Aurora Australis Marine Science Cruise AU9407 - Oceanographic Field Measurements and Analysis MARK ROSENBERG Antarctic CRC, GPO Box 252C, Hobart, Australia RUTH ERIKSEN Antarctic CRC, GPO Box 252C, Hobart, Australia STEVE BELL Antarctic CRC, GPO Box 252C, Hobart, Australia NATHAN BINDOFF Antarctic CRC, GPO Box 252C, Hobart, Australia STEVE RINTOUL Antarctic CRC, GPO Box 252C, Hobart, Australia; CSIRO Division of Oceanography, Hobart, Australia ABSTRACT Oceanographic measurements were conducted in January 1994 along WOCE Southern Ocean meridional section SR3 between Tasmania and Antarctica, and along a northward section lying between 82 and 86oE and crossing the Princess Elizabeth Trough. Additional measurements were made at mooring locations, and at a time series station. A total of 102 CTD vertical profile stations were taken, most to near bottom. Over 2000 Niskin bottle water samples were collected for the measurement of salinity, dissolved oxygen, nutrients, dissolved inorganic and organic carbon, carbon 13, dimethyl sulphide/dimethyl sulphoniopropionate, iodate/iodide, and biological parameters, using a 24 bottle rosette sampler. Measurement and data processing techniques are described, and a summary of the data is presented in graphical and tabular form. 1 INTRODUCTION Marine science cruise AU9407 of the Cooperative Research Centre for the Antarctic and Southern Ocean Environment (Antarctic CRC) was conducted aboard the Australian Antarctic Division vessel RSV Aurora Australis from January to March 1994. The first major constituent of the cruise was the collection of oceanographic data relevant to the Australian Southern Ocean WOCE Hydrographic Program, along WOCE section SR3 (Figure 1). The primary scientific objectives of this program are: 1. to estimate the interbasin exchange of heat, freshwater and other properties south of Australia, and the seasonal and interannual variability of this exchange; 2. to investigate the mechanisms responsible for the formation of deep and intermediate water masses in the Southern Ocean, and to identify the ventilation pathways that newly formed water masses follow into the ocean interior; 3. in conjunction with current meter data, to determine the importance of eddy heat and momentum fluxes in the dynamics and thermodynamics of the Antarctic Circumpolar Current south of Australia. Section SR3 was occupied twice previously, in the spring of 1991 (Rintoul and Bullister, in prep.), and in the autumn of 1993 (Rosenberg et al., 1995). The second major constituent of the cruise was oceanographic measurements along the Princess Elizabeth Trough section (PET). This is a short hydrographic section from the West Ice Shelf, at approximately 85oE, across the trough and along the ridge of the Kerguelen Plateau, connecting the earlier hydrographic section of Speer and Forbes (1994) to the Antarctic continent, and passing over an array of five current meter moorings deployed by MAFF (Dickson, 1993) (Figure 1). The primary aims of the PET measurements are: 1. to estimate the interbasin transport through the Princess Elizabeth Trough and the effect of the Kerguelen Plateau on the Circumpolar Current; 2. to investigate the distribution and properties of Antarctic Bottom waters; 3. in conjunction with other sections, provide additional constraints on the transport of heat and freshwater. Additional measurements were made at two upward looking sonar mooring sites, a single bottom pressure recorder mooring location, and at a time series station. This latter station is to be occupied several times per year in collaboration with French and Japanese scientists (P. Treguer, D. Mackey, H. Marchant, pers. comms). The cruise discussed in this report is the second in a series of Southern Ocean marine science cruises, the first being described in Rosenberg et al. (1995). The report describes the collection of oceanographic data from the two transects, and the chemical analysis and data processing methods employed. Brief comparisons are also made with existing historical data. All information required for use of the data set is presented in tabular and graphical form. 2 CRUISE ITINERARY The cruise commenced with a north to south traverse of section SR3. A bottom pressure recorder mooring (Table 4) was recovered along the way, at the northern end of the section, and replaced by a new instrument. An attempt to recover two bottom pressure recorders at the southern end of the section was unsuccessful; an additional recorder was deployed in the vicinity. Following completion of SR3, the ship steamed west to deploy two upward looking sonar moorings, then continued west to the Princess Elizabeth Trough; the PET section was traversed from south to north. Following occupation of the time series station at the end of January, the major planned oceanographic component of the cruise was completed. Through most of February, personel transfers and cargo operations were conducted at the Australian Antarctic bases (Table 1). The ship returned east to Dumont D'Urville in late February to collect trawl gear, delivered by the French supply ship L'Astrolabe from Hobart to the French base. The old bottom pressure recorders at the southern end of the SR3 section were relocated, and several unsuccessful attempts were made to trawl for the moorings. The ship then returned to Hobart. 3 CRUISE SUMMARY 3.1 CTD casts In the course of the cruise, 102 CTD casts were completed at 83 different sites along the SR3 and PET sections (Figure 1) (Table 2), plus additional locations, at a typical spacing of 30 nautical miles between sites along the sections, and with most casts reaching to within 15 m of the sea floor (Table 2). Sea ice and weather conditions did not restrict the southern extent of SR3, and the section was successfully closed off on the Antarctic continental shelf. The section was continued into the D'Urville Trough on the continental shelf, where three additional casts were taken to trace possible formation sites for Antarctic bottom water. Of the 58 different locations along SR3 where casts were taken, a shallow and deep cast pair were taken at 13 of the locations, both to increase the vertical resolution of Niskin bottle sample data, and to ensure a sufficient volume of water was obtained for the biology programs. Prior to starting the PET section, a bathymetric survey was taken up the Antarctic continental slope onto the shelf, in order to determine the best locations of casts over the shelf and slope on the northward transect. CTD casts were taken at 21 different locations along the PET transect, commencing approximately 19 nautical miles north of the West Ice Shelf. Shallow/deep cast pairs were performed at 3 of the locations. Additional CTD casts were performed at the two upward looking sonar mooring sites midway between SR3 and PET, and at the unretrieved bottom pressure recorder site at the southern end of SR3 (Figure 1). A shallow/deep CTD cast pair was taken at the time series station at approximately 63oS 71oE (Figure 1). Table 1: Summary of cruise itinerary. Expedition Designation Cruise AU9407 (cruise acronym SHAM), encompassing WOCE section SR3 and Princess Elizabeth Trough section (PET) Chief Scientists: Bronte Tilbrook, CSIRO Nathan Bindoff, Antarctic CRC Ship: RSV Aurora Australis Ports of Call Mawson Law Base Davis Casey Dumont D'Urville Cruise Dates: January 1 to March 1, 1994 Figure 1: CTD station positions for RSV Aurora Australis cruise AU9407 along WOCE transect SR3, and along the PET transect. Additional CTD sites are shown at the time series station (TS), at the 2 upward looking sonar mooring locations (ULS), and at the unretrieved bottom pressure recorder site (BPR). 3.2 Water samples from CTD casts Over 2000 Niskin bottle water samples were collected for the measurement of salinity, dissolved oxygen, nutrients, dissolved inorganic carbon, 13C, dimethyl sulphide/dimethyl sulphoniopropionate, dissolved organic carbon, iodate/iodide, and biological parameters, using a 24 bottle rosette sampler. Test samples at a few stations were drawn for the analysis of 18O, and alkalinity. Table 3 provides a summary of samples drawn at each station. Principal investigators for the various water sampling programmes are listed in Table 6a. For all stations, the different samples were drawn in a fixed sequence, as discussed in section 4.1.3. The methods for drawing the salinity, dissolved oxygen and nutrient samples are discussed in section 4.1.4. Salinity, dissolved oxygen and nutrients: Samples were drawn from most stations for salinity, dissolved oxygen and nutrient analyses. Salinity and dissolved oxygen hydrology data was further used for the calibration of CTD salinity and dissolved oxygen data; nutrient samples were analysed for concentration of orthophosphate, nitrate plus nitrite, and reactive silicate. Dissolved inorganic carbon: Samples were drawn for total dissolved inorganic carbon analysis approximately every second station. In general, salinity and oxygen properties determined the Niskin sampling strategy, thus the sampling depths were not always best suited to the resolution of dissolved inorganic carbon gradients in the top 300 m of the water column. Results from these analyses are reported elsewhere (Tilbrook, pers. comm.), and are not discussed further in this report. 13C and 18O, and alkalinity: Samples were drawn for 13C analysis from multiple depths on the SR3 transect and at the time series station; surface 13C samples were taken on both SR3 and PET. Samples for 18O analysis were drawn on the PET transect only; alkalinity samples were taken only on PET and at the time series station. These sample sets are not discussed further in this report. Dimethyl sulphide/dimethyl sulphoniopropionate (i.e. DMS and DMSP): These samples were drawn on the SR3 transect only. The data is not discussed further in this report. Dissolved organic carbon: Six SR3 locations, plus the time series station, were sampled for dissolved organic carbon. Additional surface samples were taken at several of the PET stations. Iodate/iodide: These samples were drawn on both the SR3 and PET transects; results are not discussed in this report. Primary productivity: For casts taken during daylight hours on the SR3 transect (and at the time series station), samples were drawn for analysis of primary productivity and suspended particle size. These samples were taken from the shallowest four Niskin bottles. At most primary productivity sites, a Seabird "Seacat" CTD was deployed to obtain vertical profiles of photosynthetically active radiation (p.a.r.) and fluorescence from the top part of the water column. These data are not discussed further in this report. Biological sampling: Several different analyses were performed on the biological water samples, as follows: (i) pigments (using high performance liquid chromatography) (ii) algal counts (lugols iodine fixed) (iii) cyanobacteria counts (iv) total algal counts (v) osmicated samples, for looking at ultrastructure of algae by electron microscopy (vi) cultured samples, growing various flagellates (vii) coccolith counts (viii) parmales cultures (ix) flavobacteria Biological samples were usually drawn from the shallowest four or five Niskin bottles, except for flavobacteria (up to 40 litres collected from a single depth). These data are not discussed further in this report. Table 2 (following 3 pages): Summary of station information for RSV Aurora Australis cruise AU9407. The information shown includes time, date, position and ocean depth for the start of the cast, at the bottom of the cast, and for the end of the cast. The maximum pressure reached for each cast, and the altimeter reading at the bottom of each cast (i.e. elevation above the bed) are also included. Missing ocean depth values are due to noise from the ship's bow thrusters, as discussed in Appendix 2, section A2.3. For casts which do not reach to within 100 m of the bed (i.e. the altimeter range), there is no altimeter value. For station names, TEST is a test cast, ULS is an upward looking sonar mooring site, TS is the time series station, and BPR is a bottom pressure recorder mooring location. Note that all times are UTC (i.e. GMT). CTD unit 6 (serial no. 2568) was used for all stations. station START maxP BOTTOM END number time date latitude longitude depth (dbar) time latitude longitude depth altimeter time latitude longitude depth(m) (m) (m) (m) -------------------------------------------------------------------------------------------------------------------------------------- 1 TEST 1954 1-JAN-94 44:07.13S 146:12.68E 994 1002 2030 44:07.33S 146:12.97E 1034 35.0 2114 44:07.65S 146:13.17E 1066 2 TEST 0229 2-JAN-94 44:07.44S 146:12.94E 1041 906 0312 44:07.59S 146:12.76E 1040 36.5 0406 44:07.76S 146:12.34E 1060 3 SR3 0733 2-JAN-94 43:59.97S 146:19.04E 248 212 0745 43:59.93S 146:19.18E 218 10.0 0814 43:59.82S 146:19.48E 188 4 SR3 1020 2-JAN-94 44:07.03S 146:13.35E 1015 1038 1100 44:07.14S 146:13.71E 1045 6.9 1222 44:06.61S 146:13.95E 1026 5 SR3 1437 2-JAN-94 44:22.85S 146:10.57E 2351 2346 1553 44:22.45S 146:09.88E - 16.0 1712 44:22.00S 146:09.80E - 6 SR3 1952 2-JAN-94 44:43.08S 146:02.64E 3211 3254 2105 44:43.30S 146:02.70E 3232 4.9 2230 44:43.43S 146:02.76E 3237 7 SR3 0144 3-JAN-94 45:12.92S 145:51.27E 2859 206 0156 45:12.96S 145:51.21E - - 0218 45:12.91S 145:51.17E - 8 SR3 0312 3-JAN-94 45:12.82S 145:51.28E 2859 2888 0434 45:12.76S 145:51.06E 2879 15.0 0551 45:12.45S 145:50.88E - 9 SR3 0901 3-JAN-94 45:41.87S 145:39.65E 2020 2006 0956 45:42.08S 145:40.20E 2020 15.0 1109 45:42.31S 145:41.01E 2040 10 SR3 1424 3-JAN-94 46:10.33S 145:27.67E 2776 2752 1546 46:10.28S 145:27.83E 2745 15.0 1712 46:10.36S 145:27.96E 2745 11 SR3 2030 3-JAN-94 46:39.16S 145:15.38E 3366 206 2039 46:39.15S 145:15.40E - - 2054 46:39.18S 145:15.35E - 12 SR3 2138 3-JAN-94 46:39.07S 145:15.16E 3366 3388 2247 46:38.88S 145:15.08E - 18.1 0021 46:38.84S 145:15.03E 3366 13 SR3 0324 4-JAN-94 47:08.67S 145:03.06E 4610 4526 0515 47:08.38S 145:02.19E - 22.0 0718 47:08.20S 145:01.76E - 14 SR3 1123 4-JAN-94 47:28.15S 144:54.28E 4413 4438 1258 47:28.17S 144:55.08E - 13.5 1458 47:28.03S 144:55.68E 4330 15 SR3 1727 4-JAN-94 47:48.25S 144:44.48E 3957 3972 1857 47:48.12S 144:46.07E 3915 16.0 2038 47:47.81S 144:46.68E 3853 16 SR3 0008 5-JAN-94 48:18.83S 144:32.14E 4092 206 0025 48:18.70S 144:32.25E - - 0051 48:18.41S 144:32.34E 4092 17 SR3 0206 5-JAN-94 48:18.90S 144:31.73E 4143 4132 0358 48:18.27S 144:32.50E 4143 15.0 0530 48:17.77S 144:32.95E 4206 18 SR3 0835 5-JAN-94 48:46.93S 144:19.11E 4164 156 0840 48:46.84S 144:18.99E - - 0852 48:46.75S 144:18.82E 4164 19 SR3 0912 5-JAN-94 48:46.53S 144:18.73E 4174 4150 1044 48:45.68S 144:18.40E 4192 17.0 1221 48:44.79S 144:18.15E 4040 20 SR3 1549 5-JAN-94 49:16.17S 144:05.64E 4237 4278 1715 49:15.97S 144:04.98E 4216 11.0 1850 49:15.61S 144:04.65E 4226 21 SR3 2203 5-JAN-94 49:45.09S 143:52.10E 3553 158 2213 49:45.02S 143:52.09E - - 2225 49:44.90S 143:52.14E 3553 22 SR3 2313 5-JAN-94 49:43.88S 143:52.41E 3553 3676 0020 49:43.80S 143:52.12E - 17.0 0152 49:43.53S 143:51.80E - 23 SR3 1811 6-JAN-94 50:14.06S 143:38.95E 3729 3802 1922 50:14.25S 143:39.48E 3729 13.4 2048 50:14.12S 143:40.30E 3729 24 SR3 0043 7-JAN-94 50:45.75S 143:24.88E 3853 4054 0215 50:46.03S 143:26.38E - 15.0 0348 50:46.23S 143:27.58E - 25 SR3 0604 7-JAN-94 51:01.99S 143:14.29E 3729 3842 0724 51:02.53S 143:15.04E 3833 17.6 0900 51:03.09S 143:15.73E 3755 26 SR3 1142 7-JAN-94 51:25.88S 143:02.40E 3729 3772 1322 51:26.28S 143:03.33E 3781 15.1 1457 51:26.55S 143:04.09E - 27 SR3 1724 7-JAN-94 51:50.62S 142:49.69E 3159 56 1727 51:50.62S 142:49.69E - - 1731 51:50.68S 142:49.68E 3159 28 SR3 1752 7-JAN-94 51:51.07S 142:49.83E 3729 3648 1915 51:51.65S 142:49.95E 3605 17.0 2046 51:51.93S 142:50.12E 3688 29 SR3 2338 7-JAN-94 52:15.46S 142:37.53E 3470 3458 0112 52:15.86S 142:37.21E - 15.0 0246 52:15.72S 142:37.38E - 30 SR3 0509 8-JAN-94 52:38.34S 142:23.24E 3522 154 0515 52:38.41S 142:23.24E - - 0532 52:38.68S 142:23.08E - 31 SR3 0614 8-JAN-94 52:39.41S 142:22.88E 3522 3456 0727 52:40.25S 142:22.35E - 14.7 0853 52:40.68S 142:22.08E - 32 SR3 1129 8-JAN-94 53:07.54S 142:07.94E 3170 3126 1308 53:08.97S 142:06.30E 3107 18.0 1440 53:10.02S 142:05.43E 3159 33 SR3 1708 8-JAN-94 53:34.86S 141:51.95E 2382 2516 1805 53:34.91S 141:52.01E 2538 13.0 1923 53:34.83S 141:52.36E 2486 34 SR3 2240 8-JAN-94 54:03.85S 141:35.79E 2693 2754 0002 54:03.39S 141:35.46E 2693 17.7 0117 54:03.15S 141:35.25E 2693 35 SR3 0424 9-JAN-94 54:31.99S 141:19.39E 2848 104 0428 54:31.95S 141:19.45E - - 0431 54:31.92S 141:19.42E 2848 36 SR3 0447 9-JAN-94 54:31.79S 141:19.31E 2797 2822 0549 54:31.18S 141:19.19E 2859 17.3 0711 54:30.40S 141:18.92E 2745 37 SR3 1030 9-JAN-94 55:01.16S 141:00.58E 3263 3284 1149 55:00.73S 141:00.27E 3315 17.0 1309 55:00.54S 141:00.33E 3366 38 SR3 1611 9-JAN-94 55:29.83S 140:43.65E 3988 4112 1727 55:29.23S 140:43.33E 4092 18.6 1918 55:28.42S 140:42.33E 4195 39 SR3 0118 10-JAN-94 55:55.72S 140:24.42E 3729 154 0122 55:55.72S 140:24.40E - - 0138 55:55.72S 140:24.40E - 40 SR3 0231 10-JAN-94 55:55.90S 140:24.06E 3729 3602 0352 55:56.05S 140:23.79E 3729 9.0 0524 55:56.08S 140:23.55E 3729 41 SR3 0821 10-JAN-94 56:26.28S 140:06.01E 4143 4166 1002 56:26.38S 140:05.84E - 15.8 1154 56:27.00S 140:06.61E 4143 42 SR3 1514 10-JAN-94 56:55.46S 139:50.94E 4143 4188 1639 56:55.82S 139:52.16E - 12.1 1820 56:55.92S 139:53.11E - 43 SR3 2055 10-JAN-94 57:23.02S 139:50.86E 4143 206 2105 57:23.04S 139:50.87E - - 2120 57:23.05S 139:50.84E - 44 SR3 2213 10-JAN-94 57:22.42S 139:51.08E 4143 4038 2345 57:22.02S 139:50.38E - 17.9 0125 57:21.76S 139:49.29E - 45 SR3 0411 11-JAN-94 57:51.52S 139:51.40E 4040 4172 0555 57:51.76S 139:51.84E - 14.3 0730 57:52.24S 139:51.96E 4143 46 SR3 1029 11-JAN-94 58:20.61S 139:51.16E 3988 4030 1204 58:21.13S 139:52.02E - 15.8 1344 58:21.49S 139:53.37E 3936 47 SR3 1901 11-JAN-94 58:51.28S 139:50.58E 3936 3998 2040 58:52.02S 139:50.35E - 13.9 2227 58:52.88S 139:49.48E 3853 48 SR3 0107 12-JAN-94 59:20.87S 139:50.81E 4221 4216 0256 59:21.58S 139:50.16E 4169 7.1 0420 59:22.11S 139:50.65E 4169 49 SR3 0701 12-JAN-94 59:51.58S 139:51.28E 4485 204 0709 59:51.64S 139:51.43E 4485 - 0730 59:51.89S 139:51.85E 4485 50 SR3 0827 12-JAN-94 59:51.45S 139:50.90E 4485 4532 1009 59:51.71S 139:51.72E 4485 16.0 1156 59:52.35S 139:52.05E 4485 51 SR3 1459 12-JAN-94 60:21.36S 139:50.59E 4433 4492 1653 60:22.35S 139:49.60E 4444 15.7 1849 60:22.63S 139:48.03E - 52 SR3 2141 12-JAN-94 60:51.05S 139:50.83E 4408 154 2150 60:51.10S 139:50.75E 4408 - 2206 60:51.10S 139:50.63E 4408 53 SR3 2243 12-JAN-94 60:51.14S 139:50.63E 4408 4450 0030 60:51.37S 139:50.61E 4402 16.8 0221 60:51.43S 139:51.13E 4408 54 SR3 0514 13-JAN-94 61:20.99S 139:50.92E 4351 4392 0714 61:20.89S 139:49.66E 4371 15.1 0909 61:21.28S 139:47.59E 4371 55 SR3 1250 13-JAN-94 61:51.11S 139:50.95E 4309 4342 1424 61:51.25S 139:50.85E - 12.2 1550 61:51.26S 139:51.41E 4299 56 SR3 1930 13-JAN-94 62:21.22S 139:50.83E 3967 3988 2112 62:20.80S 139:50.68E - 14.9 2301 62:20.37S 139:51.82E 3967 57 SR3 0311 14-JAN-94 62:50.94S 139:50.80E 3211 106 0314 62:50.92S 139:50.73E - - 0320 62:50.86S 139:50.63E - 58 SR3 0340 14-JAN-94 62:50.69S 139:50.24E 3211 3230 0451 62:50.18S 139:49.29E 3211 12.9 0628 62:49.66S 139:47.82E 3211 59 SR3 0937 14-JAN-94 63:21.10S 139:50.70E 3812 3838 1059 63:21.07S 139:49.94E 3812 8.9 1224 63:20.83S 139:49.32E 3812 60 SR3 1541 14-JAN-94 63:52.03S 139:51.10E 3739 3756 1730 63:51.98S 139:50.29E 3739 13.9 1904 63:51.88S 139:50.21E 3739 61 SR3 2139 14-JAN-94 64:16.89S 139:52.08E 3470 3478 2310 64:17.14S 139:53.26E 3470 13.9 0042 64:17.00S 139:55.22E 3470 62 SR3 0414 15-JAN-94 64:49.26S 139:50.65E 2610 154 0418 64:49.26S 139:50.64E 2610 - 0424 64:49.24S 139:50.62E 2610 63 SR3 0444 15-JAN-94 64:49.16S 139:50.61E 2610 2586 0540 64:49.07S 139:50.99E 2610 8.4 0654 64:49.01S 139:51.63E 2620 64 SR3 0849 15-JAN-94 65:04.98S 139:50.91E 2766 2766 1001 65:04.65S 139:50.93E 2693 8.8 1114 65:04.39S 139:50.99E 2558 65 SR3 1354 15-JAN-94 65:24.14S 139:50.91E 2455 2438 1456 65:24.05S 139:51.15E - 11.0 1558 65:24.01S 139:51.20E 2455 66 SR3 1711 15-JAN-94 65:26.00S 139:51.07E 1854 1936 1811 65:25.98S 139:51.21E 1823 7.5 1914 65:25.86S 139:51.54E 1792 67 SR3 0151 16-JAN-94 65:31.84S 139:51.13E 1274 1264 0229 65:31.74S 139:50.97E 1280 9.0 0319 65:31.66S 139:50.62E 1305 68 SR3 0443 16-JAN-94 65:34.09S 139:50.66E 866 880 0510 65:34.05S 139:50.76E 889 6.9 0549 65:34.05S 139:50.86E 892 69 SR3 0731 16-JAN-94 65:42.49S 139:51.13E 307 294 0744 65:42.42S 139:50.88E 309 8.0 0812 65:42.15S 139:50.55E 320 70 SR3 0916 16-JAN-94 65:48.39S 139:51.04E 220 206 0927 65:48.35S 139:51.02E 221 8.6 0950 65:48.24S 139:50.86E 221 71 SR3 1705 16-JAN-94 66:27.27S 140:12.29E 1082 998 1745 66:27.51S 140:11.81E 994 10.8 1816 66:27.64S 140:11.61E 1004 72 SR3 1953 16-JAN-94 66:22.91S 140:21.73E 824 734 2025 66:22.72S 140:21.53E 761 11.1 2051 66:22.60S 140:21.66E 751 73 SR3 2350 16-JAN-94 66:19.70S 140:28.18E 398 364 0003 66:19.77S 140:28.15E 393 16.1 0024 66:19.87S 140:28.08E 398 74 ULS 1418 21-JAN-94 65:07.37S 107:46.15E 542 518 1439 65:07.36S 107:45.96E 543 13.9 1508 65:07.37S 107:45.73E 543 75 ULS 0000 23-JAN-94 63:18.23S 107:49.63E 3304 3304 0111 63:17.84S 107:49.37E 3304 13.0 0229 63:17.84S 107:49.43E 3304 76 PET 0438 26-JAN-94 66:19.31S 84:43.02E 569 550 0506 66:19.33S 84:42.93E 569 9.0 0532 66:19.44S 84:42.86E 574 77 PET 0802 26-JAN-94 66:08.85S 85:00.54E 563 540 0827 66:08.85S 85:00.41E 564 9.3 0854 66:08.97S 85:00.63E 564 78 PET 1107 26-JAN-94 65:59.22S 85:25.88E 258 242 1115 65:59.25S 85:25.91E 258 7.1 1138 65:59.34S 85:26.07E 258 79 PET 1247 26-JAN-94 65:53.61S 85:24.69E 1253 1284 1330 65:53.85S 85:24.98E 1305 5.8 1420 65:54.04S 85:25.31E - 80 PET 1520 26-JAN-94 65:49.41S 85:25.37E 1735 1912 1619 65:49.25S 85:26.24E 1787 9.2 1718 65:49.32S 85:26.73E - 81 PET 1819 26-JAN-94 65:44.83S 85:24.48E 2517 2522 1929 65:44.22S 85:23.72E 2539 9.5 2038 65:43.63S 85:22.98E 2517 82 PET 2207 26-JAN-94 65:32.95S 85:24.67E 2942 2932 2321 65:32.71S 85:24.51E 2947 10.8 0030 65:32.59S 85:24.34E 2900 83 PET 0316 27-JAN-94 65:05.55S 85:18.81E 3107 306 0336 65:05.61S 85:19.15E - - 0402 65:05.79S 85:19.52E 3128 84 PET 0452 27-JAN-94 65:05.70S 85:18.87E 3107 3112 0603 65:05.77S 85:19.17E 3123 7.2 0722 65:05.79S 85:19.86E 3159 85 PET 1028 27-JAN-94 64:37.05S 84:59.92E 3612 3624 1154 64:37.49S 85:00.16E 3625 5.1 1309 64:37.96S 84:59.77E 3615 86 PET 1628 27-JAN-94 64:09.81S 84:35.62E 3688 206 1645 64:09.92S 84:35.70E 3688 - 1702 64:10.02S 84:35.65E 3781 87 PET 1748 27-JAN-94 64:10.21S 84:36.00E 3688 3710 1915 64:10.18S 84:37.80E 3693 10.8 2043 64:10.42S 84:38.46E 3693 88 PET 2350 27-JAN-94 63:43.22S 84:08.56E 3729 3736 0116 63:43.19S 84:09.42E 3721 13.1 0244 63:43.09S 84:10.35E 3723 89 PET 0530 28-JAN-94 63:17.42S 83:44.76E 2786 256 0536 63:17.42S 83:44.83E 2797 - 0602 63:17.29S 83:45.34E 2797 90 PET 0646 28-JAN-94 63:17.20S 83:45.12E 2776 2770 0748 63:16.60S 83:45.61E 2766 8.5 0853 63:16.15S 83:46.23E 2745 91 PET 1153 28-JAN-94 62:44.89S 83:28.59E 2507 2496 1303 62:44.63S 83:28.45E 2507 8.7 1414 62:44.43S 83:28.47E 2507 92 PET 1734 28-JAN-94 62:09.89S 83:16.70E 2672 2666 1842 62:10.00S 83:16.93E 2672 16.9 1951 62:10.10S 83:17.20E 2683 93 PET 2209 28-JAN-94 61:48.12S 83:07.39E 2310 2298 2306 61:48.06S 83:07.41E 2320 10.0 0001 61:47.75S 83:07.54E 2320 94 PET 0208 29-JAN-94 61:26.43S 82:58.29E 1833 1850 0314 61:26.68S 82:58.18E 1875 10.0 0412 61:26.70S 82:58.21E 1885 95 PET 0704 29-JAN-94 60:56.51S 82:46.17E 2517 2478 0807 60:56.12S 82:45.80E 2486 9.2 0916 60:55.84S 82:45.87E 2465 96 PET 1200 29-JAN-94 60:26.79S 82:34.45E 1673 1644 1243 60:26.71S 82:34.63E 1662 7.4 1335 60:26.70S 82:35.23E 1709 97 PET 1622 29-JAN-94 59:58.36S 82:22.28E 1647 1624 1715 59:58.27S 82:22.78E 1647 15.8 1754 59:58.14S 82:22.89E 1647 98 PET 2117 29-JAN-94 59:27.43S 82:12.33E 1657 1630 2204 59:27.24S 82:12.73E 1657 10.4 2250 59:27.14S 82:12.88E 1657 99 PET 0143 30-JAN-94 58:57.72S 82:01.64E 1310 1290 0225 58:57.54S 82:02.07E 1315 11.5 0309 58:57.38S 82:02.20E 1315 100 TS 1431 31-JAN-94 63:00.06S 70:59.71E 4014 306 1453 63:00.13S 70:59.07E 4014 - 1522 63:00.21S 70:58.36E 4014 101 TS 1632 31-JAN-94 63:00.01S 70:59.79E 4014 4044 1805 63:00.18S 70:59.66E 4014 1.0 1943 63:00.15S 70:59.79E 4014 102 BPR 1538 23-FEB-94 65:26.04S 139:11.02E 890 952 1630 65:25.68S 139:10.84E 963 15.0 1717 65:25.42S 139:10.05E 1061 Table 3: Summary of samples drawn from Niskin bottles at each station, including salinity (sal), dissolved oxygen (do), nutrients (nut), dissolved inorganic carbon (dic), 13C, 18O, alkalinity (alk), dissolved organic carbon (doc), iodate/iodide (ii), dimethyl sulphide/dimethyl sulphoniopropionate (dms), primary productivity (pp), "Seacat" casts (cat), and the following biological samples: pigments (pig), lugols iodine fixed algal counts (lug), cyanobacteria counts (cya), total algal counts (alg), osmicated samples (os), flagellate cultures (fc), coccolith counts (coc), parmales cultures (pc), flavobacteria (flv). Note that 1=sample taken, 0=no sample taken, 2=surface sample only (i.e. from shallowest Niskin bottle). ---------------biology---------- station sal do nut dic 13C 18O alk doc ii dms pp cat pig lug cya alg os fc coc pc flv 1 TEST 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 TEST 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 SR3 1 1 1 1 1 0 0 0 1 1 1 0 1 1 1 1 0 1 0 0 1 4 SR3 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 5 SR3 1 1 1 1 0 0 0 0 0 1 0 0 1 0 1 1 0 1 0 0 0 6 SR3 1 1 1 0 0 0 0 1 1 0 1 0 1 0 1 0 0 0 0 0 0 7 SR3 1 1 1 1 2 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 8 SR3 1 1 1 1 0 0 0 0 1 0 0 0 1 1 0 0 0 0 0 0 0 9 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 1 1 0 1 0 0 0 10 SR3 1 1 1 1 1 0 0 0 1 0 0 0 1 0 1 1 1 1 0 0 0 11 SR3 1 1 1 0 0 0 0 0 0 1 1 0 1 1 1 1 0 1 0 0 1 12 SR3 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 SR3 1 1 1 1 0 0 0 0 1 0 0 0 1 0 1 1 0 1 0 0 0 14 SR3 1 1 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 1 0 0 0 15 SR3 1 1 1 1 1 0 0 1 1 0 1 1 1 0 1 0 0 0 0 0 0 16 SR3 1 1 1 1 0 0 0 0 0 1 0 0 1 1 1 1 1 1 1 0 1 17 SR3 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 18 SR3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 19 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 1 1 1 1 0 0 0 20 SR3 1 1 1 1 2 0 0 0 1 1 1 0 1 0 1 0 0 0 0 0 0 21 SR3 1 1 1 0 0 0 0 0 0 0 1 0 1 1 1 1 0 0 1 0 1 22 SR3 1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 23 SR3 1 1 1 1 1 0 0 0 1 0 1 1 1 0 1 1 0 0 0 0 0 24 SR3 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1 1 0 0 0 0 0 25 SR3 1 1 1 1 2 0 0 0 1 0 0 0 1 0 1 1 0 0 0 0 0 26 SR3 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 27 SR3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 28 SR3 1 1 1 1 1 0 0 1 1 0 1 1 1 0 1 0 0 0 0 0 0 29 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 30 SR3 1 1 1 1 2 0 0 0 1 0 0 0 1 1 1 1 0 1 1 0 1 31 SR3 1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 32 SR3 1 1 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 1 0 0 0 33 SR3 1 1 1 1 1 0 0 0 1 0 1 1 1 0 1 1 0 1 0 0 0 34 SR3 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 35 SR3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 36 SR3 1 1 1 1 0 0 0 0 1 0 0 0 1 0 1 1 0 1 0 0 0 37 SR3 1 1 1 0 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1 0 0 38 SR3 1 1 1 1 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 39 SR3 1 1 1 0 0 0 0 0 0 0 1 1 1 0 1 1 0 0 1 0 1 40 SR3 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 41 SR3 1 1 1 1 0 0 0 0 0 0 0 0 1 0 1 1 0 1 0 0 1 42 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 0 43 SR3 1 1 1 1 1 0 0 0 1 0 1 1 1 0 1 1 0 1 1 0 1 44 SR3 1 1 1 1 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 45 SR3 1 1 1 0 0 0 0 1 0 0 1 1 1 1 1 1 0 1 0 0 1 46 SR3 1 1 1 1 2 0 0 0 0 1 0 0 1 0 0 1 0 1 0 0 0 47 SR3 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 48 SR3 1 1 1 1 1 0 0 0 1 0 1 1 1 1 0 0 0 0 0 0 1 49 SR3 1 1 1 1 0 0 0 0 0 1 1 1 1 0 0 1 1 1 0 1 1 50 SR3 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 51 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 1 0 0 52 SR3 1 1 1 1 0 0 0 0 1 0 1 1 1 1 0 1 0 0 0 1 1 53 SR3 1 1 1 1 2 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 54 SR3 1 1 1 0 0 0 0 0 0 1 1 1 1 0 0 1 0 0 1 0 0 55 SR3 1 1 1 1 1 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 56 SR3 1 1 1 0 0 0 0 1 0 1 1 1 1 0 0 1 0 0 0 0 0 57 SR3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 58 SR3 1 1 1 1 2 0 0 0 1 0 0 0 1 1 0 1 0 0 0 0 0 59 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 1 0 0 60 SR3 1 1 1 1 1 0 0 0 1 1 0 0 1 0 0 1 0 0 0 0 0 61 SR3 1 1 1 0 0 0 0 0 0 1 1 1 1 0 0 1 0 0 0 0 0 62 SR3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 63 SR3 1 1 1 1 2 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0 0 64 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 0 1 0 65 SR3 1 1 1 1 2 0 0 0 1 0 0 0 1 1 0 0 0 0 0 0 0 66 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 0 0 1 1 0 0 1 0 67 SR3 1 1 1 1 2 0 0 1 1 0 1 1 1 0 0 1 0 1 0 1 0 68 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 1 69 SR3 1 1 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 70 SR3 1 1 1 1 2 0 0 0 1 0 0 0 1 0 0 1 0 0 0 1 0 71 SR3 1 1 1 1 0 0 0 0 0 0 0 0 1 0 0 1 0 1 0 0 0 72 SR3 1 1 1 0 0 0 0 0 0 0 1 1 1 1 0 1 0 0 0 1 0 73 SR3 1 1 1 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 0 74 ULS 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 75 ULS 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 76 PET 1 1 1 1 2 1 1 2 1 0 0 0 1 1 0 1 0 0 0 1 1 77 PET 1 1 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 78 PET 1 1 1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 79 PET 1 1 1 1 2 1 0 2 0 0 0 0 1 0 0 1 0 0 0 0 0 80 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 81 PET 1 1 1 1 2 0 0 2 1 0 0 0 1 0 0 1 0 0 0 0 0 82 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 83 PET 1 1 1 1 2 0 0 2 0 0 0 0 1 1 0 1 0 0 0 1 0 84 PET 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 85 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 86 PET 1 1 1 1 0 0 0 2 0 0 0 0 1 1 0 0 0 0 1 1 0 87 PET 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 PET 1 1 1 0 0 0 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 89 PET 1 1 1 1 2 0 0 2 0 0 0 0 1 1 0 1 0 0 0 0 1 90 PET 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 91 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 92 PET 1 1 1 1 2 1 0 2 0 0 0 0 1 0 0 0 0 0 1 1 0 93 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 0 94 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0 95 PET 1 1 1 1 2 0 1 2 0 0 0 0 1 1 0 0 0 0 0 0 0 96 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 97 PET 1 1 1 1 0 1 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 98 PET 1 1 1 0 0 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 0 99 PET 1 1 1 1 2 1 0 2 1 0 0 0 1 0 0 1 0 0 0 0 1 100 TS 1 1 1 1 1 0 1 1 0 0 1 1 1 1 0 1 0 0 1 0 0 101 TS 1 1 1 1 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 102 BPR 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3.3 Additional drifters and moorings deployed/recovered Two bottom pressure recorders (principal investigators Tom Whitworth, University of Texas, A&M, and Dale Pillsbury, Oregon State University) were deployed near either end of the SR3 section, and a single recorder was recovered from the northern end. An additional two bottom pressure recorders near the southern end of SR3 could not be recovered (Table 4). Two upward looking sonar moorings (Bush, 1994), including current meters, were deployed in the vicinity of Casey (Table 5). Several meteorological buoys were also deployed throughout the cruise - these are not discussed further. 3.4 XBT/XCTD deployments A total of 16 model T-7 Sippican XBT deployments were made along both the SR3 and PET transects. The data were processed further by CSIRO Division of Oceanography (R. Bailey, pers. comm.). Results are not reported here. 3.5 Principal investigators The principal investigators for the CTD and water sample measurements are listed in Table 6a. Cruise participants are listed in Table 6b. Table 4: Bottom pressure recorder moorings. Note that for the instruments not recovered, positions given are where the moorings were located at the indicated time. Recovery attempts were made on 17/01/94, and in late Feb. deployment deployment/recovery latitude longitude CTD number time (UTC) station no. ---------------------------------------------------------------------------- instruments deployed Hobart94 02:07, 02/01/94 44o 07.18'S 146o 13.13'E 4 SR3 Dumont94 03:45, 16/01/94 65o 33.68'S 139o 51.15'E 68 SR3 instruments recovered Hobart91a 23:21, 01/01/94 44o 06.70'S 146o 13.08'E 4 SR3 unsuccessful recovery attempts Dumont92a 21:00, 15/01/94 65o 25.95'S 139o 11.63'E 102 SR3 Dumont92b 21:00, 15/01/94 65o 25.97'S 139o 11.63'E 102 SR3 Dumont92a & b 24/02/94-25/02/94 7 bottom trawls around site 102 SR3 Table 5: Upward looking sonar (ULS) moorings deployed (including current meters [CM]). ------------------------------------------------------------------------------------ site deployment bottom latitude longitude instrument CTD name time (UTC) depth (m) depths (m) station no. SONEAR 16:45, 21/01/94 520 65o 07.350'S 107o 46.130'E 170 (ULS) 74 220 (CM) 420 (CM) SOFAR 04:36, 23/01/94 3260 63o 17.746'S 107o 49.429'E 150 (ULS) 75 200 (CM) Table 6a: Principal investigators (*=cruise participant) for water sampling programmes. measurement name affiliation ---------------------------------------------------------------------- CTD, salinity, O2, nutrients Steve Rintoul CSIRO D.I.C., 13C, 18O, alkalinity *Bronte Tilbrook CSIRO primary productivity John Parslow CSIRO biological sampling Harvey Marchant Antarctic Division iodate/iodide Ed Butler CSIRO D.M.S. Graham Jones James Cook University D.O.C. Tom Trull Antarctic CRC flavobacteria *Sue Dobson University of Tasmania Table 6b: Scientific personnel (cruise participants). name measurement affiliation --------------------------------------------------------------------------------------- Nathan Bindoff CTD Antarctic CRC Jeremy Harris CTD Antarctic CRC Ian Knott CTD, electronics Antarctic CRC Dennis Root BPR's, CTD Oregon State University Mark Rosenberg CTD Antarctic CRC Steve Bell salinity, oxygen, nutrients Antarctic CRC Ruth Eriksen salinity, oxygen, nutrients Antarctic CRC Anna Brandao iodate/iodide, D.I.C. Antarctic CRC Roger Dargaville D.I.C., isotopes, alkalinity Melbourne University Mark Pretty D.I.C., isotopes, alkalinity CSIRO Bronte Tilbrook D.I.C., isotopes, alkalinity CSIRO Andrew Broadbent D.M.S. James Cook University Don McKenzie primary productivity, CTD CSIRO Sue Dobson flavobacteria University of Tasmania Naomi Parker biological sampling Antarctic CRC Alison Turnbull biological sampling Antarctic Division David James ornithology Royal Australasian Ornithologists Union Paul Scofield ornithology Royal Australasian Ornithologists Union Clive Rapier radiometry Antarctic CRC Pia Geijsel krill biology Antarctic Division Rob King krill biology Antarctic Division Greg Bush ULS moorings Curtin University Wayne Galbraith ULS moorings Curtin University Gordon Bain deputy voyage leader, CTD Antarctic Division Pamela Brodie computing Antarctic Division Jo Jacka voyage leader Antarctic Division Michael Sexton doctor Antarctic Division Tim Stevenson computing Antarctic Division *Andrew Tabor gear officer Antarctic Division Mark Underwood electronics Antarctic Division * retrieved with trawl gear from Dumont D'Urville 4 FIELD DATA COLLECTION METHODS 4.1 CTD and hydrology measurements In this section, CTD and hydrology data collection methods are discussed. CTD data processing techniques are referred to in Appendix 2, while hydrology laboratory analysis methods are described in Appendix 3. Preliminary results of the CTD data calibration, along with data quality information, are presented in Section 6. 4.1.1 CTD Instrumentation A General Oceanics (formerly E.G.&G.) Mark IIIC (i.e. WOCE upgraded) CTD unit, together with a model 1401 deck unit, were used for CTD measurements (Table 7). The raw data stream was logged by two separate IBM compatible PC's, using the General Oceanics data aquisition software CTDACQ. The duplication of the data logging PC's allowed data to be viewed simultaneously (in real time) as column formatted numbers on one screen, and in graphical format on the other; the second PC also provided a backup log of the data. Table 7: CTD manufacturer specifications. parameter sensor accuracy resolution -------------------------------------------------------------------------------------- Pressure Paine Model 211-36-390-02 1500 ohm +1.2 dbar 0.1 dbar bonded titanium strain gauge bridge, tube type Temperature Rosemount Model 171 BJ platinum thermometer +0.003 oC 0.0005oC Conductivity Neil Brown Instruments 4 electrode cell +0.0003 mS/cm 0.0001 mS/cm (0.4cm x 0.4cm x 3.0 cm long) Oxygen Sensormedics polarographic oxygen sensor - - Altimeter Benthos Model 2110 +5% 0.1 m Fluorometer Sea Tech - - P.A.R. sensor Li-Cor model LI-192SA underwater quantum +/-5% in air - sensor The same CTD unit (serial no. 2568) was used throughout the entire cruise. For the electronic and data stream configuration of the instrument, see Table 8. Two different General Oceanics 24-bottle model 1015 rosette pylons were used during the cruise, together with 10 litre General Oceanics Niskin bottles; several CSIRO manufactured 10 litre Niskin bottles were also used for some casts. Deep sea reversing thermometers (Gohla-Precision) were used to keep track of CTD temperature sensor performance. In general, two protected thermometers were mounted on the shallowest Niskin bottle, while three thermometers (two protected and one unprotected) were mounted on the second deepest bottle. The manufacturer specified accuracy of the protected thermometers is to within (0.01oC for the main thermometer, and (0.1oC for the auxiliary. Readings can be resolved to the third decimal place for the main on the protected thermometers, and to the second decimal place for auxiliary and unprotected readings. Table 8: CTD electronic and data stream configuration, and data processing parameters. Note that all parameters are assigned 2 bytes in the raw data stream. For the CTD upcast burst data, the first nstart and the last nend data scans are ignored for calculation of burst statistics; the first jfilt data scans are ignored each time the data lagging recursive filter is restarted. _T is the time constant of the temperature sensor. jmin is the minimum number of values required in a 2 dbar pressure bin. All the above constants and calculations are described in detail in Appendix 2. CTD unit serial scanning bytes per bytes per nstart nend jfilt TauT jmin number number frequency (Hz) record scan (s) -------------------------------------------------------------------------------------- 6 2568 25.00 129 25 7 4 13 0.205 10 Scan byte layout: synch. byte, pressure, temperature, conductivity, fast temperature, utility byte, altimeter, pressure temperature, oxygen current, oxygen temperature, fluorescence, photosynthetically active radiation (P.A.R.), end bytes 4.1.2 CTD instrument calibrations Complete calibration information for the CTD pressure, platinum temperature and pressure temperature sensors are presented in Appendix 1. Formulae used for parameter calculations are presented in Appendix 2. Pre and post cruise calibrations were available for the pressure sensor, the former supplied by the manufacturer, the latter done at the CSIRO Division of Oceanography Calibration Facility. The post cruise CSIRO pressure sensor calibration was used for the cruise data. This calibration was performed using a Budenberg Deadweight Tester (accurate to (0.05% of the pressure being measured) over the range 0 to 6203 dbar. The titanium strain gauge pressure sensors used in the Mark IIIC CTD's display reduced hysteresis effect compared to the older stainless steel type pressure sensors (Millard et al., 1993). Calibration points from the increasing pressure only were used to produce a single third order fit for pressure (Appendix 1). Pre and post cruise calibrations were also available for the platinum temperature sensor, as for the pressure sensor. However, a significant change was noted between the two temperature calibrations. In particular, a shift of - 0.407oC was noted in the post cruise temperature offset relative to pre cruise value. Comparison of cruise CTD temperature data with reversing thermometer readings (Figure 2a) confirmed that the pre cruise calibration (supplied by the manufacturer) was applicable to the entire cruise. Later testing of the instrument indicated that the post cruise temperature calibration was faulty, thus the pre cruise calibration was used. A pre cruise manufacturer supplied calibration of the pressure temperature sensor was used for the cruise data. Note that readings from this sensor are applied in a correction formula to pressure data (Appendix 2). CTD conductivity measurements were calibrated from the in situ salinity samples collected at each station. As a rule, this enables CTD salinity values to be calculated to a much higher accuracy than by the bulk application of a single set of laboratory determined calibration coefficients. Thus there are no laboratory calibrations for the conductivity sensor. Checks were made prior to the cruise to ensure the conductivity sensor was functioning correctly. Similarly, CTD dissolved oxygen measurements were calibrated from the in situ dissolved oxygen samples. The complete conductivity and oxygen in situ calibrations are presented in a later section. Manufacturer supplied calibrations were applied to the fluorescence and p.a.r. data (Appendix 1). These calibrations are not expected to be correct - correct scaling of fluorescence and p.a.r. data awaits linkage with primary productivity and Seacat (section 3.2) data. 4.1.3 CTD and hydrology data collection techniques When on deck, the rosette package was housed in a closed laboratory space. Thus all samples were drawn "indoors". The package was deployed through an outward opening hatch, which doubles as a gantry, and was lowered/raised at the following speeds: 0 to 500 m depth - 20 m/min 500 to 1000 m depth - 40 m/min below 1000 m depth - 60 m/min Winch speeds were maintained by constantly adjusting the winch wire tension, and thus are approximate average values only. The altimeter output was used to guide the instrument to within 20 m (in most cases) of the bed. CTD data was logged continuously for the entire down and upcast, while Niskin bottles were fired on the upcast only. At each station, the firing depths for the Niskin bottles were decided on using the graphical output of the CTD downcast data. Typically, the deepest bottle was fired at the bottom of the cast. The rosette package was stopped at each level prior to firing; bottles with reversing thermometers were allowed to equilibrate for 5 min before firing. A fixed sequence was followed for the drawing of water samples on deck, as follows: first sample: dissolved oxygen d.i.c. 13C 18O alkalinity productivity d.o.c. salinity nutrients d.m.s. iodate/iodide last sample: biology (see Table 3 for a summary of which samples were drawn at each station). Reversing thermometers were read after the sampling was complete (or nearing completion), typically within one hour of the raising of the rosette package onto the deck. In between stations, the Niskin bottles were only emptied when resetting the bottles for the next station. This helped prevent the crystallization of salt in o-ring seats and spiggots. 4.1.4 Water sampling methods The methods used for drawing the various water samples from the Niskin bottles are described here. Laboratory analysis techniques are described in later sections. Dissolved oxygen: sample bottle volume = 300 ml Bottles are washed and dried before use. As dissolved oxygen samples are drawn first, the Niskin is first tested for obvious leakage by opening the spiggot before opening the air valve. Tight fitting silicon tubing is attached to the Niskin spiggot for sample drawing. Pickling reagent 1 is 3 M MnCl2 (2.0 ml used); reagent 2 is 8 N NaOH/4 M NaI (2.0 ml used); reagent 3 is 10 N H2SO4 (2.0 ml used). * start water flow through tube for several seconds, making sure no bubbles remain in tube * pinch off flow in tube, and insert into bottom of sample bottle * let flow commence slowly into bottle, gradually increasing by releasing tubing, at all times ensuring no bubbles enter the sample and that turbulence is kept to a minimum * fill bottle, overflow by at least one full volume * pinch off tube and slowly remove so that bottle remains full to the brim, then rinse glass stopper * immediately pickle with reagents 1 then 2, inserting reagent dispenser at least 1 cm below water surface * insert glass stopper, ensuring no bubbles are trapped in sample * thoroughly shake sample (at least 30 vigorous inversions) * store samples in the dark until analysis * acidify samples with reagent 3 immediately prior to analysis Dissolved inorganic carbon: sample bottle volume = 250 ml Tight fitting silicon tubing is attached to the Niskin spiggot for sample drawing. Samples are poisoned with 100 _l of a saturated solution of HgCl2. * drain remaining old sample from the bottle * start water flow through tube for several seconds, making sure no bubbles remain in tube * insert tube into bottom of inverted sample bottle, allowing water to flush bottle for several seconds * pinch off flow in tube, and invert sample bottle to upright position, keeping tube in bottom of bottle * let flow commence slowly into bottle, gradually increasing, at all times ensuring no bubbles enter the sample * fill bottle, overflow by one full volume, and rinse cap * shake a small amount of water from top, so that water level is between threads and bottle shoulder * insert tip of poison dispenser just into sample, and poison * screw on cap, and invert bottle several times to allow poison to disperse through sample Salinity: sample bottle volume = 300 ml * drain remaining old sample from the bottle (bottles are always stored approximately 1/3 full with water between stations) * rinse bottle and cap 3 times with 100 ml of sample (shaking thoroughly each time); on each rinse, contents of sample bottle are poured over the Niskin bottle spiggot * fill bottle with sample, to bottle shoulder, and screw cap on firmly At all filling stages, care is taken not to let the Niskin bottle spiggot touch the sample bottle. Nutrients: sample tube volume = 12 ml Two nutrient sample tubes are filled simultaneously at each Niskin bottle. * rinse tubes and caps 3 times * fill tubes * shake out water from tubes so that water level is at or below marking line 2 cm below top of tubes (10 ml mark), and screw on caps firmly After sampling, one set of tubes are refrigerated for analysis within 12 hours; the duplicate set of tubes are placed in a freezer until required. Carbon Isotopes: These are sampled and poisoned in the same fashion as dissolved inorganic carbon, except that 500 ml glass stoppered vacuum flasks are used, and vacuum grease is placed around the stopper before inserting. Iodate: same as for nutrients Iodide: same as for nutrients, except 100 ml plastic bottle used. DMS and DMSP: Sample containers are quickly rinsed, then filled. For shallow samples only, a 750 ml amber glass bottle are used. For full profile sampling, 250 ml polyethylene screwcap jars are used. Subsamples (acidified with 1 ml of concentrated HCl) are taken in the laboratory for DMS and DMSP analysis. Dissolved organic carbon: Sample jar volume = 250 ml (jars baked for 12 hours at 550oC) During d.o.c. sampling, polyethylene gloves were worn by the sampler. The gloves were changed every second sample. * clean spiggot with lint free tissue sprayed with acetone * rinse spiggot copiously with sample water * rinse sample jar twice * fill jar with ~200 ml and screw cap on tightly After sampling, the jars are stored in the dark in a freezer at -18oC. Alkalinity: same as for d.i.c. samples, except 500 ml bottle used. 18O: Sample bottle volume = 20 ml Sample bottles given 3 quick rinses, then filled. 4.2 Underway measurements Throughout the cruise, the ship's data logging system continuously recorded bottom depth, ship's position and motion, surface water properties and meteorological information. All measurements were quality controlled during the cruise, to remove bad data (Ryan, 1995). After quality controlling of the automatically logged GPS data set, gaps (due to missing data and data flagged as bad) are automatically filled by dead-reckoned positions (using the ship's speed and heading). Positions used for CTD stations are derived from this final GPS data set. Bottom depth is measured by a Simrad EA200 12 kHz echo sounder. A sound speed of 1498 ms-1 is used for all depth calculations, and the ship's draught of 7.3 m has been accounted for in final depth values (i.e. depths are values from the surface). Seawater is pumped on board via an inlet at 7 m below the surface. A portion of this water is diverted to the thermosalinograph (Aplied Microsystems Ltd, model STD-12), and to the fluorometer (Turner Design, peak sensitivity for chlorophyll-a). Sea surface temperatures are measured by a sensor next to the seawater inlet at 7 m depth. The underway measurements for the cruise are contained in column formatted ascii files (Appendix 4). The two file types are as follows (see Appendix 4 for a complete description): (i) 10 second digitised underway measurement data, including time, latitude, longitude, depth and sea surface temperature; (ii) 15 minute averaged data, including time, latitude and longitude, air pressure, wind speed and direction, air temperature, humidity, quantum radiation, ship speed and heading, roll and pitch, sea surface salinity and temperature, average fluorescence, and seawater flow. 5 MAJOR PROBLEMS ENCOUNTERED The most serious problem on the cruise was failure to recover either of the two bottom pressure recorder moorings near the southern end of the SR3 section (Table 4). The site was visited initially on 15th January 1994 - no recovery attempt was made at this time due to the 70% sea ice cover. The ship returned to the site on 17th January. Distance ranges were easily obtained from the acoustic releases on each of the two moorings; however, neither mooring would release from the bottom. A final recovery attempt was made when the ship returned to the site in late February with bottom trawling gear. Seven unsuccessful trawl attempts were made, then the ship returned to Hobart. Problems with some of the CTD and laboratory equipment resulted in some data loss and/or compromise to data quality. The most significant of these problems was failure of all three YeoKal Mk IV salinometers used for the analysis of salinity samples.The first salinometer developed a large drift and was difficult to standardise. The replacement salinometers proved faulty, and were unusable. A significant number of salinity samples were analysed before the problem was fully appreciated, and as a result, the bottle salinity data for stations 69 to 86 are considered unusable. All salinity samples following station 86 were retained for analysis back in port. The two CTD logging PC's crashed on numerous occasions. In most cases, this occurred while the instruments were still on deck, thus no data was lost. On one occasion however (station 2), the PC's crashed during the upcast, resulting in loss of the entire downcast CTD data, and half of the upcast CTD burst data. The Antarctic CRC-owned rosette pylon developed problems early in the cruise, misfiring on many occasions, resulting in missed bottles. The unit was replaced with the spare pylon following station 15. The original pylon was refurbished and reinstalled for station 75 onwards, and performed well for the remainder of the cruise. Comparison of CTD temperature data with reversing thermometer measurements (Figure 2a) revealed a problem with the CTD temperature sensor calibration for sub-zero water temperatures, most noticeable for stations 61 to 82. Sensor output appears to deviate significantly from a linear response at these lower temperatures, resulting in temperature offsets of the order 0.02oC from the expected calibrated values. This is discussed further in section 6. 6 RESULTS This section details information relevant to the creation and the quality of the final CTD and hydrology data set. For actual use of the data, the following is important: CTD data - Tables 16, 17 and 18, and section 6.1.2; hydrology data - Tables 21 and 22. Historical data comparisons are made in Appendix 6. 6.1 CTD measurements 6.1.1 Creation of CTD 2 dbar-averaged and upcast burst data Information relevant to the creation of the calibrated CTD 2 dbar-averaged and upcast burst data is tabulated, as follows: * Surface pressure offsets calculated for each station (Appendix 2, section A2.6.1) are listed in Table 11. * Missing 2 dbar data averages are listed in Table 12. * CTD conductivity calibration coefficients, including the station groupings used for the conductivity calibration, are listed in Tables 13 and 14. * CTD raw data scans flagged for special treatment (Appendix 2, section A2.11.1) are listed in Table 15. * Suspect 2 dbar averages are listed in Tables 16 and 17. Table 18 lists 2 dbar averages which are linear interpolations of the surrounding 2 dbar averages. * CTD dissolved oxygen calibration coefficients are listed in Table 19. The starting values used for the coefficients prior to iteration, and the coefficients varied during the iteration, are listed in Table 20. * Upcast CTD burst data automatically flagged with the code -1 (rejected for conductivity calibration) or 0 (questionable value, but still used for conductivity calibration) (see Appendix 2, section A2.7.4) are listed in Appendix 5, Table A5.1. * Stations containing fluorescence and photosynthetically active radiation data are listed in Appendix 5, Table A5.3. * The different protected and unprotected thermometers used for the stations are listed in Appendix 5, Table A5.4. 6.1.2 CTD data quality The final calibration results for conductivity/salinity and dissolved oxygen, along with the performance check for temperature, are plotted in Figures 2 to 5. For temperature, salinity and dissolved oxygen, the respective residuals (Ttherm - Tcal), (sbtl - scal) and (obtl - ocal) are plotted. For conductivity, the ratio cbtl/ccal is plotted. Ttherm and Tcal are respectively the protected thermometer and calibrated upcast CTD burst temperature values; sbtl, scal, obtl, ocal, cbtl and ccal , and the mean and standard deviation values in Figures 2 to 5, are as defined in Appendix 2. CTD data quality cautions for the various parameters are discussed below. Table 9 contains a summary of these cautions. Pressure The titanium strain gauge pressure sensors used in the Mark IIIC CTD's display a higher noise level than the older stainless steel strain gauge models, with an rms of ~(0.2 dbar (Millard et al., 1993). A small error is therefore introduced to the surface pressure offset values, noting that the offsets are derived by taking spot values from the pressure record (Appendix 2, section A2.6.1). For stations 95 and 96 in particular, offset values fell on small pressure spikes, thus the final surface pressure offsets were estimated from a manual inspection of the pressure data. Note that any noise in the pressure signal is ultimately removed by the 2 dbar-averaging. The surface pressure offset values for stations 2, 7, 9, 10, 70 and 88 were estimated from the surrounding stations (Table 11). Any resulting additional error in the CTD pressure data is judged to be small (no more than 0.2 dbar). For the additional digital channels (including pressure temperature, oxygen current, oxygen temperature, fluorescence and photosynthetically active radiation) on the Mark III CTD's, a problem with one digital channel generally transfers to the other digital channels. Thus for stations 6, 8, 9 and 10, flooding of the dissolved oxygen sensor with seawater resulted in bad oxygen current data, which in turn resulted in bad data (Table 12) for the other digital channels, including pressure temperature. To allow accurate calculation of pressure in dbar (Appendix 2), pressure temperature data from station 12 were used in pressure calculations for stations 6, 8, 9 and 10, as follows: station 6: station 12 pressure temperature data used for entire downcast, and for upcast burst data; station 8: station 12 pressure temperature data used for 25(p(65 and p>1355 for downcast, and for all upcast burst data (where pressure p and all numerical values are in dbar); station 9: station 12 pressure temperature data used for entire downcast and for upcast burst data; station 10: station 12 pressure temperature data used for entire downcast and for upcast burst data. Note that the pressure temperature profile for station 12 provides the closest match to the assumed pressure temperature profiles for stations 6, 8, 9 and 10. From Millard et al. (1993), a pressure temperature error of 0.1oC produces a maximum pressure error of less than 0.05 dbar. Thus any resulting error in pressure data for these stations is judged to be small (<0.3 dbar). Salinity The conductivity ratios for all bottle samples are plotted in Figure 3, while the salinity residuals are plotted in Figure 4. The final standard deviation values for the salinity residuals (Figure 4) indicate the CTD salinity data is accurate to within (0.002 psu, except for stations 69 to 86 and station 1 (as discussed below). Station 1 was a test cast, with all bottles fired at a single depth. The calibration of station 1 conductivity uses salinity samples from this single depth only, thus CTD salinity for this station can only be considered accurate to ~0.01 psu. No bottle samples were taken for the shallow casts at stations 18, 27, 35, 57 and 62. These stations are grouped with surrounding stations for conductivity calibration (Table 13). No salinity bottle data was available for stations 69 to 86 due to salinometer problems, as discussed in section 5. For calibration of CTD conductivity from salinity bottle data (Appendix 2), stations 69 to 73 were grouped with the calibration of stations 64 to 68, while stations 74 to 86 were grouped with the calibration of stations 87 to 88 (Table 13). An appreciable variation in conductivity cell response is likely to have occurred over such a large span of stations. Therefore as no in situ conductivity calibrations were possible for stations 69 to 86, CTD salinity data for these stations can only be regarded as accurate to, at best, 0.005 psu. For station 64, analysis results were bad for salinity samples from rosette positions 15 to 22. These bottles were rejected for the conductivity calibration. For station 101, the conductivity sensor was fouled for the entire upcast. All upcast burst data was rejected for the conductivity calibration, and the station was grouped with the calibrations applied to stations 100 and 102 (Table 13). Temperature The temperature residuals are shown in Figure 2a, along with the mean offset and standard deviation of the residuals. The thermometer value used in each case is the mean of the two protected thermometer readings (protected thermometers used are listed in Appendix 5, Table A5.4). Note that in the figures, the "dubious" and "rejected" categories refer to corresponding bottle samples and upcast CTD bursts in the conductivity calibration. As discussed in section 5, a temperature calibration problem exists for sub-zero water temperatures, most significant for stations 61 to 82 (Figure 2b). For these stations, accuracy of the CTD temperature data is diminished to 0.02oC. Dissolved Oxygen The dissolved oxygen residuals are plotted in Figure 5. The final standard deviation values are within 1% of full scale values (where full scale is approximately equal to 250 (mol/l for pressure > 750 dbar, and 350 (mol/l for pressure < 750 dbar). In general, good calibrations of the CTD dissolved oxygen data were obtained using the in situ bottle data, however some atypical values were found for the calibration coefficients (Tables 19 and 20) (see Appendix 2 for full details of calibration formulae). For most stations, the best calibration was achieved using large values of the order 6.0 for the coefficient K1 (i.e. oxygen current slope), and large negative values of the order -0.7 for the coefficient K3 (i.e. oxygen current bias). This, however, is not considered relevant to actual data quality. The approximate magnitude of K1 and K3 values is sensitive to the oxygen current and oxygen temperature values as determined by eqns A2.9 and A2.10 (Appendix 2): these initial oxygen current and oxygen temperature values are in approximate engineering units only, as there is typically no laboratory calibration of individual oxygen sensors. In addition, the following unusual coefficient values were found (for typical values, see Millard and Yang, 1993, and Millard, 1991): station 7: K6 < 0 (usually expect a positive value); station 11: K5 > 1 (usually expect 0 1 (usually expect 0 1 (usually expect 0 1 (usually expect 0 0 (usually expect a negative value); station 76: K5 > 1 (usually expect 01355 for downcast, and for all upcast burst data 9 pressure surface pressure offset estimated from surrounding stations 9 pressure station 12 pressure temperature profile used for pressure calculation 10 pressure surface pressure offset estimated from surrounding stations 10 pressure station 12 pressure temperature profile used for pressure calculation 18 salinity CTD conductivity calibrated with bottles from surrounding stations 27 salinity CTD conductivity calibrated with bottles from surrounding stations 35 salinity CTD conductivity calibrated with bottles from surrounding stations 57 salinity CTD conductivity calibrated with bottles from surrounding stations 61 to 82 temperature temperature accuracy diminished 62 salinity CTD conductivity calibrated with bottles from surrounding stations 64 salinity bottles 15 to 22 not used in CTD conductivity calibration 69 to 86 salinity no salinity bottle samples - CTD conductivity calibrated with bottles from stations 64 to 68 and 87 to 88; salinity accuracy reduced 70 pressure surface pressure offset estimated from surrounding stations 88 pressure surface pressure offset estimated from surrounding stations 101 salinity CTD conductivity calibrated with bottles from surrounding stations 1 to 102 fluorescence/p.a.r. fluorescence and p.a.r. sensors (where active) are uncalibrated Nutrients For the phosphate analyses, it was found that the autoanalyser peak height of a sample which was run immediately after a series of wash solution vials (low nutrient sea water) was suppressed by, on average, 2%. It is suspected that this was due to sorption of the phosphomolybdate complex produced by the presence of phosphate in the sample onto the walls of the instrument tubing, after having been exposed to the cleaning action of the low nutrient sea water wash. This effect is best illustrated by running a series of replicate samples: autoanalyser peak heights gradually increase and stabilise to a constant value as successive replicate samples are analysed. The same effect has also been observed for phosphate analyses using Technicon Autoanalysers (D. Terhell, pers. comm.). Flushing with sodium hydroxide reduced the severity of the effect, but did not eliminate it. The effect was most noticeable for phosphate analyses from stations 25 onwards. Phosphate samples thus effected (in most cases from rosette positions 12 and 24) were deleted from the hydrology data set. No substantial error was noted for stations 1 to 24, so no phosphate samples were deleted from these stations. The same suppressed peak height effect was noted for data from the previous cruise (Rosenberg et al., 1995). For future cruises, additional "dummy" samples drawn from the Niskin bottles will be inserted in autoanalyser runs immediatley following wash solution vials to artificially mask the suppression effect on subsequent samples. It is expected that the resulting vertical phosphate profiles will appear "neater"; however the method used for phosphate analysis needs to be scrutinised for a more permanent solution to the problem. Note that an alternative phosphate chemistry using hydrazine instead of ascorbic acid as the reductant has been trialed (Gordon et al., 1993), with no apparent improvement. For all near-surface silicate samples (i.e. above ~200 dbar), the autoanalyser silicate peaks were spiked, causing problems in the automatic peak integration performed by the software DAPA (see Appendix 3). The peaks in question were measured manually using an interactive graphics option within the DAPA software. The cause of the spikes is unknown - the samples coincided with high levels of diatoms (A. Turnbull, pers. comm.), however filtering of the samples produced no change. The following notes also apply to the nutrient data: * For stations 1 and 2 (test casts), and for station 102, no nutrient samples were collected. * For the following stations, nutrient concentrations were derived from manual measurements of autoanalyser peak heights, using the strip chart recordings: station 67 - nitrate+nitrite data station 71 - silicate data station 74 - all nutrient data station 75 - all nutrient data 6.2.2 Hydrology sample replicates The accuracy and precision of bottle data are considered relative to the full scale deflection of measurement for nutrients, and relative to the maximum data value for dissolved oxygen (Table 10). Table 10: Maximum values for dissolved oxygen analyses, and full scale deflection values for nutrient analyses. dissolved oxygen: ~350 µmol/l for pressure < 750 dbar ~250 µmol/l for pressure > 750 dbar phosphate: 3.0 µmol/l nitrate+nitrite: 35.0 µmol/l silicate: 140 µmol/l In general, no organised sample replication was carried out, thus the replicate data set discussed here is small. Most replicate data were obtained opportunistically, from multiple fired Niskin bottles taken during bottle test casts, or from depths sampled in both casts of shallow/deep cast pairs. Three types of replicate data were obtained from the hydrology data set, as follows. Replicate samples drawn from the same Niskin bottle A series of repeat dissolved oxygen samples were drawn from 4 different Niskin bottles at station 102 (Figure 6a). A standard deviation about the mean of 0.218 µmol/l was found for the sample set of 15 values, representing a precision level of better than 0.1% of full scale. No other data were available for this class of replicates. Replicate samples drawn from different Niskin bottles tripped at same depth At several stations, multiple Niskin bottles were fired at a single depth. Salinity samples were drawn from all multiple fired Niskins, while dissolved oxygen samples were drawn from only some. No nutrient samples were drawn for this class of replicates. For each set of Niskin bottles tripped at a single depth, a mean value mx was calculated for the sample set and the differences x- mx formed, where x is the salinity or dissolved oxygen bottle value; the standard deviation of all x-mx values for the replicate data was calculated. Note that 10 samples were rejected from the analysis, as they were drawn from leaking Niskin bottles. Absolute values of the differences x-mx are shown in Figure 6b. The results are summarised as follows: parameter standard deviation number of number of of x-mx samples sample groups -------------------------------------------------------------- salinity 0.0012 psu 51 13 dissolved oxygen 0.199 (mol/l 24 11 It is assumed that these precision values would be further reduced if sample groups were drawn from the same Niskin bottle. Replicate samples drawn from equivalent positions at different stations For some shallow/deep cast pairs i.e. consecutive shallow and deep stations taken at approximately the same location (Table 2), certain depths were sampled for both casts (in most cases, the shallowest depth only). For all such double sampled positions, sample pairs were formed for each parameter measured. For each sample pair, the quantity 0.5(x1-x2) was calculated, where x1 and x2 are the two parameter values from the pair; the 0.5 factor is included so that the data are comparable to the other classes of replicate data, where a standard deviation about depth mean values was calculated. A quality control element was introduced by rejecting pairs for which the difference of upcast CTD burst temperatures was (0.1oC (i.e. 0.5 times the difference (0.05oC). One additional sample pair was rejected due to a leaking Niskin bottle; two salinity sample pairs were also rejected as the samples were analysed during the time of salinometer malfunction (see section 5). The results (Figure 6c) are summarised as follows: parameter standard deviation number of of 0.5(x1-x2) samples ----------------------------------------------- salinity 0.0029 psu 18 dissolved oxygen 1.966 (mol/l 22 phosphate 0.0098 (mol/l 14 nitrate+nitrite 0.3854 (mol/l 26 silicate 1.5740 (mol/l 24 The larger precision values found for this class of replicates is an expected result - significant additional error has been introduced by comparing data from different casts. The locations for cast pairs are often separated by several hundred meters (Table 2), while the common depths sampled for cast pairs are not identical. These factors are increasingly important in regions of significant horizontal gradients in parameter values. In addition, some of the precision values (dissolved oxygen, nitrate+nitrite and silicate) would be significantly improved by the rejection of only one or two outliers (Figure 6c). Figure 2a: Temperature residual (Ttherm - Tcal) versus station number for cruise au9407. The solid line is the mean of all the residuals; the broken lines are ( the standard deviation of all the residuals (as defined in Appendix 2, section A2.14). Note that the "dubious" and "rejected" categories refer to the conductivity calibration. Figure 2b: Water temperature at top Niskin bottle from CTD upcast burst data. Figure 3: Conductivity ratio cbtl/ccal versus station number for cruise au9407. The solid line follows the mean of the residuals for each station; the broken lines are ( the standard deviation of the residuals for each station (as defined in Appendix 2, section A2.14). Figure 4: Salinity residual (sbtl - scal) versus station number for cruise au9407. The solid line is the mean of all the residuals; the broken lines are ( the standard deviation of all the residuals (as defined in Appendix 2, section A2.14). Figure 5: Dissolved oxygen residual (obtl - ocal) versus station number for cruise au9407. The solid line follows the mean residual for each station; the broken lines are ( the standard deviation of the residuals for each station (as defined in Appendix 2, section A2.14). Figure 6: Absolute value of parameter differences for replicate samples, for replicates drawn from (a) the same Niskin bottle, (b) different Niskins tripped at the same depth, and (c) Niskins fired at equivalent positions during different stations. For (a) and (b), differences are between parameter values and depth mean; differences are between sample pairs times 0.5 for (c). Table 11: Surface pressure offsets (as defined in Appendix 2, section A2.6). ** indicates that value is estimated from surrounding stations, or else determined from manual inspection of pressure data. station surface p station surface p station surface p station surface p number offset (dbar) number offset (dbar) number offset (dbar) number offset (dbar) ---------------------- ---------------------- ---------------------- ---------------------- 1 TEST -1.62 27 SR3 -2.82 53 SR3 -3.55 79 PET -4.28 2 TEST -1.62** 28 SR3 -3.23 54 SR3 -3.53 80 PET -3.50 3 SR3 -1.18 29 SR3 -2.93 55 SR3 -3.63 81 PET -3.67 4 SR3 -1.34 30 SR3 -2.60 56 SR3 -2.78 82 PET -3.40 5 SR3 -1.13 31 SR3 -2.81 57 SR3 -3.91 83 PET -3.27 6 SR3 -1.41 32 SR3 -3.27 58 SR3 -4.57 84 PET -3.86 7 SR3 -1.41** 33 SR3 -3.03 59 SR3 -3.61 85 PET -3.78 8 SR3 -1.72 34 SR3 -3.35 60 SR3 -3.71 86 PET -3.61 9 SR3 -1.73** 35 SR3 -3.14 61 SR3 -4.07 87 PET -3.63 10 SR3 -1.73** 36 SR3 -3.49 62 SR3 -3.88 88 PET -3.63** 11 SR3 -1.73 37 SR3 -3.45 63 SR3 -5.02 89 PET -3.05 12 SR3 -2.44 38 SR3 -3.28 64 SR3 -4.01 90 PET -3.48 13 SR3 -1.89 39 SR3 -3.26 65 SR3 -3.64 91 PET -3.17 14 SR3 -1.89 40 SR3 -2.96 66 SR3 -3.94 92 PET -3.81 15 SR3 -1.92 41 SR3 -2.85 67 SR3 -4.62 93 PET -3.55 16 SR3 -1.65 42 SR3 -2.66 68 SR3 -3.63 94 PET -3.71 17 SR3 -1.90 43 SR3 -3.14 69 SR3 -3.66 95 PET -3.60** 18 SR3 -1.64 44 SR3 -3.12 70 SR3 -3.80** 96 PET -3.80** 19 SR3 -2.18 45 SR3 -2.89 71 SR3 -3.78 97 PET -3.68 20 SR3 -1.90 46 SR3 -2.62 72 SR3 -4.27 98 PET -3.41 21 SR3 -1.91 47 SR3 -3.00 73 SR3 -3.87 99 PET -3.06 22 SR3 -2.29 48 SR3 -3.30 74 ULS -3.27 100 TS -3.66 23 SR3 -2.29 49 SR3 -2.96 75 ULS -3.38 101 TS -3.62 24 SR3 -2.68 50 SR3 -3.20 76 PET -3.76 102 BPR -2.50 25 SR3 -2.37 51 SR3 -3.31 77 PET -3.86 26 SR3 -2.53 52 SR3 -3.08 78 PET -3.81 Table 12: Missing data points in 2 dbar-averaged files (i.e. *.all files). "1" indicates missing data for the indicated parameters (T=temperature; S=salinity, (T , specific volume anomaly and geopotential anomaly; O=dissolved oxygen; PAR=photosynthetically active radiation; F/PAR=fluorescence and photosynthetically active radiation). Note that jmin is the minimum number of data points required in a 2 dbar bin to form the 2 dbar average (Table 8). station pressures (dbar) F/ reason number where data missing T S O PAR PAR --------------------------------------------------------------------------------------- 1 entire profile 1 bad oxygen sensor 1 62-70, 96-104, 216-218 1 bad digital channel data 1 232-236, 424-426 1 bad digital channel data 1 10, 12, 16 1 bad digital channel data 2 entire profile 1 1 1 1 downcast data lost 3 188 to 212 1 bad oxygen temperature data 6 entire profile 1 1 oil drained from oxygen sensor 8 entire profile 1 1 oil drained from oxygen sensor 9 entire profile 1 1 oil drained from oxygen sensor 10 entire profile 1 1 oil drained from oxygen sensor 18 entire profile 1 no oxygen bottle data for calibration 20 2-56 1 bad oxygen data (deleted) 27 entire profile 1 no oxygen bottle data for calibration 35 entire profile 1 no oxygen bottle data for calibration 37 186-202 1 bad oxygen data (deleted) 48 134-164 1 bad oxygen data (deleted) 49 154 1 bad oxygen data (deleted) 49 156 1 1 1 no. of data pts in 2 dbar bin < jmin 51 62-200 1 bad oxygen data (deleted) 54 1402-1434 1 1 1 1 fouling of conductivity cell 55 274-360 1 bad oxygen data (deleted) 56 2, 12 1 bad oxygen data (deleted) 57 entire profile 1 no oxygen bottle data for calibration 60 54-70 1 bad oxygen data (deleted) 62 entire profile 1 no oxygen bottle data for calibration 65 228-256, 298-320 1 bad oxygen data (deleted) 71 42-72 1 bad oxygen data (deleted) 77 32-76 1 bad oxygen data (deleted) 81 2-24 1 bad oxygen data (deleted) 82 2-30 1 bad oxygen data (deleted) 82 72-112 1 1 1 1 fouling of conductivity cell 89 entire profile 1 no oxygen bottle data for calibration 102 2-12 1 1 1 bad data (deleted 2-10) Table 13: CTD conductivity calibration coefficients. F1 , F2 and F3 are respectively conductivity bias, slope and station-dependent correction calibration terms. n is the number of samples retained for calibration in each station grouping; _ is the standard deviation of the conductivity residual for the n samples in the station grouping (eqn A2.19). station grouping F1 F2 F3 n sigma ---------------------------------------------------------------------------------- 001 to 002 SR3 0.36218645E-02 0.96061420E-03 -0.26311155E-06 27 0.000586 003 to 004 SR3 -0.79432793E-02 0.96056305E-03 -0.25317777E-07 26 0.001233 005 to 006 SR3 -0.22839899E-01 0.96087041E-03 -0.71206286E-08 38 0.001247 007 to 009 SR3 -0.26490370E-01 0.96077609E-03 0.21866512E-07 42 0.001213 010 to 014 SR3 -0.25781659E-01 0.96079196E-03 0.10695138E-07 84 0.001266 015 to 016 SR3 -0.29190034E-01 0.96018556E-03 0.63939699E-07 27 0.001195 017 to 019 SR3 -0.33231401E-01 0.96240143E-03 -0.64696331E-07 40 0.001025 020 to 023 SR3 -0.27351840E-01 0.96051767E-03 0.25721478E-07 74 0.001974 024 to 025 SR3 -0.30299213E-01 0.96067744E-03 0.18476654E-07 42 0.001161 026 to 028 SR3 -0.42514665E-01 0.96120679E-03 0.12865039E-07 45 0.001309 029 to 031 SR3 -0.50353096E-01 0.96279223E-03 -0.31623779E-07 49 0.001088 032 to 033 SR3 -0.45130955E-01 0.95878717E-03 0.87353864E-07 43 0.001058 034 to 038 SR3 -0.26467805E-01 0.96070695E-03 0.10383061E-07 93 0.001443 039 to 043 SR3 -0.40217876E-01 0.96192558E-03 -0.93443880E-08 76 0.001547 044 to 048 SR3 -0.25328764E-01 0.96000326E-03 0.21846514E-07 112 0.001417 049 to 051 SR3 0.59190565E-02 0.96019179E-03 -0.28442240E-08 48 0.001926 052 to 054 SR3 0.26730532E-02 0.95881628E-03 0.25804226E-07 51 0.001744 055 to 056 SR3 -0.69062123E-02 0.95954156E-03 0.17440878E-07 42 0.001954 057 to 059 SR3 -0.11119307 0.96722355E-03 -0.58908391E-07 43 0.001861 060 to 063 SR3 -0.66439878E-02 0.95750857E-03 0.49025023E-07 65 0.001911 064 to 073 SR3 0.21312741E-01 0.95976433E-03 -0.20776732E-08 79 0.001890 074 to 088 PET 0.77402498E-01 0.95470654E-03 0.36176548E-07 43 0.001747 089 to 090 PET -0.80990196E-02 0.95899591E-03 0.17099928E-07 30 0.000962 091 to 095 PET 0.42051964E-01 0.95938559E-03 -0.45634442E-08 100 0.001398 096 to 099 PET 0.53432623E-01 0.95899046E-03 -0.35852866E-08 60 0.001041 100 to 102TS/BPR -0.48512144E-02 0.95829965E-03 0.21607866E-07 25 0.000738 Table 14: Station-dependent-corrected conductivity slope term (F2 + F3 . N), for station number N, and F2 and F3 the conductivity slope and station- dependent correction calibration terms respectively. station (F2 + F3 . N) station (F2 + F3 . N) station (F2 + F3 . N) number number number ----------------------- ---------------------- ----------------------- 1 TEST 0.96035109E-03 35 SR3 0.96107036E-03 69 SR3 0.95962097E-03 2 TEST 0.96008798E-03 36 SR3 0.96108075E-03 70 SR3 0.95961890E-03 3 SR3 0.96048709E-03 37 SR3 0.96109113E-03 71 SR3 0.95961682E-03 4 SR3 0.96046177E-03 38 SR3 0.96110151E-03 72 SR3 0.95961474E-03 5 SR3 0.96083481E-03 39 SR3 0.96156115E-03 73 SR3 0.95961266E-03 6 SR3 0.96082769E-03 40 SR3 0.96155181E-03 74 ULS 0.95738360E-03 7 SR3 0.96092915E-03 41 SR3 0.96154246E-03 75 ULS 0.95741978E-03 8 SR3 0.96095102E-03 42 SR3 0.96153312E-03 76 PET 0.95745596E-03 9 SR3 0.96097289E-03 43 SR3 0.96152377E-03 77 PET 0.95749213E-03 10 SR3 0.96089892E-03 44 SR3 0.96096450E-03 78 PET 0.95752831E-03 11 SR3 0.96090961E-03 45 SR3 0.96098635E-03 79 PET 0.95756449E-03 12 SR3 0.96092031E-03 46 SR3 0.96100820E-03 80 PET 0.95760066E-03 13 SR3 0.96093100E-03 47 SR3 0.96103004E-03 81 PET 0.95763684E-03 14 SR3 0.96094170E-03 48 SR3 0.96105189E-03 82 PET 0.95767302E-03 15 SR3 0.96114466E-03 49 SR3 0.96005242E-03 83 PET 0.95770919E-03 16 SR3 0.96120860E-03 50 SR3 0.96004958E-03 84 PET 0.95774537E-03 17 SR3 0.96130159E-03 51 SR3 0.96004673E-03 85 PET 0.95778155E-03 18 SR3 0.96123689E-03 52 SR3 0.96015810E-03 86 PET 0.95781772E-03 19 SR3 0.96117220E-03 53 SR3 0.96018391E-03 87 PET 0.95785390E-03 20 SR3 0.96103210E-03 54 SR3 0.96020971E-03 88 PET 0.95789008E-03 21 SR3 0.96105782E-03 55 SR3 0.96050081E-03 89 PET 0.96051780E-03 22 SR3 0.96108354E-03 56 SR3 0.96051825E-03 90 PET 0.96053490E-03 23 SR3 0.96110926E-03 57 SR3 0.96386578E-03 91 PET 0.95897032E-03 24 SR3 0.96112088E-03 58 SR3 0.96380687E-03 92 PET 0.95896576E-03 25 SR3 0.96113935E-03 59 SR3 0.96374796E-03 93 PET 0.95896119E-03 26 SR3 0.96154128E-03 60 SR3 0.96045007E-03 94 PET 0.95895663E-03 27 SR3 0.96155415E-03 61 SR3 0.96049909E-03 95 PET 0.95895207E-03 28 SR3 0.96156701E-03 62 SR3 0.96054812E-03 96 PET 0.95864627E-03 29 SR3 0.96187514E-03 63 SR3 0.96059714E-03 97 PET 0.95864269E-03 30 SR3 0.96184352E-03 64 SR3 0.95963136E-03 98 PET 0.95863910E-03 31 SR3 0.96181190E-03 65 SR3 0.95962928E-03 99 PET 0.95863551E-03 32 SR3 0.96158250E-03 66 SR3 0.95962721E-03 100 TS 0.96046044E-03 33 SR3 0.96166985E-03 67 SR3 0.95962513E-03 101 TS 0.96048204E-03 34 SR3 0.96105998E-03 68 SR3 0.95962305E-03 102 BPR 0.96050365E-03 Table 15: CTD raw data scans, mostly in the vicinity of artificial density inversions, flagged for special treatment. Note that the pressure listed is approximate only; possible actions taken are either to ignore the raw data scans for all further calculations, or to apply a linear interpolation over the region of the bad data scans. Causes of bad data, listed in the last column, are detailed in Appendix 2 (section A2.11.1). For the raw scan number ranges, the lowest and highest scans numbers are not included in the ignore or interpolate actions. station approximate raw scan action reason number pressure (dbar) numbers taken --------------------------------------------------------------------------------------- 1 TEST 16; 54; 54 6675-6688; 9873-9898; 9951-9958 ignore bad press. temp. data 1 TEST 54; 62; 68 10791-10798;11889-11896;11967-11974 ignore bad press. temp. data 1 TEST 95;102;206 13059-13090;13329-13372;18657-18688 ignore bad press. temp. data 1 TEST 209;214;215 18849-18874;19263-19270;19287-19318 ignore bad press. temp. data 1 TEST 233; 424 20385-20452; 27864-27931 ignore bad press. temp. data 7 SR3 0; 1 1225-1250; 1261-1382 ignore bad press. temp. data 8 SR3 1164 65925-66036 ignore fouling of cond. cell 26 SR3 812 61152-61281 ignore fouling of cond. cell 39 SR3 16 2154-2159 ignore wake effect 49 SR3 155 9707-9884 ignore wake effect 50 SR3 29 2415-2633 ignore fouling of cond. cell 54 SR3 1397-1433 73257-74347 ignore fouling of cond. cell 56 SR3 63 13483-13639 ignore wake effect 56 SR3 731 50347-50470 ignore fouling of cond. cell 59 SR3 65; 73 7065-7073; 7425-7431 ignore fouling of cond. cell 70 SR3 0 1-5000 ignore seawater in sensor cap 76 PET 72; 93; 97 5789-5791; 7548-7550; 7780-7783 ignore suspect pressure value 82 PET 70-113 4552-7156 ignore fouling of cond. cell 87 PET 61 3845-3847 ignore suspect pressure value 102 BPR 0-50 1838-15394 ignore preliminary dip to 50 dbar Table 16: Suspect 2 dbar averages. station suspect 2 dbar values (dbar) number bad questionable reason --------------------------------------------------------------------------------------- Suspect salinity values 30 SR3 - 76 salinity spike in steep local gradient 30 SR3 78 - salinity spike in steep local gradient 31 SR3 - 80 salinity spike in steep local gradient 32 SR3 - 58-62 salinity spike in steep local gradient 33 SR3 - 44-50; 72-74 salinity spike in steep local gradient 34 SR3 - 36 salinity spike in steep local gradient 39 SR3 - 16-20 salinity spike in steep local gradient 40 SR3 - 8; 18-20 salinity spike in steep local gradient 41 SR3 - 14; 104 salinity spike in steep local gradient 42 SR3 - 16 salinity spike in steep local gradient 43 SR3 - 34-40 salinity spike in steep local gradient 44 SR3 - 18 salinity spike in steep local gradient 45 SR3 - 30-34 salinity spike in steep local gradient 56 SR3 - 60 salinity spike in steep local gradient 56 SR3 62 - salinity spike in steep local gradient 94 PET - 36 salinity spike in steep local gradient Suspect dissolved oxygen values 31 SR3 - 76-282 no nearby oxygen bottles to confirm values Table 17a: Suspect 2 dbar-averaged data from near the surface (applies to all parameters other than dissolved oxygen, except where noted). Note that for station 102, suspect near surface values have been deleted from the data. station suspect 2 dbar station suspect 2 dbar number values (dbar) number values (dbar) bad questionable comment bad questionable comment ----------------------------------------- --------------------------------------------------------- 6 SR3 - 2-20 temperature ok 71 SR3 - 2-4 temperature ok 11 SR3 - 2 73 SR3 - 2-4 temperature ok 23 SR3 - 2 74 ULS - 2-4 temperature ok 26 SR3 - 2 76 PET 2-20 30 SR3 - 2-4 temperature ok 77 PET - 2 temperature ok 48 SR3 - 2 temperature ok 78 PET - 2 temperature ok 49 SR3 - 2 80 PET - 2 temperature ok 52 SR3 - 2 temperature ok 82 PET - 2-4 temperature ok 54 SR3 - 2 temperature ok 83 PET - 2 temperature ok 60 SR3 - 2-4 temperature ok 86 PET - 2 temperature ok 61 SR3 - 2-4 temperature ok 87 PET - 2 temperature ok 63 SR3 - 2 temperature ok 91 PET - 2 temperature ok 64 SR3 - 2-4 temperature ok 98 PET - 2 temperature ok 66 SR3 - 2 temperature ok 99 PET - 2 temperature ok 68 SR3 - 2 temperature ok 102 BPR 2-10(deleted) Table 17b: Suspect 2 dbar-averaged dissolved oxygen data from near the surface. station suspect dissolved oxygen station suspect dissolved oxygen number 2 dbar values (dbar) number 2 dbar values (dbar) bad questionable bad questionable --------------------------------- --------------------------------- 3 SR3 - 2-10 50 SR3 - 2-14 5 SR3 - 2-6 54 SR3 - 2-6 14 SR3 - 2-12 56 SR3 - 4 15 SR3 - 2-10 66 SR3 - 2 17 SR3 - 2-10 67 SR3 - 2 19 SR3 - 2 71 SR3 - 2 21 SR3 - 2-10 72 SR3 - 2 22 SR3 - 2-10 73 SR3 - 2-12 24 SR3 - 2-18 77 PET - 2-12 26 SR3 - 2-6 86 PET - 2-12 33 SR3 - 2-6 94 PET - 2-10 40 SR3 - 2-6 99 PET - 2-14 41 SR3 - 2-12 100 TS - 2-18 44 SR3 - 2-10 102 BPR 2-10(deleted) 14-16 Table 18: 2 dbar averages interpolated from surrounding 2 dbar values, for the indicated parameters (T=temperature; S=salinity, (T , specific volume anomaly and geopotential anomaly; O=dissolved oxygen; PAR=photosynthetically active radiation; F=fluorescence). station interpolated 2 dbar parameters station interpolated 2 dbar parameters number values (dbar) interpolated number values (dbar) interpolated ------------------------------------------ ------------------------------------------ 1 TEST 102;216;234 T,S,O,PAR,F 8 SR3 1164-1168 S 60 SR3 3486 T,S,O,PAR 14 SR3 1726 T,S,O,PAR 23 SR3 1726 T,S,O,PAR 65 SR3 56-58 O 25 SR3 1726 T,S,O,PAR 26 SR3 812-814 T,S,O,PAR 66 SR3 1726 T,S,O,PAR 28 SR3 3486 T,S,O,PAR 29 SR3 1726 T,S,O,PAR 71 SR3 288 T,S,O,PAR 31 SR3 1726 T,S,O,PAR 34 SR3 2158 T,S,O,PAR 47 SR3 1726 T,S,O,PAR 48 SR3 3486 T,S,O,PAR 49 SR3 156 PAR 85 PET 3486 T,S,O,PAR 50 SR3 1126-1130;1134 T,S,O,PAR 87 PET 1726;3266;3486 T,S,O,PAR 51 SR3 2380;3486 T,S,O,PAR 88 PET 1726 T,S,O,PAR 53 SR3 1726;3486;4370 T,S,O,PAR 90 PET 1726 T,S,O,PAR 92 PET 474 T,S,O,PAR 54 SR3 3486 T,S,O,PAR 93 PET 454-456 T,S,O,PAR 55 SR3 2270 T,S,O,PAR 94 PET 1726 T,S,O,PAR 101 PET 3222;3486;3838 T,S,O,PAR 56 SR3 732 T,S,O,PAR 102 BPR 12 PAR 59 SR3 3486 T,S,O,PAR Table 19: CTD dissolved oxygen calibration coefficients. K1, K2, K3, K4, K5 and K6 are respectively oxygen current slope, oxygen sensor time constant, oxygen current bias, temperature correction term, weighting factor, and pressure correction term. dox is equal to 2.8( (for ( defined as in eqn A2.24, Appendix 2); n is the number of samples retained for calibration in each station or station grouping. station number K1 K2 K3 K4 K5 K6 dox n ----------------------------------------------------------------------------------------------------- 1 - - - - - - - - 2 - - - - - - - - 3 4.6380 6.0000 -0.499 -0.35246E-01 0.85692 0.19557E-03 0.09458 15 4 5.7430 6.0000 -0.694 -0.45441E-01 0.63331 0.49313E-04 0.19995 17 5 5.2373 7.0000 -0.662 -0.42852E-01 0.98652 0.16922E-03 0.17248 19 6 - - - - - - - - 7 4.1457 8.5000 0.089 -0.44330E-01 0.52732 -0.40804E-03 0.18887 4 8 - - - - - - - - 9 - - - - - - - - 10 - - - - - - - - 11 2.3532 8.0000 0.189 -0.33572E-01 1.80280 0.24611E-03 0.13388 6 12 6.1765 9.0000 -0.732 -0.63687E-01 1.07130 0.13644E-03 0.19449 20 13 6.0734 7.0000 -0.754 -0.31820E-01 0.65404 0.14486E-03 0.17334 18 14 5.9180 7.0000 -0.726 -0.28478E-01 0.74514 0.14197E-03 0.13436 21 15 6.5298 7.0000 -0.839 -0.48408E-01 0.85422 0.16013E-03 0.16797 20 16 2.7427 6.0000 0.172 -0.10445E-01 0.82533 0.22646E-05 0.22420 6 17 6.9610 6.0000 -0.903 -0.40488E-01 0.35658 0.14081E-03 0.14863 23 18 - - - - - - - -- 19 6.5143 7.0000 -0.816 -0.40945E-01 0.66320 0.14422E-03 0.17335 21 20 6.0785 8.0000 -0.734 -0.41127E-01 0.99372 0.14462E-03 0.21642 22 21 3.6227 7.0000 0.074 -0.35112E-01 0.53569 0.45722E-03 0.25442 5 22 6.3627 6.0000 -0.800 -0.37144E-01 0.74879 0.15083E-03 0.18242 24 23 4.5865 7.0000 -0.489 -0.20154E-01 0.85461 0.12940E-03 0.20779 21 24 6.7039 6.0000 -0.925 -0.32788E-01 0.10467 0.16169E-03 0.16233 20 25 6.3919 7.0000 -0.843 -0.33998E-01 0.09458 0.14892E-03 0.17711 24 26 6.5035 7.0000 -0.825 -0.43069E-01 0.05608 0.12397E-03 0.22038 23 27 - - - - - - - - 28 5.3339 8.0000 -0.605 -0.22513E-01 0.96866 0.12891E-03 0.13132 24 29 6.6119 7.0000 -0.824 -0.57820E-01 0.91699 0.13465E-03 0.15826 23 30 3.4514 8.0000 0.015 -0.31230E-01 0.98815 0.11837E-03 0.01867 4 31 6.6362 7.0000 -0.840 -0.41346E-01 0.09029 0.11522E-03 0.17252 23 32 6.3479 7.0000 -0.800 -0.51777E-01 0.79205 0.15087E-03 0.15665 24 33 6.4601 7.0000 -0.787 -0.54719E-01 0.63788 0.11984E-03 0.14759 20 34 6.3667 8.0000 -0.800 -0.46579E-01 0.75000 0.14241E-03 0.14771 24 35 - - - - - - - - 36 6.4085 8.0000 -0.797 -0.49059E-01 0.68932 0.12963E-03 0.15741 24 37 6.3854 7.0000 -0.794 -0.52244E-01 0.65159 0.13701E-03 0.14836 23 38 6.0999 8.0000 -0.800 -0.14073E-01 0.48763 0.16022E-03 0.21646 24 39 4.5162 8.0000 -0.300 -0.36802E-01 1.85790 0.77629E-03 0.18002 5 40 6.8948 6.0000 -0.848 -0.79608E-01 0.66005 0.11927E-03 0.15073 23 41 6.1060 7.0000 -0.731 -0.39274E-01 0.62488 0.12322E-03 0.19740 24 42 6.1393 8.0000 -0.733 -0.59422E-01 0.20872 0.11615E-03 0.22714 19 43 5.8006 6.0000 -0.689 -0.39875E-01 0.99597 0.51551E-03 0.01830 6 44 6.1588 7.0000 -0.752 -0.35296E-01 0.09952 0.11839E-03 0.13997 23 45 6.3478 7.0000 -0.764 -0.66340E-01 0.33186 0.11850E-03 0.19849 22 46 5.9345 7.0000 -0.669 -0.59129E-01 0.48042 0.10859E-03 0.21122 20 47 5.9509 7.0000 -0.699 -0.39938E-01 0.18616 0.11603E-03 0.14874 23 48 6.2069 9.0000 -0.718 -0.86073E-01 0.38118 0.11437E-03 0.21979 23 49 5.1940 7.0000 -0.445 -0.51910E-01 0.22550 -0.35579E-03 0.07725 6 50 6.2883 7.0000 -0.763 -0.44615E-01 1.42890 0.14517E-03 0.10690 24 51 6.2979 8.0000 -0.716 -0.10704E+00 0.35117 0.10695E-03 0.23228 20 52 5.9395 8.0000 -0.681 -0.37050E-01 0.91636 0.50784E-03 0.06294 6 53 6.1534 7.0000 -0.759 -0.29688E-01 0.67398 0.13619E-03 0.16369 24 54 4.8639 7.0000 -0.400 -0.10146E+00 0.23846 0.73753E-04 0.24248 21 55 6.3252 8.0000 -0.800 -0.32462E-01 0.61883 0.13900E-03 0.18171 23 56 6.1977 8.0000 -0.800 -0.12419E-01 0.83712 0.15614E-03 0.13539 24 57 - - - - - - - - 58 6.2700 9.0000 -0.750 -0.49034E-01 0.99695 0.13949E-03 0.21450 23 59 6.4541 7.0000 -0.806 -0.32198E-01 0.64776 0.13141E-03 0.17712 24 60 6.7973 7.0000 -0.898 -0.24175E-01 0.81216 0.15090E-03 0.18743 23 61 6.6318 8.0000 -0.778 -0.77340E-01 0.72495 0.11865E-03 0.20031 24 62 - - - - - - - - 63 6.3574 7.0000 -0.694 -0.80017E-01 0.73133 0.10015E-03 0.19322 23 64 7.2582 7.0000 -0.832 -0.13165E+00 0.75470 0.10743E-03 0.22842 23 65 6.9347 7.0000 -0.807 -0.11184E+00 0.74299 0.12646E-03 0.28536 21 66 6.4475 8.0000 -0.699 -0.11351E+00 0.57400 0.10346E-03 0.17570 19 67 6.2322 7.0000 -0.526 -0.15564E+00 0.73666 0.92372E-04 0.20346 18 68 4.3597 8.0000 -0.271 -0.59517E-01 0.11228 0.49944E-04 0.18126 10 69 5.0115 8.0000 -0.542 0.16156E+00 0.53560 0.29590E-03 0.33524 9 70 5.2768 8.0000 -0.370 -0.11754E+00 0.76528 0.31268E-03 0.24593 5 71 7.0209 8.0000 -0.748 -0.11598E+00 0.71450 0.11669E-03 0.18306 11 72 6.2069 7.0000 -0.688 -0.10458E+00 0.32446 0.11254E-03 0.21289 12 73 6.7986 8.0000 -0.778 -0.33701E-01 0.98052 0.19701E-03 0.09915 7 74 5.4635 6.0000 -0.389 -0.62671E-01 0.90499 0.30995E-04 0.16128 10 75 6.4537 8.0000 -0.756 -0.35922E-01 0.19910 0.12203E-03 0.19733 22 76 7.1131 8.0000 -0.719 -0.38633E-01 1.44620 0.14820E-03 0.18317 11 77 3.9327 8.0000 -0.138 -0.14997E+00 0.37492 0.75369E-04 0.20512 10 78 5.7709 8.0000 -0.495 -0.75655E-01 0.70279 0.32271E-03 0.21500 9 79 7.2419 7.0000 -0.714 -0.14947E+00 0.73222 0.52000E-04 0.16924 15 80 6.7614 8.0000 -0.712 -0.97309E-01 0.57134 0.73130E-04 0.17648 20 81 8.2262 7.0000 -0.951 -0.14529E+00 0.78983 0.10095E-03 0.24020 22 82 2.5566 8.0000 0.162 -0.12975E+00 0.22082 0.19988E-04 0.30134 23 83 6.4409 7.0000 -0.669 -0.28786E-01 0.76568 0.20110E-03 0.12518 10 84 7.6570 7.0000 -0.846 -0.13857E+00 0.74234 0.10002E-03 0.19037 23 85 6.7930 9.0000 -0.712 -0.92281E-01 0.78289 0.96905E-04 0.22427 22 86-87 6.3847 7.0000 -0.715 -0.51692E-01 0.76584 0.12598E-03 0.13637 32 88 6.6636 8.0000 -0.798 -0.34515E-01 0.04419 0.11580E-03 0.19381 24 89 - - - - - - - - 90 7.0042 8.0000 -0.871 -0.37846E-01 0.07843 0.12325E-03 0.07300 24 91 6.1781 8.0000 -0.627 -0.91690E-01 0.06101 0.75328E-04 0.24590 23 92 6.9241 7.0000 -0.855 -0.34158E-01 0.29276 0.13071E-03 0.15088 24 93 6.2057 7.0000 -0.707 -0.35806E-01 0.13871 0.11699E-03 0.15338 24 94 7.5742 6.0000 -0.993 -0.35650E-01 0.54185 0.14447E-03 0.17418 18 95 7.0757 6.0000 -0.893 -0.32040E-01 0.80723 0.14461E-03 0.17630 23 96 6.8257 7.0000 -0.831 -0.33619E-01 0.90412 0.12997E-03 0.22452 24 97 6.2544 6.0000 -0.707 -0.38825E-01 0.07702 0.11080E-03 0.20367 17 98 6.2725 6.0000 -0.712 -0.36046E-01 0.34643 0.11232E-03 0.17540 17 99 7.6084 9.0000 -0.972 -0.46474E-01 0.99575 0.14924E-03 0.21023 16 100 6.3074 8.0000 -0.747 -0.44495E-01 0.29023 0.14280E-03 0.24388 10 101 6.8670 8.0000 -0.872 -0.26029E-01 0.87188 0.14902E-03 0.13024 21 102 6.0076 5.0000 -0.548 -0.10098E-01 0.89303 0.30454E-05 0.35115 22 Table 20: Starting values for CTD dissolved oxygen calibration coefficients prior to iteration, and coefficients varied during iteration (sections A2.12.1 and A2.12.3). Note that coefficients not varied during iteration are held constant at the starting value. Station coefficients number K1 K2 K3 K4 K5 K6 varied --------------------------------------------------------------------------------- 1 - - - - - - - 2 - - - - - - - 3 4.6000 6.0000 -0.500 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 4 5.7000 6.0000 -0.700 -0.400E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 5 5.0500 7.0000 -0.700 -0.360E-01 0.900 0.15000E-03 K1 K3 K4 K5 K6 6 - - - - - - - 7 4.2500 8.5000 0.020 -0.360E-01 0.700 0.15000E-03 K1 K3 K4 K5 K6 8 - - - - - - - 9 - - - - - - - 10 - - - - - - - 11 2.3000 8.0000 0.190 -0.400E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 12 6.1000 9.0000 -0.750 -0.750E-01 0.900 0.15000E-03 K1 K3 K4 K5 K6 13 6.5000 7.0000 -0.600 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 14 5.9000 7.0000 -0.710 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 15 6.2000 7.0000 -0.850 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 16 2.6000 6.0000 0.180 -0.110E-01 0.650 0.80000E-03 K1 K3 K4 K5 K6 17 6.9000 6.0000 -0.900 -0.400E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 18 - - - - - - - 19 6.5000 7.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 20 5.9000 8.0000 -0.760 -0.500E-01 0.650 0.15000E-03 K1 K3 K4 K5 K6 21 3.3300 7.0000 0.000 -0.370E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 22 6.3000 6.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 23 4.5000 7.0000 -0.500 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 24 6.8000 6.0000 -0.920 -0.300E-01 0.900 0.15000E-03 K1 K3 K4 K5 K6 25 6.2000 7.0000 -0.880 -0.420E-01 0.700 0.15000E-03 K1 K3 K4 K5 K6 26 6.4000 7.0000 -0.820 -0.360E-01 0.800 0.15000E-03 K1 K3 K4 K5 K6 27 - - - - - - - 28 5.2000 8.0000 -0.630 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 29 6.5000 7.0000 -0.810 -0.500E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 30 3.4000 8.0000 0.000 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 31 6.5900 7.0000 -0.810 -0.420E-01 0.950 0.15000E-03 K1 K3 K4 K5 K6 32 2.2800 7.0000 -0.800 -0.480E-01 0.800 0.15000E-03 K1 K4 K5 K6 33 6.4000 7.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 34 6.3000 8.0000 -0.800 -0.400E-01 0.750 0.15000E-03 K1 K4 K6 35 - - - - - - - 36 6.3300 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 37 6.3000 7.0000 -0.800 -0.360E-01 0.740 0.15000E-03 K1 K3 K4 K5 K6 38 6.1500 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 39 4.3200 8.0000 -0.400 -0.360E-01 0.950 0.15000E-03 K1 K3 K4 K5 K6 40 6.8000 6.0000 -0.880 -0.400E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 41 5.9000 7.0000 -0.710 -0.370E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 42 5.9500 8.0000 -0.720 -0.360E-01 0.200 0.15000E-03 K1 K3 K4 K5 K6 43 5.8000 6.0000 -0.700 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 44 6.1000 7.0000 -0.750 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 45 6.1500 7.0000 -0.800 -0.400E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 46 5.8000 7.0000 -0.700 -0.400E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 47 5.8200 7.0000 -0.700 -0.380E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 48 6.3000 9.0000 -0.780 -0.700E-01 0.600 0.15000E-03 K1 K3 K4 K5 K6 49 5.1000 7.0000 -0.450 -0.500E-01 0.350 0.10000E-03 K1 K3 K4 K5 K6 50 6.2000 7.0000 -0.790 -0.500E-01 0.900 0.15000E-03 K1 K3 K4 K5 K6 51 6.6000 8.0000 -0.700 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 52 5.9000 8.0000 -0.700 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 53 6.1000 7.0000 -0.750 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 54 4.1000 7.0000 -0.400 -0.360E-01 0.600 0.15000E-03 K1 K4 K5 K6 55 6.4000 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K4 K5 K6 56 6.5000 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K4 K5 K6 57 - - - - - - - 58 6.7000 9.0000 -0.750 -0.110E+00 0.900 0.15000E-03 K1 K4 K5 K6 59 6.4000 7.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 60 6.9500 7.0000 -0.850 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 61 6.5500 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 62 - - - - - - - 63 6.1500 7.0000 -0.700 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 64 7.2000 7.0000 -0.890 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 65 6.9000 7.0000 -0.860 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 66 5.9700 8.0000 -0.750 -0.420E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 67 6.0300 7.0000 -0.580 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 68 4.1300 8.0000 -0.300 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 69 4.7400 8.0000 -0.600 -0.360E-01 0.740 0.15000E-03 K1 K3 K4 K5 K6 70 5.3800 8.0000 -0.400 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 71 6.7400 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 72 5.9600 7.0000 -0.850 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 73 6.7000 8.0000 -0.820 -0.360E-01 0.660 0.15000E-03 K1 K3 K4 K5 K6 74 5.5000 6.0000 -0.350 -0.500E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 75 6.3900 8.0000 -0.740 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 76 7.2000 8.0000 -0.700 -0.350E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 77 4.0700 8.0000 -0.350 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 78 5.3600 8.0000 -0.660 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 79 6.3100 7.0000 -0.750 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 80 5.7700 8.0000 -0.700 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 81 6.4800 7.0000 -0.900 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 82 5.8400 8.0000 -0.790 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 83 6.4000 7.0000 -0.650 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 84 7.4000 7.0000 -0.880 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 85 6.7000 9.0000 -0.750 -0.360E-01 0.770 0.15000E-03 K1 K3 K4 K5 K6 86-87 6.4000 7.0000 -0.730 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 88 6.6500 8.0000 -0.800 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 89 - - - - - - - 90 6.9000 8.0000 -0.900 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 91 6.5000 8.0000 -0.550 -0.400E-01 0.700 0.15000E-03 K1 K3 K4 K5 K6 92 7.4000 7.0000 -0.730 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 93 6.3000 7.0000 -0.680 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 94 7.7000 6.0000 -0.910 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 95 7.0000 6.0000 -0.900 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 96 6.7000 7.0000 -0.870 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 97 6.6000 6.0000 -0.600 -0.360E-01 0.850 0.15000E-03 K1 K3 K4 K5 K6 98 6.3000 6.0000 -0.700 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 99 7.7000 9.0000 -0.920 -0.450E-01 0.800 0.15000E-03 K1 K3 K4 K5 K6 100 6.2000 8.0000 -0.630 -0.400E-01 0.650 0.15000E-03 K1 K3 K4 K5 K6 101 7.3000 8.0000 -0.750 -0.360E-01 0.750 0.15000E-03 K1 K3 K4 K5 K6 102 5.8000 5.0000 -0.450 -0.750E-01 0.900 0.20000E-03 K1 K3 K4 K5 K6 Table 21: Questionable dissolved oxygen Niskin bottle sample values (not deleted from hydrology data file). station rosette number position ----------------- 6 _ 16 Table 22: Questionable nutrient sample values (not deleted from hydrology data file). PHOSPHATE NITRATE SILICATE station rosette station rosette station rosette number position number position number position ------------------ ------------------ ------------------ 4 23 5 2 6 2 8 8 10 4 14 8 14 8,9 22 3,8 24 8 24 8 24 4,8 25 2 31 10 31 10 31 10 32 4 37 3 38 5 41 10 41 10 42 3 46 1 54 5,7 55 9 60 7 61 4,7 66 19 66 12,14 67 6 67 6 68 5 68 5 73 3 75 4 75 4 87 4 Table 23: Laboratory temperatures Tl at the times of nutrient analyses, used for conversion to gravimetric units for WOCE format data (Appendix 7). The lower temperatures for stations 6 to 11, and station 16, were due to malfunction of the laboratory air-conditioning system. stn Tl stn Tl stn Tl stn Tl stn Tl no. (oC) no. (oC) no. (oC) no. (oC) no. (oC) ----------- ----------- ----------- ----------- ------------------ 1 TEST - 22 SR3 22.5 43 SR3 22.5 64 SR3 22.0 85 PET 23.0 2 TEST - 23 SR3 22.5 44 SR3 23.5 65 SR3 22.0 86 PET 22.0 3 SR3 22.0 24 SR3 21.5 45 SR3 21.0 66 SR3 22.0 87 PET 22.0 4 SR3 22.0 25 SR3 22.5 46 SR3 24.5 67 SR3 22.0 88 PET 22.0 5 SR3 22.0 26 SR3 22.5 47 SR3 22.0 68 SR3 22.0 89 PET 22.0 6 SR3 20.0 27 SR3 - 48 SR3 22.5 69 SR3 22.5 90 PET 22.0 7 SR3 20.0 28 SR3 22.5 49 SR3 23.0 70 SR3 22.5 91 PET 23.0 8 SR3 20.0 29 SR3 22.5 50 SR3 23.0 71 SR3 22.3 92 PET 20.5 9 SR3 18.0 30 SR3 22.5 51 SR3 23.0 72 SR3 22.3 93 PET 22.0 10 SR3 18.0 31 SR3 22.5 52 SR3 23.0 73 SR3 22.3 94 PET 23.5 11 SR3 20.0 32 SR3 22.5 53 SR3 23.0 74 ULS 23.5 95 PET 23.0 12 SR3 22.5 33 SR3 23.5 54 SR3 23.0 75 ULS 23.5 96 PET 23.0 13 SR3 22.5 34 SR3 23.5 55 SR3 23.0 76 PET 20.5 97 PET 23.0 14 SR3 22.5 35 SR3 - 56 SR3 23.0 77 PET 20.5 98 PET 23.0 15 SR3 23.0 36 SR3 24.5 57 SR3 - 78 PET 20.5 99 PET 23.0 16 SR3 20.0 37 SR3 22.5 58 SR3 22.5 79 PET 21.0 100 TS 23.0 17 SR3 24.0 38 SR3 22.0 59 SR3 23.0 80 PET 22.0 101 TS 23.5 18 SR3 - 39 SR3 22.5 60 SR3 23.5 81 PET 22.0 102 BPR - 19 SR3 24.0 40 SR3 20.5 61 SR3 22.5 82 PET 22.0 20 SR3 23.5 41 SR3 22.5 62 SR3 - 83 PET 22.0 21 SR3 22.5 42 SR3 23.0 63 SR3 22.5 84 PET 23.0 ACKNOWLEDGEMENTS Thanks to all scientific personnel who participated in the cruise, and to the crew of the RSV Aurora Australis. The work was supported by the Department of Environment, Sport and Territories through the CSIRO Climate Change Research Program, the Antarctic Cooperative Research Centre, and the Australian Antarctic Division. REFERENCES Bush, G., 1994. Deployment of upward looking sonar buoys. Centre for Marine Science and Technology, Curtin University of Technology, Western Australia, Report No. C94-4 (unpublished). Dickson, B.R., 1993. RRS Discovery Cruise 200: the Deployment Phase of the Antarctic Deep Outflow Experiment (ADOX-I). In Sigma, the UK WOCE Newsletter, Issue No. 10, July 1993 (David Cotton and Louise Fuger editors), James Rennell Centre, Southampton. Gordon, L.I., Jennings, J.C. Jr., Ross, A.A., and Krest, J.M., 1993. A Suggested Protocol for Continuous Flow Automated Analysis of Seawater Nutrients (Phosphate, Nitrate, Nitrite and Silicic Acid) in the WOCE Hydrographic Program and the Joint Global Ocean Fluxes Study. Oregon State University College of Oceanic and Atmospheric Sciences, Corvallis, Oregon. Technical Report No. 93-1. 51 pp. Millard, R.C., 1991. CTD Oxygen Calibration Procedure - in WOCE Operations Manual, 1991. WHP Office Report WHPO 91-1, WOCE Report No. 68/91, Woods Hole, Mass., USA. Millard, R.C. and Yang, K., 1993. CTD calibration and processing methods used at Woods Hole Oceanographic Institution. Woods Hole Oceanographic Institution Technical Report No. 93-44. 96 pp. Millard, R., Bond, G. and Toole, J., 1993. Implementation of a titanium strain gauge pressure transducer for CTD applications. Deep-Sea Research I, Vol. 40, No. 5, pp1009-1021. Rintoul, S.R. and Bullister, J.L. (in preparation). A late winter section between Tasmania and Antarctica: Circulation, transport and water mass formation. Rosenberg, M., Eriksen, R. and Rintoul, S., 1995. Aurora Australis marine science cruise AU9309/AU9391 - oceanographic field measurements and analysis. Antarctic Cooperative Research Centre, Research Report No. 2, March 1995. 103 pp. Ryan, T., 1995. Data Quality Manual for the data logged instrumentation aboard the RSV Aurora Australis.. Australian Antarctic Division, unpublished manuscript, second edition, April 1995. Speer, K.G. and Forbes, A., 1994. A deep western boundary current in the South Indian Basin. Deep-Sea Research I, Vol. 41, No. 9, pp. 1289-1303. APPENDIX 1 CTD Instrument Calibrations Table A1.1: Calibration coefficients and calibration dates for CTD serial number 2568 (unit no. 6) used during RSV Aurora Australis cruise AU9407. Note that an additional pressure bias term due to the station dependent surface pressure offset exists for each station (eqn A2.1, Appendix 2). Also note that platinum temperature calibrations are for the ITS-90 scale. coefficient value of coefficient pressure calibration coefficients (after terminology of eqn A2.1, Appendix 2) CSIRO Calibration Facility - 13/09/1994 pcal0 -3.825752e+01 pcal1 1.075168e-01 pcal2 -9.623422e-10 pcal3 1.921987e-14 platinum temperature calibration coefficients (after terminology of eqn A2.4, Appendix 2) General Oceanics - July 1993 Tcal0 2.146878e-01 Tcal1 5.011125e-04 Tcal2 -1.881981e-12 pressure temperature calibration coefficients (after terminology of eqn A2.3, Appendix 2) General Oceanics - July 1993 Tpcal0 6.059389e+01 Tpcal1 -1.703108e-03 Tpcal2 -3.010541e-09 Tpcal3 0.000000 coefficients for temperature correction to pressure (after terminology of eqn A2.2, Appendix 2) General Oceanics - July 1993 T0 21.65 (laboratory temperature for CSIRO pressure sensor calibration) S1 -3.5198e-06 S2 -2.3380e-01 preliminary polynomial coefficients applied to fluorescence (fl) and photosynthetically active radiation (par) raw digitiser counts (supplied by manufacturer) f0 -2.699918e+01 f1 8.239746e-04 f2 -2.071294e-24 par0 -4.499860e+02 par1 1.373000e-02 par2 -3.452000e-21 APPENDIX 2 CTD and Hydrology Data Processing and Calibration Techniques ABSTRACT Complete details are presented of the calibration and data processing techniques used to generate calibrated and quality controlled CTD 2 dbar-averaged data, and hydrology data. Attention is given to the order in which the various calculations and corrections are applied, as any variation will affect the final data values produced. A2.1 INTRODUCTION This Appendix details the data processing and calibration techniques employed in the production of the final CTD data set on shore. Logging of the data at sea is discussed in the main text. The different sections in this Appendix, and the description within each section, are ordered to match the steps in the data processing flow. Most of the data processing software is written in FORTRAN. Data sets for different cruises may vary in the specifications of the CTD (Tables 7 and 8 in the main text), and in the parameters generated. The generality of this description is retained so that it will be applicable to future data sets. Thus, the processing of a CTD raw data stream which includes pressure, temperature, conductivity, pressure temperature, oxygen current, oxygen temperature, and additional digitiser channels (e.g. fluorescence, photosynthetically active radiation, etc.) (Table 8) is detailed here. For future cruise data sets, any variation in the processing and calibration techniques described here will be detailed in the data report attached to the data set. Changes to calibration techniques from previous cruise Note that this Appendix is for the most part reproduced from Rosenberg et al. (1995). Several changes to the calibration techniques from the previous cruise have however occurred, and these changes are incorporated into the text and equations of this Appendix. The changes are summarised as follows: 1. Two pressure bias terms are applied in the pressure calibration (eqn A2.1) - the surface pressure offset term poff , which varies for each station; and the pressure offset calibration coefficient pcal0, derived from calibration of the pressure sensor (Appendix 1), which applies to the entire cruise (these two terms were previously incorporated into a single station varying coefficient). Calibrated pressure values (including application of the pressure temperature correction) are used in the determination of poff values (section A2.6.1) (uncalibrated pressures were used previously). 2. The same set of pressure calibration coefficients (eqn A2.1) now apply to both downcast and upcast data (section A2.6.2). These coefficients are applied to the pressure in raw digitiser counts. 3. The pressure sensor calibration (eqn A2.1) is no longer to the fifth order (third order only for this cruise). 4. A new correction is made to pressure data for the temperature at the pressure sensor (eqns A2.2 and A2.3). 5. The temperature calibration is now second order (eqn A2.4), and the calibration coefficients are applied to the temperature in raw digitiser counts. 6. In the conductivity cell deformation correction (eqn A2.7), the standard temperature and pressure values subtracted from T and p respectively have new values. 7. The temperature sensor time constant _T used in sensor lagging corrections (section A2.7.2) has a value of 0.205 (was set to 0.175 previously). 8. Problems with one of the additional digital channels (e.g. oxygen current) can result in bad pressure temperature data, which in turn yield bad pressure values (section A2.11.1). In such cases, pressure data must be reconstructed using pressure temperature profiles from other stations. 9. In determining an approximate time base for calculation of the oxygen current derivative with respect to time, the denominators in eqn A2.23 have new values (they are set according to the data recording frequency of the CTD). 10. When calculating the variance of the dissolved oxygen residuals (obtl - ocal) in eqn A2.24, the denominator is now (n-1) (was previously n) - this follows the strict definition of a variance. A2.2 DATA FILE TYPES The various data files used throughout the data calibration procedure on shore (and produced by it) are outlined below. A complete description of final calibrated data files is given in Appendix 4. A2.2.1 CTD data files Throughout this report, three types of CTD file are referred to: (i) raw CTD files, which contain the complete CTD data prior to removal of pressure reversals, and prior to averaging; note that a data scan refers to one complete data line containing all the logged parameters - thus the raw data is logged at N data scans per second, where N is the scanning frequency (Table 8); (ii) intermediate CTD files prior to 2 dbar averaging, despiked and with sensor lags applied, and with pressure reversals removed for downcast data; (iii) 2 dbar-averaged CTD files, which contain the CTD data averaged over 2 dbar bins. The CTD filenames are of the form vyyccusss.xxx:n (e.g. a94076046.raw:1) where v = vessel (e.g. "a" for Aurora Australis) yy = year (e.g. 94) cc = cruise number (e.g. 07) u = CTD unit number (i.e. instrument number) (e.g. 6) sss = station number (e.g. 046) xxx = file type (e.g. "raw" for raw data file) n = dip number (i.e. 1 for downcast data, 2 for upcast burst data) (does not apply to 2 dbar-averaged files) The various file suffixes (xxx in the above naming convention) are raw = raw data file cda = intermediate data file, which is the raw data file despiked and with pressure reversal removed, and with appropriate data lagging applied between parameters unc = uncalibrated 2 dbar-averaged file ave = calibrated (except for dissolved oxygen) 2 dbar-averaged file oxy = same as ave, but including the oxygen current derivative with respect to time (for the calibration of dissolved oxygen) all = final calibrated 2 dbar-averaged file (with or without dissolved oxygens) A2.2.2 Hydrology data files The final hydrology data file produced on shore contains the Niskin bottle data, output from the hydrology data processing program "HYDRO" (Appendix 3), merged with averages calculated from upcast CTD burst data. The file is named vyycc.bot (e.g. a9407.bot), where v, yy and cc are as above in the CTD file naming convention. During the CTD calibration procedure, intermediate hydrology data files are produced, named calib.dat:nn (e.g. 01), where "nn" is the version number. In general, the later version numbers are for more advanced stages in the quality control of Niskin bottle data. A2.2.3 Station information file This file contains station information, including position, time, depth etc. The file is named vyycc.sta (e.g. a9407.sta), where v, yy and cc are as above. A2.3 STATION HEADER INFORMATION Position: All station position information is derived from the quality controlled GPS underway measurement data set (section 4.2, and Appendix 4). Bottom depth: On the Aurora Australis, bow thrusters are used to maintain station. Unfortunately, the turbulence caused by the thrusters interferes with the echo sounder readings, so that the digital output from the sounder is unusable while thrusters are engaged on station. Depths while on station (Table 2) are obtained by reading the echo sounder printout, and are entered manually to the CTD data logging PC at sea. The automatically logged underway depth measurements immediately before and after station (i.e. when the bow thrusters are not in operation) are later used to check the plausibility of the manually entered values. Times: All start and end times recorded in the header information are stamped automatically by the CTD data acquisition program at the start and end of CTD data logging. Times are derived from the internal clock on the logging PC; this clock is independent of the ship's main time log, but is checked prior to each station. Bottom times (i.e. time at the bottom of the CTD cast) are as recorded manually at the bottom of each cast during data logging. A2.4 CONVERTING SHIP-LOGGED RAW DATA FILES FOR SHORE-DATA PROCESSING For the CTD instruments used on the Aurora Australis, the raw binary data files (as logged by the PC system on board the ship) are binary files consisting of data scans with length n bytes, where the value of n is fixed for each CTD instrument (Table 8). All further CTD data processing on shore is carried out on a Unix system. For each station, the raw binary data is split into two binary files: vyyccusss.raw:1 (also known as the "dip 1" file) e.g. a94076046.raw:1 vyyccusss.raw:2 (also known as the "dip 2" file) e.g. a94076046.raw:2 The dip 1 file contains the CTD data (uncalibrated), where only the downcast data has been preserved (down to the maximum pressure value recorded by the pressure sensor prior to the first Niskin bottle firing.) The dip 2 file contains CTD data bursts extracted from the upcast portion of the data at times corresponding to Niskin bottle firings. At each bottle firing, the 5 seconds of CTD data previous to the firing is stored in the dip 2 file. A2.5 PRODUCING THE DATA PROCESSING MASTER FILE A master file named "ctdmaster.sho" is created as a template from CTD header information. This file stores all data processing and calibration information, including station header details (e.g. positions, times, maximum pressure etc.), calibration coefficients, calibration status, and digitiser channel information. The master file is automatically updated by the data processing and calibration programs at all stages of the calibration procedure. A2.6 CALCULATION OF PARAMETERS The CTD pressure (including pressure temperature) and temperature sensor calibration coefficients (Appendix 1) are written to the master file - pressure and temperature data used at all further stages of the data processing are calibrated values. The conductivity and dissolved oxygen sensors are calibrated entirely from cruise Niskin bottle data, thus final conductivity and dissolved oxygen calibration coefficients are not included till a later stage in the processing. Note that for pressure, temperature, conductivity, salinity and parameters for additional digitiser channels, calculations (including application of calibration coefficients) are performed on the raw data prior to averaging into 2 dbar intervals. The calibration of dissolved oxygen data is performed on the 2 dbar averaged data only. A2.6.1 Surface pressure offset The point at which the CTD enters the water is found by identifying the first conductivity value greater than 10 mS/cm. The second data scan after this is then nominated as the first "in water" value. The value of the pressure for this scan is usually slightly greater than or less than zero, due both to atmospheric pressure variation, and to small calibration drift in the pressure sensor. The surface pressure offset value poff, equal to -1 times the pressure reading when the CTD enters the water, is retained for each station (Table 11), and each offset is added to all pressure values for the station. Note that calibrated pressures, including the pressure temperature correction (section A2.6.2), are used when finding poff values. A2.6.2 Pressure calculation As discussed in the main text, hysteresis displayed by the new titanium strain gauge pressure sensors is greatly reduced compared to the older stainless steel type pressure sensors. Calibration points from the increasing pressure calibration run only were used to produce a single third order fit for pressure (Appendix 1), applicable to both increasing and decreasing pressure (note that the difference between increasing and decreasing pressure calibrations due to hysteresis was, at greatest, 1 dbar). A further correction is made for the temperature at the pressure sensor (Millard et al., 1993), as measured by the pressure temperature sensor. Final calibrated pressure p is given by p = (pcal0 + poff) + pcal1.praw + pcal2.praw2 + pcal3.praw3 + pcorr (eqn A2.1) where pcal0 to pcal3 are the pressure sensor calibration coefficients, and praw is the raw pressure output by the CTD in digitiser counts. pcorr is the pressure temperature correction term, given by pcorr = (S2 + S1.praw) . (Tp - T0) (eqn A2.2) (Millard et al., 1993). S1 and S2 (supplied by the manufacturer) are pressure temperature correction slope and bias terms respectively, while T0 is the room temperature at which the pressure sensor calibration was obtained (i.e. where pcal0 to pcal3 were obtained) (Appendix 1). The calibrated pressure temperature Tp is given by Tp = Tpcal0 + Tpcal1.Tpraw + Tpcal2.Tpraw2 + Tpcal3.Tpraw3 (eqn A2.3) where Tpcal0 to Tpcal3 are the pressure temperature sensor calibration coefficients (Appendix 1), and Tpraw is the raw pressure temperature output by the CTD in digitiser counts. The CTD pressure was calibrated over the range 0 to 6203 dbar (Appendix 1). No greater pressures were reached during the cruise. A2.6.3 Temperature calculation CTD temperature values are in terms of the International Temperature Scale of 1990 (ITS-90). A quadratic fit is used for calibration of the temperature data, as follows: T = Tcal0 + Tcal1.Traw + Tcal2.Traw2 (eqn A2.4) where T is the calibrated temperature, Tcal0, Tcal1 and Tcal2 are temperature calibration coefficients (Appendix 1), and Traw is the raw temperature output by the CTD in digitiser counts. When conversion of temperature as ITS-90 to temperature expressed on the International Practical Temperature Scale of 1968 (IPTS-68) is required (e.g. for salinity PSS-78 calculation), the following conversion factors are used (Saunders, 1990): T68 = 1.00024 T90 (eqn A2.5) T90 = 0.99976 T68 (eqn A2.6) A2.6.4 Conductivity cell deformation correction Conductivity cell geometry is effected by temperature and pressure. The correction applied for this cell deformation is c = gctd . [1 - 6.5e-6 (T - 2.8) + 1.5e-8 (p - 3000)] (eqn A2.7) for conductivity c, calibrated temperature and pressure T and p respectively, and where gctd is the raw conductance graw as measured by the CTD and converted to approximate engineering units by gctd = graw / 1000 (eqn A2.8) A2.6.5 Salinity calculation Salinity is calculated from the conductivity, temperature and pressure using the practical salinity scale of 1978 (PSS-78), via the algorithm SAL78 (Fofonoff and Millard, 1983). Note that temperatures expressed on the ITS-90 scale must first be converted to IPTS-68 temperatures (eqn A2.5) for input into the salinity PSS- 78 routine. A2.6.6 Oxygen current and oxygen temperature conversion The raw oxygen current and oxygen temperature, ocraw and otraw respectively as measured by the CTD, are converted to occtd and otctd in approximate engineering units by occtd = ocraw / 80000 (eqn A2.9) otctd = otraw / 6000 (eqn A2.10) Calibration of the dissolved oxygen using these parameters is performed on 2 dbar averages only. A2.6.7 Additional digitiser channel parameters Manufacturer supplied polynomial fit coefficients are applied to digitiser channel parameters (Appendix 1). No further calibration is applied to these values. A2.7 CREATION OF INTERMEDIATE CTD FILES, AND AUTOMATIC QUALITY FLAGGING OF CTD BURST DATA Several processing steps take place when the intermediate CTD files are produced (section A2.7.5). Briefly, the parameters are despiked, sensor lagging corrections are applied, and pressure reversals are removed. For the upcast CTD burst data, individual bursts are automatically assigned a quality code. A2.7.1 Despiking Spurious data points are replaced by the previous data point. This preserves the equal time spacing between data points, required for the sensor lagging corrections discussed below. The criteria used to reject data values are shown in Table A2.1. Note that these criteria are unchanged over the entire water column. For pressure, temperature, conductivity and salinity, if any one of these parameters falls outside the criteria for acceptable data (Table A2.1), then the entire data scan is replaced by the previous data scan (i.e. all parameters are replaced by the previous value), and the scan replacement counter nrep is incremented by 1. If more than 3 consecutive data scans require replacement by the previous scan (i.e. nrep > 3), then all parameters are reset to their current value (i.e. the scan is not replaced by the previous scan) and nrep is reset to 0. Table A2.1: Criteria used to determine spurious data values. The low and high limits are respectively the minimum and maximum allowable values for the parameter. The maximum allowable step is the maximum difference permitted between consecutive values. parameter units low limit high limit maximum allowable step ------------------ ------- --------- ---------- ---------------------- pressure dbar 0 6203 1.0 temperature oC -5 32 1.0 conductivity mS.cm-1 5 80 1.0 salinity psu 10 50 0.25 oxygen current µA 0 2 0.25 oxygen temperature oC -5 32 1.0 For oxygen current oc and oxygen temperature ot, if either of these parameters falls outside the criteria in Table A2.1, then the current oc and ot values are replaced by null data points; the other parameters are unaffected, and nrep is not incremented. Note that when oc and ot are replaced by null values, then the maximum allowable step criterion (Table A2.1) is not applied to the next oc and ot values; however the low and high limit tests (Table A2.1) are still applied. For any parameters from the additional digitiser channels, no automatic check is made for spurious data values. A2.7.2 Sensor lagging corrections Lag corrections are required to compensate for the different response times of the sensors. Data from the faster sensors (pressure and conductivity) are slowed down to match the slowest sensor (temperature). A recursive filter (Millard, 1982) is used to lag the pressure and conductivity data, of the form y( t ) = y( t - dt ) . W0 + x( t ) . W1 (eqn A2.11) where y( t ) = output lagged conductivity or pressure at time t dt = recording interval of the instrument x( t ) = input conductivity or pressure prior to lagging W0 = exp( -dt / tau ) W1 = 1 - W0 The time constant tau is obtained as follows. The response of the pressure sensor is assumed to be instantaneous; the response time of the conductivity cell is taken as 0.03 seconds, which is equal to the flushing time of the 3 cm conductivity cell at a lowering rate of 1 m.s-1. Thus for tauT equal to the response time of the temperature sensor, we have tau = tauT when pressure is being lagged, and tau = tauT - 0.03 when conductivity is being lagged. tau T is obtained by performing a cross-correlation between the temperature and conductivity data to determine the response difference between the two sensors. Typically, a value of 0.205 s is used for tauT (Table 8). The same recursive filter (eqn A2.11) is applied to the oxygen current and oxygen temperature, as well as to data in the additional digitiser channels. For all these parameters, the value tau = tauT is used for the time constant. A2.7.3 Pressure reversals After despiking and application of the lagging correction, for downcast data all pressure reversals are removed. Stepping through the data scans, the maximum pressure value is updated each time the pressure increases, and the scan is written to the intermediate CTD file (including the case where pressure does not change); data scans with a pressure value less than the current maximum pressure value are not written to the intermediate file. Thus for downcast data, the intermediate CTD file contains data for non-decreasing pressure. For upcast burst data, pressure reversals are not removed. A2.7.4 Upcast CTD burst data A burst of CTD data is associated with each firing of a Niskin bottle, each burst consisting of the 5 seconds of CTD data prior to the bottle firing. For each burst, the mean and standard deviation of the parameters are calculated: for these calculations, the first nstart and last nend data scans (Table 8) in each burst are ignored. The range of the parameters in each burst is also found (equal to the difference of the maximum and minimum values). The mean values from the burst data are used for comparison with the salinity and dissolved oxygen bottle samples, for the subsequent calibration of the conductivity and dissolved oxygen sensors. Table A2.2: Criteria for automatic flagging of upcast CTD burst data. The subscripts std and range refer respectively to the standard deviation and range of the parameter over the data burst. The data quality code iqual has the following values: iqual=1 acceptable value, used for conductivity calibration iqual=0 questionable value, but still used for conductivity calibration iqual=-1 bad value, not used for conductivity calibration Note that setting iqual to -1 takes precedence over setting iqual=0, which in turn takes precedence over setting iqual=1. STANDARD DEVIATION CRITERIA RANGE CRITERIA --------------------------------------- ------------------------------------------ set iqual = -1 for set iqual = 0 for set iqual = -1 for set iqual = 0 for following cases following cases following cases following cases 4.00 < pstd 2.00 < pstd 4.00 (Trange)/(crange) < 0.5 0.04 < Tstd 0.02 < Tstd 0.04 (Trange)/(crange) > 2.0 0.04 < cstd 0.02 < cstd 0.04 crange = 0 0.01 < sstd 0.005 < sstd 0.01 0.02 < srange 0.01 < srange 0.02 0.40 < ocstd 0.20 < ocstd 0.40 0.40 < otstd 0.20 < otstd 0.40 1998 < adstd 999 < adstd 1998 The standard deviations and ranges of the burst data are used to assign a quality code to each burst (Table A2.2). Note that there is only one quality code assigned to each data burst and associated Niskin bottle sample in the hydrology data file: this code refers to values used in the calibration of the CTD conductivity. For the criteria in Table A2.2, setting of the quality code to -1 takes precedence over setting to 0. If none of the criteria are met, the quality code is set to 1 i.e. value accepted for calibration of the conductivity. The standard deviation xstd of parameters x in each data burst is calculated from n-nend _ xstd = { [ ( xi - x )2 ] / [n - (nstart+nend+1)] } 1/2 (eqn A2.12) i=nstart where n is the total number of data points xi in the burst, and the mean value x for each burst is given by _ n-nend x = ( xi ) / (n-nstart-nend) (eqn A2.13) i=nstart A2.7.5 Processing flow Stepping through the raw data scans one scan at a time, the parameters in the scan first have the calculations and corrections applied, as described in section A2.6. The data is then despiked (section A2.7.1); spurious values are replaced by the previous data scan, up to a maximum of 3 consecutive scans, after which time the scan is reset to the current value. The sensor lagging correction is then applied via the recursive filter (section A2.7.2). When the filter is started, the first jfilt scans (Table 8) are ignored. Note that whenever nrep > 3 (section A2.7.1), the filter is restarted, and the first jfilt scans are again ignored. Salinity is recalculated for each data scan, after all lagging corrections have been applied. Data is then written to the intermediate CTD file, removing pressure reversals for the case of downcast data (section A2.7.3). For upcast burst data, statistical calculations are performed and a quality code assigned for each burst (section A2.7.4). The mean values and quality codes for the bursts are written to a template intermediate hydrology data file. A2.8 CREATION OF 2 DBAR-AVERAGED FILES Data scans from the intermediate CTD files are sorted into 2 dbar pressure bins, with each bin centered on the even integral pressure value, starting at 2 dbar, as follows. A data scan is placed into the ith 2 dbar pressure bin if pmidi - 1 < p pmidi + 1 (eqn A2.14) where pmidi is the ith 2 dbar pressure bin centre, and p is the pressure value for the data scan. After sorting, the temperature, conductivity, oxygen current, oxygen temperature and additional digitiser channel values in each 2 dbar bin are averaged and written to the 2 dbar-averaged file. There is no pressure centering of these parameters i.e. for the ith 2 dbar pressure bin, the parameters are assigned to the even integral pressure value at the centre of the bin. Note that if the number of points in a bin is less than jmin (Table 8), no averages are calculated for that bin. The salinity sav for each 2 dbar bin is calculated from Tav, cav and pmid, where Tav and cav are respectively the temperature and conductivity averages for the bin. Note that Tav is first converted from the ITS-90 scale to the IPTS-68 scale using eqn A2.5 (this also applies to the calculations below for sigmaT, delta and ). The following quantities are also calculated for each 2 dbar bin, and are written to the 2 dbar-averaged file: sigmaT: sigma-T is equal to (rho - 1000), where the density rho is calculated at the surface, and at the in situ temperature and salinity Tav and sav respectively, using the 1980 equation of state for seawater (Millero et al., 1980; Millero and Poisson, 1981). delta: specific volume anomaly (units x108 m3.kg-1), calculated with Tav, sav and pmid, using the 1980 equation of state for seawater (Millero et al., 1980; Millero and Poisson, 1981). Delta theta: geopotential anomaly (units J.kg-1), calculated relative to the sea surface (p=0), from p=pmid -- = delta . dp (eqn A2.15) p=0 nbin: number of points in the 2 dbar bin Tbinstd: standard deviation of all temperature values in the bin cbinstd: standard deviation of all conductivity values in the bin When 2 dbar averages are calculated for oxygen current and oxygen temperature, an additional test is made to exclude suspect oxygen data, as follows. For a 2 dbar bin, if we have either standard deviation of binned oc > 0.1 or standard deviation of binned ot > 0.5 then the following 2 conditions must be met for a scan to be included in the averaging of oc and ot for the bin: 0 < oc 2.047 (eqn A2.16) | ot - T | 5 (eqn A2.17) After this test has been made, if the number of scans in the bin has been reduced by more than half, then no oc or ot data is included for the bin. A2.9 HYDROLOGY DATA FILE PROCESSING An intermediate hydrology data file is formed by merging the results from the salinity, dissolved oxygen and nutrient laboratory analyses with the averages calculated from the upcast CTD burst data (section A2.7.4). Prior to calibration of the CTD conductivity and dissolved oxygen data, the Niskin bottle data undergo preliminary quality control. Salinity bottle data which are obviously bad are given the quality code -1 (i.e. bottle not used for calibration of CTD conductivity) in the intermediate hydrology data file. Reasons for rejecting salinity bottle data at this stage include bad samples due to leaking or incorrectly tripped Niskin bottles, mixed up samples due to misfiring rosette pylon, samples drawn out of sequence from Niskin bottles, etc. Dissolved oxygen bottle data pass through an initial quality control similar to salinity bottle data, except that bad dissolved oxygen bottle values are deleted from the hydrology data file. Questionable dissolved oxygen bottle values (not deleted) are noted (Table 21). Suspect reversing thermometer readings are also deleted at this stage. Nutrient data are quality controlled at a later stage, following calibration of all the CTD data. A2.10 CALIBRATION OF CTD CONDUCTIVITY For the CTD conductivity data, calibrations are carried out by comparing the upcast CTD burst data with the hydrology data, then applying the resulting calibrations to the downcast CTD data. The conductivity calibration follows the method of Millard and Yang (1993). For groups of consecutive stations, a conductivity slope and bias term are found to fit the CTD conductivity from the upcast burst data to the hydrology data; a linear station-dependent slope correction (Millard and Yang, 1993) is applied to account for calibration drift of the CTD conductivity cell. Note that data from the entire water column are used in the conductivity calibration. Also note that no correction is made for the vertical separation of the Niskin bottles and the CTD sensors (of the order 1 m). A2.10.1 Determination of CTD conductivity calibration coefficients The following definitions apply for the conductivity calibration: cctd = uncalibrated CTD conductivity from the upcast burst data ccal = calibrated CTD conductivity from the upcast burst data cbtl = 'in situ' Niskin bottle conductivity, found by using CTD pressure and temperature from the burst data in the conversion of Niskin bottle salinity to conductivity F1 = conductivity bias term F2 = conductivity slope term F3 = station-dependent conductivity slope correction N = station number CTD conductivities are calibrated by the equation ccal = (1000 cctd) . (F2 + F3 . N) + F1 (eqn A2.18) Niskin bottle salinity data are first converted to 'in situ' conductivities cbtl. The ratio cbtl/ccal for all bottle samples is then plotted against station number, along with the mean and standard deviation of the ratio for each station (Figure 3 is the version of this plot for the final calibrated data). Groups of consecutive stations are selected to follow approximately linear trends in the drift of the station-mean cbtl/ccal (Table 13). For each of these groups, the three calibration coefficients F1, F2 and F3 are found by a least squares fit: F1, F2 and F3 in eqn A2.18 are all varied to minimize the variance sigma2 of the conductivity residual (cbtl-ccal), where sigma2 is defined by sigma2 = (cbtl - ccal)2 / (n - 1) (eqn A2.19) for n equal to the total number of bottle samples in the station grouping. Note that samples with a previously assigned quality code of -1 (sections A2.7.4. and A2.9) are excluded from the above calculations. In addition, samples for which | (cbtl - ccal) | > 2.8 sigma (eqn A2.20) are also flagged with the quality code -1, and excluded from the final calculation of the conductivity calibration coefficients F1, F2 and F3. Samples rejected at this stage often include those collected in steep vertical temperature and salinity gradients, and not already rejected. A2.10.2 Application of CTD conductivity calibration coefficients The set of coefficients F1, F2 and F3 found for each station (Tables 13 and 14) are first used to calibrate the upcast CTD conductivity burst data in the hydrology data file. The conductivity calibration is applied to the mean value for each burst only (as opposed to each raw data scan in the burst). Similarly, upcast CTD salinity burst values are recalculated from the calibrated CTD burst mean values of conductivity, temperature and pressure. Next, the intermediate CTD files are reproduced (as per section A2.7) for the downcast data only. Note that on this occasion, following application of the conductivity cell deformation correction (eqn A2.7), the coefficients F1, F2 and F3 are used to calibrate the raw conductivity data scans. The 2 dbar-averaged CTD downcast data are then recalculated, as in section A2.8. A2.10.3 Processing flow The intermediate hydrology file data, containing upcast CTD burst data means and Niskin bottle data, are used to determine the conductivity calibration coefficients F1, F2 and F3. Station groupings are determined from the bias drift of the conductivity cell with time (section A2.10.1). For each station group, the following occurs: 1. 3 iterations are made of the least squares fitting procedure (section A2.10.1) to calculate F1, F2 and F3, each iteration beginning with the latest value for the coefficients; 2. bottles are rejected according to the criterion of eqn A2.20; 3. steps 1 and 2 are repeated until no further bottle rejection occurs. For each station group, there is a single value for each of the 3 coefficients F1, F2 and F3 (Table 13); following the station-dependent correction, an individual corrected slope term (F2 + F3.N) (as in eqn A2.18) applies to each station (Table 14). When final values of the coefficients have been obtained, the conductivity calibration is applied to both the upcast CTD burst data and the downcast CTD data (section A2.10.2). Finally, plots are made of both the ratio cbtl/ccal and the residual (sbtl - scal) versus station number (Figures 3 and 4), where sbtl is the Niskin bottle salinity and scal is the calibrated CTD salinity from the upcast burst data (section A2.10.2). Following calibration of the CTD conductivity, the mean of the salinity residuals (sbtl - scal) for the entire data set is equal to 0. The standard deviation about 0 of the salinity residual (section A2.14) provides an indicator for the quality of the data set. To meet WOCE specifications, this standard deviation should be less than or equal to 0.002 psu (Joyce et al., 1991). A2.11 QUALITY CONTROL OF 2 DBAR-AVERAGED DATA Two levels of quality control are undertaken for the 2 dbar-averaged data. Suspicious raw data scans, indicated by suspicious 2 dbar averages, are flagged for later action (Table 15); and remaining suspect 2 dbar averages are noted (Tables 16 and 17) (suspect 2 dbar averages are rarely removed directly). A2.11.1 Investigation of density inversions The calibrated 2 dbar-averaged data are searched automatically for density inversions i.e. for instances where the in situ density (calculated from in situ pressure, temperature and salinity) decreases with depth. Raw CTD data in the vicinity of the density inversions are then examined for anything which might artificially cause the inversions. The most commonly encountered problems are (a) water from the wake of the moving instrument package catching up to the CTD sensors during rolls induced by surface waves; (b) fouling of the CTD sensors; (c) salinity spikes caused by mismatching of the temperature and conductivity data in very steep vertical gradients, where the sensor lagging corrections (section A2.7.2) are not adequate; (d) bad pressure temperature data caused by a malfunctioning dissolved oxygen sensor, as described in section 6.1.2 of the main text (note that in this case, density inversions do not always result). If these or any other problems are identified in the raw CTD data, one of two possible actions follow: (i) the relevant data scans are ignored for all further calculations - a counter preserves the constant scanning frequency required for application of the sensor lagging corrections; note that for cases where the ignoring of raw data scans results in missing 2 dbar averages, a linear interpolation is applied between surrounding 2 dbar averages to fill any data gaps (Table 18); (ii) a linear interpolation is applied over the region of bad data, in which case the interpolation is applied to the raw CTD data scans prior to any calibration calculations. The status of data scans flagged for special treatment (Table 15) is updated in the data processing master file (section A2.5). For the case of bad pressure temperature data, pressure temperature profiles from other stations (see section 6.1.2 in the main text for an example) may be used to allow calculation of the pressure from eqns A2.1 to A2.3. A2.11.2 Manual inspection of data Data plots of the 2 dbar-averaged data are inspected to identify any additional suspicious data. Suspect values remaining are most commonly due to the following: (a) large salinity spikes (as in section A2.11.1) in very steep gradients in the thermocline - for these large salinity spikes, 2 dbar averages are flagged instead of raw data scans (Table 16); (b) suspect data near the surface due to transient effects of the sensors entering the water (e.g. bubbles trapped on sensors, or fouling) (Table 17). 2 dbar-averaged data regarded as suspicious for these or any other reasons are flagged accordingly. A2.12 CALIBRATION OF CTD DISSOLVED OXYGEN For the CTD dissolved oxygen data, the calibration procedure is carried out using the downcast uncalibrated CTD data. Downcast CTD data is matched with the Niskin bottle dissolved oxygen samples on equivalent pressures. The calibration is based on the method of Owens and Millard (1985). A2.12.1 Determination of CTD dissolved oxygen calibration coefficients The following definitions apply for the dissolved oxygen calibration: ocal = calibrated CTD dissolved oxygen oc = CTD oxygen current ot = CTD oxygen temperature T = CTD temperature s = CTD salinity p = CTD pressure 6oc/6t = oxygen current derivative with respect to time K1 = oxygen current slope K2 = oxygen sensor time constant K3 = oxygen current bias K4 = temperature correction term K5 = weighting factor of ot relative to T K6 = pressure correction term obtl = Niskin bottle dissolved oxygen value All the above CTD parameters are 2 dbar-averaged data. CTD dissolved oxygen is calibrated using the sensor model of Owens and Millard (1985), as follows: ocal = [ K1 . ( oc + K2 . (oc/(t + K3 ) ] . oxsat(T,s) . exp{ K4 . [ T + K5 . (ot - T) ] + K6 . p } (eqn A2.21) where the oxygen saturation value oxsat is calculated at T and s using the formula of Weiss (1970): oxsat(T,s) = exp{ A1 + A2.(100/TK) + A3.ln(TK/100) + A4.(TK/100) + s.[B1 + B2.(TK/100) + B3.(TK/100)2] } (eqn A2.22) for TK equal to the CTD temperature in degrees Kelvin (=T+273.16), and the additional coefficients having the values (Weiss, 1970): A1 = -173.4292 B1 = -0.033096 A2 = 249.6339 B2 = 0.014259 A3 = 143.3483 B3 = -0.0017 A4 = -21.8492 Note that the CTD temperature T in equations A2.21 and A2.22 is first converted from the ITS-90 scale to the IPTS-68 scale using eqn A2.5. (oc/(t in eqn A2.21 is calculated as follows. A time base is first estimated from the 2 dbar averaged data by assigning the time tk in seconds at the kth 2dbar value equal to k-1 tk = [ ( nbinj / 25 ] + (nbink / 50) (eqn A2.23) i=1 where nbink is the number of data scans in the kth 2 dbar bin (for bins with no data points, nbin is set to 25). Note that this time base is an approximation only, as nbin does not include data scans in pressure reversals (sections A2.7.3 and A2.8), and in addition, a constant lowering rate of the instrument package is being assumed. (oc/(t is then calculated at the kth 2 dbar value by applying a linear regression over a 16 dbar interval centered on the kth 2dbar value: (oc/(t is the slope of the linear best fit line of the oxygen currents (ock-4, ock-3, ock-2, ock-1, ock, ock+1, ock+2, ock+3, ock+4) to the times (tk-4, tk-3, tk-2, tk-1, tk, tk+1, tk+2, tk+3, tk+4). If there is no data for either of ock or otk (section A2.8), a null value is assigned to ((oc/(t)k . In most cases, CTD dissolved oxygen is calibrated for individual stations; station groupings (as in the CTD conductivity calibration) may be formed to cover casts with few Niskin samples, or else for deep/shallow cast pairs at a single location. For each individual station, or each station grouping, the calibration coefficients K1 to K6 in eqn A2.21 are found by varying some or all of the 6 coefficients in order to minimize the variance sogma2 of the dissolved oxygen residual obtl - ocal, where sigma2 is defined by sigma2 = (obtl - ocal)2 / (n-1) (eqn A2.24) for n equal to the total number of bottle samples at the station (or in the station grouping). A non-linear least squares fitting routine, utilising the subroutines MRQMIN, MRQCOF, COVSRT and GAUSSJ in Press et al. (1986), is applied to find K1 to K6. In application of the routine, convergence is judged to have occurred when ( (obtl - ocal)2 / (0.6)2 < 0.96 n (eqn A2.25) or else after a maximum of 5 iterations. Note that when calculating sigma2 for each Niskin bottle sample, the pressure from the upcast CTD burst data (i.e. the pressure assigned to the bottle sample) is used in eqn A2.21, while all other parameters are from the downcast data (at the nearest equivalent 2 dbar pressure value). Downcast CTD pressure is used in eqn A2.21 when the resulting calibration is being applied to finalise the entire 2 dbar dissolved oxygen data. Also note that there is no automatic rejection of dissolved oxygen bottle data analogous to eqn A2.20 in the conductivity calibration. A2.12.2 Application of CTD dissolved oxygen calibration coefficients The set of coefficients K1 to K6 found for each station or station grouping (Table 19) are used in eqn A2.21 to calculate CTD dissolved oxygen 2 dbar data from the existing 2 dbar pressure, temperature, salinity, oxygen current and oxygen temperature data. A2.12.3 Processing flow * The .oxy files (section A2.2.1), which include values of (oc/(t (calculated as in section A2.12.1) as well as all the other downcast 2 dbar data, are first created from the existing calibrated 2 dbar-averaged files. * For each station, the upcast CTD burst pressure values from the hydrology data file (sections A2.7.4 and A2.7.5) are matched to the closest 2 dbar pressure values in the .oxy file; then for each Niskin bottle sample, the following data are written to the file oxydwn.dat: p (upcast CTD burst value) T, s, oc, ot, (oc/(t (all 2 dbar downcast values) obtl obtl quality code The -1 bottle quality code (sections A2.7.4 and A2.9) is not relevant to the dissolved oxygen calibration. Instead, a code of -9 in the oxydwn.dat file indicates that the bottle is not used for the dissolved oxygen calibration calculations. * All calibration calculations are performed on dissolved oxygen (i.e. Niskin bottle and CTD dissolved oxygen values, and oxygen saturation values) in units of ml/l; all values are reported in units of (mol/l. The conversion factor used is ( (mol/l ) = 44.6596 . ( ml/l ) (eqn A2.26) * The fitting routine is applied to find values of the coefficients K1 to K6 (section A2.12.1), using the data in the oxydwn.dat file. The number of coefficients varied may be chosen, as well as the starting values for the coefficients prior to iteration (Table 20). Starting values are typically close to the following: K1 = 2.50 K4 = -0.036 K2 = 8.0 K5 = 0.75 K3 = 0.0 K6 = 0.00015 With successive attempts at fitting the CTD data to the Niskin bottle data, bottles which are suspect are flagged manually with the quality code -9 in oxydwn.dat, and are rejected for further calibration attempts (Appendix 5, Table A5.2). The number of coefficients chosen to vary, and the coefficient starting values, are varied to achieve the best fit of the CTD to the bottle data. In general, the fit for a station (or group of stations) is not considered satisfactory until 2.8( < 0.3 (for ( defined as in eqn A2.24) (Table 19). * Following calibration of the CTD dissolved oxygen, the residuals (obtl - ocal) are plotted against station number (Figure 5). The mean of the residuals for the entire data set is very close to 0. The standard deviation about the mean of the residuals (section A2.14) provides an indicator for the quality of the data set. To meet WOCE specifications (Joyce et al., 1991), this standard deviation should be less than 1% of the maximum data value i.e. approximately < 2.5 (mol/l below 750 dbar, and approximately < 3.5 (mol/l above 750 dbar, for the data set presented in this report. A2.13 QUALITY CONTROL OF NUTRIENT DATA Nutrient data which are obviously bad are removed from the hydrology data file. Causes of bad samples include leaking or incorrectly tripped Niskin bottles, and errors occurring during sampling or analysis. On occasion, autoanalyser errors may necessitate the flagging of an entire station as suspect. The data are checked by overlaying vertical profiles of groups of consecutive stations, looking at bulk plots (e.g. nitrate versus phosphate) of large numbers of stations, and by comparing values to any available historical data. Questionable nutrient data (not obviously bad, and therefore not deleted from the hydrology data file) are noted (Table 22). A2.14 FINAL CTD DATA RESIDUALS/RATIOS The final residuals (Ttherm - Tcal), (sbtl - scal) and (obtl - ocal) are plotted (Figures 2, 4 and 5) for temperature, salinity and dissolved oxygen (Ttherm and Tcal are respectively the protected thermometer and calibrated upcast CTD burst temperature values); for conductivity, the ratio cbtl/ccal is plotted (Figure 3). The plots include mean and standard deviation values, as follows: temperature, salinity and dissolved oxygen: The standard deviations of the residuals for temperature, salinity and dissolved oxygen are calculated from n xstd = { [ ( xi - xmean )2 ] / (n - 1) } 1/2 (eqn A2.27) i=1 where xstd is the standard deviation of x (for x equal to the temperature, salinity or dissolved oxygen residual). For both temperature and salinity, the summation in eqn A2.27 does not include points rejected for the CTD conductivity calibration. Similarly for dissolved oxygen, the summation does not include points rejected for the CTD dissolved oxygen calibration. Thus n is equal to the total number of data points xi not rejected for the relevant calibration, with mean value xmean of the xi values (xmean is the mean for all the stations in the plot). conductivity: The standard deviation of the conductivity ratio is calculated as in eqn A2.27, except that in the summation, for each point xi the value xmean is the mean for the particular station to which xi belongs. x in eqn A2.27 is equal to the conductivity ratio. The summation in eqn A2.27 does not include points rejected for the CTD conductivity calibration. A2.15 CONCLUSIONS A complete description is presented of the CTD data calibration methods. Sufficient details are supplied to minimize the need for cross-referencing, and to provide a useful reference for comparison with the calibration methods used by other institutions. Any variation in the techniques employed at each stage of the processing, and the order in which the various techniques are applied, ultimately affect the final data values produced. As such, all CTD data sets need to be considered in conjunction with the calibration details. ACKNOWLEDGEMENTS Many thanks go to Neil White and Dave Vaudrey at CSIRO Division of Oceanography, who created the bulk of the CTD calibration software, and familiarised me with the contents. REFERENCES Fofonoff, N.P. and Millard, R.C., Jr., 1983. Algorithms for computation of fundamental properties of seawater. UNESCO Technical Papers in Marine Science, No. 44. 53 pp. Joyce, T., Corry, C. and Stalcup, M., 1991. Requirements for WOCE Hydrographic Programme Data Reporting. WHP Office Report WHPO 90-1, Revision 1, WOCE Report No. 67/91, Woods Hole Oceanographic Institution. 71 pp. Millard, R.C., Jr., 1982. CTD calibration and data processing techniques at WHOI using the 1978 Practical Salinity Scale. Proceedings of the International STD Conference and Workshop. Millard, R.C. and Yang, K., 1993. CTD calibration and processing methods used at Woods Hole Oceanographic Institution. Woods Hole Oceanographic Institution Technical Report No. 93-44. 96 pp. Millard, R., Bond, G. and Toole, J., 1993. Implementation of a titanium strain gauge pressure transducer for CTD applications. Deep-Sea Research I, Vol. 40, No. 5, pp1009-1021. Millero, F.J., Chen, C.-T., Bradshaw, A. and Schleicher,K., 1980. A new high- pressure equation of state for seawater. Deep-Sea Research. 27a: 255-264. Millero, F.J. and Poisson, A., 1981. International one-atmosphere equation of state of seawater. Deep-Sea Research. 28a: 625-629. Owens, W.B. and Millard, R.C., Jr., 1985. A new algorithm for CTD oxygen calibration. Journal of Physical Oceanography. 15: 621-631. Press, W.H., Flannery, B.P., Teukolsky, S.A. and Vetterling, W.T., 1986. Numerical Recipes. The Art of Scientific Computing. Cambridge University Press. 818 pp. Rosenberg, M., Eriksen, R. and Rintoul, S., 1995. Aurora Australis marine science cruise AU9309/AU9391 - oceanographic field measurements and analysis. Antarctic Cooperative Research Centre, Research Report No. 2, March 1995. 103 pp. Saunders, P.M., 1990. The International Temperature Scale of 1990. ITS-90. WOCE Newsletter, 10, IOS, Wormley, UK. Weiss, R.F., 1970. The solubility of nitrogen, oxygen and argon in water and seawater. Deep-Sea Research. 17: 721-735. APPENDIX 3 Hydrology Analytical Methods This Appendix covers the analytical techniques and data processing routines employed in the Hydrographic Laboratory onboard the RSV Aurora Australis for cruise AU9407, January 1 to March 1, 1994. All analysis results are merged with station details in the program "HYDRO" (CSIRO Division of Oceanography). Output from HYDRO is ultimately used for merging with CTD data. A limited replicate sample set was obtained from several stations. Estimates of nutrient, dissolved oxygen and salinity precision derived from these data are discussed in section 6.2.2 of the main text. All methods are fully documented in a hydrology manual (Eriksen, in preparation). A3.1 NUTRIENT ANALYSES A3.1.1 Equipment and technique Nutrient analyses were performed by two analysts from the Antarctic CRC. An Alpkem "Flow Solution" Autoanalyser was used for the simultaneous analysis of reactive silicate, nitrate plus nitrite, and orthophosphate in seawater. All analyses were carried out in the Segmented Flow Analysis (SFA) mode, although the instrument can be configured for Flow Injection Analysis. Data output from the autoanalyser was processed by the commercial software package "DAPA" (Data Acquisition Processing Analysis Scientific Version 1.43, Curtin University, Box 58 Kalamunda Western Australia 6070). The 510 Monochromator Detectors in the Alpkem instrumentation were mounted on foam pads 8 cm thick, to insulate them from the constant vibration caused by the ship's engines, ice breaking etc. Previous voyages had shown that the 510 detectors are very susceptible to air bubbles lodging in the flow cell when the unit is subjected to high frequency vibration (Rosenberg et al., 1995). In addition, the waste lines on all three channels were increased in length from 30 to 50 cm, using 1.0 mm ID tubing. The additional tubing prevents bubbles from passing into the flow cell, by maintaining slight back pressure on the system. A new air-conditioning system was installed in the laboratory for this cruise, to provide a more constant temperature environment for chemical analyses. For most of the time, laboratory temperature was maintained at 22 ( 2oC (see Table 23 in the main text), a significant improvement on previous cruises. A3.1.1.1 Silicate Reactive silicate was analysed in accordance with the method provided for seawater analysis in the Alpkem Manual (Alpkem Corp, 1992). The silica in solution as silicic acid or silicate reacts with a molybdate reagent in acid media to form -molybdo silicic acid. The complex is then reduced to a highly coloured molybdenum blue following mixing with ascorbic acid. Interference from phosphate is suppressed by the addition of oxalic acid. Absorbance is measured at 660 nm. Two modifications to the Alpkem methodology were introduced. Firstly, the reaction cartridge was thermostated to 37oC, to avoid the effects of any ambient temperature fluctuations on the sensitivity of the chemistry (Smythe-Wright et al., 1992, report that there may be up to 3% deviation in peak height with a 1oC change in temperature when using this chemistry). Secondly, the EVA tubing carrying the reaction product from the cartridge to the detector was fed between two layers of TygonTM tubing, effectively forming an insulation jacket around the line. A3.1.1.2 Nitrate plus nitrite Nitrate plus nitrite was analysed using an Imidazole buffer chemistry in place of the Alpkem methodology. A 12" Open Tubular Cadmium Reactor (OTCR) supplied by Alpkem is used for quantitative reduction of nitrate to nitrite. The nitrite due to nitrate, plus the nitrite originally present in the sample, then undergoes diazotization with sulphanilamide and subsequent coupling with N-1- napthylethylene-diamine dihydrochloride. The azo dye is detected at 540 nm. A standard nitrite solution is used frequently to check the reduction efficiency of the column. Efficiencies over 95% are commonly achieved. The columns are re- activated with a 2% copper sulphate solution after every second analysis run. Details of the chemistry and procedures for nitrate plus nitrite analysis follow. Methodology for nitrate plus nitrite analysis in seawater All reagents are analytical grade (AR), unless otherwise specified. All volumetric glassware for reagent preparation is A grade dedicated glassware, and cleaned with acid prior to each voyage. Glassware is stored full of deionised water when not in use. Reagent chemistry Start-up solution: Add 0.5 ml of 30% w/v Brij-35 to 200 ml of deionised water. Mix thoroughly. This reagent is refreshed daily. Imidazole buffer pH 7.8: Dissolve 4.25 g of Imidazole buffer in 800 ml of deionised water. Add 11.25 ml of 10% HCl to adjust the final pH to 7.8. Make up to a litre and mix well. Add 1 ml of 30% w/v Brij-35 after decanting liquid to reagent container. Store at 4oC when not in use. Replenish every 2 to 3 days. N-1 napthylethylene-diamine dihydrochloric acid (NEDD): Dissolve 0.31 g of NEDD in 1 l of deionised water. Add 1 ml of 30% w/v Brij-35 after decanting to reagent container. Store at 4oC when not in use. Sulphanilamide: Dissolve 3.12 g of sulphanilamide in 800 ml of deionised water in a 1 l volumetric flask. Add 31 ml of concentrated HCl carefully, and make up to the mark. Pump configuration Reagent Pump tube Flow rate at 50% pump speed ---------------- ------------- --------------------------- NEDD Orange/yellow 0.18 ml/ min Sulphanilamide Orange/yellow 0.18 ml/min Imidazole Buffer Black/black 0.32 ml/min Nitrogen Orange/white 0.25 ml/min Sample Black/black 0.32 ml/min Activation of the OTCR The activation and installation of the OTCR is performed in accordance with the method in the Alpkem Manual (Alpkem Corp, 1992). A separate batch of Imidazole buffer, that does not contain Brij-35, is used for the activation of the OTCR. When not in use, OTCR's are stored filled with deionised water, or high purity nitrogen. Figure A3.1: Cartridge configuration for nitrate + nitrite analysis. A3.1.1.3 Phosphate Phosphate analysis was carried out using the methodology supplied by Alpkem (Alpkem Corp, 1992). The chemistry involves reaction with an acidified molybdate reagent and potassium antimonyl tatrate. The compound produced is then reduced by ascorbic acid to a highly coloured molybdenum blue complex. The chemistry uses a single colour reagent, which has a shelf life of 8 hours and must be renewed accordingly. The pump tubing was optimised to reduce the effect of carryover on samples. The monochromator detector was modified to increase the upper wavelength selection limit from 800 to 900 nm. It was found that using 880 nm as the detection wavelength, instead of 660 nm as recommended by Alpkem, increased the sensitivity of the method by 30%. Pump configuration Reagent Pump tube Flow rate at 50% pump speed -------------------- ------------ --------------------------- Dowfax Red/red 0.71 ml/ min Air Black/black 0.32 ml/min Sample Red/red 0.71 ml/min Mixed colour reagent Orange/green 0.10 ml/min A3.1.2 Sampling procedure Nutrients were sampled after dissolved gases and salinity samples had been drawn. Typically, 30 to 45 minutes lapsed between the arrival of the CTD on deck and sampling for nutrients. Duplicate samples were collected in 12 ml polypropylene screw cap tubes with a 10 ml mark to prevent overfilling. Tubes and caps were rinsed three times with approximately half the volume of the tube before drawing the final sample (see section 4.1.4 in the main text). For the SR3 transect, pairs of tubes were placed into polystyrene trays, and snap frozen without any chemical preservation. When required, samples were thawed, mixed thoroughly and placed directly into the autosampler, so that no sample transfers were necessary. The racks of the autosampler had been specially modified by Alpkem to take the 12 ml sample tubes. Experiments conducted at CSIRO Division of Oceanography (R. Plaschke, unpublished notes) have shown that with careful thawing procedures, silicate samples processed within one week of freezing undergo no significant loss of silicate by polymerisation. On the PET transect, conditions were such that samples could be analysed directly, without freezing. Samples were left capped until equilibrated to room temperature, and were mixed thoroughly before analysis. If a delay was experienced, the samples were refrigerated until required. The duplicate was frozen as an emergency backup. All frozen duplicate samples were returned to Hobart and retained until data processing was completed. It is intended that immediate analysis will become a routine part of the analysis procedure on future cruises, to overcome any potential problems encountered with freezing and thawing of seawater samples. A3.1.3 Calibration and standards Standard ranges used for nutrient analyses are shown in Table A3.1. Combined standards were prepared daily using an Eppendorf Multipette and dedicated A grade volumetric glassware, using artificial seawater made from high purity reagents as a diluent. For the northernmost shallow stations of the SR3 transect, where silicate levels are depleted, a low range of silicate standards was used to increase the precision of the determinations. The calibration standards were run prior to analysing each station, in order to check the linearity of detector response, and to calculate the calibration factor required to convert peak height of an unknown sample to a concentration in mol/l. Stock standards were prepared from analytical grade reagents prior to departure of the cruise. The new batch of stock standard nutrient solutions were compared to the previous batch of stock standards as a QC check. Table A3.1: Range of calibration standards and concentration of QC standards used for analysis of nutrients on SR3 and PET transects. Nutrient Range of standards used QC standard (mol/l) (mol/l) -------------------------------------------------------------------------------- Reactive silicate (high range) as Na2SiF6 0, 28, 56, 84, 112, 140 140 Reactive silicate (low range) as Na2SiF6 0, 7, 14, 21, 28, 35 35 Orthophosphate as KH2PO4 0, 0.6, 1.2, 1.8, 2.4, 3.0 3 Nitrate plus nitrite as KNO3 0, 7, 14, 21, 28, 35 35 A3.1.4 Low Nutrient Sea Water (LNSW) LNSW is prepared from high purity NaCl, and used as a diluent for standard solutions and as the wash solution in the analytical manifold. If pure water were used as a wash solution, each peak on the phosphate and nitrate channels would be accompanied by a significant spike as the interface between pure water and seawater alternately refracts and focuses light on the photodiode. The data processing software DAPA cannot be programmed to ignore the refractive index spike, and so erroneous concentrations would be reported. By using artificial seawater, of similar salinity to the samples, the refractive index disturbance that occurs when a pure water baseline is used is eliminated. Even the highest purity NaCl, however, can be significantly contaminated with respect to phosphate. A background colour reagent is used to correct for traces of phosphate present in the wash solution and also in the analytical reagents. A3.1.5 Temperature effects and corrections During the previous cruise (Rosenberg et al., 1995) there was no temperature regulation in the hydrographic laboratory, resulting in fluctuations in sensitivity of the silicate channel of up to 20% in one day. It was not possible to maintain a stable environment, so the worst analysis runs were rejected and repeated. Those stations still showing a drift in silicate sensitivity were corrected for drift by applying a linear gain adjustment available in the data processing software DAPA. During the course of an analytical run, quality control standards are interspersed at regular intervals. These QC standards are equivalent in concentration to the top standard for each nutrient, and are used to check for drift, carryover etc. Adjacent pairs of QC standards were measured and compared, and the heights of sample peaks that fell between them were corrected by linear interpolation. (The concentration of calibration and QC standards are shown in Table A3.1.) During this cruise, there was still some evidence of fluctuating sensitivity of the silicate channel. The frequency and scale of the resulting drift was dramatically reduced relative to the previous cruise, however a correction was still required for some stations. The drift correction was applied to silicate data for all stations, rather than arbitrarily selecting the worse runs for reprocessing. In addition, a drift correction was applied to the nitrate+nitrite data for station 61. On future cruises, all nutrient data will be automatically corrected for any gain or loss of sensitivity during the course of the analytical run. When data processing in DAPA is completed, the data is imported into the program HYDRO where it is merged with the relevant cruise and station data. A3.2 DISSOLVED OXYGEN ANALYSIS A3.2.1 Equipment and technique The methodology for dissolved oxygen analysis used by the Antarctic CRC was reviewed after the previous cruise (AU9309/AU9391), and found to be in need of significant improvement. As a result, an automated dissolved oxygen system was commissioned to replace the manual titration method that had been used for previous cruises. The new methodology is based on the automated system developed by Woods Hole Oceanographic Institution (WHOI) (Knapp et al., 1990). The new method has the following advantages over the manual system: * It utilises the Carpenter (1965) reagent chemistry specified by WOCE (WOCE Operations Manual, 1991). * It requires the measurement of a reagent blank, to correct for reducing impurities. * There is a significant improvement in precision and accuracy. * The system only requires standardisation every second day, instead of daily. * The standardisation method is such that any loss of volatile iodine incurred in the process of collecting an aliquot of sample is taken into account. * There is a reduction in operator error, as the whole titration process is controlled by computer. * Data is directly acquired by computer, reducing the data processing time. * The system is extremely seaworthy and can be used reliably in rough weather, when manual titrations can be exceptionally difficult. * The method is used by a number of organisations participating in the WOCE program. The new system was trialled at sea and on shore before being adopted for routine analysis. The results of these trials, and the equations used for the calculation of dissolved oxygen concentration, are detailed in Eriksen and Terhell (in prep.) There were several ammendments made to the Knapp et al. (1990) automated method, to accomodate the equipment available at the Antarctic CRC. 300 ml sample bottles were used in place of the recommended 150 ml bottles. As a result, 2.0 ml of each of the pickling reagents was added at the time of sampling, and 2.0 ml of 10N H2SO4 was used to acidify samples prior to analysis. The calulation value for total amount of oxygen added with reagents was then 0.034 ml, as opposed to 0.017 ml in the Knapp et al. (1990) method. Table A3.2 summarises the details of the automated dissolved oxygen method. Note that duplicate titrations were performed every 6 samples as a check on the reproducibility of titrations. Table A3.2: Summary of details of WHOI automated oxygen method (Knapp et al., 1990). Modifications to the WHOI automated method include: (a) 300 ml sample bottles are used rather than 150 ml (note a in the table), and subsequently (b) 2 ml of reagents are added to the sample bottle rather than 1 ml (note b in the table). Endpoint: Amperometric Bottle volume: 300 ml (note a) Aliquot volumes: 50 ml Size of burette: 10 ml Smallest measurable volume increment (l): 1 Standard solution: 0.01N KH(IO3)2 Standard preparation: Vacuum dried Standard volume: 15 ml Blank determined: Yes Blank tests for: Redox species in reagents plus bias in measured endpoint Blank result used in calculations: Yes Scope for negative blank: Yes Mn reagent in standards: Yes Standardise daily: No Thiosulphate normality: 0.01 N Reagent chemistry: 3 M MnCl2 (2 ml) (note b) 8 N NaOH/4 M NaI (2 ml) (note b) 10 N H2SO4 (2 ml) (note b) Reagent filtering: All double filtered Final sample pH: 2 Specified reaction time: 2-4 hours Correction for DO in reagents: Yes Standard and sample handling procedures the same: Yes Average sample processing time: 1.5-2 minutes A3.2.2 Sampling procedure Samples were drawn in accordance with the protocols documented in section 4.1.4 of the main text. A3.3 SALINITY ANALYSIS A3.3.1 Equipment and technique Salinity analysis was conducted using a YeoKal Mark 4 Inductively Coupled Salinometer (Yeokal Electronics, Sydney Australia). The manufacturer claims that with sufficient care, and in a constant temperature environment, an experienced operator should be able to attain an accuracy of 0.003 psu. The salinometer was standardised daily using IAPSO P-series salinity standards, in accordance with WOCE guidelines. Immediately after the standardisation procedure was completed, the conductivity ratio of a bulk seawater "substandard" was measured. The substandard was then measured in triplicate every 10 samples, to monitor the electronic drift of the instrument. If the drift exceeded 0.00005 conductivity units, then another vial of IAPSO International seawater was used to check the calibration of the instrument. Samples were left for 12 to 24 hours to equilibrate to room temperature before analysing. The station to be analysed next was always positioned beside the substandard and international standard, to ensure that all three fell within the same temperature compensation bandwidth. The YeoKal salinometers do not have a thermostated bath around the conductivity cell, thus the temperature at which conductivity ratios are determined is also measured, and must be confined to a narrow range. A3.3.2 Sampling procedure Samples were collected in accordance with the protocol detailed in section 4.1.4 of the main text. A3.3.3 Data processing Conductivity ratios were entered manually into the HYDRO program, which calculates salinity (PSS-78) from the conductivity and calibration data acquired on the salinometer. The program also calculates and corrects for any instrument drift by linear interpolation between pairs of substandard observations. REFERENCES Alpkem Corporation, 1992. "The Flow Solution" Operation Manual. Alpkem Corporation 9445 SW Ridder Rd Wilsonville, OR 97070 USA. Carpenter, J.H., 1965. The Chesapeake Bay Institute technique for the Winkler Dissolved Oxygen method. Limnology and Oceanography. 10: 141-143. Eriksen, R. (in prep.). A Practical Manual for the Determination of Salinity, Dissolved Oxygen and Nutrients in Seawater. Antarctic CRC Technical Report, Hobart. Eriksen, R. and Terhell, D., (in prep.). A Comparison of Manual and Automated Methods for the Determination of Dissolved Oxygen in Seawater. Antarctic CRC Technical Report, Hobart. Knapp, G.P., Stalcup, M.C., and Stanley, R.J., 1990. Automated Oxygen Titration and Salinity Determination. Woods Hole Oceanographic Institution Technical Report WHOI-90-35. Rosenberg, M., Eriksen, R. and Rintoul, S., 1995. Aurora Australis marine science cruise AU9309/AU9391 - oceanographic field measurements and analysis. Antarctic Cooperative Research Centre, Research Report No. 2, March 1995. 103 pp. Smythe-Wright, D., Paylor, R. and Holley, S., 1992. Chemical tracer studies as IOSDL-3. The measurement of silicate, nitrate and phosphate in seawater. Institute of Oceanographic Sciences Deacon Laboratory, Report No. 302, 1992. 73 pp. WOCE Operations Manual, 1991. WHP Office Report WHPO 91-1, WOCE Report No. 68/91, Woods Hole, Mass., USA. APPENDIX 4 Data File Types A4.1 UNDERWAY MEASUREMENTS The underway measurements for the cruise, as logged automatically by the ship's data logging system, and quality controlled by human operator (Ryan, 1995), are contained in column formatted ascii files. The two file types contain 10 sec digitised data, and 15 min averaged data. In both cases, missing data or data flagged as bad are replaced by the null value -999. The files are padded out to commence on the first digitising interval of the first day in the file, and ending at the last digitising interval on the last day in the file. A4.1.1 10 second digitised underway measurement data Data at the minimum digitised interval of 10 sec. are contained in files named *.alf (Table A4.1), where the data filename prefix corresponds to the cruise acronym ("sham"). A two line header is followed by the data as follows: column parameter ------ -------------------------------------------------------------------------- 1 decimal time (0.0=midnight on December 31st, therefore, for example, 1.5=midday on January 2nd) 2 day 3 month 4 year 5 hour 6 minute 7 second 8 latitude (decimal degrees, +ve=north, -ve=south) 9 longitude (decimal degrees, +ve=east, -ve=west) 10 depth (m) 11 sea surface temperature (oC) (measured at the seawater inlet at 7 m depth) Note that all times are UTC. Table A4.1: Example 10 sec digitised underway measurement file (*.alf file). Aurora Australis data - GPS pos. (deg), depth (m), sea surface temp (deg C) decimaltime day mn yr hr m s lat lon depth SST --------------------------------------------------------------------- 70.00000004 12 3 1993 0 0 0 -999.0000 -999.0000 -999.0 -999.00 70.00011578 12 3 1993 0 0 10 -999.0000 -999.0000 -999.0 -999.00 70.00023148 12 3 1993 0 0 20 -44.0044 146.3534 284.6 15.20 70.00034722 12 3 1993 0 0 30 -44.0044 146.3529 -999.0 15.20 70.00046296 12 3 1993 0 0 40 -44.0044 146.3530 283.5 15.20 70.00057870 12 3 1993 0 0 50 -44.0044 146.3523 287.4 15.20 70.00069444 12 3 1993 0 1 0 -44.0043 146.3519 282.2 15.20 70.00081019 12 3 1993 0 1 10 -44.0044 146.3515 282.4 15.20 70.00092593 12 3 1993 0 1 20 -44.0044 146.3511 283.3 15.20 70.00104167 12 3 1993 0 1 30 -44.0044 146.3507 286.0 15.20 70.00115741 12 3 1993 0 1 40 -44.0044 146.3507 286.3 15.20 70.00127315 12 3 1993 0 1 50 -44.0044 146.3502 286.8 15.20 70.00138889 12 3 1993 0 2 0 -44.0043 146.3498 287.4 15.20 70.00150463 12 3 1993 0 2 10 -44.0043 146.3493 291.0 15.20 A4.1.2 15 minute averaged underway measurement data 15 minute averaged data are contained in files named *.exp (Table A4.2), where the data filename prefix corresponds to the cruise acronym ("sham"). Note that wind direction and ship's heading are instantaneous values. All times represent the centre of the averaging interval. A two line header is followed by the data as follows: column parameter ------ --------------------------------------------------------------------------- 1 decimal time (as for 10 sec digitised files) 2 latitude (as for 10 sec digitised files) 3 longitude (as for 10 sec digitised files) 4 air pressure (hecto Pascals) 5 wind speed (knots) 6 wind direction (deg. true) 7 port air temperature (oC) 8 starboard air temperature (oC) 9 port relative humidity (%) 10 starboard relative humidity (%) 11 quantum radiation (µmol/s/m2) 12 ship speed (knots) (speed through the water) 13 ship heading (deg. true) 14 ship roll (deg.) 15 ship pitch (deg.) 16 sea surface salinity (parts per thousand) (from seawater inlet at 7 m depth) 17 sea surface temperature (oC) (at seawater inlet, 7 m depth) 18 average fluorescence (arbitrary units) (from seawater inlet at 7 m depth) 19 seawater flow (l/min) (flow rate at seawater inlet) Note that all times are UTC. Table A4.2: Example 15 min averaged underway measurement file (*.exp file). Aurora Australis DLS data: dumped by EXPORT. Column units: days,°,°,hPa,knots,°True,°C,°C,%,%,umol/s/m2,knots,°True,°,°,ppt,°C,-,l/min decimaltime lat long airP windsp windd poairT stairT pohum sthum qrad shipspd shiphdg roll pitch ssSAL ssT avfluo seaflow -------------------------------------------------------------------------------------------------------------------------------------------------- 70.00520833 -44.00310 146.33583 1022.2 19.6 293 14.2 14.2 93 88 -999 6.56 235.5 1.185341 0.486591 35.175 15.20 -999.000 9.95 70.01562500 -44.00076 146.31305 1022.3 22.1 290 14.2 14.3 92 87 -999 1.15 235.5 1.295333 0.346111 35.165 15.10 -999.000 9.97 70.02604167 -44.00056 146.31239 1022.3 20.6 305 14.0 14.0 94 89 -999 0.00 235.5 2.568000 0.287667 35.159 15.10 -999.000 9.98 70.03645833 -44.00036 146.31232 1022.2 20.6 298 14.1 14.0 94 89 -999 0.00 235.5 1.303000 0.274444 35.165 15.10 -999.000 9.99 70.04687500 -44.00000 146.31136 1022.2 20.1 298 14.0 14.0 95 90 -999 0.00 234.5 1.380111 0.433667 35.166 15.10 -999.000 9.99 70.05729167 -43.99958 146.31143 1022.2 20.7 288 14.1 14.1 94 89 222 0.00 234.5 1.801667 0.464667 35.165 15.10 -999.000 9.97 70.06770833 -43.99918 146.31229 1022.3 18.5 295 13.8 14.1 96 90 170 0.00 234.5 1.619333 0.398334 35.164 15.20 -999.000 9.99 A4.2 2 DBAR AVERAGED CTD DATA FILES The final format in which CTD data is distributed is as 2 dbar averaged data, contained in column formatted ascii files, named *.all (Table A4.3) (the filename prefix is discussed in Appendix 2). Averaging bins are centered on even pressure values, starting at 2 dbar. A 15 line header is followed by the data, as follows: column parameter ------ --------------------------------------------------------------------------- 1 pressure (dbar) 2 temperature (oC) (ITS-90) 3 salinity (psu) 4 sigmaT = density-1000 (kg.m-3) 5 specific volume anomaly x 108(m3.kg-1) 6 geopotential anomaly (J.kg-1) 7 dissolved oxygen (µmol.l-1) 8 number of data points used in the 2 dbar averaging bin 9 standard deviation of temperature values in the 2 dbar bin 10 standard deviation of conductivity values in the 2 dbar bin 11 fluorescence (mg.m-3) (uncalibrated) 12 photosynthetically active radiation (µmol.s-1.m2) (uncalibrated) All files start at the 2 dbar pressure level, incrementing by 2 dbar for each new data line. Missing data are filled by blank characters (this most often applies to dissolved oxygen data). Table A4.3: Example 2 dbar averaged CTD data file (*.all file). SHIP R.V. Aurora Australis ----------------- --------------------------- STATION NUMBER 4 DATE 02-JAN-1994 (DAY NUMBER 2) START TIME 1020 UTC = Z BOTTOM TIME 1100 UTC = Z FINISH TIME 1222 UTC = Z CRUISE Au94/07 START POSITION 44:07.03S 146:13.35E BOTTOM POSITION 44:07.14S 146:13.71E FINISH POSITION 44:06.61S 146:13.95E MAXIMUM PRESSURE 1038 DECIBARS BOTTOM DEPTH 1015 METRES PRESS TEMP (T-90) SAL SIGMA-T S.V.A. G.A. D.O. fluorescence p.a.r. ------------------------------------------------------------------------------------- 2.0 11.899 34.773 26.432 158.69 0.032 277.6 30 0.001 0.007 0.95569E+01 -0.49498E+00 4.0 11.899 34.778 26.436 158.41 0.063 280.3 30 0.001 0.001 0.10817E+02 -0.63459E+00 6.0 11.903 34.779 26.436 158.46 0.095 281.1 45 0.001 0.002 0.90911E+01 -0.60488E+00 8.0 11.903 34.778 26.435 158.55 0.127 278.0 41 0.000 0.000 0.80700E+01 -0.58265E+00 10.0 11.903 34.778 26.435 158.60 0.159 278.6 32 0.001 0.001 0.75122E+01 -0.66496E+00 12.0 11.904 34.778 26.435 158.66 0.190 280.2 32 0.001 0.001 0.72758E+01 -0.55944E+00 14.0 11.905 34.778 26.435 158.72 0.222 281.5 40 0.000 0.000 0.73697E+01 -0.62194E+00 16.0 11.907 34.779 26.435 158.76 0.254 277.5 34 0.002 0.002 0.69932E+01 -0.56719E+00 18.0 11.908 34.780 26.435 158.77 0.286 275.7 25 0.002 0.002 0.68356E+01 -0.63807E+00 20.0 11.909 34.779 26.435 158.90 0.317 276.0 30 0.002 0.002 0.69607E+01 -0.54045E+00 22.0 11.911 34.780 26.435 158.93 0.349 275.3 63 0.003 0.003 0.69971E+01 -0.59554E+00 24.0 11.917 34.781 26.435 158.95 0.381 265.2 47 0.004 0.006 0.69678E+01 -0.58176E+00 26.0 11.923 34.783 26.435 159.01 0.413 268.1 31 0.002 0.003 0.70108E+01 -0.63026E+00 28.0 11.909 34.779 26.434 159.11 0.444 272.3 26 0.002 0.002 0.68240E+01 -0.57804E+00 A4.3 HYDROLOGY DATA FILES Files named *.bot (where the filename prefix is the the cruise code e.g. a9407) are column formatted ascii files containing the hydrology data, together with CTD upcast burst data (Table A4.4). The columns contain the following values: column parameter ------ -------------------------------------------------- 1 station number 2 CTD pressure (dbar) 3 CTD temperature (oC) 4 reversing thermometer temperature (oC) 5 CTD conductivity (mS.cm-1) 6 CTD salinity (psu) 7 bottle salinity (psu) 8 ortho phosphate concentration (µmol.l-1) 9 nitrate + nitrite concentration (µmol.l-1) 10 reactive silicate concentration (µmol.l-1) 11 bottle dissolved oxygen concentration (µmol.l-1) 12 bottle quality flag (-1=rejected, 0=suspect, 1=good) 13 niskin bottle number Missing data values are filled by a decimal point (surrounded by blank characters). Parameters 2,3,5 and 6 are mean values from the upcast CTD burst data at the time of bottle firing, where each burst contains the data 5 sec previous to the time of bottle firing. Parameters 7 to 11 are laboratory values for the hydrology analyses. Parameter 12, the bottle quality flag, is relevant to the calibration of CTD salinities - bottles flagged 1 and 0 are used for calibration, while those flagged -1 are rejected. Criteria for flagging of the bottle data are discussed elsewhere (Appendix 2). Parameter 13, the niskin bottle number, is a unique identifier for each bottle. Note that the bottle number does not always correspond with rosette position. Table A4.4: Example hydrology data file (*.bot file). 2 8.556 15.155 15.154 43.109 35.032 35.031 0.29 8.80 7.7 247.10 1 11 2 25.593 15.111 . 43.076 35.034 35.035 0.28 0.20 3.7 248.50 1 9 2 50.992 15.105 . 43.085 35.038 35.038 0.27 0.30 2.2 249.10 1 8 2 73.718 14.188 . 42.227 35.068 35.077 0.48 4.40 2.8 228.70 -1 7 2 98.376 12.840 . 40.910 35.055 35.051 0.66 7.70 2.5 227.60 -1 6 2 123.524 12.490 . 40.618 35.089 35.081 0.76 9.60 3.0 223.10 -1 5 2 148.516 11.904 . 40.025 35.052 35.067 0.85 11.10 3.4 223.30 -1 4 2 200.278 11.085 . 39.174 34.963 34.965 0.90 13.30 4.0 226.40 -1 3 2 247.807 10.678 10.691 38.758 34.914 34.914 1.02 13.90 4.1 230.40 0 2 2 289.188 9.625 . 37.640 34.769 34.794 1.13 15.80 4.8 232.40 -1 1 3 8.609 15.984 15.958 44.199 35.274 35.275 . 0.20 1.6 270.80 1 16 3 21.504 15.975 . 44.198 35.276 35.275 0.25 0.20 1.5 266.60 1 15 3 48.210 15.935 . 44.171 35.277 35.276 0.25 0.40 0.7 264.60 1 14 3 73.795 15.897 . 44.140 35.273 35.270 0.27 0.80 1.6 238.30 -1 13 3 98.905 14.011 . 42.238 35.229 35.236 0.63 7.50 2.3 . -1 12 3 148.674 12.557 . 40.763 35.155 35.155 0.81 10.90 4.1 216.00 0 11 3 197.813 11.432 . 39.575 35.033 35.033 0.92 12.80 3.9 227.30 1 10 3 298.658 10.110 . 38.158 34.828 34.831 1.10 15.40 4.6 230.70 1 9 3 396.295 9.214 . 37.238 34.702 34.703 1.28 18.70 6.0 226.20 -1 8 3 496.675 8.371 . 36.405 34.604 34.603 1.52 22.50 9.3 210.60 1 7 3 597.207 7.385 . 35.469 34.524 34.524 1.71 25.90 14.6 199.30 1 6 3 697.115 6.587 . 34.751 34.487 34.486 1.90 28.30 20.6 195.30 1 5 3 778.707 5.739 . 33.995 34.458 34.458 2.05 30.50 27.8 . 1 4 3 900.509 4.315 . 32.710 34.381 34.382 2.20 32.70 33.6 198.50 1 3 3 1000.091 4.027 4.029 32.574 34.471 34.471 2.34 34.30 49.6 171.00 1 2 3 1113.395 3.403 . 32.110 34.517 34.522 2.42 35.40 61.3 169.90 -1 1 4 23.926 15.341 . 43.397 35.121 35.120 0.26 0.10 0.6 230.60 1 23 4 49.736 15.198 . 43.231 35.088 35.087 0.26 0.30 0.6 229.10 1 22 4 99.651 13.388 . 41.599 35.202 35.200 0.77 9.00 2.6 200.60 1 21 4 148.952 12.164 . 40.341 35.114 35.122 0.86 12.90 3.8 221.80 -1 20 4 196.847 11.114 . 39.222 34.985 34.980 0.95 11.40 3.6 233.30 -1 19 4 298.033 9.997 . 38.028 34.804 34.803 1.02 13.80 . 254.10 -1 18 4 384.198 9.235 . 37.228 34.676 34.677 . . . 256.20 -1 17 4 495.853 8.452 . 36.455 34.578 34.577 1.43 20.70 8.1 232.70 -1 16 A4.4 STATION INFORMATION FILES Station information files, named *.sta (Table A4.5) (where the filename prefix is the cruise code), contain position, time, bottom depth and maximum pressure of cast for CTD stations. The CTD instrument number is specified in the file header. Position and time (UTC) are specified at the start, bottom and end of the cast, while the bottom depth is for the start of the cast. Note that small inconsistencies may exist between bottom depth and maximum pressure, due to drift of the vessel between the start and bottom of the cast. In addition, a single value is assumed for the sound velocity in seawater for echo sounder calculations (1498 m.s-1), which may cause small errors in water depth values. Table A4.5: Example CTD station information file (*.sta file). ===================================================================== RSV Aurora Australis Cruise: Au93/09 CTD station list (CTD unit 4) ===================================================================== bottom| | stat| start depth|max P | bottom | end no. |time date latitude longitude (m)|(dbar)|time latitude longitude |time latitude longitude ----|----------------------------------------|------|-------------------------|------------------------- 1 |2032 11-MAR-93 44:06.73S 146:14.35E 1000| 956 |2118 44:06.37S 146:14.35E|2154 44:06.19S 146:14.60E 2 |0027 12-MAR-93 44:00.06S 146:18.61E 300| 289 |0042 44:00.03S 146:18.77E|0115 43:59.97S 146:18.64E 3 |0513 12-MAR-93 44:07.51S 146:14.89E 1100| 1115 |0549 44:07.48S 146:15.06E|0632 44:07.39S 146:15.23E 4 |0854 12-MAR-93 44:27.89S 146:07.94E 2340| 2335 |0938 44:27.52S 146:07.30E|1028 44:27.32S 146:07.51E 5 |1437 12-MAR-93 44:56.71S 145:56.67E 3380| 3465 |1606 44:56.10S 145:56.52E|1727 44:55.56S 145:56.36E REFERENCES Ryan, T., 1995. Data Quality Manual for the data logged instrumentation aboard the RSV Aurora Australis.. Australian Antarctic Division, unpublished manuscript, second edition, April 1995. APPENDIX 5 Data Processing Information Table A5.1: Upcast CTD bursts automatically flagged during creation of intermediate CTD files (Appendix 2). station rosette position station rosette position number flag=-1 flag=0 number flag=-1 flag=0 -------------------------------------- ----------------------------------------------------- 1 SR3 19,22,24 56 SR3 23 2 SR3 14 15,17,22 57 SR3 21 22,24 3 SR3 1,2,9,11,12 5,8,10,13 58 SR3 21,22 20,23,24 4 SR3 1,19,20 3,7,8,15,17 59 SR3 22,24 5 SR3 11,14,15,17,19 60 SR3 21,23 6 SR3 20,21 11,12,13 61 SR3 22 4,5,7,13,14,16,17,18,23 7 SR3 4 63 SR3 20,23 22 8 SR3 12,14,21,22 15,16,17,20 64 SR3 19 22,23 9 SR3 22 15,20,21 65 SR3 19 6,16,17,18,20 10 SR3 21 19,20 66 SR3 5,6,10,12,23 11 SR3 16 14 67 SR3 11,12,17,21,24 2,9,10,14,16,20 12 SR3 13,22 68 SR3 6,17 11,13,14,20 13 SR3 19,20 10,11,17,18 69 SR3 13,14 3,4,5 14 SR3 13,20 70 SR3 13 4,5,7 15 SR3 20 12,13,14 71 SR3 9 16 SR3 13,14 72 SR3 14 13,15 17 SR3 20 17,18 73 SR3 8 19 SR3 18,19,22 13,15,17 74 ULS 3,11 4,17,21 20 SR3 14 75 ULS 19,23 21 SR3 18 76 PET 19,21 15,17,20 22 SR3 18 77 PET 19,21 23 SR3 14,21 78 PET 10 12,14,15,18,19,21 24 SR3 14 11,12,15,21 79 PET 16 9,14,21,22,23 25 SR3 17,22 3,21 80 PET 19 12,20,22,23,24 26 SR3 14,23 17,19 81 PET 22,23 14,15 30 SR3 15 16,17,18,19,20 82 PET 15,18,19,23 32 SR3 18,21,23 22 83 PET 17,19,21 15,18,20,22 33 SR3 19,22 18,21 84 PET 22 21,24 34 SR3 21 85 PET 19,21,22,23,24 35 SR3 20,21 86 PET 9,11,12,13,15 3,5,6,7,10,16,20 36 SR3 24 19,20,21,22,23 87 PET 22 15 37 SR3 19,20,21,24 88 PET 21,22 19,23 39 SR3 15,21,23,24 10,16,17,18 89 PET 5,7,9,11,13,14 10,12 40 SR3 24 91 PET 21,22 41 SR3 24 20,22 92 PET 21 22,23 42 SR3 18,20 22,23 93 PET 22 43 SR3 11,13,17,19,22 14,15,21 94 PET 3 44 SR3 20 95 PET 21 20,22,23,24 45 SR3 21 20 96 PET 18,19 17,21,22 46 SR3 20 97 PET 16,17,21,23 14,15,19,20,24 48 SR3 20 21,23 98 PET 14,15,17 49 SR3 11,12,23 14,18,19,20,22 99 PET 15,16,17,19 13,20 51 SR3 21,22 100 TS 11,12,13,15,16,17,18 5,14,19 52 SR3 8,10 9 101 TS 6,7,23 53 SR3 23,24 102 BPR 13,14 15,20 54 SR3 21,24 55 SR3 21,22 20 Table A5.2: Dissolved oxygen Niskin bottle samples flagged as -9 for dissolved oxygen calibration. Note that this does not necessarily indicate a bad bottle sample - in many cases, flagging is due to bad CTD dissolved oxygen data. station rosette station rosette station rosette number position number position number position ----------------- -------------------- ----------------- 3 1 42 17,18,19,20 71 11 4 19 45 20,21 74 3 5 16,21 46 18,19,20 75 19 7 2 47 21 76 20 13 14 48 21 77 17,19 15 12,17 51 21,22 79 8,13 17 22 54 21 81 14,24 19 15,16,17 55 19 83 8,18 20 23,24 58 21 84 22 23 16 60 22 86 18 24 15,21,24 63 20 91 21 26 18 64 21 98 16 30 16 65 18,19 99 14,20 33 18,19,20 68 11,15,17 100 12,18 39 18 69 13,15 102 13,14 40 23 70 11 Table A5.3: Stations containing fluorescence (fl) and photosynthetically active radiation (par) 2 dbar-averaged data. stations with fl data stations with par data ------------------------------------ ---------------------- 1 1 3,4,5 3,4,5 7 7 11 11 to 102 16 21 39 43 76,77,78,79,80,81 83 86 89,90,91,92,93,94,95,96,97,98,99,100 Table A5.4: Protected and unprotected reversing thermometers used for cruise AU9407 (serial numbers are listed). protected thermometers station rosette position 24 rosette position 2 rosette position 18 numbers thermometers thermometers thermometers -------- ------------------- ------------------ ------------------- 1 to 101 12095,12096 12094,11973 - 102 12095,12096 12094,11973 12120,12119 unprotected thermometers station rosette position 24 rosette position 2 rosette position 18 numbers thermometers thermometers thermometers -------- ------------------- ------------------ ------------------- 1 to 101 11992 - 102 11992 11993 APPENDIX 6 Historical Data Comparisons A6.1 INTRODUCTION In this Appendix, a brief comparison is made between the au9407 cruise data, and data from the previous cruise au9309; data from the SR3 transect only is discussed. The SR3 transect was occupied during the autumn of 1993, and summer of 1993/94, for cruises au9309 and au9407 respectively. For data prior to 1993, see Appendix 6 in Rosenberg et al. (1995). The following terminology is used for the discussion in this Appendix (taken from Patterson and Whitworth, 1990): Subantarctic Zone - lying between the Subtropical and Subantarctic Fronts; Subantarctic Front - as defined by Gordon et al. (1977); marked by a rapid southward decrease in surface temperature and salinity; Polar Frontal Zone - transition zone between the Subantarctic and Polar Fronts; Polar Front - as defined by Emery (1977); marked by the northern terminus of the well defined subsurface temperature minimum layer; Antarctic Zone - south of the Polar Front; TS curves and vertical profiles of dissolved oxygen and nutrients from a series of locations along the SR3 transect are compared for the two cruises (Figure A6.1). Although some nomenclature differences exist in the literature when discussing the Southern Ocean, particularly for meridional transitions in water characteristics, the above definitions are used as convenient references for selecting representative stations for the inter-cruise comparison. Positions for all stations referred to in Figure A6.1 are listed in Table A6.1. Note that the stations selected for graphing in Figure A6.1 display most, but not all, of the general trends discussed below: it is difficult to simultaneously display all these trends without graphing the entire data set. Table A6.1: Positions for all stations referred to in Figure A6.1. au9309 au9407 ----------------------- ----------------------- stn lat. S long. E stn lat. S long. E Subantarctic Zone 12 48:18.91 144:32.00 17 48:18.90 144:31.73 Subantarctic Zone 14 49:16.18 144:05.26 20 49:16.17 144:05.64 Polar Frontal Zone 21 52:15.27 142:37.50 29 52:15.46 142:37.53 Polar Frontal Zone 22 52:38.18 142:23.56 31 52:39.41 142:22.88 Antarctic Zone 27 55:01.15 141:00.75 37 55:01.16 141:00.58 Antarctic Zone 30 56:26.22 140:06.15 41 56:26.28 140:06.01 Antarctic Zone 42 60:21.22 139:50.86 51 60:21.36 139:50.59 Antarctic Zone 43 60:21.34 139:50.91 Antarctic Zone 48 61:50.76 139:51.22 55 61:51.11 139:50.95 Antarctic Zone 49 61:51.06 139:51.58 Antarctic Zone 56 63:50.89 139:51.75 60 63:52.03 139:51.10 Antarctic Zone 57 63:47.35 139:54.20 Antarctic Zone 62 65:05.06 139:51.08 64 65:04.98 139:50.91 Antarctic Zone 63 65:04.89 139:51.27 A6.2 RESULTS A6.2.1 CTD temperature and salinity Comparison of TS diagrams for the two cruises (Figure A6.1a to e) reveals a relative decrease in salinities for au9407 data over most of the water column. This unexpected result is particularly surprising for the Circumpolar Deep Water. Comparison of meridional variation of the salinity maximum for the two cruises i.e. for Lower Circumpolar Deep Water (as defined by Gordon, 1967) (Figure A6.2) reveals a large consistent relative decrease in salinity for au9407 data, of the order 0.006 psu. Note that in Figure A6.2, property differences are only formed between station pairs (i.e. corresponding au9309 and au9407 stations) which are separated by less than 1.5 nautical miles of latitude; north of ~47.5o south, station locations for the two cruises do not correspond spatially. On initial inspection, the consistent deep water salinity difference between the two cruises suggests some systematic error in the salinity data. In particluar, temperature and pressure differences at the salinity maxima are randomly scattered about zero (Figure A6.2), suggesting that there is a biasing of the salinity measurements for one of the cruises (no biasing exists for either the temperature or pressure data). However, at the time of writing, comparison of au9309 data with the latest salinity data from the SR3 transect in January 1995 (unpublished) reveals a comparable decrease in salinity at many locations i.e. agreement between au9407 and the January 1995 data. Comparison of the au9309 data with earlier data from cruise au9101 (Figure A6.3) shows a possible salinity increase for au9309 south of ~49o south. This increase is however not comparable to the large salinity difference between cruises au9309 and au9407. The International Standard Seawater (ISS) batches used for salinity analyses during the different cruises are listed in Table A6.2. Note that the same batch was used for au9309 and for part of au9404. In particular, degradation of ISS batches with time would result in salinity increase of the ISS, in turn yielding a positive salinity bias for salinometer analyses. This is not consistent with the observed decrease in salinities from au9309 to au9407 and au9404. Thus the ISS is unlikely to account for the large salinity differences. The only other possible source of a systematic salinity error is the salinometers. Different YeoKal Mk IV salinometers were indeed used for cruises au9309 and au9407 (Table A6.2). In addition, temperature control of the hydrology laboratory was only introduced just prior to cruise au9407. As a result, for cruise au9407 and later, salinity analyses were conducted at laboratory temperatures of ~22 to 23oC ( 2oC; whereas for cruise au9309, the average laboratory temperature was ~19 to 20oC (see Figure 2 in Rosenberg et al., 1995), and much larger fluctuations in temperature were experienced (up to ~( 10oC). Some as yet unknown inconsistency of salinometer behaviour, either a function of the instrument used or else the ambient temperature at which analyses occur, may indeed contribute to the observed salinity differences between cruises. Whether the entire difference can be attributed to systematic instrument error is, at this stage, inconclusive. Table A6.2: International Standard Seawater (ISS) batches and salinometers used for different cruises. cruise station nos. ISS date of ISS salinometer no. batch # ------ ------------------ ------- -------------- ----------------------- au9309 1-63 (SR3) P121 8th Sept. 1992 601003 (stations 1-63) au9407 1-79 (SR3 and PET) P123 10th June 1993 601855 (stations 1-86) au9407 80-102 (PET) P121 8th Sept. 1992 601003 (stations 87-102) au9404 1-85 (S4 and SR3) P123 10th June 1993 601855 (stations 1-107) au9404 86-107 (SR3) P121 8th Sept. 1992 For the Polar Frontal Zone and more northerly parts of the Antarctic Zone (Figures A6.1b to d), surface waters for the au9309 data are fresher as a result of relatively higher precipitation during the autumn months. Further south (Figure A6.1e), surface waters are saltier for the au9309 data, due to removal of fresh water by the formation of sea ice. Also worth noting is the temperature difference in the sub surface temperature minimum layer south of the Polar Front (Figures A6.1c to e). For cruise au9407, the minimum temperature in this layer is lower by up to ~1oC, attributable to seasonal variablility. A6.2.2 Dissolved oxygen Au9407 dissolved oxygen concentrations, measured using the WHOI automated method (see Appendix 3), are consistently higher than au9309 values, measured using the CSIRO manual method (Appendix 3 in Rosenberg et al., 1995), over the entire water column (Figures A6.1a to e). This observation is consistent with results found by Eriksen and Terhell (in prep.) when comparing the automated and manual analysis methods: concentrations analysed using the manual method are consistently lower by approximately 1%. A6.2.3 Nutrients Phosphate and nitrate+nitrite concentrations are in general consistent for the au9407 and au9309 data, revealed by comparison of the nitrate+nitrite to phosphate ratio (Figure A6.4). For more northerly stations (in the Subantarctic Zone), nitrate+nitrite concentrations are typically higher for au9309 data. These data can be seen (Figure A6.4) as a cluster of higher au9309 nitrate+nitrite values at the high concentration end of the scale. For phosphate data, au9309 concentrations are frequently lower in the more northerly parts of the Antarctic Zone, contributing to the clustering of au9309 data to the left of the best fit line at the higher concentration end of the scale in Figure A6.4. At latitudes other than those just mentioned, there is no consistent offset between phosphate and nitrate+nitrite data for the two cruises, with the exception of surface waters subject to high seasonal variability. For stations south of ~49o south, silicate concentrations are more often higher for the au9309 data by, on average, ~3 to 5 _mol/l. Exceptions to this pattern are for surface waters, where silicate values are usually lower for the au9309 data; and for data near the bottom in the Antarctic Zone, where no consistent silicate concentration offset between the two cruises is evident. REFERENCES Emery, W.J., 1977. Antarctic Polar Front Zone from Australia to the Drake Passage. Journal of Physical Oceanography. 9: 456-468. Eriksen, R. and Terhell, D., (in prep.). A Comparison of Manual and Automated Methods for the Determination of Dissolved Oxygen in Seawater. Antarctic CRC Technical Report, Hobart. Gordon, A.L., 1967. Structure of Antarctic waters between 20oW and 170oW. Antarctic Map Folio Series, Folio 6, Bushnell, V. (ed.). American Geophysical Society, New York. Gordon, A.L., Taylor, H.W. and Georgi, D.T., 1977. Antarctic oceanography zonation. In Polar Oceans, Dunbar, M.J. (ed.). Proceedings of the Polar Ocean Conference, McGill University, Montreal. Arctic Institute of North America, Calgary. Patterson, S.L. and Whitworth, T., 1990. Chapter 3 Physical Oceanography in Antarctic Sector of the Pacific (G.P. Glasby editor). Elsevier Oceanography Series, 51. Rosenberg, M., Eriksen, R. and Rintoul, S., 1995. Aurora Australis marine science cruise AU9309/AU9391 - oceanographic field measurements and analysis. Antarctic Cooperative Research Centre, Research Report No. 2, March 1995. 103 pp. Figure A6.1a: TS diagrams, and dissolved oxygen and nutrient vertical profile data, for comparison of au9407 and au9309 data: stations north of the Subantarctic Front. Note that all dissolved oxygen data is CTD 2 dbar-averaged data. Figure A6.1b: TS diagrams, and dissolved oxygen and nutrient vertical profile data, for comparison of au9407 and au9309 data: stations between the Subantarctic and Polar Fronts. Note that all dissolved oxygen data is CTD 2 dbar-averaged data. Figure A6.1c: TS diagrams, and dissolved oxygen and nutrient vertical profile data, for comparison of au9407 and au9309 data: stations south of the Polar Front. Note that all dissolved oxygen data is CTD 2 dbar-averaged data. Figure A6.1d: TS diagrams, and dissolved oxygen and nutrient vertical profile data, for comparison of au9407 and au9309 data: stations south of the Polar Front. Note that dissolved oxygen data is CTD 2 dbar-averaged data for au9407, and Niskin bottle data for au9309. Figure A6.1e: TS diagrams, and dissolved oxygen and nutrient vertical profile data, for comparison of au9407 and au9309 data: stations south of the Polar Front. Note that dissolved oxygen data is CTD 2 dbar-averaged data for au9407, and Niskin bottle data for au9309. Figure A6.2: Variation with latitude south along the SR3 transect of properties at the deep salinity maximum (marking the Lower Circumpolar Deep Water): property differences are between cruise au9309 and cruise au9407 i.e. au9309 value minus au9407 value. Note that differences are formed only between stations from the two cruises which are separated by no more than 1.5 nautical miles of latitude. Figure A6.3: Variation with latitude south along the SR3 transect of properties at the deep salinity maximum (marking the Lower Circumpolar Deep Water): property differences are between cruise au9309 and cruise au9101 i.e. au9309 value minus au9101 value. Note that au9309 values are obtained by linearly interpolating between au9309 station latitudes to correspond with au9101 station latitudes. Figure A6.4: Bulk plot of nitrate+nitrite versus phosphate for all au9309 and au9407 data along the SR3 transect, together with linear best fit lines. APPENDIX 7: WOCE Data Format Addendum A7.1 INTRODUCTION This Appendix is relevant only to data submitted to the WHP Office. For WOCE format data, file format descriptions as detailed earlier in this report should be ignored. Data files submitted to the WHP Office are in the standard WOCE format as specified in Joyce et al. (1991). A7.2 CTD 2 DBAR-AVERAGED DATA FILES * CTD 2 dbar-averaged file format is as per Table 3.12 of Joyce et al. (1991), except that measurements are centered on even pressure bins (with first value at 2 dbar). * CTD temperature and salinity are reported to the third decimal place only. * Files are named as in Appendix 2, section A2.2.1, except that for WOCE format data the suffix ".all" is replaced with ".ctd". * The quality flags for CTD data are defined in Table A7.1. Data quality information is detailed in earlier sections of this report. A7.3 HYDROLOGY DATA FILES * Hydrology data file format is as per Table 3.7 of Joyce et al. (1991), with quality flags defined in Tables A7.2 and A7.3. * Files are named as in Appendix 2, section A2.2.2, except that for WOCE format data the suffix ".bot" is replaced by ".sea". * The total value of nitrate+nitrite only is listed. * Silicate and nitrate+nitrite are reported to the first decimal place only. * CTD temperature (including theta), CTD salinity and bottle salinity are all reported to the third decimal place only. * CTD temperature (including theta), CTD pressure and CTD salinity are all derived from upcast CTD burst data; CTD dissolved oxygen is derived from downcast 2 dbar-averaged data (see Appendix 2). * Raw CTD pressure values are not reported. * SAMPNO is equal to the rosette position of the Niskin bottle. A7.4 CONVERSION OF UNITS FOR DISSOLVED OXYGEN AND NUTRIENTS A7.4.1 Dissolved oxygen Niskin bottle data For the WOCE format files, all Niskin bottle dissolved oxygen concentration values have been converted from volumetric units (mol/l to gravimetric units (mol/kg, as follows. Concentration Ck in (mol/kg is given by Ck = 1000 Cl / (((,s,0) (eqn A7.1) where Cl is the concentration in (mol/l, 1000 is a conversion factor, and (((,s,0) is the potential density at zero pressure and at the potential temperature (, where potential temperature is given by ( = ((T,s,p) (eqn A7.2) for the in situ temperature T, salinity s and pressure p values at which the Niskin bottle was fired. Note that T, s and p are upcast CTD burst data averages (see Appendix 2, section A2.7.4). CTD data In the WOCE format files, CTD dissolved oxygen data are converted to (mol/kg by the same method as above, except that T, s and p in eqns A7.1 and A7.2 are CTD 2 dbar-averaged data. A7.4.2 Nutrients For the WOCE format files, all Niskin bottle nutrient concentration values have been converted from volumetric units (mol/l to gravimetric units (mol/kg using Ck = 1000 Cl / ((Tl,s,0) (eqn A7.3) where 1000 is a conversion factor, and ((Tl,s,0) is the water density in the hydrology laboratory at the laboratory temperature Tl and at zero pressure. Tl values used for each station are listed in Table 23 of the main text. Upcast CTD burst data averages are used for s. Table A7.1: Definition of quality flags for CTD data (after Table 3.11 in Joyce et al., 1991). These flags apply both to CTD data in the 2 dbar-averaged *.ctd files, and to upcast CTD burst data in the *.sea files. flag definition ---- --------------------------------- 1 not calibrated with water samples 2 acceptable measurement 3 questionable measurement 4 bad measurement 5 measurement not reported 6 interpolated value 7,8 these flags are not used 9 parameter not sampled Table A7.2: Definition of quality flags for Niskin bottles (i.e. parameter BTLNBR in *.sea files) (after Table 3.8 in Joyce et al., 1991). flag definition ---- -------------------------------------------------------------------- 1 this flag is not used 2 no problems noted 3 bottle leaking, as noted when rosette package returned on deck 4 bottle did not trip correctly 5 bottle leaking, as noted from data analysis 6 bottle not fired at correct depth, due to misfiring of rosette pylon 7,8 these flags are not usedinterpolated value 9 samples not drawn from this bottle Table A7.3: Definition of quality flags for water samples in *.sea files (after Table 3.9 in Joyce et al., 1991). flag definition ---- ------------------------ 1 this flag is not used 2 acceptable measurement 3 questionable measurement 4 bad measurement 5 measurement not reported 6,8 these flags are not used 9 parameter not sampled A7.5 STATION INFORMATION FILES * File format is as per section 2.2.2 of Joyce et al. (1991), and files are named as in Appendix 2, section A2.2.3, except that for WOCE format data the suffix ".sta" is replaced by ".sum". * All depths are calculated using a uniform speed of sound through the water column of 1498 ms-1. Reported depths are as measured from the water surface. Missing depths are due to interference of the ship's bow thrusters with the echo sounder signal, as described in Appendix 2, section A2.3. * An altimeter attached to the base of the rosette frame (approximately at the same vertical position as the CTD sensors) measures the elevation (or height above the bottom) in metres. The elevation value at each station is recorded manually from the CTD data stream display at the bottom of each CTD downcast. Motion of the ship due to waves can cause an error in these manually recorded values of up to (3 m. * Lineout (i.e. meter wheel readings of the CTD winch) were unavailable. REFERENCES Joyce, T., Corry, C. and Stalcup, M., 1991. Requirements for WOCE Hydrographic Programme Data Reporting. WHP Office Report WHPO 90-1, Revision 1, WOCE Report No. 67/91, Woods Hole Oceanographic Institution. 71 pp.