-CTD oxygen differences resulted in the statistics below: Mean oxygen Cruise Pressure difference Standard #values leg range(dbars) (bottle-CTD ml/1) deviation in mean ------ ------------- ----------------- --------- ------- SAVE-4 all pressures -.0009 .0813 1883 allp (4,2rej) +.0001 .0391 1740 press < 1500 -.0044 .1095 963 press > 1500 +.0027 .0312 920 SAVE-5 all pressures -.0032 .0780 2361 allp (4,2rej) +.0003 .0382 2193 press < 1500 -.0071 .1036 1196 press > 1500 +.0009 .0356 1165 ------------------------------------------------------------- * 4,2 standard deviation data rejection 3.2.4. ADDITIONAL PROCESSING A software filter was used on a third of the casts to remove larger conductivity or temperature spiking problems. Additionally, oxygen spikes were filtered out of nearly half of the casts. Pressure did not require filtering 0.35% of the time-series data were affected by the filter. Two-thirds of the spiking problems were concentrated in the CTD- 2/PRT-2 casts, where the distance between the secondary PRT and the conductivity sensor resulted in poor signal match-up in high-gradient areas, and station 202, with its fouled-sensor problem. After filtering, the downcast (upcast for station 202) portion of each time-series was pressure-sequenced into 2-decibar pressure intervals. A ship-roll. filter was applied to each cast to disallow pressure reversals. The remaining density inversions in high-gradient regions cannot be accounted for by a mis-match of pressure, temperature and conductivity sensor response. Detailed examination of the raw data shows significant mixing occurring in these areas because of ship roll. The ship-roll filter resulted in a reduction in the amount and size of density inversions. 3.3. GENERAL COMMENTS/PROBLEMS There were 212 CTD rosette casts: three of these (station 178 casts 2 and 3, plus station 328 cast 1) were aborted because of various computer problems, but another CTD cast was done immediately afterward at the same locations. There is one pressure-sequenced CTD data set, to near the ocean floor, for each of 209 stations. The data reported is all from downcasts, excepting station 202, which is an upcast. The effect of the initial sensor fouling on station 202's downcast was not possible to filter out, while the upcast problems were mostly resolvable. The 0-decibar level of some casts was extrapolated using a quadratic fit through the next three deeper levels. Recorded surface values were rejected only when it appeared that the drift was caused by sensors adjusting to the in-water transition; if there were any question that the that the surface values might be real, the original data was reported. Several shipboard time-series data sets had areas of missing or noisy data. These casts were recovered by redigitizing the raw signal from analog tape. A total of 9 data levels were interpolated in 8 casts: the pressures for these interpolated data frames are listed in the CTD Processing Notes as well as other shipboard or processing comments regarding individual casts. During conductivity data calibrations it was noticed that there was an apparent +.002 conductivity offset in the deep data for some stations. This apparent offset appeared in both the downcasts and upcasts for the affected stations. A similar phenomenon has been noticed before in this and other CTDs where the raw conductivity value crosses from 32.768 to 32.767. On some stations, a -.002 shift back appears yet deeper, where the raw conductivity value crosses back over to 32.767 from the other direction. This is a problem specific to certain instruments, and currently there is no correction. It is most noticeable in Atlantic Ocean data because this particular conductivity crossover" value typically occurs in deep water, where salinity is stable over many hundreds of meters. The CTD oxygen sensor often requires several seconds in the water before being wet enough to respond properly; this is manifested as low or high CTD oxygen values at the start of some casts. Flow-dependence problems occur when the lowering rate varies, or when the CTD is stopped, as at the cast bottom or bottle trips, where depletion of oxygen at the sensor causes lower oxygen readings. CTD downcast oxygen data are usually smoother than upcast data because of the more constant lowering rate. Station 202, the only upcast, demonstrates the typical oxygen depletion effect at each bottle stop. Any delays or stops during the downcasts that may have similarly affected the CTD oxygen data are documented in the CTD Processing Notes. There were various winch, wire or rosette problems throughout the cruise. These resulted in occasional stops, pauses or yoyos during casts. Only those problems that may have affected the data continuity have been noted. As mentioned above, these changes in the lowering rate can affect oxygen in particular. A16S (SAVE 5) • Smethie/McCartney • R/V Melville • 1989 __________________________________________________________________________ __________________________________________________________________________ 4. DATA TABLES AND PLOTS 4.1. STATION AND CAST DESCRIPTION Latitudes and longitudes were read from the ship's navigation system. The time (GMT) is the beginning of the CTD up cast or the time the messenger was sent for the Gerard casts. The bottom depth was read from the PGR, and corrected according to Carter Tables (Carter, 1980). The value (in meters) in the column "DAB" (Distance above bottom) was obtained from the PGR, unless the "Comments" column reports altimeter, as was the ocean depth. Note that the maximum sampling depth for the Rosette casts is calculated from the corrected CTD data since the deepest bottle may not have been closed at the deepest CTD sampling point. If less than 36 bottles were sampled for the Rosette casts or 9 barrels for the Gerard casts, the number of samples taken are reported in the comments. The complementary programs are reported in the "Samples" column. 4.2. TABULAR DATA, CTD, ROSETTE AND GERARD Station numbers are consecutive from the beginning to end of the cruise, without interruption. Cast numbers are consecutive at each station, including aborted casts. Meteorological data were collected by the ship's officers and were copied from the Melville's Bridge Log. If more than one CTD cast were done on a station, the deepest cast is reported. The headings in both the CTD data and ROSETTE and GERARD bottle data have been abbreviated to PRESS, TEMP, and O2 for pressure (decibars), temperature (degrees Celsius), and oxygen which is reported in both (milliliters per liter) and (micromoles per kilogram). In the CTD data listings, specific volume anomaly (centiliters/ton) was abbreviated SVA and calculated according to Millero et al. (1980) and Fofonoff et al. (1983), Sound Velocity (meters per second) to SVEL, (Chen and Millero, 1977), Dynamic Height (dynamic meters) to DYN HT, (Sverdrup et al., 1942), Vaisla Frequency (cycles per hour) to VAIS FREQ which uses a subroutine by Bob Millard modified by Lynne Talley to incorporate Gaussian weighting after the formulation of Breck Owens and N. P. Fofonoff. In the bottle data listings, the headings have been abbreviated to SAW NUM and SALT for Sample Number and Salinity (Practical Salinity Units), AOU, PO4, NO3, SiO3, and NO2 for apparent oxygen utilization, phosphate, nitrate, silicate, and nitrite (micromoles/kilogram), respectively. Density anomalies in sigma-notation follow the usual practice; e.g. sigma-theta (or sigma-0) is the potential density in kg/m3 referenced to pressure=O, from which 1000 has been subtracted. Potential temperature, 0, (degrees Celsius) has been calculated according to Fofonoff (1977) and Bryden (1973) and depth (meters) by Saunders (1981) and Mantyla (1982-1983). Throughout the bottle data report alphabetic characters may be found in the tabular data. These characters have the following meaning: D A salinity value, normally from a bottle sample, has been taken from CTD records. G Data value appears to be high or low as compared with the station profile, however as compared with CTD trace the sample appears good. H A pressure or temperature value, normally from CTD records, has been taken from reversing thermometers. L The sample bottle appears to have leaked. This usually refers to Gerard barrels or the piggy-back Niskin and indicates that the samples may be contaminated. P The sampler either pre or post-tripped. This usually refers to Gerard barrels and indicates that the samples may be contaminated. U A data value is suspect, although no obvious reason has been found. Comments and investigation of these values are reported in the appropriate leg's Bottle Data Processing Notes section. Surface samples from the ship's underway pump line were assigned a separate cast number and reported in the data table under the heading GERARD. This was not the caw on Leg 4, these samples may have the rosette data cast number. These surface samples are distinguished by an "H" footnote on the pressure and reported in the GERARD section. 4.3. STATION PLOTS The hydrographic station plots provide a visualization of the data that is not possible from listings. The first few overlay plots represent the entire leg. Then for each station, the upper plots on the even numbered pages are CTD data, the lower two plots are bottle data. The odd numbered pages are single station representation of potential temperature vs. oxygen, phosphate, nitrate and silicate. 5. ACKNOWLEDGEMENTS This data set was acquired under National Science Foundation Grant OCE 86-13330. The assistance provided by Dr. Neil Anderson is gratefully acknowledged. 5. REFERENCES Armstrong, F. A. J., C. R. Steam, and J. D. H. Strickland, 1967. The measurement of upwelling and subsequent biological processes by means of the Technicon AutoAnalyzer and associated equipment, Deep-Sea Research 14,381-389. Atlas, E. L., S. W. Hager, L. 1. Gordon and P. K. Park, 1971. A Practical Manual for Use of the Technicon AutoAnalyzer in Seawater Nutrient Analyses; Revised. Technical Report 215, Reference 71-22. Oregon State University, Department of Oceanography. 49 pp. Bernhardt, H. and A. Wilhelms, 1967. The continuous determination of low level iron, soluble phosphate and total phosphate with the Auto- Analyzer, Technicon Symposia, Volume 1, 385-389. Brewer, P. G. and G. T. F. Wong, 1974. The determination and distribution of iodate in South Atlantic waters. Journal of Marine Research, 32,1:25-36. Bryden, H. L., 1973. New Polynomials for Thermal Expansion, Adiabatic Temperature Gradient, Deep-Sea Research 20, 401-408. Carpenter, J. H., 1965. The Chesapeake Bay Institute technique for the Winkler dissolved oxygen method, Limnology and Oceanography 10, 141-143. Carter, D. J. T., 1980 (Third Edition). Echo-Sounding Correction Tables, Hydrographic Department, Ministry of Defense, Taunton Somerset. Chen, C.-T. and F. J. Millero, 1977. Speed of sound in seawater at high pressures. Journal Acoustical Society of America, Volume 62, No. 5, 1129-1135. Fofonoff, N. P., 1977. Computation of Potential Temperature of Seawater for an Arbitrary Reference Pressure. Deep-Sea Research 24, 489-49 1. Fofonoff, N. P. and R. C. Millard, 1983. Algorithms for Computation of Fundamental Properties of Seawater. UNESCO Report No. 44,15-24. Hager, S. W., E. L. Atlas, L. D. Gordon, A. W. Mantyla, and P K. Park, 1972. A comparison at sea of manual and AutoAnalyzer analyses of phosphate, nitrate, and silicate. Limnology and Oceanography 17 931-937. Lewis, E. L., 1980. The Practical Salinity Scale 1978 and Its Antecedents. IEEE Journal of Oceanographic Engineering, OE-5,3-8. Mantyla, A. W., 1982-1983. Private correspondence. Millero, F. J., C.-T. Chen, A. Bradshaw and K. Schleicher, 1980. A New High Pressure Equation of State for Seawater. Deep-Sea Research 27A, 255-264. Saunders, P. M., 198 1. Practical Conversion of Pressure to Depth. Journal of Physical Oceanography 11, 573-574. STS/ODF, 1992. South Atlantic Ventilation Experiment (SAVE) Chemical, Physical, and CTD Data Report, Legs 1, 2 and 3. SIO Reference #92-9. STS/ODF, 1992. HYDROS Chemical, Physical, and CTD Data Report, Leg 4. SIO Reference #92-12. Sverdrup, H. U., M. W. Johnson, and R. H. Fleming, 1942. The Oceans, Their Physics, Chemistry and General Biology, Prentice-Hall, Inc., Englewood Cliff, NJ. UNESCO, 1981. Background papers and supporting data on the Practical Salinity Scale, 1978. UNESCO Technical Papers in Marine Science, No. 37, 144 p. A16S (SAVE 5) • Smethie/McCartney • R/V Melville • 1989 __________________________________________________________________________ __________________________________________________________________________ CTD PROCESSING NOTES Stn Cst Remarks --- --- ---------------------------------------------------------------- 236 1 ODF-CTD #1 with TAMU transmissometer #100-D 237 1 CTD #1 238 1 CTD #1 239 1 CTD #1 240 1 CTD #1 241 1 CTD #1 242 1 CTD #1 243 3 CTD #1; -.002 psu salinity offset 3748-3980 dbars 244 1 CTD #1 245 1 CTD #1 246 1 CTD #1 247 1 CTD #1; winch problem/stop at 4090 dbars down 248 1 CTD #1; only 11 oxygen samples: added bottles from stations 247/249 for CTD oxygen fitting 249 1 CTD #1 250 1 CTD #1 251 1 CTD #1; 7-minute delay in cast at 2945 dbars down 252 2 CTD #1 253 1 CTD #1; 7-minute delay in cast at 2750 dbars down 254 1 CTD #1; new end termination 255 1 CTD #1; transmissometer cleaned before cast; multiplexed channels (transmissometer altimeter, elapsed-time, trip-confirm) not working: replaced 3 chips in CTD after cast; rosette hit bottom; found 2 kinks in wire 256 1 CTD #1; new end termination, pylon leads crossed: 12 bottles open, no bottle data 510-2500 dbars; CTD oxygen fitting used bottle data from this cast only 257 1 CTD #1; pylon leads switched back 258 1 CTD #1 259 2 CTD #1; pylon problems, top 10 bottles not closed; CTD oxygen fitting used cast 3/Gerard bottle oxygens for top 500 dbars 260 1 CTD #1; 1370-dbar data level interpolated 261 1 CTD #1 262 1 CTD #1 263 1 CTD #1; new end termination 264 1 CTD #1; 1202-dbar data level interpolated 265 1 CTD #1 266 1 CTD #1 267 1 CTD #1 268 1 CTD #1; new end termination following CTD cable repair 269 1 CTD #1; yoyo 3655 dbars back to 3645 dbars on downcast 270 1 CTD #1; -.004 psu salinity dropout 3524-3556 dbars, then residual -.001 psu offset 3556-4906 dbars 271 2 CTD #1; 8-minute delays in cast at 1180 + 1280 dbars, down 272 1 CTD #1; new end termination following CTD flag block replacement; 15-minute delay in cast at 11-15 dbars 273 1 CTD #1; CTD wire rubbing on cheeks of flag block 274 1 CTD #1 275 1 CTD #1 276 1 CTD #1 277 1 CTD #1 278 1 CTD #1 279 1 CTD #1; new CTD oxygen sensor installed/failed: no CTD oxygen data this cast 280 1 CTD #1; original CTD oxygen sensor re-installed 281 1 CTD #1 282 1 CTD #1; 3-/7-minute delays in cast at 1785/1850 dbars, down 283 1 CTD #1 284 1 CTD #1 285 1 CTD #1 286 1 CTD #1 287 1 CTD #1 288 1 CTD #1 289 1 CTD #1 290 1 CTD #1 291 2 CTD #1 292 1 CTD #1 293 1 CTD #1 294 1 CTD #1 295 1 CTD #1 296 1 CTD #1; yoyo 5125 dbars back to 5114 dbars on downcast; rinsed CTD connector cables after cast 297 1 CTD #1 298 2 CTD #1 299 1 CTD #1 300 1 CTD #1 301 1 CTD #1 302 2 CTD #1 303 1 CTD #1 304 1 CTD #1 305 1 CTD #1 306 1 CTD #1; 36-hour delay for engine repairs en route to station 307 1 CTD #1 308 1 CTD #1 A16S (SAVE 5) • Smethie/McCartney • R/V Melville • 1989 __________________________________________________________________________ __________________________________________________________________________ BOTTLE DATA PROCESSING NOTES Remarks for deleted or missing samples and footnoted data from SAVE Leg 5. Uncertainty of data (footnoted data) results in values that may not fit the station profile, but are within the accuracy of the measurement. Investigation of data may include comparison of bottle salinity and oxygen with CTD data, comparison of Niskin/Gerard salinity, review of data plots of station profile and adjoining stations, rereading of charts (i.e., nutrients). CTD data is reported instead of bottle salinity when comments refer to deleted salinity samples or missing for various reasons. Oxygen and nutrient values are referred to in ml/l and µm/l, respectively. On Gerard casts, if there is a comment regarding the Niskin bottle the corresponding Gerard barrel number follows the Niskin number (i.e., 153/183 is Niskin 53 and Gerard 83). STATION 236 104 Nutrients: NO2-"Bubbly" Junk in water. Analyst having trouble with NO2, this sample @41db was affected. 108 Nutrients: NO2-"Bubbly" Junk in water. NO2 missing-had no readable peak @187db. 133 delta-S .004 low at 4618db. Calc ok. CTD trace smooth. No notes. Same value as sample 135. Probably bad draw or run. Delete hydro salt (34.727). STATION 237 1all Salinity: general comment: Autosal drifting - Reruns indicate drift not linear and rerun values too ragged to use. Delete bottle salinities. STATION 240 101 Sample Log: Missing salts and nuts. No salts or nutrients drawn @1db. 132 Hydro O2 @3401db .1 high compared to CTDO. Calc ok. No notes. Delete bottle oxygen (5.66). STATION 241 260/287 Sample Log: Bottom leaker. Bad spring. No other water samples from Niskin#60. Niskin-Gerard S=.014 at 776db. Delete bottle salinity (34.272). STATION 242 124 delta-S .058 low at 2100db. Calc ok. Same value as 123 salt at 1900db. Apparent dupe draw or run. Delete bottle salinity (34.757). 126 delta-S .026 low at 2500db. Calc ok. Same value as 125 salt at 2300db. Apparent dupe draw or run. Delete bottle salinity (34.847). 127 delta-S .017 low at 2700db. Calc ok. Normal CTD gradient. Apparent bad draw or run. Delete bottle salinity (34.869). 134 Salinity: "Dirty bottle" delta-S .019 high at 3949db. Calc ok. Same value as 133 at 3801db. Apparent dupe draw or run. Delete bottle salinity (34.813). STATION 243 192 Sample Log: Fork jumped wire, rest of cast did not trip. Untripped barrels sent down again as cast 2. 260 Sample Log: Bottom leaker. Niskin-Gerard S=.021 low at 2997db. Delete Niskin bottle salinity (34.865). 309 Sample Log: Helium drawn between freon; vent leaks. Oxygen:" stopper 797 switched with flask 798. Deleted discrete O2 until flask 798 recalibrated with stopper 797. Flasks 798 & 797 broken on station 248 when O2 box 9 fell apart, recalibration could not be performed, Delete bottle oxygen (5.33) @300db. 310 Delete O2 until flask 797 recalibrated with stopper 798. Deleted discrete O2 until flask 797 recalibrated with stopper 798. Flasks 755 & 756 broken on station 248 when O2 box 9 fell apart, recalibration could not be performed. Delete bottle oxygen (5.47) @349db. 451/481 Temperature: Thermometer Rack#5 did not flip. No temperature @524db. 454 Sample Log: Didn't close. No salinity @948db. Barium drawn from Gerard. STATION 244 121 Sample Log: O2 flask 756, lid 755. Oxygen: "stoppers switched again" Deleted discrete O2 until flask 756 recalibrated with stopper 755. Flasks 755 & 756 broken on station 248 when O2 box 9 fell apart, re calibration could not be performed. Delete bottle oxygen (4.35) @1394db. 122 Sample Log: O2 flask 755, lid 756. Deleted discrete O2 until flask 755 recalibrated with stopper 756. Flasks 755 & 756 broken on station 248 when O2 box 9 fell apart, recalibration could not be performed. Delete bottle oxygen (4.20) @1498db. STATION 246 120 PO4 missing @1298db. Original and rerun peaks both bad. 132 Sample Log: Missing drain. No water samples @3400db. STATION 248 101 Sample Log: Drain fell apart on opening. Assume means disk came off spigot. Oxygen flasks broken, no oxygen samples @2db. 102 Oxygen flasks broken, no oxygen samples @25db. 103 Oxygen flasks broken, no oxygen samples @54db. 104 Sample Log: Spigot opened while bringing aboard. No gases sampled @77db. Oxygen flasks broken, no oxygen samples @77db. 105 Oxygen flasks broken, no oxygen samples @98db. 106 Oxygen flasks broken, no oxygen samples @152db. 112 Sample Log: Retake helium. Oxygen flasks broken, no oxygen samples @499db. 114 Sample Log: Bottom leaker. delta-S .000 at 697db. Oxygen flasks broken, no oxygen samples. 116 Oxygen flasks broken, no oxygen samples @894db. 118 Oxygen flasks broken, no oxygen samples @1096db. 120 Oxygen flasks broken, no oxygen samples @1293db. 121 Oxygen flasks broken, no oxygen samples @1293db. 122 Oxygen flasks broken, no oxygen samples @1496db. 123 Oxygen flasks broken, no oxygen samples @1596db. 125 Oxygen flasks broken, no oxygen samples @1897db. 126 Oxygen flasks broken, no oxygen samples @2099db. 127 Oxygen flasks broken, no oxygen samples @2300db. 128 Oxygen flasks broken, no oxygen samples @2498db. 129 Oxygen flasks broken, no oxygen samples @270ldb. 131 Oxygen flasks broken, no oxygen samples @3099db. 132 Oxygen flasks broken, no oxygen samples @3299db. 133 Oxygen flasks broken, no oxygen samples @3498db. 134 Oxygen flasks broken, no oxygen samples @3702db. 135 Oxygen flasks broken, no oxygen samples @3899db. 136 Sample Log: Retake helium. Oxygen flasks broken, no oxygen samples @3989db. STATION 249 101-136 Delete all bottle salinities; problem with temperature in salinity lab. STATION 250 104 Sample Log: Drain broken, probably broken on pinger. No water samples @100db. STATION 251 111 delta-S .183 high at 378db. Calc ok. Normal gradient. No notes. Salt value .007 lower than 110 salt value which could indicate possible dupe draw. Other water samples ok. Delete bottle salinity (34.973). 171 Sample Log: Bottom endcap open. No water samples 49db. STATION 252 160 Sample Log: Empty. No salt sample for Gerard 89 check @4352db. Gerard 89 salt agrees with rosette salts and rest of Cast 1 salts. 222 delta-S .051 at 1651db. Calc ok. Normal gradient, no notes. No apparent reason for large difference here. Other water samples ok (02 min). Delete bottle salinity (34.538). 3all Niskin salinities were not drawn. No explanation given, must just have been sampling oversight. STATION 253 116 Nutrients high @897db, same values as 117, assume dupe draw. Delete nutrients (PO4=2.08, NO2=0.00, NO3=31.4, S103=32.1). 130 Discrete O2 appears .1 low at 3196db. Agrees with CTDO, conductivity and temperature feature this level. Footnote "G" indicating sample appears low, but verifies with CTD trace. 136 delta-S .162 low at 4322db. Calc ok. Salt bottle #PP47 on Sample Log & #4277 on Salinity Data Sheet. Assume mixup with sample bottles. Delete bottle salinity (34.843). 256 Sample Log: Hung up, no sample. Gerard 86 salt @802db agrees with other salts this level so Gerard ok. 260 Niskin-Gerard S=.023 at 826db. Calc ok. Gerard 89 salt agrees with other salts this level so assume Niskin leaked and Gerard sample ok. No other samples from Niskin 60. Delete Niskin bottle salinity (34.247). STATION 254 101 Nutrients: Evidently too much junk in water to resolve peaks on NO2. No NO2 @6db. 136 Discrete O2 about .1 high at 4368db. CTDO increasing at bottom but not this much. Only 2 other stations have unexplained high O2 for Niskin #36. Possibly drawer forgot to flush pickling syringes. Delete bottle oxygen (5.16). STATION 256 108 delta-S .012 low at 3897db. Calc ok. Normal gradient. Same value as sample 109. Assume dupe draw. Delete bottle salinity (34.681). 124 No O2 drawn @10db, Keeling CO2 sample. STATION 259 181/152 Sample Log: Too shallow. Pre-tripped at 1631db vs. 3800db intended depth. Niskin-Gerard S=.008 due poor flush Cast not repeated due bad weather. Samples drawn & analyzed. Suggested footnote: "Pretrip, depth uncertain by +/- 50 meters." "P" 182/153 Sample Log: Too shallow, leaking but vent is tight. Pretripped at 2127db vs. 4200db intended depth. Niskin-Gerard S=.024 due poor flush and probable Gerard leak. Unable to open Gerard 82 lid after this cast. Cast not repeated due bad weather. Samples drawn & analyzed. Suggested footnote: "Pretrip, depth uncertain by +/- 100 meters." "P" 183/154 Sample Log: Too shallow. Pretripped at 2489db vs. intended depth of 4440db. Niskin-Gerard S=.009 due poor flush. Cast not repeated due bad weather. Samples drawn & analyzed except no Argon this station. Suggested footnote: "Pretrip, depth uncertain by +/- 50 meters." "P" 185 Sample Log: Did not trip and no messenger on top. It did drop its messenger. No water samples. Cast not repeated due bad weather. 186 Sample Log: Tripped at the surface. No water samples drawn. Cast not repeated due bad weather. 187 Tripped at the surface per DSRTs. No water samples drawn. Cast not repeated due bad weather. 188 Tripped at the surface per DSRTs. No water samples drawn. Cast not repeated due bad weather. 189 Tripped at the surface per DSRTs. No water samples drawn. Cast not repeated due bad weather. 151 Delete low PO4 (2.09 kg, 2.14 1) @3371db. 201 Sample Log: Did not trip. Bad Pylon 202 Sample Log: Did not trip. Bad Pylon 203 Sample Log: Did not trip. Bad Pylon 204 Sample Log: Did not trip. Bad Pylon 205 Sample Log: Did not trip. Bad Pylon 206 Sample Log: Did not trip. Bad Pylon. No water samples @178db. 207 Sample Log: Did not trip. Bad Pylon. No water samples @248db. 208 Sample Log: Did not trip. Bad Pylon. No water samples @299db. 209 Sample Log: Did not trip. Bad Pylon. No water samples @371db. 210 Sample Log: Did not trip. Bad Pylon. No water samples @427db. 211 No salt drawn @522db per sample log. Other water samples ok. No notes. Forgotten? All three (silicate, phosphate and nitrate) do not fit at all with the vertical section and are pretty clearly erroneous. - L. Talley 920409 Bottle appears to have tripped at ~300db, Delete all water samples (O2=5.91, PO4=1.77, NO2=0.01, NO3=25.9, SiO3=13.3). 212 No salt drawn @623db per sample log. Other water samples ok. No notes. Forgotten? 215 delta-S .075 low @988db. Calc ok. Same value as 214. Assume dupe draw. Delete bottle salinity (34.266). 216 delta-S .016 high at 1096db. All water samples appear to be from about 100db. No notes. Assume bad trip or bad leak. Delete all water samples (S=34.441, O2=5.16, PO4=1.00, NO2=0.08, NO3=12.3, SiO3=12.8). 222 L. Talley indicated PO4 high. Investigation of the data shows good comparison with CTDO2 even though bottle oxygen appears low. Silicate appears high, but only fits lower in the water column. NO3 appears high, also fits lower in the water column. Salinity agrees with CTD .003 so both salinity and oxygen show that something is different here, and appears to be real. ODF has suggested that the samples be footnoted either "U" for uncertain or "G" for good and verified by other data parameters. Foot note PO4 "U" for uncertain. 228 Salinity: "loose top" delta-S .004 high at 3112db. Delete bottle salinity (34.741). 234 Sample Log: Lanyard bung up. No water samples @4298db. 356 Niskin-Gerard S=.312 at 421db. Calc ok. No notes. 356 water samples appear to be surface water. Gerard 86 salt agrees with CTD salt this level so Gerard sample good. Delete all Niskin water samples (S=34.526, O2=5.91, PO4=0.78, NO2=0.09, NO3=9.0, SiO3=2.9). 360 Sample Log: Leaked. Details not noted. Salinity: "jumpy" Niskin-Gerard S=-.007 at 1325db. Calc ok. Normal gradient small leak in Niskin 60 likely and Gerard 89 looks ok. Freon & barium. also drawn from Niskin. Suggested footnote: "Sample Log note and .007 difference between Niskin and Gerard salts indicate slight leak in Niskin sample 360." "L" STATION 260 102 No discrete O2 @7db, Keeling CO2 samples. 171 Sample Log: Did not fire. No further explanation. No notes Conops or Deck Log. No water samples @79db. 116 Sample Log: Lanyard hung on frame during recovery, lifted top stopper. Discrete O2 @1059db about .12 high compared to CTDO. No freon or helium. drawn this sample. Salt & nutrients ok. Delete bottle oxygen (4.62). 128 PO4 @3200db low. Footnote PO4 "U" uncertain. 129 PO4 @3402db low. Footnote PO4 "U" uncertain. 130 PO4 .05 low at 3599db. Calc ok, peaks fair. Air in PO4 line before 130 and after 136. Footnote PO4 "U" uncertain. 131 PO4 .04 low at 3799db. Calc ok, peaks fair. Air in PO4 fine before 130 and after 136. Footnote PO4 "U" uncertain. 132 PO4 .04 low at 3999db. Calc ok, peaks fair. Air in PO4 line before 130 and after 136. Footnote PO4 "U" uncertain. 133 PO4 .03 low at 4198db. Calc ok, peaks fair. Air in PO4 line before 130 and after 136. Footnote PO4 "U" uncertain. 134 PO4 .03 low at 4349db. Calc ok, peaks fair. Air in PO4 line before 130 and after 136. Footnote PO4 "U" uncertain. 135 PO4 .O2 low at 4499db. Calc ok, peaks fair. Air in PO4 line before 130 and after 136. Footnote PO4 "U" uncertain. 136 PO4 .04 low at 4550db. Calc ok, peaks fair. Air in PO4 line before 130 and after 136. Footnote PO4 "U" uncertain. STATION 263 106 Salinity: "Lost" No reason given. No hydro salt 249db. STATION 265 131 delta-S .006 high at 4402db. Calc ok. Smooth CTD gradient. Probable bad draw or run. Delete bottle salinity (34.680). STATION 266 111 Salinity: "may leak a little" delta-S .004 high at 601db. Same salt bottle used subsequent station with no problem. Delete bottle salinity (34.224). 171 Sample Log: Lanyard hung, no sample. No water samples @98db. STATION 267 101 Oxygen: "broken flask, no sample" No oxygen sample @4db. 125 O2 .04 high at 3199db compared to CTDO. Calc ok. Possibly drawn from Niskin 26?. Delete bottle oxygen (4.84). 127 Sample Log: Vent leaks. delta-S .001 low at 3600db. Other water samples also ok. NO3 @3600db 1.0 higher than adjacent stations. Rerun (between 130 & 131) was used which was definitely higher than original, poorer, peak. All subsequent NO3 values also higher than adjacent stations. Footnote high NO3 as uncertain "U". 128 NO3 @3802db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 129 NO3 @4004db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 130 NO3 @4200db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 131 NO3 @4399db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 132 NO3 @4596db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 133 NO3 @4802db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 134 NO3 @5000db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 135 NO3 @5198db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". 136 NO3 @5253db about 1.0 higher than adjacent stations. Peak not perfect but definitely high. Footnote high NO3 as uncertain "U". STATION 268 117 Sample Log: Valve pushed in. delta-S .071 low at 1498db. Calc ok. Other water samples also indicate leak. Delete all water samples (S=34.684, O2=4.31, PO4=2.22, NO2=0.00, NO3=32.8, SiO3=81.1). 127 NO3 high @3401 db, footnote "U". 128 NO3 high @3608db, footnote "U". 129 NO3 high @3802db, footnote "U". 130 NO3 high @4001db, footnote "U". 136 Oxygen: "Forgot to write down value, lost" No oxygen data @5146db. STATION 270 136 Sample Log: Tritium drawn after nuts. O2 .4 high at 5638db compared to CTDO and adjacent stations. Calc ok. Assume draw or titration. Delete bottle oxygen (6.01). STATION 271 1all No temperatures, forgot to read DSRTs before next Gerard cast. Calculating barrel depths based on corrected wire length, pinger distance from bottom, and corrected bottom depth gives good agreement between Gerard salts and rosette salts. 211 No nutrients @605db per nutrient data sheet. Ok per sample log. Assume not drawn. 217 O2 .4 high compared to CTDO at 1658db. Calc ok. Delete bottle oxygen (4.83). 223 O2 .04 high compared to CTDO at 2796db. Calc ok. Delete bottle oxygen (4.88). 228 delta-S .005 high at 3799db. Calc ok. Smooth CTD gradient. Note on salt data sheet indicates possible salt crystal contamination. Delete bottle salinity (34.677). 3all Sample Log: Cast aborted, double pinging after two barrels. No samples. 451/487 DSRTs indicate posttrip; TP 1505db vs. intended depth of 2900m. Delete Niskin and Gerard level. Post trip. See 451. 452/488 Sample Log: Rigged wrong, didn't close. No Niskin salts or temp. DSRTs on barrels above & below indicate posttrip. Delete Niskin and Gerard level. Posttrip. See 452. 453/489 DSRTs indicate posttrip; TP 1746db vs. intended depth of 3800m. Delete Niskin and Gerard level. Post trip. See 453. 456 Sample Log: Rigged wrong, didn't close. No Salt for Gerard check @2366db. See 486. Gerard 86 salt agrees with rosette salt this level. Gerard sample ok. 459/485 Sample Log: Rigged wrong, did not close. No salt for Gerard check @2O27db. See 485. Gerard 85 salt is .005 higher than rosette salt this level. Gerard sample probably ok. 553/589 DSRTs indicate pre or posttrip; TP 1337db vs. 3600m intended depth. Niskin-Gerard S=.014 indicating poor flush. Delete Niskin and Gerard level. Post?-trip. See 553. STATION 272 113 Salinity: "top loose, leaks" delta-S .005 high at 801db. Normal CTD gradient. Delete bottle salinity (34.663). 134 delta-S .005 high at 4753db. Calc ok. Same value as 132. Possible dupe draw. Delete bottle salinity (34.659). 171 Sample Log: Smashed drain. No water samples @148db. STATION 273 113 Sample Log: No sample, end caps lost. No water samples @801db. 131 Sample Log: 6 pCO2 samples, nothing else. No water samples @4000db. STATION 274 132 NO3 1.5 high @4200db. Poor peak but high. Rerun after standards had worse peak. Delete nitrate (36.2). STATION 275 122 Sample Log: Bottom end cap hung up. No sample @1803db. 135 O2 .12 high @3938db. Calc ok. CTDO shows smooth gradient. Delete bottle oxygen (5.40). STATION 279 120 Keeling & LDGO CO2 only. No hydro S, O2 or nutrients @4db. STATION 280 125 PCO2 only. No other water samples @903db. STATION 281 101 No Salt, O2 or nutrients @4db, Keeling CO2. 106 No Salt, O2 or nutrients @200db, Keeling CO2. 108 No Salt, O2 or nutrients @247db, Keeling CO2. 111 No Salt, O2 or nutrients @347db, Keeling CO2. 114 No Salt, O2 or nutrients @499db, Keeling CO2. 116 No Salt, O2 or nutrients @597db, Keeling CO2. 118 No Salt, O2 or nutrients @748db, Keeling CO2. 120 No Salt, O2 or nutrients @998db, Keeling CO2. 123 No Salt, O2 or nutrients @1497db, Keeling CO2. 126 No Salt, O2 or nutrients @1999db, Keeling CO2. 129 No Salt, O2 or nutrients @2484db, Keeling CO2. 132 No Salt, O2 or nutrients @3010db, Keeling CO2. 135 No Salt, O2 or nutrients @3628db, Keeling CO2. STATION 281 (continued) 136 S103 6. too high. Delete S103 (140.6). NO2.03 high, not sure what caused this. Footnote NO2 "U" for uncertain. 171 No Salt, O2 or nutrients @126db, Keeling CO2. STATION 282 115 All water samples appear to be from about 2000db vs. 5 db when tripped. Top lanyard found broken on recovery per Deck Log. Delete all water samples (S=34.709, O2=4.67, PO4=2.18, NO2=0.00, NO3=33.9, SiO3=119.1). 129 delta-S .004 high at 14OOdb. Calc ok. CTD trace smooth and indicates sal max closer to 1500db. No notes. No apparent problems with other water samples. Delete bottle salinity (34.720). STATION 283 135 Sample Log: Lid not closed. 5m above 136. Suspended particles checked on sample log. Deck Log note: "top of Niskin 35 did not close completely slightly askew." No other samples drawn @1714db so sample integrity unknown. STATION 284 131 Discrete O2 .2 high at 1297db. Calc ok. CTDO down & up traces show min not max this level. O2 values similar to 132 value so possibly drawn from Niskin32 in error. Delete bottle oxygen (4.39). 287 Sample Log: Went down without washer on valve arm. Assume means no o-ring on valve cap. Niskin Gerard S=.003 at 889db. Probable slight leak. Suggested footnote: "No o-ring on valve cap. Probable slight leak." "L" STATION 286 110 Sample Log: Air leak. delta-S .000 at 551db. Other water samples also ok. 122 Discrete O2 .1 high. Calc ok. No notes. CTDO shows feature this level both down & up but only about 1/2 as high. Nutrients & salt also have feature but salt is proportional to CTD down & up traces. Suggest: FOOTNOTE "U" STATION 287 112 Sample Log: Air leak, sample leaked out on deck. No water samples drawn @801db. 113 delta-S .014 high at 948db. Calc ok. High gradient. Nutrients ok (NO3 & PO4 max). O2 problem apparently unrelated is described below. O2 value appears about .07 high at 948db compared to CTDO down & up traces. Oxygen analyst indicated sample was over-titrated, but sample does not give good results. Delete bottle oxygen (4.00). 259 Sample Log: No salts drawn. Sampling error @5203db, they should have been. Freon & barium were drawn. Gerard (#289) salt agrees well with CTD salt trace. 260 Sample Log: No salts drawn. Sampling error @5380db, they should have been. Freon & barium were drawn. Gerard (290) salt agrees well with CTD salt trace. 353 Salinity: "top loose" Niskin-Gerard S=.009 at 208db. Gerard salt looks ok. Delete Niskin salinity (34.147). STATION 291 160 Niskin-Gerard S=.005 at 5449db. Calc ok. No notes. Gerard (190) salt agrees with CTD and other Argon cast salts this level. No other samples taken Niskin 160. Delete Niskin salinity (34.671). STATION 292 130 SIL 1.6 higher than 131 SIL tripped at same level (4198db). 131 value in better agreement with adjacent values and 130 has poor peak. Delete silicate value (129.3). STATION 293 116 Sample Log: Hung up. No water samples @1502db. STATION 294 121 delta-S .018 low at 2198db. Calc ok. Same value as 120 salt so possible dupe draw. CTDs noisy but not this much. O2 & PO4 look ok. NO3 1.0 high and SIL 2.0 high. NO3 peak poor, SIL peak ok. Calc ok. Delete bottle salinity (34.773). 136 Discrete O2 @5242db .33 ml/L high compared to adjacent samples and CTDO traces. Calc ok. Assume bad draw or titration. Deleted bottle oxygen (5.51). 257/288 Thermometer: Lanyard hang-up, therms did not reverse. No temperature @4716db. Sample Log: Lid not completely closed. Niskin-Gerard S=-.001 at 4699db. Gerard sample ok. 259/289 Sample Log: Air leak. Salinity: "Perhaps drawn quite late" Niskin-Gerard S=-.002 but both DSRTs and salinities indicate Niskin and Gerard tripped at about 2550db instead of 5080db as intended. Probably post-tripped. See 290. Suggested footnote: "Post-trip, poor flush.". 260/290 Sample Log: Air leak. Salinity: "Perhaps drawn quite late" Niskin-Gerard S=.008 at 5241db. Gerard salt agrees with rosette & CTD salts. Assume Niskin leaked. Freon & barium also drawn from Niskin. Delete Niskin salinity (34.673). Gerard samples looks good. 360/392 Niskin salt low at 1476db indicating leak. No notes but same Niskin leaked last cast. Barium also drawn from Niskin #60. Delete bottle salinity (34.596). Sample Log: Air leak. Niskin- Gerard S=-.013 at 1476db. Gerard salt agrees with rosette & CTD salts. Assume Gerard ok. STATION 295 112 Sample Log: Hung up/no sample. No water samples @849db. STATION 296 133 delta-S .004 low at 4797db. Calc ok. Normal CTD gradient. Same value as 134, possible dupe draw. Delete bottle salinity (34.667). STATION 297 130 delta-S .006 high at 3898db. Calc ok. Normal CTD gradient. Possible salt crystal from cap got into sample. Nutrients ok. Delete bottle salinity (34.708). STATION 298 154/185 Sample Log: Didn't trip. No water @3022db. Gerard (185) salt agrees with CTD salt. 157/188 Sample Log: Didn't trip. No water @4515db. Gerard (188) salt agrees with CTD salt. STATION 300 131 delta-S .004 high at 3899db. Calc ok. Normal gradient. No notes. Other water samples ok. Deleted bottle salinity (34.700). STATION 302 210 delta-S .051 high at 473db. Calc ok. Normal gradient. No notes or explanation Other water samples ok (02 min). Delete bottle salinity (34.880). 235 PO4.42 high at 4550db. Calc & peak ok. Assume contamination problem. Delete phosphate (2.78). 236 delta-S .002 to .003 high at 4643db. Steady increase in each Autosal run indicating possible salt crystal contamination from bottle cap. Delete bottle salinity (34.671). STATION 303 113 Sample Log: Drained for salts. No salinity sample @651db. 115 Sample Log: Empty for salts. No hydro S @803db. 117 Sample Log: Low for salts. delta-S .002 high at 1002db. Hydro O2 low at 1003db per CTDO calibration. Calc ok, no notes. This is Keeling CO2 bottle so bottle half empty when O2 drawn. O2 .22 lower than Niskin 16 tripped at same level. Delete bottle oxygen (5.19). 124 Sample Log: Empty for salts, barely sampled nuts. No bottle salinity @2110db. 131 Sample Log: Freon drawn after TCO2. Does not effect samples @355ldb. delta-S .008 high at 3550db. Calc ok. Small spike in raw CTD S trace this level but block averaged data has smooth normal gradient. No notes. Delete bottle salinity (34.785). STATION 304 134 delta-S .080 high at 3696db. Calc ok. All water samples appear to be from about 3475db. Delete all water samples (S=34.779, O2=5.19, PO4=2.00, NO2=0.00, NO3=28.8, SiO3=86.1) STATION 305 113 Oxygen @10OOdb is low. Appears to have drawn from bottle 14. Delete oxygen (5.33). 171 Sample Log: Sample cock sheared off during recovery, sample lost. No water samples @129db. STATION 306 119 Oxygen data sheet "bubble - large variety" O2 .2 high @1954db. Same value as 120. Smooth CTDO trace. Delete bottle oxygen (4.75). 453 Salinity data sheet: "Junk in water" Niskin-Gerard S=.006 at 5049db. Calc ok. Niskin is high and Gerard (88) agrees with CTD, rosette and other samples this Argon cast. No other sample drawn this Niskin. Delete Niskin salinity (34.676). 457 Niskin-Gerard S=.011 at 4862db. Calc ok. Niskin is high and Gerard (83) agrees with CTD, rosette and other samples this cast. Freon and Barium also drawn this Niskin. Niskin salt is similar to 454 salt which was the adjacent bottle in the rack. Possibly 457 was misdrawn and the other samples are ok. No notes. Delete Niskin salinity (34.681). 459 Sample Log: Empty - didn't trip. No Niskin salt or hydro T & P at 5025db. Gerard salt (485) agrees with CTD, rosette and other samples this Argon cast. STATION 307 128 Sample Log: Thermometer lanyard caught in bottle top. Sample discarded. Nutrients & salinity run; both indicate severe leak@3301db. Delete all water samples (S=34.893, O2=1.35, PO4=1.35, NO2=0.01, NO3=19.3, SiO3=37.4). 186 Sample Log: Not latched, water squirting out of lid on recovery. Niskin-Gerard S=-.007. Niskin sample (56) agrees with other samples this level and Gerard salt is high. After cast found wrong size air vent plug had been used. (New {WHOI} barrels have slightly different diameter hole in lid for air vent plug than the old (yellow- hooded) barrels.) Sample Log indicates Argon and radium 228 may have been taken from Gerard. Leave bad Gerard salt in report to indicate problem, footnote "U". STATION 308 106 delta-S .258 low at 199db. Calc ok. Same value at 107 salt. Other water samples ok. Assume 106 salt drawn from Niskin 7. Delete bottle salinity (34.626). A16S (SAVE 5) • Radiocarbon • Östlund/Grall • 1989 __________________________________________________________________________ __________________________________________________________________________ UNIVERSITY OF MIAMI ROSENSTIEL SCHOOL OF MARINE AND ATMOSPHERIC SCIENCE MIAMI, FL 33149 S.A.V.E. RADIOCARBON SOUTH ATLANTIC VENTILATION EXPERIMENT 1987-1989 TRITIUM LABORATORY DATA REPORT # 18 by H. Göte Östlund and Charlene Grall This report constitutes the Final Report for the following grant: NSF OCE-86-13329 September 1992 TRITIUM LABORATORY DATA REPORTS #1 TRITIUM IN THE TROPOSPHERE AND SURFACE WATER OF NORTH ATLANTIC OCEAN 1964-70, H.G. Östlund, R.D. Stearns, and R. Brescher. July 1971. #11 ANTARCTIC TRITIUM 1977-1979, A.S. Mason and H.G. Östlund. November 1981. #12 GEOSECS TRITIUM, Atlantic Ocean 1972-73, Pacific Ocean 1973-74, Indian Ocean 1977-78, Station 347 Revisits, H.G. Östlund and R. Brescher. December 1982. #13 NAGS TRITIUM, North Atlantic Gyre Studies and Associated Projects, H.G. Östlund. June 1984. #14 ATMOSPHERIC TRITIUM 1968-1984, H.G. Östlund and A.S. Mason. April 1985. #15 EQUATORIAL PACIFIC TRITIUM, H.G. Östlund, C. Grall, and R.E. Brescher. April 1986. #16 TTO NORTH AND TROPICAL ATLANTIC TRITIUM AND RADIOCARBON, H.G. Östlund and C. Grall. February 1987. #17 INDIGO 1985-1987, INDIAN OCEAN RADIOCARBON. H.G. Östlund and C. Grall. April 1988. #18 S.A.V.E. RADIOCARBON, South Atlantic Ventilation Experiment, 1987- 1989. H.G. Östlund and C. Grall. November 1992. THIS REPORT. NOTE: Data Report #s 2,3,4,5,6,7,8,9,10 have been superseded by #s 12,13 and 14. The measurement results of #12 are also available in Vol. 7 of the GEOSECS ATLAS. PREFACE This report presents all the radiocarbon measurements produced from the six S.A.V.E Cruises 1987-1989. The following informal Data Releases include all measurement results previously reported. Additional quality control has slightly changed some earlier data which are hereby superseded. #89-05 S.A.V.E. Radiocarbon Results, Legs 1 and 2. #89-38 S.A.V.E. Radiocarbon Results, Legs 2 and 3. #90-32 S.A.V.E. Radiocarbon Results, Legs 4,5 and HYDROS Leg 4. #91-56 S.A.V.E. Radiocarbon Results, Legs 3,4,5 and HYDROS Leg 4. The data in this report are hereby in the public domain to, be used by anyone. Conventional reference is appreciated. Copies of this Data Report, or any other Tritium Laboratory Data Report, may be obtained by addressing the Tritium Laboratory, RSMAS, University of Miami, 4600 Rickenbacker Causeway, Miami, Florida 33149, U.S.A. A 5-1/4" floppy disk, containing all radiocarbon data in this report in addition to some hydrographic data, in ASCII format, is available upon request. We hope that this data set will be used extensively in many modeling studies in the years to come. ACKNOWLEDGMENTS The measurement of all the radiocarbon samples and the production of this report were carried out under National Science Foundation Grant NSF-OCE-86-13329. We are very grateful for their long term support. We also thank the technical staff of the Ocean Data Facility at Scripps Institution of Oceanography for the radiocarbon sample collection and shipboard extraction. We sincerely appreciate the efforts of the chief scientist to include as many sample collections as they did especially after the untimely loss of the Gerard barrels on the first cruise of this project. M. Stuiver and T. Saling at Quaternary Research Office, University of Washington, made the numerous and excellent 13C measurements which is hereby gratefully acknowledged. TABLE OF CONTENTS TRITIUM LABORATORY DATA REPORTS PREFACE ACKNOWLEDGEMENTS (S.A.V.E.) RADIOCARBON SOUTH ATLANTIC VENTILATION EXPERIMENT, 1987-1989. I INTRODUCTION II CRUISE DESCRIPTIONS III TECHNICAL PROCEDURES IV NOTES ON SCALES V ACCURACY AND ERRORS VI HYDROGRAPHIC DATA . . VII COMMENTS TO FIGURES AND TABLES VIII REFERENCES APPENDIX SECTIONS (SAVE 5 - A16S only) S.A.V.E. RADIOCARBON SOUTH ATLANTIC VENTILATION EXPERIMENT, 1987-1989 I. INTRODUCTION The project called Transient Tracers in the Ocean, TTO was initiated it 1980 with the intent to study the Atlantic Ocean. The North Atlantic Study (NAS) and Tropical Atlantic Study (TAS) scientific cruises continued only as far south as 10°S. The South Atlantic Ventilation Experiment (SAVE) expeditions traversed as far south as 53°S and east to Africa. Through the use of various tracers, the program investigated circulation, mixing and ventilation of the South Atlantic Ocean. Radiocarbon (14C) is considered an important tracer to study because it helps provide information on rates of internal oceanic mixing as well as the rate of gas exchange between ocean surface and atmosphere. The coverage of 14C measurements from SAVE is much denser than that of its predecessor, GEOSECS, and can provide a much better base upon which future C02 studies can build. It also provides a view at the evolution of bomb 14C distribution over the last decade since GEOSECS, as several GEOSECS stations were reoccupied during SAVE. II. CRUISE DESCRIPTIONS The SAVE expedition consisted of two cruises subdivided into six legs. A. YEAR 1 The first leg aboard the R/V Knorr, began 23 November 1987 from Recife, Brazil. The Knorr crossed the Atlantic Ocean on a northeasterly course and arrived into Abidjan, Ivory Coast on 13 December 1987. A total of 121 radiocarbon samples was collected from ten large volume stations. During a deep Gerard cast on station 32, all barrels were lost when the winch cable snapped. Two spare Gerard barrels were used during the remainder of Leg 1, which limited sample collection to the upper 1000m of depth. The Knorr departed Abidjan on 18 December 1987, heading southwest and arrived in Rio de Janeiro, Brazil on 23 January 1988. During Leg 2, 163 14C samples were collected from 21 large volume stations. With the exceptions of stations 95 and 99, where deep samples were collected from the core of the Antarctic Bottom Water and North Atlantic Deep Water, maximum sample depth was 1000m. On Leg 3, once again armed with full complement of Gerard barrels, the expedition set out on 28 January 1988, traveled southeast to about O°W then heading north back to Abidjan. Before the cruise ended on 7 March 1988, 349 radiocarbon samples were taken at 30 large volume stations. B. YEAR 2 The second year of SAVE began with departure of the R/V Melville from Punta Arenas, Chile on 7 December 1988. She crossed the South Atlantic Ocean along 47°S to 20°W, headed northeast to 30°S before steaming due east to Capetown, South Africa, arriving on 15 January 1989. During Leg 4, 299 radiocarbon samples were collected from 27 large volume stations. Leg 5 cruise departed Capetown on 23 January 1989. The cruise track consisted of two meridional sections: the first traversing from 24°W south to South Georgia Island; the second section going north along 41°W. A small northwest portion of the track heading toward Montevideo, Uruguay ended on March 8, 1989. A total of 173 14C samples was taken on 20 stations. The remainder of the cruise was a meridional section called HYDROS, Leg 4, which ran north from SAVE station 237 to the equator along 25°W. During this leg, 107 radiocarbon samples were collected from eight stations. The program ended 19 April 1989. The table below summarizes the sample collection and measurement. | # Radiocarbon Samples | --------------------- Cruise Leg | Collected | Measured ---------- | --------- | -------- SAVE 1 | 121 | 117 2 | 163 | 111 3 | 349 | 296 4 | 299 | 256 5 | 173 | 159 HYDROS 4 | 107 | 100 --------- | --------- | -------- TOTAL | 1,212 | 1,039 C. SAMPLES STILL AVAILABLE FOR ANALYSIS Although most of the samples collected during SAVE have been analyzed, 139 samples are still unmeasured and are archived at our laboratory. Stations not yet analyzed are presented as tables located in the last section of this report, starting on page 72. We hope that the sample selection by various investigators and us has covered the major and relevant ocean features explored during the two years of SAVE. In the event that some questions remain that could be answered with additional analyses, we intend to store the remaining samples for a few more years. III. TECHNICAL PROCEDURES A. SAMPLING Water samples were collected on board ship using 270 liter stainless steel Gerard barrels. Shipboard extraction of C02 was performed by acidifying 200 liters of seawater with 50ml of 95% sulfuric acid, and bubbling through an inert C02-free gas. The released C02 was then stripped from the carrier gas and re-dissolved into a 450ml bottle of 4N sodium hydroxide (NaOH). The sample bottles were then shipped to the Tritium Laboratory for measurement. Resulting yield, under favorable conditions, was about 9 lit-atm of C02 or 95 - 98% of the total dissolved C02 in the sample. B. PREPARATION OF SAMPLING GAS In the laboratory, C02 was released from the NaOH solution by adding phosphoric acid. Nitrogen gas, the carrier, transferred C02 and any other evolved gases to collection traps maintained at -196°C by liquid nitrogen. After pumping on the solid C02, this gas was then passed through a purification system consisting of a series of cold traps to remove water vapor, and two copper-silver ovens to remove gaseous electronegative impurities, mainly chlorine and oxygen. Remaining impurities were removed by pumping on condensed C02 at a temperature of -196°C. It is our experience that yields are about 99.7% of the total C02 contained in the NaOH solution. Occasionally, yields were less than optimal for counting in the large radiocarbon counters. These samples were saved when possible by counting in a smaller counter for twice the length of time required for a normal size sample. At this point, a small aliquot of purified gas was removed for massspectrometric delta-13C analysis. The remaining gas was stored in a stainless steel high pressure cylinder for 14 to 21 days before counting to allow any radon to decay (half-life is 4 days). The radon apparently originated from the phosphoric acid, and occasionally from the glass of the NaOH bottles. Below follows a description of our standard 14C measurement technique and special modification for this project. C. COUNTING We have four 2.5 liter and one 1 liter low-level gas proportional counting tubes for 14C, using C02 gas samples. The counters are shielded by 2.5 cm of selected lead, a ring of anti-coincidence Geiger counters, 10cm of paraffin wax, boric acid and/or borated polyethylene, and at least 20 cm of iron, plus the walls and ceiling of the building. The counter is filled with the sample to a working pressure of 45 psi. The proper operating voltage, dependent on temperature and pressure in the counter, is adjusted to produce the most efficient and stable 14C detection. This is done by an external radioactive source, usually 60- CO. Gas amplification is continually monitored by the distribution of meson counts in selected energy channels. The occurrence of radon is monitored in a separate channel with each counter. Each sample is counted for at least two separate periods of about 20 hours each, in different counters, with an interim waiting time of at least 7 days. Inconsistency between the two counts prompts an additional measurement in a third counter. For the short samples in this series, two counters were temporarily standardized to work at low pressure. These samples were counted for a total of 4 to 6 days. D. BACKGROUND AND STANDARDS Background count rates, typically 3.5 cpm, are determined weekly by measuring 14C-free C02 gas. The standard material for 14C measurements is the NBS oxalic acid standard, RM49 and SRM4990C, for radiocarbon dating. The C02 prepared from this standard is counted for two days every week in each counter, to determine counting efficiency; typical count rate for an ocean surface sample is 43 cpm above background. E. UPDATE Periodically, usually every five weeks, all measurements in all counters for the preceding time period are recomputed, statistical tests are applied, and results scrutinized for flaw in quality of sample counts and stability of efficiency and background. This includes all measurements of standards, unknown and backgrounds. Only after this step is the measurement result considered final. Contribution from minute quantities of C02 in the original NaOH solutions is measured for each batch of NaOH and appropriate corrections have been applied. IV. NOTES ON SCALES The measurement results in the ratio, r, between the measured specific activity of the sample C02 and that Of C02 prepared from the NBS standard. The activity value of the standard is corrected to a hypothetical delta-13C value of -19‰ and age-corrected from today back to AD1950, all according to international agreement. DELTA-14C is the deviation, in ‰ (per mil), from unity, of the activity ratio, isotope- corrected to a sample delta-13C value of -25‰ Normally there is very little (± 0.3‰) isotope separation in our preparation procedure. Within these limits, our delta-13C values represent the state of isotope fraction- ation of the original seawater C02. According to this convention, each of our reported DELTA-14C values is, however, normalized to the delta value it would have if its delta-13C changed to -25‰, noting that the change is equal to twice the change in 13C . Note that the mass of 14C is always proportional to the r-value, not to DELTA-14C See Appendix 1. Dr. Minze Stuiver at the Quaternary Lab, University of Washington measured the delta-13C values for this project by precision mass-spectrometry. For further explanation of these features, and information on standards, cf. preface to each issue of Radiocarbon, and papers by Broecker and Olson (1), Stuiver and Robinson (2) and by Stuiver and Polach (3). Also refer to Appendix. V. ACCURACY AND ERRORS Errors for delta-14C are not listed in the tables. For a normal size sample the errors are 3.0 to 3.5‰ for precision and 3.5 to 4.0‰ for accuracy, including all conceivable uncertainties. The small samples, due to the very long counting times should be about as good. VI. HYDROGRAPHIC DATA The hydrographic data were supplied by the Ocean Data Facility (ODF), Scripps Institution of Oceanography. Total C02 data were supplied by Taro Takahashi, Lamont Doherty Geological Observatory at Columbia University. They are presented in µmol/kg. We calculated potential temperature from in situ temperature, pressure, and salinity by the equations of Bryden (4). The density (sigma-theta) values were calculated according to the International one-atmosphere equation of state of seawater according to Millero and Poisson (5), which generates absolute densities in units of kg/m3. For serious use of the hydrographic, data, please refer to the Scripps Institution of Oceanography SAVE Data Reports; S.I.0. Reference 92-9, 92-10 and 92-12. For total C02 data please contact Taro Takahashi. VII. COMMENTS TO FIGURES AND TABLES A. SECTIONS Radiocarbon is expressed as DELTA-14C The horizontal scale distance is given in kilometers with accompanying latitudes or longitudes. The vertical depth scale is in meters. Dots indicate measured samples. Due to scant data coverage in several areas, some subjectivity enters in drawing the isopleths. B. PROFILES Radiocarbon is plotted as A14C ‰ versus depth in meters. The bottom depth is marked by a horizontal line when it falls inside graph limits. Occasionally, some of the data were not reliable and are designated by "++" or small letters. Explanations are found in the tables of data. C. TABLES Positions are stated in degrees and minutes. SMPL# is the cast and Niskin bottle number. GER represents the cast and Gerard barrel number. DC14 is DELTA-14C ‰ and dC13 is delta-13C ‰ TC02 is total dissolved carbonate in µmol/kg. All collected sample's are presented in the tables. Those samples marked with an asterisk, * are still available for analysis. VIII. REFERENCES Broeker, W.S. , and E.A. Olson, 1961. Lamont radiocarbon measurements VIII. Radiocarbon, 3, 176-274. Stuiver, M. , and W.S. Robinson, 1974. University of Washington GEOSECS North Atlantic carbon-14 results. Earth Planet, Sci. Lett., 23, 87-90. Stuiver, M. , and H.A. Polach, 1977. Discussion reporting of 14 C data. Radiocarbon, 19, (3), 355-363. Bryden, H.L., 1973. New polynomials for thermal expansion, adiabatic temperature gradient and potential temperature of sea water. Deep-Sea Res., 20, 401-408. Millero, F.J. and A. Poisson, 1981. International one-atmosphere equation of state of seawater. Deep-Sea Res., 28, (6A), 625-629. APPENDIX (Revised August 15, 1992) ON THE MEANING AND USE OF DELTA-14C AND DELTA-14C IN OCEANOGRAPHY by H. Göte Östlund Tritium Laboratory, RSMAS, University of Miami Summary OOPS! Don't use delta-14C for calculating quantity of 14C. ARGUMENT delta Beta counting of 14C produces results in form of count rates, which can be converted to the relative specific activity r of the sample defined by r = Asx/Ast = Isx/Ist (1) where Ast is the adjusted specific activity of the NBS oxalic acid 14C standard, 13.56 dpm/g of carbon, and Asx is the specific activity of the sample in dpm/g. By applying the decay rate of 14C this ratio can also be expressed in isotope 14C/Ctot-abundance ratios I, where Ist is 1.176 - 10-12. Often a 'per mil deviation', delta-14C is calculated according to delta-14C = 1000 · (r - 1) (2) See Remark. Assuming that isotope fractionation is negligible during sample preparation, the r-value (and thereby the delta-14C) is the proper parameter to use for computing the absolute, true quantity of 14C as number of atoms or moles. Thus the amount of 14C in moles, in a kg of seawater or air is -12 SIGMA-14C = 1.176 - 10 · r · SIGMA (3) where SIGMA-C is the absolute assay of carbon in moles/kg. Mass inventories of 14C, as well as macroscopic mixing and transports by currents and turbulent diffusion call for the use of Eq 3. If there is fractionation during collection and processing of a sample, the true, original r-value can be calculated from the sample r-value if delta-13C data are available for both. By experience the standard ship-board C02 extraction from seawater and laboratory processing cause isotope shifts that are less than 0.5‰ and thus insignificant. ARGUMENT DELTA: By (an unfortunate) convention environmental 14C data are usually expressed as DELTA-14C or R according to DELTA-14C = delta-14C - 2 (delta-13C + 25) · (1 + 0.001delta-14C) (4) R = 1 + 0. 001.DELTA-14C (5) These equations describe what the 14C/Ctot isotope ratio would be if a hypothetical chemical fractionation process brings the delta-13C value from its actual value to -25 per mil. The use of the DELTA-14C and R is a carry-over from the radiocarbon dating procedures, where isotope fractionation in the biological cycle necessitates this correction to allow calculation of best 'radiocarbon age' of a sample, most often organic material. The delta-13C reference value -25 was originally chosen as standard value, being typical for terrestrial wood, first used for 14C standard material. In geochemical considerations, conversion to DELTA-14C is needed for calculations of "mummy" ages and for the proper handling of chemical equilibria, rate of chemical reactions and phase transfers. Examples: Mummy age = (-ln R) / lambda years before AD 1950. -1 where lambda is 14C decay rate in years In air-sea exchange of 14C02, assume Pair and Pdia are partial pressures of atmospheric and dissolved C02 and Rair and Rdis the isotope adjusted R-values according to Eq. 5, then there should be chemical equilibrium (no net transport) of 14C02 if Pair Rair - Pdis Rdis CONCLUSION: The r-values (delta-14C) are needed for use in the calculation of mass of 14C. If only DELTA-14C values are available, this is accomplished by solving for r in the equations 5,4 and 2 above and putting the value into Eq 3. With some approximation: Oceanic dissolved C02 delta-13C = 0.0 r = 1.050 R Atmospheric C02 delta-13C = - 8.5 r = 1.033 R Wood (cellulose) delta-13C = -25 r = 1.000 R The factors in column 3, above express the errors obtained in 14C inventories if DELTA-14C, or R, is used instead of delta-14C, or r. When one compares 14C in various parts of the ocean, a systematic error of 5% may be insignificant. Also measured transfer coefficients have larger uncertainties than 5%. However, include a production term for 14C (cosmic, industrial), or the atmospheric inventory, and there is cause for concern. REMARK In the exhaustive, maybe exhausting, paper by Stuiver and Polach (1977) delt-14C is indeed defined accurately and properly, with age corrections of standard material activity to AD 1950, and for the time elapsed between collection and measurement of the unknown sample. It is the first equation (un-numbered) on pg. 361 in that paper. The activity and isotope ratio values of the NBS oxalic acid standard are listed in several papers, among them one by Stuiver et al, (1981), viz pg. 208. The inventory equation on that page should use delta-14C , not DELTA-14C. POSTSCRIPT: This error was pointed out in a recent paper by Bard et al, (1988), and is discussed by Östlund and Rooth (1990), as well as by Lassey (1989). In this pioneering paper on global 14C, Craig (1957), correctly used the unadjusted r-values, and so did Kuc (1986 and 1987) on atmospheric 14C inventories. Most often, however, DELTA-14C is erroneously used for absolute quantities of 14C. __________________________________________________________________________ __________________________________________________________________________ FIRST PRESENTED AT THE MODELING WORKSHOP, SEATTLE, WA - AUGUST, 1989 __________________________________________________________________________ __________________________________________________________________________ REFERENCES: Bard, E., M. Arnold, H.G. Östlund, P. Maurice, P. Monfray and J-C Duplessy. Penetration of bomb radiocarbon in the tropical Indian Ocean measured by means of accelerator mass spectrometry, Earth and Planet. Sci. Lett., 87, 379-389, 1988, viz pg. 380. Craig, H. The Natural Distribution of Radiocarbon and the Exchange Time of Carbon Dioxide between Atmosphere and Sea, Tellus, 9(1), 1-17, 1957. Kuc, T. Carbon isotopes in atmospheric C02 Of the Krakow Region: A two- year record, Radiocarbon, 28, 649-654, 1986. Kuc, T. 14C traced in Krakow after the Chernobyl accident, Radiocarbon, 29, 319-322, 1987. Lassey, K.R., M.R. Manning and B.J. O'Brien. An Overview of oceanic radiocarbon, its inventory and dynamics, Aquatic Sciences, 3, 117-146, 1990. Östlund, H.G. and C.G.H. Rooth. The North Atlantic tritium and radiocarbon transients, 1972-1983. J. Geophys. Res., 95, 20,147-20,165, 1990. Stuiver, M. and H.A. Polach. Discussion Reporting of 14C Data, Radiocarbon, 19, 355-363, 1977. Stuiver, M., H.G. Östlund and T.A. McConnaughey. GEOSECS Atlantic and Pacific 14C Distribution in Carbon Cycle Modeling, Wiley and Sons, 1981, edited by Bert Bolin. A16S (SAVE 5) • 1989 __________________________________________________________________________ __________________________________________________________________________ CCHDO-WHPO DATA PROCESSING NOTES DATE CONTACT DATA TYPE DATA STATUS SUMMARY -------- ----------- --------- ----------------------------------------- 12/14/98 Key DELC14 Key has data 01/11/99 Smethie CTD/BTL Data are public 04/16/99 Jenkins He/Tr Projected Submission Date: 1999.09.15 disk crash/must reprocess 06/03/99 Diggs CTD CTD data are still with the PI. 06/04/99 Smethie CTD Data avail. at ODF/SIO 06/07/99 Smethie CTD Data are Public I thought you had all the SAVE and you certainly have my permission to obtain the CTD data. It is my understanding as you indicate in your e-mail that it is public 03/07/00 Talley CTD/BTL Data are Public; ctd data should be on website 05/31/00 Huynh DOC Website Updated; pdf, txt versions online 06/16/00 Talley CTD Submitted; not woce formatted 07/07/00 Anderson CTD/SUM Data Reformatted Reformatted .sum file. Reformatted CTD files, added correct headers. Used 2 for all QUALT codes. * Station 279 did not have oxygen, so used quality flag of 9. 07/10/00 Bartolacci SUM Updated file online I have replaced the current sumfile with the newly reformatted sumfile (by SAnderson) and updated all table references 07/10/00 Bartolacci CTD Reformatted file online I have linked the newly reformatted CTD file (reformatted by S. Anderson) and updated all table references. 12/11/00 Uribe DOC Submitted File contained here is a CRUISE SUMMARY and NOT sumfile; Documentation is online. Files were found in incoming directory under whp_reports. This directory was zipped, files were separated and placed under proper cruise. All of them are sum files. Received 1997 August 15th. DATE CONTACT DATA TYPE DATA STATUS SUMMARY -------- ----------- --------- ----------------------------------------- 04/03/01 Diggs CFCs Submitted CFC data submitted by Peter Salameh @ SIO. Placed files in data/onetime/ atlantic/a16/ onetime 04/11/01 Muus CFCs/SUM CFC Data Merged; New SUM file generated Notes file for A16S modification and final CFC merging. SAVE Leg5/HYDROS Leg 3 R/V MELVILLE April 11, 2001 D. Muus 1. Generated new SUMMARY file from final SIO/ODF data. Now has Uncorrected as well as Corrected Bottom Depths, corrected EVENT CODEs, NAV, Height ABOVE BOTTOM, WIRE OUT, MAX PRESS, NO. OF BOTTLES, complete parameter numbers from Sample Log Sheets and COMMENTS from ODF Station and Cast Description. Neodynium was not included since it does not have a parameter number. Particulates were not included since the documentation does not say whether they were parameter number 40 or 41 (or both). 2. Generated new SEA file from final SIO/ODF data. Now has BTLNBRs, complete CTDSAL and more decimal places in data. ODF did not include CTDOXY with bottle data in pre-woce cruises so it is not available for this report. Oxygen is included with the CTD data. 3. Merged final CFC data received from Peter Salameh April 3, 2001/usr/export/html-public/data/onetime/atlantic/a16/ original/A16_SAVE_CFC/save5.txt into the new SEA file, replacing the shipboard CFCs in the ODF data. 04/16/01 Muus CFCs Reformatted by WHPO The modified a16c and a16s data are in ~dave/SDIGGS/A16. I was wrong in my April 3rd message assuming the .SEA files were from ODF. They look more like they were generated from .sd2 files. I redid everything from the final ODF data: .SUM files, .SEA files for rosette and .SEA files for large volume samplers. Freons were collected from the piggyback Niskins on many of the Large Volume casts. Many of the CFCs have quality codes of 6 meaning they are averages of replicate samples. The WOCE instructions say the details on the replicates should be in the documentation. I haven't seen any such documentation, could it be in the original CFC data directory in "a16_save_cfc.mail"? The read permission is 'owner only'. We do not have any of the actual large volume data (c14, argon-39, krypton, radium 228 & 226). The new files included them in the .SUM parameters but do not have any headers in the .SEA files. Successfully ran sumchk and wocecvt. Please let me know if you have any questions or need further clarification. DATE CONTACT DATA TYPE DATA STATUS SUMMARY -------- ----------- --------- ----------------------------------------- 06/20/01 Uribe BTL Website Updated; Exchange File Online Bottle file in exchange format has been linked to website. 06/21/01 Uribe CTD/BTL Website Updated; new exchange files The exchange bottle file name in directory and index file was modified to lower case. CTD exchange files were put online. 08/02/01 Key DELC14 Website Updated; PI names corrected The PIs for C14 during SAVE/HYDROS were Broecker/Ostlund. The summary table shows No Data for A16C or A16S. I can provide copies of the LV files for all of SAVE/HYDROS if you do not have. I do not have Jenkin's (not Smethie) H3/He3 data for these cruises. 08/27/01 Swift He/Tr HE/TR data requested by J Swift WHPO records indicate he/tr data not yet submitted. Request for earliest possible submission sent to Bill Jenkins. 09/17/01 Diggs CFCs/LV New hyd/sum/LV & exchange files online Replaced both the SUM and BOTTLE files online with the ones with CFC values (Muus 20010416). Also placed LVS file online, produced EXCHANGE file and placed on website. EXCHANGE file verified w/ JOA v3.0. 11/12/01 Key BTL/SUM BTL and SUM submitted, see note: I uploaded the SAVE+HYDROS LV files today using your form. Note that a bottle number of "99" is a surface sample obtained without using a Gerard barrel. The file I sent does NOT have the piggyback niskin bottles. The only way I have that information is in a merged ascii file with "everything". This merged file does not have flags and is closer to the original files sent out by ODF + merged data I recieved from individual investigators. I have attached a copy of this file "just in case". The attached file is in an shortened format, but your guys won't have any trouble doing a reformat. DATE CONTACT DATA TYPE DATA STATUS SUMMARY -------- ----------- --------- ----------------------------------------- 11/19/01 Key BTL Large Volume submitted, not DQE'd This is information regarding line: SAVE 1-5; HYDROS-4 ExpoCode: Cruise Date: 1987/11/23 - 1989/04/19 From: KEY, ROBERT Email address: key@princeton.edu Institution: PRINCETON Country: USA The file: SAVE.dat - 209411 bytes has been saved as: 20011112.075818_KEY_SAVEÝ1_5;ÝHYDROS_4_SAVE.dat in the directory: 20011112.075818_KEY_SAVEÝ1_5;ÝHYDROS_4 The data disposition is: Public The bottle file has the following parameters: CTDPRS, REVTMP, SALNTY, OXYGEN, SILCAT, NITRAT NITRIT, PHSPHT, C14, C14ERR, TCARBN, ALKALI The file format is: Comma Separated Values The archive type is: NONE - Individual File The data type(s) is: Summary (navigation) Bottle Data (hyd) The file contains these water sample identifiers: Cast Number (CASTNO) Station Number (STATNO) Bottle Number (BTLNBR) KEY, ROBERT would like the following action(s) taken on the data: Place Data Online Any additional notes are: WOCE flags have been assigned and are included, however, th QC was NOT thorough, especially with respect to routine hydro data. 12/06/01 Key DELC14 Data QC'd to WOCE standards The primary ref. for the SAVE/HYDROS C14 data is: S.A.V.E. Radiocarbon South Atlantic Ventilation Experiment 1987-1989 Tritium Laboratory Data Report No. 18 H. gote Ostlund and C. Grall Univ. OF Miami Rosenstiel School of Marine and Atmospheric Science Miami, FL 33149 Sept. 1992 84pp I will try to do a more complete QC of SAVE by the end of the year, but this should only amount to a footnote saying that the data have been QCed to WOCE standards. DATE CONTACT DATA TYPE DATA STATUS SUMMARY -------- ----------- --------- ----------------------------------------- 12/20/01 Uribe CTD Website Updated; Exchange file online CTD has been converted to exchange using the new code and put online. 12/20/01 Hajrasuliha CTD Data consistency check done 07/10/02 Huynh DOC Website Update; PDF & TXT cruise reports online 03/20/03 Kozyr TCO2/PCO2 Submitted I noticed that the carbon data for A16S (R/V MELVILLE CR SAVE-5/HYDROS-3 WHP-ID A16S 23 JAN 1989 - 8 MAR 1989 20010417WHPOSIODM) are missing in WHPO bottle data file. I have put the data file with TCO2 and pCO2 (and pCO2TMP) values into WHPO ftp area. 04/16/03 Anderson TCO2/PCO2 Data merged into online BTL file Merged carbon data into online file. Made new exchange file, put files online. Notes on a16s carbon merging: April 16, 2003 A. Kozyr noted that the carbon data for a16s was not in the online file. (See e-mail above) Since the file he sent only had station number and bottle number to use for merging, I retrieved the a16s carbon data from his web site. • Merged TCARBN, PCO2, and PCO2TMP into the online file 20010412WHPOSIODM. • Copied QUALT1 flags to QUALT2 flags. • Changed the missing data code for TCARBN and PCO2 from -999.9 to -9.0 and for PCO2TMP from -999.90 to -9.00. • Made new exchange file, and put files online. 09/28/04 Kappa DOC PDF & Text Cruise reports expanded/updated Additions to both versions (PDF and Text): Final ODF-SIO Data Report which includes: • Bottle Data Collection, Analyses and Processing Salinity, Oxygen, Nutrients • CTD Data Collection, Analyses and Processing • References • CTD Processing Notes • Bottle Data Processing Notes • List of Participants Radiocarbon report by H. Göte Östlund and Charlene Grall CCHDO-WHPO cruise summary page (page 1) These CCHDO-WHPO Data Processing Notes Changes to PDF version only: Added standard "WOCE" outline (page 2) with links to text Updated CCHDO-WHPO station plot with improved clarity Added station track from author's original report Added scatter plots (nutrients) Added figures for radiocarbon report