A. Cruise Narrative: A03 A.1. Highlights WHP Cruise Summary Information WOCE section designation A03 Expedition designation (EXPOCODE) 90CT40_1 Chief Scientist/affiliation Vladimir Tereschenkov, SOI* Dates 1993 SEP 11 - 1993 NOV 21 Ship R/V Professor Multanovskiy Ports of call St. Petersburg-Hamburg-Woods Hole Woods Hole-Hamburg-St. Petersburg Number of stations 133 38°14.36'N Geographic boundaries of the stations 08°31.58'W 73°40.36'W 36°11.06'N Floats and drifters deployed none Moorings deployed or recovered none Contributing Authors S. Dobroluybov Ev. Yakushev S. Borodkin V. Konnov *Shirshov Institute of Oceanography ~ Krasikova 23 ~ Moscow 117218 Russian Federation Fax: 7-095-124-5983 ~ email: rocc@sovam.com B.1 CRUISE SUMMARY Cruise Track The cruise track and station location are shown (in PDF doc.). SAMPLING ACCOMPLISHED Water sampling on the cruise included measurements of salinity both by CTD and water bottle samples, bottle sample oxygen determination, CTD and DSRT temperature, CTD and UDSRT pressure, nutrients (silicate, nitrate, nitrite and phosphate). TYPE AND NUMBER OF STATIONS During the occupation of A3 section a total of 133 CTD/rosette stations were occupied using 24-bottle rosettes. 98 XBT stations were occupied on a track along 48°N. B.2 List of Principal Investigators TABLE 1: Principal Investigators for All Measurements Name Responsibility Affiliation --------------- -------------- ----------- V. Tereschenkov CTD SOI U. Reva CTD SOI S. Dobroluybov Salt MSU V. Konnov Nutrients IO RAN Ev. Yakushev Oxygen IO RAN Nutrients S. Borodkin Oxygen IO RAN B.3 PRELIMINARY RESULTS The ship departed from St. Petersburg on September 11, 1993. On September 15-17 the ship made a stop at Hamburg. During this stay some problems concerned with scientific equipment and supplies were solved due to generous assistance of oceanographers from BSH. On 21 of September two stations were occupied near 45°N 8.5°W location to test the CTD/rosette system and to define the quality of the sampling bottles and the operating state of the ship boarded equipment. The first station on A3 section was occupied on September 24. The whole section was completed in 32 days with total of 133 stations occupied. Space resolution between the stations varied from 29 n. miles in the open ocean to 8 n. miles in the boundary regions. On each station a CTD/rosette cast was carried out, that extended as close to the bottom as it was possible, considering the bottom detection uncertainties. On the up cast up to 24 water samples were taken. After the rosette was brought on board, water samples were drawn in the following order: oxygen, nutrients, and salinity. The chemical analyses were routinely conducted soon after the samples were collected. Two CTD/rosette systems were used in the cruise for the seawater temperature and conductivity profiling and water sampling collection purposes: o EG&G NBIS Mark-III CTD together with the GO Rosette equipped with 24 1.7l GO Niskin bottles o "Hydrozond-6000" CTD together with 24 position rosette with 11 PVC bottles (manufactured by Central Construction Bureau of Hydromet Instruments, Russia). Water temperature and pressure were also measured by mercury Deep-Sea Reversing Thermometers (protected and unprotected, respectively). The conductivity of the bottle water samples was determined using Guildline Laboratory Autosal 8400A and than transformed to salinity according to equation of the Practical Salinity Scale of 1978 (UNESCO, 1981). The dissolved oxygen analyses were carried out using Winkler method. The silicate and nitrate plus nitrites analyses were performed using AKEA Flowcomp 1500 auto analyzer, phosphate was determined using KFK-3. Details of calibrations, methods, techniques and accuracies are documented below. During the ships movement along the A3 line continuous depth recording has been conducted using ELAG echo sounder. On October 27-30 the ship made a call to Woods Hole for the refill. There is a pleasant and productive communications with WHOI people had occurred. After that "Professor Multanovskiy" occupied an XBT section similar to A2 line. Total of 98 Sippican T-7 probes were dropped at synoptical space resolution. On November 14-17 the ship visited Hamburg, where the XBT data and some scientific equipment were passed to scientist from BSH and the results of the first view data analyses have been discussed. On November 21 ORV "Professor Multanovkiy" returned to St. Petersburg. B.4 PROBLEMS During an occupation of station 31 (Sept. 30 1993) the Niel Brown Mark-III CTD underwater unit together with the GO Rosette slipped off the cable and was lost. After that the backup instruments were used. The problems arise since that moment were all connected with the unstable performance off the "Hydrozond-6000" sensors. Nevertheless the natural desire to get the best of the data caused the departure of the post-cruise data processing procedure from the scheduled routine. That is why the final version of the CTD data is still not available. B.5 Other Incidents of Note none B.6 List of Cruise Participants TABLE 2: Cruise Participants. Name Responsibility Affiliation --------------- ---------------- ----------- V. Tereschenkov Chief Scientist SOI A. Sokov CTD SOI Rev.Instrum. S. Pisarev CTD SOI U. Reva CTD SOI A. Andreev CTD SOI S. Grigoriev Software SOI S. Dobroluybov Salinity MSU V. Konnov nutrients IO RAN Yu. Konnova nutrients IO RAN Ev. Yakushev Salts,oxygen IO RAN S. Borodkin oxygen IO RAN V. Bulanov Hardware VNIIRO S. Yunovidov Hardware AARI A. Tarasov Watch Stander AARI A. Nikankin Watch Stander AARI C. CTD MEASUREMENTS C.1 CHIEF SCIENTIST OVERVIEW The ship occupied total of 133 hydrographic stations along A3 (WHP-ID) section. Detailed information on stations' position can be found in 90CT40.sum file. The stations were occupied from the surface to the bottom with respect to the fact that the sounding instruments in use were not equipped with the altimeter. Thus precautions were taken to avoid the contact of the instrument with the bottom, what resulted in the absence of near bottom observations. The first 31 stations were carried out using the NBIS Mark-3B CTD system owned by the VNIIRO (Russian Fishery and Oceanography Institute, Moscow Russia). The rest of the stations were occupied with the Hydrozond-6000 CTD system, possessed by the AARI (Arctic and Antarctic Research Institute, St. Petersburg, Russia). The latter instrumentation was produced by Central Design Bureau of the Hydromet Instruments (Obninsk, Russia). The switch of the CTD occurred due to unexpected loss of the NBIS CTD device, which sledded off the cable wire on the way up and drowned. The reasons of the accident were thoroughly investigated. The conclusion indicated the malfunctioning of the "frog-type" grasping mechanism and no operator fault. In any case the following information will be concerned with both measuring systems. C.2 CTD DATA COLLECTION AND PROCESSING. C.2.1 CTD DATA ACQUISITION Three channels (pressure, temperature, conductivity) were acquired by the NBIS Mark-B CTD at a data rate of 15.85 Hz and Hydrozond-6000 at a data rate of 4 Hz. The CTD signal was demodulated by a deck unit and output to an RS-232 bus interface. A 386DX IBM PC with a 120 Mb hard disk and 1 Mb RAM was used as the primary data collection device. The data from NBIS CTD has been logged in the computer using EG&G Oceansoft MkIII/SCTD Acquisition Software package. The data from Hydrozond-6000 was logged using a software package developed by SOI computer group. Each cast data were transferred to Phillips 386DX computer with 640 Mb hard disk and 8 Mb memory for further processing. A backup of all the data was stored on magnetic cartridge tape and magnetic diskettes. C.2.2 CTD LABORATORY CALIBRATIONS The manufacturer's sensor specification is given in Table 3. The pre-cruise calibration was performed only for the NBIS Mark-3B CTD. The laboratory calibration was performed by VNIIRO group, using EG&G Ocean Products calibration stand. All the standards have been certified by both the US and Russian National Standards Bureau. The post cruise calibration of the unit was impossible due to the loss of the instrument. The Hydrozond-6000 sensors were not calibrated in the laboratory water bath at all. According to the routine adopted by the Russian Hydrometeorological Service the supervisor of the CTD owner AARI, only the scheduled check and correction of the resistant bridges of the measuring circuit were fulfilled two month prior to the cruise. TABLE 3: Manufacturer CTD sensor specification. NBIS Mark-3B Hydrozond-6000 Sensors resolution accuracy resolution accuracy ------------------- ---------- -------- ---------- -------- Pressure, dbar | % 0.1 6.5 1 0.5% Temperature, ¯C 0.0005 0.005 0.01 0.02 Conductivity, mS/cm 0.001 0.005 0.01 0.03 C.2.2.1 PRESSURE TRANSDUCER CALIBRATION NBIS CTD Paine Instruments pressure transducer was calibrated in a temperature controlled bath by comparison with the pressures generated by an EG&G Chandler Engineering 58-001J-T-1 piston pressure gage. The calibration tests showed the accuracy of the CTD sensor of ± 2 dbar in a pressure range 0-9000 psi with respect to the loading - unloading hysteresis. C.2.2.2 PRT TEMPERATURE CALIBRATION The NBIS CTD Rosemount PRT temperature sensor was calibrated in a temperature controlled bath by comparison to a standard PRT used in EG&G Ocean Products calibration system. The latter was checked against the known water and diphenyl ether triple point cells. The calibration tests showed the agreement of the two sensors readings within ± 0.002°C. C.2.3 FIELD CALIBRATION As long as Hydrozond-6000 sensors were not calibrated at the laboratory, the at sea calibration was the only one applied to the sensors readings, based on the comparison with the bottled salinity and the pressure and temperature observations obtained by thermometric method using mercury DSRT and UDSRT. Same means were used to control the NBIS pressure and temperature sensor performance and to calibrate the NBIS CTD conductivity sensor. The field calibration routine was performed by producing a polynomial fit of the CTD pressure, temperature and conductivity readings with the appropriate measurements of the onboard salinometer and reversing thermometers. The CTD values obtained by averaging the CTD sensors readings taken during 15 seconds before the bottles were fired and thermometer racks reversed. The bottles not equipped with the thermometer racks were kept at the sampling level for 30 seconds, otherwise they stayed unmoved for 3 minutes, the time interval adequate for thermometers to stabilize. All the data selected for calibration was subjected to 2.8 standard deviation rejection. These procedures require an established level of confidence in the auxiliary observational tools. C.2.3.1 DSRT AND UDSRT During the cruise the DSRT and UDSRT produced by Gohla Precision (Kiel, Germany) were implemented (see Table 4). The possibility to utilize the temperature and pressure values determined by these instruments for needs of CTD calibration is based on the research of Quadfasel, Verch and Langhof (1991). One of the advantages of the mercury thermometers is their highly stable time behavior. The time drift of the measurements is shown to be only ± 1.4 mK/year. Calibration of the Gohla Precision instruments is done with an accuracy of 0.001°C. TABLE 4: Gohla Precision DSRT and UDSRT summary. Instrument Serial # Range Etching Calibration date ---------- -------- -------- ------- ---------------- DSRT 11459 -2 + 6 0.01 17.07.90 11709 -2 + 6 0.01 17.01.92 11818 -2 + 6 0.01 25.11.91 11817 -2 + 6 0.01 25.11.91 11816 -2 + 6 0.01 25.11.91 11738 -2 + 16 0.02 16.01.92 11739 -2 + 16 0.02 21.11.91 11741 -2 + 16 0.02 21.11.91 11835 -2 + 16 0.02 21.11.91 11959 -2 + 35 0.1 19.11.91 11960 -2 + 35 0.1 19.11.91 11626 -2 + 35 0.1 09.11.90 12067 -2 + 35 0.1 19.11.91 UDSRT 11519 +30 + 60 0.1 16.05.91 11520 +30 + 60 0.1 16.05.91 11521 +30 + 60 0.1 16.05.91 11428 -1 + 35 0.1 30.05.90 11427 -1 + 35 0.1 30.05.90 11426 -1 + 35 0.1 30.05.91 11492 -2 + 60 0.2 07.06.90 11491 -2 + 60 0.2 07.06.90 11695 -2 + 60 0.2 27.11.91 The manufacturer claims the accuracy of the thermometers to be 0.5 of the etching interval. But with the experienced observers we believe it to be 0.2 of the etching interval. The field test of the thermometers was performed on a test station. All the thermometers were reversed on the same level within the homogeneously stratified water layer. The standard deviation of the high resolution DSRT was 0.005°C. These instruments were mainly used for calibration purposes. The scatter of the low-resolution thermometers was within ± 0.01°C. These DSRT were used in 1-2 racks put above the thermocline. Same tests were performed several times during the cruise with similar results. At the first 31 station the pairs of DSRT were changed all the time. And in all cases the difference between the measured temperatures was less than the warranted accuracy. After that the combinations of the DSRT stayed permanent. The differences between the reading of the DSRT and UDSRT in the same rack didn't change with time confirming the time stability of the instruments. The variations between the pressure measurements carried out by different UDSRT agreed within 0.3% of the depth. The thermometric measurements of pressure and temperature were supervised by A. Sokov (now at P.P. Shirshov Institute of Oceanology). C.2.3.2 BOTTLE SALINITY The water sample salinity were measured with a Guildline Autosal Model 8400A salinometer N54403 that was standardized daily (usually twice) with IAPSO Standard Sea Water Batch 115. The last calibration of the salinometer was carried out by Guildline representation on May 1991. All measurements, initial quality control and shipboard data-processing were performed by S.A. Dobroliubov (Moscow State University). A total of 3094 water samples from 132 stations (more than 23 samples per each) were analyzed for salinity including 80 replicates. Salinity samples were collected strictly according to WHP Manual (Stalcup, 1991). The published accuracy of the 8400A salinometer according to Operating Manual is 0.003 psu. But if measurements are made in a laboratory with quasi-constant temperature ±1°C (as in our case) it should be possible to reach the precision better than 0.001 psu. All the procedures to attain highest accuracy described in the Operating Manual (section 6.5) were performed. Mean difference between standardizations was 0.0004 psu. These shifts were linearly interpolated between the first and the last sample readings. The salinometer readings were inverted to salinity in accordance with Guildline Operational Manual formula and Practical Salinity Scale-1978. Autosal calibration control was performed weekly with a simple suppression switch check described by Stalcup (1987). This check showed the absence of any discontinuities. Another data control included replicate analysis. 21 pair of samples were collected from the same 1.8 l Niskin bottles at different stations. Module mean difference between the salinities of these pairs was 0.0007 psu with standard deviation 0.0006 psu. The same procedure was promoted with Hydrozond- 6000 1.2 l bottles. 59 replicates were collected including two calibration casts and its analysis revealed module mean salinity difference of 0.0009 psu with standard deviation 0.0007 psu. From this point of view the difference in measurements' accuracy with both types of bottles was considered to be statistically insignificant. C.2.3.3 NBIS MARK-3B VS. THERMOMETRIC PRESSURE AND TEMPERATURE Statistics of NBIS Mark-3B temperature and pressure comparison with DSRT (temperatures) and UDSRT (pressure) observations is presented in Table 4. The good agreement is a justification for DSRT and UDSRT measurements application to Hydrozond-6000 calibration. C.2.3.4 HYDROZOND-6000 PRESSURE CORRECTION The Hydrozond-6000 CTD uses three separate pressure sensors each working in a certain pressure range without any overlaps. The switch between the sensors occurs at prescribed pressure readings that are 320 and 1780. The switch between the sensors was accompanied by a jump in pressure readings. So for each station these jumps were determined and eliminated by applying a local pressure offset to the preceding readings in order to ensure the continuous records. The analyses of the initial pressure records also revealed consistent peculiarities in the sensor behavior. All the records had a noticeable "silence" zone. That is by reaching certain pressure (reading) 1000 and 3100 the readings didn't change for approximately 40 and 250 seconds consequently. To avoid this effect and maintain the gradual pressure change, the averaged pressure time increment was calculated for periods of 30 seconds before and after the "silence" zone. Than the difference between the time increment was linearly distributed for time interval corresponding to the constant pressure segment and used to model the real pressure behavior. The pressure offset that occurred at the end of the "silence" zone was applied to the preceding pressure readings. C.2.3.5 HYDROZOND-6000 PRESSURE CALIBRATION As long as all the bottle samples were drawn during the CTD up cast and without a lab pressure calibration we had to trip some bottles with the thermometers racks on the way down in order to determine the pressure loading calibration. This has been done during the occupation of three special pressure calibration stations and several times with one bottle fired on a downcast during the routine work. All together 39 observations were selected to perform the loading pressure calibration. Fitting of (both loading and unloading) CTD and thermometric pressures was done by a third order polynomial. Special attention was paid to the detected Hydrozond-6000 pressure hysteresis. To obtain the up cast pressure the following formula was used: Pcor = Pmes - (ofs dn + (ofs up-ofs dn)/Pmax*(Pmax-Pmes) ) where: Pcor - corrected pressure Pmes - measured pressure Pmax - maximum downcast pressure ofs dn - downcast pressure offset ofs dn - upcast pressure offset Summary of CTD and thermometric pressure comparison can be found in Table 6. C.2.3.6 HYDROZOND-6000 TEMPERATURE CALIBRATION The CTD temperatures were calibrated against the DSRT measurements using the 3- order polynomial. The time drift of the CTD sensor was detected and taken care of by applying a linear correction. Summary of CTD and thermometric temperature comparison can be found in Table 6. C.2.3.7 NBIS MARK-3B AND HYDROZOND-6000 CONDUCTIVITY CALIBRATION The Hydrozond-6000 CTD conductivity measurements were calibrated against the Autosal bottle conductivity measurements using the 3-order polynomial. For NBIS Mark-3B the second order polynomial has been used. The both CTD conductivities were corrected for station and pressure dependence. Summary of CTD and bottle salinities comparison can be found in Tables 5 and 6. TABLE 5: Bottle vs NBIS Mark-3B CTD statistical summary. pressure Mean Standard #values Parameter range (dbar) difference deviation in mean ----------- ------------- ---------- --------- ------- all pressures -0.07822 4.03609 23 Pressure press < 1500 -0.95554 5.45300 9 press > 1500 0.48477 2.89719 14 all pressures -0.00040 0.00577 26 Temperature press < 1500 -0.00069 0.00664 12 press > 1500 -0.00016 0.00514 14 all pressures -0.00022 0.00307 471 Salinity press < 1500 0.00045 0.00377 220 press > 1500 -0.00081 0.00213 251 TABLE 6: Bottle vs Hydrozond-6000 CTD statistical summary. pressure Mean Standard #values Parameter range (dbar) difference deviation in mean ----------- ------------- ---------- --------- ------- all pressures -0.45583 13.87512 120 Pressure press < 1500 2.94694 9.30399 49 press > 1500 -2.80423 15.94445 71 all pressures -0.00053 0.00943 103 Temperature press < 1500 -0.00251 0.01110 28 press > 1500 0.00021 0.00859 75 all pressures 0.00036 0.00981 1756 Salinity press < 1500 0.00050 0.00997 832 press > 1500 0.00013 0.00966 924 C.3 CTD PROCESSING C.3.1 NBIS MARK-3B PROCESSING The NBIS Mark-3B station records were processed by procedures adopted in BSH, Germany. o the pressure offset (on deck the pressure sensor readings) is subtracted from the record. o starting cycles corresponding to time period of conductivity sensor adaptations to water conditions are eliminated from the record o pressure records are smoothed by Bezier cubic splines in order to digitize pressure beyond the sensor resolution and to evaluate the package speed o time lag correction is applied to pressure and conductivity due to inertion of the PRT sensor. The time lag was determined by deriving the least spiking salinity profile. o cycles corresponding to CTD upward or extreme movements (less are than 0.4 dbar/sec and greater than 2.5 dbar/sec) are eliminated o temperature and conductivity are twice subjected to median filter (window 31) o temperature and conductivity are subjected to running mean filter (window 21) o 2 dbar averaging is performed in accordance to the scheme suggested by R. Millard and Keqi Yang (1992). C.3.2 HYDROZOND-6000 PROCESSING The routines applied to HydrozondÄ6000 CTD data were based on traditional procedures of signal digital processing adopted for specific features of the measuring device. The peculiarity of the Hydrozond-6000 is the three pressure sensors working at different pressure ranges. The sensors differ by sensitivity and as a consequence, they are characterized by different spectral structure of the "noise" and different "signal" to "noise" ratio. That instantly leads to a necessity of independent treatment of each sensor record and smooth merge of the records. So, at the first stage the segments of the records corresponding to each sensor were detected. Within the each segment the spike 2.5 standard deviation control against the local second order polynomial was performed for all measured parameters. The noise elimination was done by a running mean filter, with utilization of second order polynomial values as weights. The filter windows were 21,51,121 for consequent pressure sensor and 21 for both temperature and salinity sensors. Application of the polynomial has a good advantage compared to usual equal weighted running mean filters. It allows us to consider the evident temporal trends of parameters. To avoid the energy "alias-ing", that is common for the running mean filters, the second filter run has been done, with changed window, which was chosen as 3/4 of the initial. This changes the frequency characteristic of the filter in such a manner, that now the extremes match the zeroes of a first run filter. The rest of the processing was in an agreement with R. Millard and K. Yang (1992). 10 dbar averaging has been selected to maintain statistics within the averaging bin, considering the sampling rate of 4Hz. C.4 GENERAL COMMENTS/PROBLEMS The temperatures measured by CTD are all in ITS-68. On station 5, the sounding was performed with unremoved cover of the conductivity sensor. The station was repeated at the same location (station 6). At stations 23, 24, 25, 26, 27, 28, 29, 30, 31 artificial staircases were detected at depth 1200 - 2000 db. For these data digital smoothing polynomial filter of 2-order was applied. On several stations conductivity sensor of Hydrozond-6000 didn't perform well. This was detected by TS relation analyses and comparison with historical data (RV Atlantic-II, 1981 and RV Chain, 1959). For these reasons data from stations 70 and 71 was rejected completely. Observations at station 55 in 981-1350 dbar layer were omitted, as well as the observations at station 69 in a layer from the surface to 3570 level. Near bottom records at stations 32, 38, 62, 84, 121 were eliminated due to same argumentation. Some suspicions exist concerning the observations in 2000 - 2300 dbar layer of station 101, although the data is reported. Special caution must be attributed to all the Hydrozond-6000 data in the deep layers. Some low-scale salinity variations were spotted in almost all the profiles. The origin of these fluctuations are still under consideration, so no corrections were applied to the data. Therefore it must be kept in mind that the data must be expertly treated in accordance with the specific individual research goals. C.5 MEASUREMENT TECHNIQUES AND CALIBRATION C.5.1. Salinity Analyses (S. Dobroluybov / Moscow State University). EQUIPMENT AND TECHNIQUE The water sample salinity were measured with a Guildline Autosal Model 8400A salinometer N54403 that was standardized daily (usually twice) with IAPSO Standard Sea Water Batch 115. The last calibration of the salinometer was carried out by Guildline representation on May 1991. All measurements, initial quality control and shipboard data-processing were performed by S. Dobroliubov (Moscow State University). A total of 3094 water samples from 132 stations (mean value - more than 23 samples per each) were analyzed for salinity including 80 replicates. SAMPLING PROCEDURE Salinity samples were collected strictly according to WHP Manual (Stalcup, 1991). BSH silicate-glass sample bottles with a capacity of 200 ml and separate cones for sealing the caps were thoroughly rinsed three times, filled to the shoulders, dried and moved to the laboratory with controlled temperature. Time lag between sampling and analysis usually varied from 4 to 20 hours, but sometimes it was decreased to 2-3 hours by immersing the sample bottles to the water bath. The number of samples exceeded analyzed ones on 520 bottles transported to BSH, Hamburg for Autosal intercalibration purposes. SAMPLE MEASUREMENTS Salinometer laboratory was the best temperature controlled room onboard though this control wasn't automatic. Nevertheless attempts to maintain the constant temperature during the section operation (24.09-25.10.1993) were successful with one exception 12.10 when air conditioning was remounted. Laboratory temperature during salinometer operation ranged mainly between 26.0 and 27.3°C with mean value 26.8°C and standard deviation 0.5°C. The published accuracy of the 8400A salinometer according to Operating Manual is 0.003 psu. But if measurements are made in a laboratory with quasi-constant temperature (+1°C) it should be possible to reach the precision better than 0.001 psu. All the procedures to attain highest accuracy described in the Operating Manual (section 6.5) were performed. Only twice the shift between the two daily standardizations exceeded 0.001 psu at the time of lowest and highest laboratory temperatures. Mean difference between standardizations was 0.0004 psu. This shifts linearly interpolated between the first and the last sample readings. All the samples were measured with Autosal bath temperature 27°C. The salinometer readings were inverted to salinity in accordance with Guildline Operational Manual formula and Practical Salinity Scale-1978. DATA QUALITY CONTROL Control operations contained four parts. The first one - Autosal calibration control that performed weekly with a simple suppression switch check described by Stalcup (1987). This check showed absence of any discontinuities. Full calibration procedure was not completed during the cruise. Another type of data control consisted of replicate analysis. 21 pairs of samples were collected from the same 1.8 l Niskin bottles at different stations. Module mean difference between the salinities of these pairs was 0.0007 psu with standard deviation 0.0006 psu. The same procedure was promoted with Gydrozond 1.2 l bottles. 59 replicates were collected including two calibration casts and its analysis revealed module mean salinity difference of 0.0009 psu with standard deviation 0.0007 psu. From this point of view the difference in measurement accuracy with both types of bottles was considered to be statistically indistinguishable. The third type of quality control included plotting scatter potential temperature-salinity diagrams and comparisons with historical data from International Geophysical Year and Long Lines 36°N sections. All measurements sharply deviated from the subsequent climatic mean deepwater trend were assigned low quality flags. In order to intercalibrate salinity measurements 520 sample bottles were collected and carried to BSH, Hamburg. These samples were randomly distributed below 2000 m. Such procedures allow checking Autosal calibration by indirect method. REFERENCES Guildline Instruments, 1984. Operating and Technical Manual for 'Autosal' Laboratory Salinometer Model 8400 A. Stalcup M.C.,1991. Salinity Measurements. WHP Operations and Manuals. WOCE Office. D. Bottle Data D.1 OXYGEN ANALYSIS (Ev. Yakushev and S. Borodkin) P.P. Shirshov Institute of Oceanology Russian Academy of Science The dissolved oxygen analysis was performed by analysts from P.P. Shirshov Institute of Oceanology, Moscow (IORAN) following the Winkler method. The procedures for the calibration of volumetric glassware, for the preparation of reagents and for water sampling was similar to corresponding IORAN technique (Chernyakova, 1991). The procedure for calculations of oxygen concentrations based on whole bottle formulas considering the oxygen pickling temperature presented in WHPO publication (Culberson, 1991). The parallel measurements of the oxygen concentrations with reagents prepared by Culberson and Chernyakova technique hadn't shown the significant differences. Therefore the IORAN technique of reagent preparation was accepted. The sampling procedure hadn't differences in both modifications. EQUIPMENT AND TECHNIQUE Reagents The manganese chloride reagent was prepared by dissolving 500 g of analytical grade MnCl2 x 4H2O in 600 ml of distilled water. This solution was filtered through the paper filter and made up to 1 liter. The alkaline iodate reagent was prepared by dissolving 700 g of analytical grade KOH in 700 ml of distilled water. If necessary, the solution was filtered. 300 grams of KI were separately dissolved in 450 ml of distilled water. Both solutions were mixed together. The solutions of MnCl2 and KOH/KI reagents were stored in the laboratory. Small portions of these reagents were added before the stations to the plastic bottles were taken on deck. The sulfuric acid (20%) solution was prepared by adding of one volume of analytical grade H2SO4 (density 1.84 g/cm3) to four volumes of distilled water. 0.02N sodium thiosulfate was prepared by dissolving of 25.0 g of Na2S2O3 x 5H2O in 5 liter of fresh prepared distilled water. The starch solution (0.5 %) was prepared by adding 0.5 g of soluble starch to 100 ml of boiling distilled water. Then the solution was boiled for a minute. The reagent blank was determined according to the following procedure. Two ml of sulfuric acid, 1 ml of KI/KOH and 1 ml of manganese chloride solutions were added to 96 ml of distilled water and thoroughly stirred after each addition. After the addition of 1 ml of starch solution a blue color didn't appear in all cases during the cruise. The titration was not necessary, and the value of blank accepted was equal to 0. INSTRUMENTS The oxygen flasks with capacity between 90-140 ml were weight calibrated with accuracy of 0.01 ml in IORAN hydrochemical laboratory (Moscow). A 10 ml weight calibrated automatic burette with scale division every 0.02 ml was used to dispense the thiosulfate. The burette constant errors changed in different ranges from 0.63 ml to 9.54 ml from -0.01 ml to + 0.04 ml. A weight calibrated 10 ml pipette (exact volume 10.00 ml) was used to dispense the potassium iodate solution during standardization. The 1 ml plastic automatic reagent dispensers for KI/KOH and MnCl2 solutions were calibrated by dispensing 1ml ten times into 10 ml graduated cylinder. The calibration checked during the work several times. A magnetic stirrer was used to thoroughly mix the sample during titration. SAMPLING PROCEDURE A six-inch piece of tygon tubing slipped over the outlet valve of the water sampler was used as the drawing tube. The oxygen flasks were rinsed three times with sample water prior to filling. The flasks were overflowed at approximately 3 bottle volumes of sample water. The MnCl2 and KOH/KaI reagents were added immediately after sampling with the dispensers. The stoppers were carefully placed in the bottles to avoid the introduction of the air bubbles. The flasks were carefully shaken (at least 15 energetic turns) and were stored in the laboratory while the precipitate settled on the bottom (about 30 minutes). The temperature of the collected water was measured using a digital thermometer connected to the tygon tubing used for the nutrient sampling just when the oxygen flask was pickling. The precipitate dissolved when 2 ml of sulfuric acid solution had been added. A magnet was placed in the flask and the flask was placed on a magnetic stirrer. The titration provided using 10-ml automatic burette. The sample was titrated with the thiosulfate solution until it became a light straw color. 1 ml of starch added as an indicator and titration continued up to elimination of the blue solution. CALCULATIONS OF OXYGEN CONCENTRATIONS The procedure of calculations of the oxygen concentrations followed one described by Culberon (Culberson, 1991). A corresponding Turbo C program had been written during the cruise. The oxygen pickling temperature and the AUTOSAL salinity determination results were used for the conversion of volumetric to weight concentrations. TABLE 7: Duplicate Samples Collected on Multanovskiy Cruise 40. Samples were collected from single Niskin bottles. The differences between the oxygen measurements made on these duplicate samples are shown. The standard deviation of the oxygen differences is 0.9 µmol/kg (0.0005 ml/l). Station Flask Oxy.Diff. Station Flask Oxy.Diff. Number Number ml/l µmol/kg Number Number ml/l µmol/kg ------- ------ ---- ------- ------- ------ ---- ------- 4 43,17 0.03 1.2 69 1,10 0.00 0.1 4 15,18 0.04 1.8 69 3,12 0.01 0.6 4 16,19 0.00 0.2 69 24,13 0.01 0.4 9 10,46 0.02 0.8 69 5,14 0.02 0.7 10 19,46 0.02 1.0 69 6,16 0.04 1.8 10 20,47 0.02 0.6 69 8,17 0.05 2.4 15 23,24 0.04 1.5 69 9,18 0.01 0.1 16 23,24 0.03 1.1 72 19, 1 0.01 0.2 19 17,46 0.06 2.7 72 20, 2 0.01 0.2 19 16,47 0.03 1.6 76 22,23 0.05 1.9 32 15, 1 0.01 0.3 77 19,13 0.00 0.1 33 20, 1 0.02 1.1 77 20, 1 0.00 0.1 35 19, 1 0.01 0.3 80 19, 1 0.01 0.7 36 18, 1 0.03 1.4 80 9,10 0.02 1.1 39 20, 1 0.03 1.2 81 19, 1 0.00 0.2 40 20, 1 0.00 0.2 81 20, 2 0.01 0.1 41 18, 1 0.02 1.0 84 20, 1 0.01 0.3 44 23, 1 0.02 1.1 85 19, 1 0.00 0.3 45 20, 1 0.01 0.6 88 19, 1 0.01 0.5 49 21, 1 0.02 0.9 88 20, 2 0.00 0.1 50 10,22 0.02 0.8 89 19, 1 0.02 1.0 50 8,21 0.02 1.0 92 10, 1 0.01 0.4 51 10,23 0.03 1.6 95 19, 1 0.01 0.3 54 20, 2 0.01 0.5 98 19, 2 0.01 0.2 54 19, 1 0.01 0.2 98 20, 1 0.01 0.3 54 21,24 0.00 0.1 98 1, 1 0.00 0.0 55 19, 1 0.02 1.1 101 19, 1 0.00 0.2 58 5,12 0.04 1.9 101 20, 2 0.01 0.4 59 19, 2 0.00 0.1 101 21, 3 0.03 1.2 62 8,16 0.03 1.3 105 19, 1 0.04 1.7 62 7,15 0.00 0.2 105 20, 2 0.00 0.1 62 6,14 0.06 2.4 108 19, 1 0.02 0.9 62 5,13 0.01 0.4 108 20, 2 0.00 0.0 62 24,12 0.03 1.2 113 19, 1 0.03 0.9 62 2,10 0.02 0.6 113 20, 2 0.01 0.4 62 1, 9 0.02 1.1 114 19, 1 0.01 0.6 63 19, 1 0.02 0.7 114 20, 2 0.02 0.6 63 20, 2 0.00 0.1 119 19, 1 0.00 0.1 66 19, 1 0.00 0.1 119 20, 2 0.01 0.5 66 20, 2 0.01 0.2 120 19, 1 0.01 0.3 CALIBRATIONS AND STANDARDS The potassium iodate standard solution (0.020 N) was prepared using 0.7134 g of twice crystallized dried at 105°C analytical grade KIO3 weighted in the IORAN Laboratory of hydrochemistry (Moscow). This KIO3 was dissolved and making up to 1 liter with distilled water using in 1-liter glass volumetric flask. The solution was stored in a glass bottle with ground glass stopper in the refrigerator. For calibration it was prepared a solution by adding of 10.00 ml of potassium iodate standard solution with weight calibrated 10 ml pipette to approximately 80-100 ml of distilled water in an oxygen flask. Two ml of sulfuric acid, 1 ml of KI/KOH and 1 ml of manganese chloride solutions were consequently added and the solution thoroughly stirred after each addition. Then the solution was titrated with thiosulfate using automatic burette. The difference between three-four parallel titer determination was during the cruise less then 0.02 ml. The thiosulfate titer checked during the cruise every second day. The replicate dissolved oxygen samples were collected from the single 1 liter Niskin bottles during the cruise every day and titrated to asses the precision of the dissolved oxygen measurements. Table 7 shows these results. The standard deviations of the two tests was 0.918 µmol/kg (0.0005 ml/l) indicated the precision was about 0.4%. On the first station of the voyage the Culberson and Chernyakova reagent preparing modifications were parallel used. The differences in reagent solution concentrations are shown in Table 7. The soviet made REACHEM reagents used in the solution's preparation. 22 measurements collected from separate Niskin bottles tripped at the same depth indicated that a difference less then 0.05 ml/l. A very short range didn't allow calculation of the deviation. COMPARISON WITH HISTORICAL DATA To check accuracy we compared our results with the data of Atlantis-2 cruise (1981). The present data set agrees with the old within our reproducibility. The maximum oxygen concentrations (greater than 6.5 ml/l) were observed in bottom layer of the western part (latitude 73°W) of the section. The minimum oxygen concentrations (less than 3.3 ml/l) were observed in the oxygen minima layer in the western part of the section. The obtained now picture was more sharp than in 1981 because the water samples were collected not in standard levels but in the hydrophysical extremum levels which depth determined before the sampling by the temperature and salinity soundings. D.2 NUTRIENT ANALYSES (Ev. Yakushev and V. Konnov) P.P. Shirshov Institute of Oceanology Russian Academy of Science EQUIPMENT AND TECHNIQUE The nutrient analyses were performed by a team of analysts from P.P. Shirshov Institute of Oceanology (Moscow) and Arctic and Antarctic Research Institute (S- Petersburg) using an ACE AutoAnalyzer model and KFK-3 photoelectric photometer. The methods for silicate acid, nitrates plus nitrite and nitrite were those given in AKEA manual (DATEX AKEA, 1978). The method for phosphate was an adaptation of Murphy-Riley method for photometers (Modern Methods, 1992). The Photometer KFK-3 (Photoelectric Photometer, 1992) was made by the Optico Mechanical Plant in Sergiev Posad (ZOMZ). KFK-3 provides measurements in spectral diapason from 315 to 990 nm with absolute error no greater than 0.15%. The 100 mm cuvets were used to analyze both phosphate and nitrite. SAMPLING PROCEDURE Sampling for nutrients followed that for dissolved oxygen on average 15-30 minutes after the casts were on deck. Samples were drawn into 500 cm3 high- density polyethylene, narrow mouse, screw-capped bottles. Then they were immediately drawn into 50 ml Nessler cylinders (for phosphate and nitrite analyses) and introduced into the AKEA sampler (for silicate acid and nitrate plus nitrite analyses) by pouring into 4 cm3 polysteren cups which fit the AKEA sampler tray. The 500 cm3 bottles, 50 ml Nessler cylinders and 4 cm3 cups were rinsed three times prior to filling. Analyses routinely were begun within 5-15 minutes after the 500 cm3 bottles were filled and completed within additional hour and a half. PROCEDURE OF ANALYSES PHOSPHATES A method based on the Koroleff (Koroleff, 1972) proposals was used. The produced color intensity is measured with KFK at 885 nm in a 100 mm cell. REAGENTS. REACHEM reagents were used: Molybdate reagent Ammonium heptamolybdate (a.g.) 15.0 g Distilled water, q.s. 500.0 ml Sulfuric Acid Sulfuric acid 5 N Antimonyl Potassium Tartrate Antimonyl potassium tartrate 0.34 g Distilled water, q.s. 250.0 ml Ascorbic Acid Ascorbic Acid 5.4 g Distilled water, q.s. 100.0 ml Calibration and Standards. Stock standard 10 000 mkg-at P /l was prepared by dissolving of salt (1.3609 g KH2PO4) in a 1000 ml volumetric flask and diluting to volume. The salt was recrystalyzed, dried at 110oC and weighted in the IORAN Laboratory of hydrochemistry (Moscow). SILICATE. The silicate method for AKEA utilized the method introduced by Grasshoff based on the formation of B - 1:12 silico molybdic acid and its partial reduction to a blue heteropoly acid. The color intensity was measured at 880 nm with a 20nm flow cell. REAGENTS. MERCK reagents were used. Molybdate reagent: Ammonium heptamolybdate (a.g.) 10.0 g Sulfuric acid 5 N 40.0 ml Distilled water, silicate-free q.s. 1000.0 ml Complexing Composition: Oxalic acid (a.g.) 7.0 g Sulfuric acid, conc. 50.0 ml Distilled water, silicate-free q.s. 1000.0 ml Reducing Reagent: Metol (p-methylaminophenol sulfate) 10.0 mg Anhydrous sodium sulfite 12.0 g Distilled water, silicate-free q.s. 1000.0 ml Wetting Agent LEVOR I 0.25 ml Wash Solution: Sodium chloride (a.g.) 20.0 g Distilled water q.s. 1000.ml CALIBRATION AND STANDARDS Stock standard 10 000 mkg-at Si /l was prepared by dissolving of salt (0.950 g Na2SiF4) in a 500 ml volumetric flask and diluting to volume. The salt was re- crystallized, dried over concentrated sulfuric acid and weighted in the IORAN Laboratory of hydrochemistry (Moscow). Working solutions, containing 1 mkg-at Si/ml were prepared with the 0.5 ml glass weight calibrated pipette and 500 ml weight calibrated flask. The solutions were prepared in distilled water. The apparent silicate contents were corrected for a salt error, which was determined separately in the beginning and the end of the cruise, and consisted percent. NITRATES PLUS NITRITES The nitrate method for AKEA utilizes the method introduced by Wood, Armstrong and Richards whereby nitrate is reduced to nitrite by a copper cadmium reductor column. The nitrite is converted to a reddish-purple azo-dye by using sulfanilamide and N-1-naphthylethylendiamine dihydrochloride for the diazotisation. The produced color intensity is measured at 520 nm in a 20 mm flowcell. REAGENTS MERCK reagents were used. Ammonium Chloride Buffer Solution: Ammonium chloride 10.0 g Distilled water q.s. 1000.0 ml Concentrated ammonia solution Triton X-100, 15 % solution 1.0 ml Sulfanilamide: Sulfanilamide 5.0 g 37 % Hydrochloric acid 35.0 ml Distilled water, q.s. 1000.0 ml N-1-Naphthylethylendiamine Dihydrochloride: N-1-Naphthylethylendiamine dihydrochloride 0.5 g Distilled water, q.s. 1000.0 ml Wash Solution: Distilled water Copper Cadmium Reductor Column: Copper cadmium Glass reductor column Calibrations and Standards Stock standard 10 000 mkg-at N /l was prepared by dissolving of salt (1.0110 g KNO3) in a 1000 ml volumetric flask and diluting to volume. The salt was recrystalyzed, dried at 110oC and weighted in the IORAN Laboratory of hydrochemistry (Moscow). NITRITES. Nitrite is converted to a reddish-purple azo-dye by using sulfanilamide and N-1-naphthylethylendiamine dihydrochloride for the diazotisation. The produced color intensity is measured with KFK at 543 nm in a 100 mm cell. REAGENTS. REACHEM reagents were used. Sulfanilac Acid: Sulfanic acid 1.0 g Acetic acid 12 %, q.s. 300.0 ml Alpfa-Naphthylamin: Alpha-naphthylamin 0.4 g Acetic acid 12 %, q.s. 300.0 ml Griss Solution: Equal parts of Alpha-naphthylamin and sulfanic acid solutions. CALIBRATIONS AND STANDARDS Stock standard 10 000 mkg-at N /l was prepared by dissolving of salt (0.6910 g NaNO2) in a 1000 ml volumetric flask and diluting to volume. The salt was dried at 110°C and weighted in the IORAN Laboratory of hydrochemistry (Moscow). CONVERSION OF VOLUMETRIC TO WEIGHT CONCENTRATIONS The obtained values of phosphate, silicate, nitrate and nitrite were converted into weight concentrations. To calculate the seawater density there were used the salinity data obtained with AUTOSAL and temperature of seawater in the moment of the sample bottles filling (the same as at which oxygen samples were pickled). The same formulas as for the oxygen conversion were used (Culberson, 1991). According to the recommendations of Gordon (Gordon et al, 1993) the temperature of the laboratory air should be used. But it doesn't appear to be optimal: for silicates analysis the + 50° bath is used while the nitrate+nitrite analysis is provided without a bath under the plastic cover (the similar for AKEA, TECHNIKON and ALPKEM). Additionally in our case phosphate and nitrites were measured using the 50-ml Nessler cylinders with relatively large volumes of water. The temperature at which the samples were drawn had been accepted because water samples from bottles of relatively large volume (about 250 ml) were drawn into AKEA cups and Nessler cylinders very quickly (from 1-2 to 15-20 minutes). The difference between the water samples from deep (less then 10°C) and surface (up to 27°C) remained even in an hour, when the phosphate samples were taken into photo-colorimeter. Nevertheless the nutrient laboratory temperature was measured and results are given in Table 8. The row of temperature at which oxygen was pickled is given in the WHP massive as an additional row. TABLE 8: Nutrient laboratory temperatures for each station. Stn Date Time T°C Stn Date Time T°C Stn Date Time T°C --- ------ ---- --- --- ------ ---- --- --- ------ ---- --- 3 092393 2305 22 47 100593 1100 24 91 101693 1720 25 4 092493 0120 21 48 100593 1535 25 92 101693 2345 25 6 092493 0520 21 49 100593 1945 25 93 101793 0720 25 7 092493 0840 22 50 100693 0015 24 94 101793 1530 25 8 092493 1130 22 51 100693 0445 24 95 101793 2350 25 9 092493 1640 22 52 100693 0920 25 96 101893 0735 25 10 092493 2245 23 53 100693 1345 26 97 101893 1415 25 11 092593 0530 23 54 100693 2030 26 98 101893 2200 25 12 092593 1140 22 55 100793 0235 26 99 101993 0530 24 13 092593 1630 22 56 100793 0940 26 100 101993 1230 24 14 092593 2215 22 57 100793 1530 26 101 101993 2030 24 15 092693 0330 22 58 100793 2130 25 102 102093 0315 23 16 092693 0850 22 59 100893 0350 24 103 102093 0945 23 17 092693 1420 22 60 100893 0850 23 104 102093 1720 23 18 092693 1830 22 61 100893 1415 23 105 102193 0005 23 19 092693 2320 22 62 100893 1925 23 106 102193 1040 24 20 092793 0450 22 63 100993 0415 23 107 102193 1715 24 21 092793 1050 22 64 100993 1030 24 108 102293 0055 24 22 092793 1645 23 65 100993 1645 25 109 102293 0445 24 23 092793 2315 22 66 100993 2325 25 110 102293 0845 25 24 092893 0540 22 67 101093 0615 25 111 102293 1245 25 25 092893 1230 22 68 101093 1230 25 112 102293 1620 24 26 092893 2000 23 69 101093 1925 25 113 102293 2030 23 27 092993 0300 23 70 101193 0100 25 114 102393 0030 22 28 092993 0830 22 71 101193 1200 26 115 102393 0420 22 29 092993 1340 23 72 101193 1855 25 116 102393 0830 22 30 092993 2350 22 74 101293 0715 26 117 102393 1215 22 31 093093 2355 23 75 101293 1315 26 118 102393 1555 22 32 100193 1740 23 76 101293 1955 26 119 102393 2015 22 33 100293 0025 23 77 101393 0230 26 120 102493 0106 22 34 100293 1015 23 78 101393 0835 25 121 102493 0540 22 35 100293 1545 23 79 101393 1445 26 122 102493 0830 22 36 100293 2130 23 80 101393 1935 26 123 102493 1020 22 37 100393 0345 23 81 101493 0100 26 124 102493 1355 22 38 100393 0935 23 82 101493 0740 26 125 102493 1630 22 39 100393 1500 24 83 101493 1345 26 126 102493 1930 22 40 100393 2030 24 84 101493 2040 26 127 102493 2230 22 41 100493 0230 24 85 101593 0325 25 128 102593 0055 22 42 100493 0825 24 86 101593 0940 25 129 102593 0315 22 43 100493 1435 24 87 101593 1545 25 130 102593 0525 22 44 100493 2000 24 88 101593 2230 25 131 102593 0730 22 45 100593 0115 24 89 101693 0455 25 132 102593 0850 22 46 100593 0640 24 90 101693 1120 25 133 102593 1025 22 COMPARISON WITH WHOI STANDARD SOLUTIONS In Woods Hole we checked our standard solutions with the WHOI ones. Dr. Zofia Molodzinska kindly presented the standard solutions taken into the analysis procedure. The WHOI phosphate standard appeared to be 4% higher than our standard. The difference between the nitrates solutions was about 3% (the WHOI standard was higher). The silicate standards haven't differed at all. The differences for phosphates and nitrates were relatively large but correspond to the methods accuracy. It's a pity, the measurements were provided at sea after we left Woods Hole, so it was not possible to repeat them and to check the standard qualities. COMPARISON WITH HISTORICAL DATA To check accuracy we compared our results with the data of Atlantis-II cruise (1981). The present data of nitrates set agrees with the old within our reproducibility. The concentrations reached 20 µmol/l in the bottom layers of western part and 22-24 µmol/l in western part. The maximum concentration (29 µmol/l) was found in the maximum layer at depth 935 m at latitude 71°W. The minimum concentrations (0 µmol/l) observed in the surface layers. The phosphate fields also agreed with the old ones. The maximum concentrations were observed in bottom layer (1.4 µmol/l in western part and 1.6 µmol/l in eastern) and in the layer of maximum in the western part (1.7 µmol/l). The minimum concentrations (0 µmol/l) were observed in the surface layer. The silicate concentrations were found some greater than in 1981. The maximum concentrations (greater than 50 µmol/l) were observed in the bottom layer at the latitude 56°W. The concentration in the maximum layer was 24 µmol/l. The minimum values (less than 1 µmol/l) were observed in the surface waters. As for oxygen, the 1993 pattern was more contrast than 1981, because the water samples were collected not at standard levels but at the hydrophysical extremum levels which were determined during the temperature and salinity downcast soundings. REFERENCES Aminot, A. and D. Kirkwood. 1995. "Report on the results of the fifth ICES intercomparison exercise for nutrients in sea water". International Council for the Exploration of the Sea. ICES Cooperative Res. Rep. No. 213. 79 pp. Bordovskiy, O.K., A.M. Chernyakova, Ed. "Modern methods of the ocean hydrochemical investigations". P.P. Shirshov Institute of Oceanology. Moscow, 200 pp. (In Russian). Chernyakova, A.M., 1992. "The Winkler method dissolved oxygen determination. Modern methods of the ocean hydrochemical investigations". Ed. O.K. Bordovskiy, A.M. Chernyakova. P.P. Shirshov Institute of Oceanology. Moscow, pp. 22-35. (In Russian). Culberson, C.H., 1991. "Dissolved oxygen. WHP Operations and Methods" - July 1991. College of Marine Studies. University of Delaware 19716 U.S.A., 15 pp. DATEX AKEA. "Automatic chemical analysis system". Technical manual. Datex Instrumentarium OY. p.o. box 387 00101 Helsinki 10 Finland. 1978. 67 pp. Fofonoff, N. P., and R. C. Millard. 1983. "Algorithms for computation of fundamental properties of seawater". UNESCO Tech. Pap. Mar. Sci. No. 44. 53 pp. Garcia, H. E. 1996. "On the large-scale characteristics, fluxes and variability of the North Atlantic deep water and its deep western boundary current deduced from nutrient and oxygen data". Ph.D. thesis, College of Oceanic and Atmospheric Sciences. 184 pp. Gordon, L.I., J.C. Jennings, Jr. A.A. Ross, J.M. Krest, 1993. "A Suggested Protocol for Continuous Flow Automatic Analysis of Seawater Nutrients (Phosphate, Nitrate, Nitrite and Silicic Acid) in the WOCE Hydrographic Program and Joint Global Ocean Fluxes Study". College of Oceanic and Atmospheric Sciences. Oregon State University. 24 August 1993. 51 pp. Koroleff, F., 1972. "Determination of dissolved inorganic phosphorus and total phosphorus. Methods for sampling and analysis of physical, chemical and biological parameters". Cooperative research report JCES, Series A., N29, pp. 44-49. Photoelectric Photometer KFK-3. Technical description and exploitation instruction. 47 pp. (In Russian). Results of an Oxygen/Salinity Comparison Cruise in the R/V Vernadsky. WHP Office Report WHPO 92-3 WOCE Report 93/92. July 1992 Woods Hole, Mass. USA, 43 pp. Whitledge, T. E., S. C. Malloy, C. J. Patton and C. D. Wirick. 1981. "Automated nutrient analysis in seawater". Brookhaven National Laboratory, U.S. Dept. of Energy and Environment, Upton, NY. 216 pp. WHPO DATA CONSISTENCY CHECK About the '_check.txt', '_sal.ps' and '_oxy.ps' files: The WHP-Exchange format bottle and/or CTD data from this cruise have been examined by a computer application for contents and consistency. The parameters found for the files are listed, a check is made to see if all CTD files for this cruise contain the same CTD parameters, a check is made to see if there is a one-to-one correspondence between bottle station numbers and CTD station numbers, a check is made to see that pressures increase through each file for each station, and a check is made to locate multiple casts for the same station number in the bottle data. Results of those checks are reported in this '_check.txt' file. When both bottle and CTD data are available, the CTD salinity data (and, if available, CTD oxygen data) reported in the bottle data file are subtracted from the corresponding bottle data and the differences are plotted for the entire cruise (not available for this cruise). Following parameters found for bottle file: EXPOCODE DATE CTDSAL SILCAT_FLAG_W SECT_ID TIME CTDSAL_FLAG_W NITRIT STNNBR LATITUDE SALNTY NITRIT_FLAG_W CASTNO LONGITUDE SALNTY_FLAG_W NO2+NO3 SAMPNO DEPTH OXYGEN NO2+NO3_FLAG_W BTLNBR CTDPRS OXYGEN_FLAG_W PHSPHT BTLNBR_FLAG_W CTDTMP SILCAT PHSPHT_FLAG_W a03_hy1.csv -> NO2+NO3_FLAG_W found without matching parameter. All ctd parameters match the parameters in the reference station. Station #11 has a CTD file, but does not exist in a03_hy1.csv. Station #27 has a CTD file, but does not exist in a03_hy1.csv. Station #31 has a CTD file, but does not exist in a03_hy1.csv. Station #106 exists in a03_hy1.csv, but does not have a corresponding CTD file. Station #107 exists in a03_hy1.csv, but does not have a corresponding CTD file. No bottle pressure inversions found. Bottle file pressures are increasing. No multiple casts found in bottle data. WHPO DATA PROCESSING NOTES: Date Contact Data Type Data Status Summary -------- ------------ ----------- ------------------------------------------- 02/28/94 Tereschenkov SUM/DOC Submitted On disk 12/11/95 Lozovatsky CTD Problem with CTD calibration 2) AR13 1992, Sokov (GOIN). This cruise was not declared in the WHP, as Dr. Alex Sokov said (to) me. But it is possible to include the data of the cruise to the WHPO data set. The problem is the CTD calibration, because the measurements were made by HYDROZOND - Russian CTD. The data will also prepare by WNDC. By the way Dr. Sokov is the head of the center. He left GOIN for the P.P. Shirshov Institute of Oceanology. He said that it will take about 2-3 months to prepare these data. 3) A03 1993, Tereschenkov (IORAS). There is the same problem like in the p.2 - the CTD calibration. Tereschenkov is finishing this procedure and as he thinks the data will be prepared for transmission in a two weeks. 01/26/96 Tereschenkov BTL/SUM/DOC Submitted for DQE On disk 03/26/96 Tereschenkov Various Data Clarifications Questions from Cindy Ruhsam/WHOI: 1) Were there any underway measurements taken during the cruise, i.e. ADCP, bathymetry that was run, Thermosalinograph and any meteorological observations observed. * Yes, we had bathymetry observations done along the cruise track 2) Is Nitrate N+N * Yes 3) What temperature scale did you use T68 or T90 ? * We used the ITS-68 12/11/96 Tereschenkov CTD Submitted for DQE On disk; no CTDOXY 04/14/99 Kappa DOC Update pdf version created 04/30/99 Kappa DOC Update Cruise Plan added 04/24/00 Kappa Cruise ID WOCE expocode updated changed from RUCT40_1 to 90CT40_1 05/10/00 Tereschenkov CTD/BTL Data are Public By this message I would like to notify you that I have no objections to make the data public. I agree the data to be distributed among the oceanographic community with no restrictions. 05/22/00 Huynh DOC Website Updated; pdf, txt versions online Date Contact Data Type Data Status Summary -------- ------------ ----------- ------------------------------------------- 06/08/00 Bartolacci CTD/BTL Website Updated; data unencrypted I have unencrypted the following onetime cruises: A03 90CT40_1: CTD, BOT (sal, O2, and nuts only) 03/20/01 Diggs CTD/BTL/SUM Files on website Updated w/ new expocodes Updated expocodes in all CTD, Bottle, SUM and (ascii) DOC files online. Changed expocode from RUCT40/1 to 90CT40_1 06/20/01 Uribe BTL Exchange file online Bottle file in exchange format has been linked to website. 06/21/01 Uribe CTD/BTL Website updated w/ exchange files The exchange bottle file name in directory and index file was modified to lower case. CTD exchange files were put online. 12/17/01 Hajrasuliha CTD/BTL Internal DQE completed The following are results from the examminer.pl and plotter.pl code that were run on this cruise. Not all of the errors are reported but rather a summery of what was found. For more information you can go to the cruise directory, and look at the NEW file called CruiseLine_check.txt. (see "Data Consistency Check" in this doc. NO _oxy.ps and _sal.ps available for this cruise because of CTDOXY unit missing from bottlefile. * A03_hy1.csv -> NO2 NO3_FLAG_W found without matching parameter. * Station #10 has a CTD file, but does not exist in A03_hy1.csv. * Station #100 has a CTD file, but does not exist in A03_hy1.csv. * Station #101 has a CTD file, but does not exist in A03_hy1.csv. 12/18/01 Uribe CTD Website Updated w/ exchange File CTD has been converted to exchange using the new code and put online. 04/08/03 Kappa DOC Update * Added WHPO-generated "Data consistency Check" * Deleted redundant sections from earlier online pdf and text docs * Updated WHPO-generated cruise track * Updated internal links in pdf doc * Reformatted text * Updated these WHPO Data Processing Notes