December 1998 Transect C Hydrolab CTD conductivity sensor problems were indicated by many salinity values that were considered too high (greater than 37ppt).
Seawater samples were collected at the same times and depths the Hydrolab values were recorded. The seawater samples were later measured in the laboratory with a PortaSal salinometer. The salinometer values were used to develop an equation to correct the Hydrolab Salinity values:
(Hydrolab Salinity - 1.5156)/1.0376 = Corrected Hydrolab Salinity
2180 pairs of Hydrolab spCond-Salinity values (considered good) which had been acquired on Transect C cruises in 1985 through 1997 were used to develop a multiple regression model of the form spCond =f(S). The result of this nonlinear regression exercise was a fourth order polynomial with a high correlation coefficient (R^2 = 0.99992).
December 1998 Transect C specific conductance was derived as follows. Derived spCond = -0.50203+1.92169*(CorrectedHydrolabSalinity)-0.02105*(CorrectedHydrolabSalinity)^2+0.000381377*(CorrectedHydrolabSalinity)^3-0.00000290903*(CorrectedHydrolabSalinity)^4
December 1998 Transect C Density sigma-t values were derived using methods described in Montgomery, R. B., and Wooster, Warren S., "Thermosteric Anomaly and the Analysis of Serial Oceanographic Data', Deep-Sea Research, Vol. 2, 1954 and values from Knudsen, Martin, Hydrographical tables, 1901. Formulas available upon request.
December 1998 Transect C Dissolved Oxygen and Percent Oxygen Saturation values logged by the Hydrolab were not corrected. Values reported for December 1998 Transect C Dissolved Oxygen and Percent Oxygen Saturation should be considered suspect.
H98Nutrients: Nutrient analyses were conducted using a Technicon (<http://www.techniconinstruments.com/>) autoanalyzer by Tom Oswald under the supervision of R. E. Turner.
H98Pigments: The Turner Designs model 10 was calibrated for chlorophyll a against a chemical supply house chlorophyll a standard measured on a spectrophotometer yearly. During cruises, the fluorometer was blanked and calibrated daily in accordance with Turner Designs recommended procedures. Pigment measurements were supervised and quality controlled by Nancy Rabalais.
H98PortaSal: The samples were analyzed in the lab by Guildline Instruments PortaSal, using Guildine methods (<http://www.guildline.ca/>). Salinity analyses were conducted by Jim Lee under the supervision of R. E. Turner.
H98Refract: The refractometer was checked with distilled water before each cruise and re-zeroed when necessary.
H98SeaBird: Sea-Bird SBE 13-01 dissolved oxygen sensors, the Paroscientific Digiquartz(r) pressure sensor, the SBE 3-01/F temperature sensors, SBE 5-01 pumps, and the SBE 4-01/0 Conductivity sensors were factory tested and calibrated at Sea-Bird (<http://www.seabird.com/>) recommended intervals and maintained and serviced by RV Pelican Electronic Technical support staff in accordance with Sea-Bird procedures. The Sea Tech 500m Fluorometer and Sea Tech 5.0 cm Transmissometer were factory tested and calibrated at Sea Tech recommended intervals and maintained and serviced by RV Pelican Electronic Technical support staff in accordance with Sea Tech procedures. Note: WET Labs acquired SEA Tech June 1, 1998, (<http://www.wetlabs.com/>). The Chelsea Fluorometer was also maintained by RV Pelican Electronic Technical support staff in accordance with Chelsea recommended procedures (<http://www.chelsea.co.uk/>).
Use of a more compact CTD carousel and extension was adopted beginning with the May 1998 cruise. The SeaBird CTD package was re-oriented from vertical to horizontal and plumbing was re-configured. The distance from the center of Niskin sample bottles and the pressure sensor decreased from 1.0 meter to 0.75 meters.
Before the start of the July shelfwide cruise, oxygen sensors were calibrated using the procedures described in SeaBird APPLICATION NOTE NO. 13-1, Rev. D. The Winkler Titration (OXFITW) oxygen value was determined chemically from replicate samples processed using a Mettler DL21 Titrator (<http://www.mt.com/>). Shipboard Winkler titrations during the cruise were used to develop a regression against CTD data in case it was necessary to correct the data. The July SeaBird data agreed with the Winkler values and no correction was necessary.
Winkler titrations and the OXFITW procedure were used to check the SeaBird oxygen sensors at the beginning of the September - December cruises.
Winkler titrations were conducted under the supervision of Nancy Rabalais. Data reductions from the SeaBird were generated by Ben Cole and quality controlled by Nancy Rabalais.
During monthly surveys conducted in other months, shipboard Winkler titrations for regressions were not made due to logistical constraints. Fewer stations were sampled and water columns were often well mixed, making the considerable effort involved in titrating an inefficient allocation of resources. A suitable range of data could not be acquired to develop a strong regression.
For the July SeaBird data, Winkler titrations were used to develop a regression which was used to correct the SeaBird oxygen data if necessary. The data for the SeaBird and the Hydrolab compare better for the shelfwide July cruise.
Jan SeaBird ~ 1mg/L lower than Hydrolab Feb SeaBird ~ 0.8mg/L lower than Hydrolab Mar SeaBird ~ 0.6-1.0mg/L lower than Hydrolab Apr SeaBird ~ 1.5-2.0mg/L too low, Note in log book not to trust May SeaBird ~ 0.5-1.0mg/L lower than Hydrolab except C6B higher than Hydrolab Jun SeaBird C9 no good, delete Jun SeaBird C6B a bit off Jun SeaBird C1 pump out, no cast Aug SeaBird is similar to Hydrolab Sep SeaBird is similar to Hydrolab Oct Acadiana Nov SeaBird is off from Hydrolab plus or minus 0.5 Dec SeaBird is similar to Hydrolab
Hydrolab data were derived from a pre- and post-calibrated instrument. Therefore, there is often disagreement between SeaBird and Hydrolab oxygen values. It is the opinion of the PI, Nancy Rabalais, that the Hydrolab data for the monthly cruises are the more reliable data.
H98Stn: Times and locations of operations were acquired from RV Pelican's Multiple Instrument Data Acquisition System (MIDAS) which was maintained by the ship's electronic staff. Navigational data were acquired using either a StarLink Invicta 210S Differential GPS System with dual output with Gyro interface (<http://www.starlinkdgps.com/>) or a Northstar 800x Differential GPS System (<http://www.northstarcmc.com>). During April and May 1998 while the Northstar was being repaired a Garmin GPS 45 with GBR 21 was used. Station depths were logged from the ship's Odom Echotrac II (<http://www.odomhydrographic.com/>) fathometer. The Secchi disk depths were measured by hand using standard protocol.
During the October cruise on the RV Acadiana, position data were logged from the ship's NorthStar 800 Loran C. Station depths were logged from Acadiana's fathometer.
H98Midas: Suspended sediment concentrations were supervised by Ben Cole or Nancy Rabalais and quality controlled by Nancy Rabalais. Times and locations of operations were acquired from the GPS component of the ship's MIDAS system (when available).
N. Rabalais performed an additional procedure in quality-control/quality-assurance by evaluating relationships between the data in H98Hydrolab, H98Midas, H98Nutrients, H98Pigments, H98PortaSal, H98Refract, and H98Stn.
Table H98SeaBird percent light transmission (Xmiss) values exceeded 100 percent at six stations: 19980414 station C9, 19980512 stations C9 and C6B, 19980610 station C6BN, C9N and C9.
SeaBird Dissolved oxygen (OxMg/L) and percent oxygen saturation data acquired 19980610 at station C9 were deleted because the data were considered unreliable.
During the July shelfwide 1998 survey the SeaBird CTD package was configured with a single fluorometer. No Chelsea Fluorometer (FlC) data were collected during the July cruise.
Suspended sediment values (SuspSed) that did not meet quality control criteria were deleted.
The navigation component of the ship's MIDAS system was not always functional. thirty-two suspended sediment latitude and longitude values (SS-Lat and SS-Lon) are missing from table H98Midas. The station latitude and longitude (Lat and Lon) values reported in table H98stn were within 0.3 nautical miles of the suspended sediment sample location.
Station positions acquired during the October cruise on RV Acadiana were logged from the ship's NorthStar 800 LORANC. The manufacturer describes the accuracy to be within 30 meters 90% of the time.
On the following dates hydrocasts were made from a small boat tied to the oilfield platform at station C6B: 19980402, 19980420, 19980619, 19980716, 19980811 and 19981012.
In order to minimize the effect of delays in oxygen sensor response time caused by temperature, sensor condition and plumbing configuration, the CTD package was lowered as close to dead slow as possible. The sensor packages were located below the Niskin bottles and rosette.
At stations where the watch chief deemed the structure of the oxygen profile contained features useful in post-processing the oxygen data (AlignCTD), the CTD package was raised at the same speed it was lowered. At all other stations, the upcast is rapid in order to save time.
The Hydrolab CTD was attached by chain to a lead weight. The weight was lowered to the bottom by hydrowire. With the weight on the bottom, the Hydrolab sonde was positioned just above the bottom. When the oxygen sensor stabilized, a data record of all the sensor values was stored electronically. The sonde was raised in approximately 0.5-meter increments, after D.O. sensor stabilization, data records were stored. After storing data for the few meters closest to the bottom, the Hydrolab was raised to two to three meters from the surface and a data record was saved. The sonde was raised, and records stored, in approximately 0.5-meter increments until finally a record was stored with the sonde submerged but as close as possible to the surface.
During the July shelfwide cruise, stations were occupied along 13 generally North-South transects across the Louisiana coastal shelf. Station depths ranged from 5.3 to 68.3 meters. The objective was to delimit and describe the area of midsummer bottom dissolved oxygen less than 2 (mg/L). Due to limitations of equipment, time and funding, this was not always possible. Northern end stations of transects were chosen based on the survey vessel's minimum depth limits for each longitude. The northern extent of hypoxia was not reached on four transects: A', A, C and F. Optimally, one or two stations south of the southern most station with bottom dissolved oxygen less than 2 (mg/L) were sampled. The limit of hypoxia on the southern ends of all transects was reached. The limit of hypoxia was reached at all stations on the western end of the survey. The limit of hypoxia was only reached at station A'6 on the eastern end of the survey. Bottom waters at stations A'1, A'2, A'3 A'4 and A'5 were hypoxic.
Depth values of 0 indicate a bucket sample collected from the surface of the water. Deepest depths of water samples were from the bottom-tripping Niskin and correspond to the deepest depth recorded from the Hydrolab. Other depths indicate the p-sensor reading for 5-l Niskin bottles on the Seabird rosette. The Seabird p-sensor was located approximately 1 meter below the mid-point of the 5-l Niskin.
Water for Suspended Sediment samples was collected by tapping into discharge of the shipâ€(tm)s seawater flow through system MIDAS. MIDAS intake was 2 -3 meters below the surface. The watch chief observed MIDAS navigational data, when distance to station was 0.3 miles, a triple-rinsed jar (1000ml) was collected from the silicon MIDAS discharge tube in the ship's bottle lab. Finally, collection time was logged.
When Seabird data were post-processed, dissolved oxygen was advanced relative to temperature and conductivity values as follows: January +3.75 seconds; February, March and August +4.0 seconds; July +4.055 seconds; April, September, November and December +4.25 seconds; May 5.5 seconds; and June +6 seconds.
Data values reported are from downcasts. Downcast scans selected for each CTD station were chosen to illustrate: 1) data values as near to the surface as possible, 2) data values at whole meter increments, and 3) data values as close to the bottom as the CTD was lowered. In certain cases where data values of a parameter changed significantly between whole meter increments, 0.1- or 0.2-meter scans were selected.