3.2.1.1 Aanderaa RCM-7/8 Current Meters The Aanderaa current meters were calibrated at E.S.I. Environmental Sensors, Inc. in Victoria, British Columbia (Canada) in July 1999 prior to initial deployment, and Aanderaa Instruments A/S in Norway in May/September 2002 following final recovery. Temperature calibration stability over the study period was within specifications (+- 0.05°C) for all but four of the instruments. Three of the four fell just outside the rated accuracy with differences on the order of +-0.07°C when compared at the high end of their measurement range (near 28°C), and the fourth instrument had drifted by as much as 0.11°C (near 22°C). One temperature sensor probe was bent and was replaced as a precaution against future leaks, and one instrument had flooded and so could not be re- calibrated. Conductivity was less stable than temperature as only two instruments were still within specifications (+-0.074 mmho/cm) at the end of the measurement program. Two instruments fell just outside the rated accuracy with deviations of 0.092 and 0.111 mmho/cm, respectively. Five sensors had deviations ranging from 0.158 to 0.237 mmho/cm, and one sensor had drifted by as much as 0.658 mmho/cm. Two sensors had failed and were replaced. A compass calibration check revealed that all but one of the compasses were operational and within specifications (+-5 Degrees) at the conclusion of the field study. The one exception was a unit that exhibited some temperature sensitivity (~7.5 degrees directional drift) over the full temperature range. Speed checks revealed that two instruments had unstable speed (some rotor turns were not counted from time to time during the tests) requiring replacement of the follower magnet, and that a third instrument had a defective rotor counter switch which produced only zero readings. This latter instrument had been identified as having had a rotor counting switch problem during the last deployment period. Data from the two instruments with follower magnet problems were reviewed and there is no obvious sign of a problem reflected in the speed data. 3.2.1.2 General Oceanics MK2 Current Meters The General Oceanics current meters were calibrated at the manufacturer’s facilities in August and October 1999. Some of these (SNs 311, 450 and 452) were used only for one deployment and were not recalibrated at the end of the program. SN 335 was calibrated in August 1999 and collected data for two deployments on the J1 mooring. Two instruments (SNs 453 and 457) were used extensively during the field study and were recalibrated in May 2002. The results of this calibration reveal that the instruments remained within specifications (+-0.25ºC, +- 1 cm/s and +- 2 degrees magnetic). 3.2.1.3 Hugrún Seamon Mini Temperature Recorders Twelve (12) Hugrún temperature recorders were calibrated at the manufacturer’s facilities in July 1999 and again at Star-Oddi in December 2002. A comparison of data processed with both the old and new calibration coefficients revealed that there was no significant change/drift in calibration over the study period. All comparisons were within each instrument’s rated accuracy (+- 0.1ºC) and varied by no more than +-0.025ºC. 3.2.1.4 InterOcean S4 Current Meters Four (4) InterOcean S4 current meters were deployed during the program. Each was equipped to measure temperature and conductivity as well as speed and direction. These instruments were calibrated at the manufacturer’s facilities in June 1999 or August 2000 prior to being deployed, and again in August 2002 after the field effort was completed. All four instruments remained within specifications for their compass and Vn readings, the Ve readings were off by less than 0.50 cm/s. Three of four temperature sensors were within or were close to sensor specification (+-0.05°C), but only one conductivity sensor remained within specification (+-0.2 mS/cm). The largest temperature error was +0.75°C and the largest conductivity error was -2.32 mS/cm. 3.2.1.4 RD Instruments ADCPs The LongRanger 75 KHz ADCP was purchased in June 1999 and first deployed in August 1999. The instrument was returned to the manufacturer in July 2002 and was serviced as part of a checkout service. The receiver board was found to have an intermittent failure in a test mode (PT5) but it was determined that this was not a problem for proper operation when deployed. It only affected testing. The manufacturer replaced the receiver board under warranty. No other instrument problems were noted and the instrument was considered to be in proper calibration (+-0.4ºC, +-5 degrees magnetic, and +- 1% of measured velocity +- 0.5 cm/s). The instrument was set up to produce a +- 1.6 cm/s standard deviation during data collection. The NarrowBand 150 KHz ADCP was serviced at the manufacturer’s facilities in June 1999 and again in August 2000. The August 2000 servicing followed the first deployment period. At this time, repairs were made to the power regulator and the backup regulator circuit. The instrument had consumed excessive sleep current, which shortened battery life, but had no impact on data quality. No other instrument problems were noted and the instrument was considered to be in proper calibration (+- 0.2ºC, +-5 degrees magnetic, and +-0.2% of measured velocity +- 0.5 cm/s). The instrument was set up to produce a +-1.8 cm/s standard deviation during data collection. Four (4) WorkHorse 300 KHz ADCPs were serviced at the manufacturer’s facilities in June 1999, and four (4) additional instruments were serviced in November 1999. Only one of these instruments (SN 214) was serviced again following deployment. This particular instrument had a manufacturing defect, a dent in the urethane of Beam 1 (caused by bubbles in the urethane). This defect caused shorting out of the beam and resulted in reduced battery life during the first deployment period. No other instrument problems were noted and the instrument was considered to be in proper calibration (+- 0.4ºC, +-5 degrees magnetic, and +-0.5% of measured velocity +-0.5 cm/s). The instrument was set up to produce a +-0.6 cm/s standard deviation during data collection. It is noted that ADCP profiling ranges vary depending on instrument frequency and power, water temperature, the nearness of an interface (the surface or bottom) and the availability of particles in the water column. As a consequence, it is not unusual to see the range of an ADCP vary significantly, even over short periods of time. In order to avoid the proliferation of short, partial-deployment-period current records, a minimum- maximum profiling range was identified for each ADCP deployment level. This minimum-maximum profiling range, from a given instrument-deployment level, was defined as the smallest, repeatable maximum range (from the transducer face) common to all instruments deployed at that level for all ensembles for the entirety of the study period (two years). In addition to the profiling range limits noted above, some beam interference was observed in the immediate vicinity of steel flotation elements on the I1 mooring for the two upward looking WorkHorse ADCPs. These instruments were deployed in inline frames at 90 meters depth (20 meters below a 41" diameter steel buoy) and 240 meters depth (40 meters below a 48" diameter steel buoy). In each case, the effect was only on data from the one or two 4-meter bins that included the depth of the flotation element. This interference is believed to be an artifact of the beam side lobe reflecting off a large, hard, slow-moving particle (the buoy) and being reflected back to the ADCP. The interference tends to bias the data from the affected level towards lower speeds and was most noticeable when the mooring was experiencing high-speed events in the upper layer. The affected bin levels were removed from the data record during processing. There was no obvious indication of ADCP to ADCP interference in the data, either for ADCPs of the same or different frequencies even when sampling the same part of the water column (see below). This was accomplished, in part, by programming a 30-second lag in the sampling sequence between adjacent WorkHorse ADCPs of the same frequency, and alternating their orientation (up or down). 3.2.1.5 Sea-Bird CT Recorders Four instruments (two each SeaCats and Microcats) were calibrated by the manufacturer in May/June 1999 prior to initial deployment, and again in June/July 2002 following final recovery. The conductivity cells were protected with anti-fouling cylinders that were replaced every 6 to 12 months. These cylinders appear to have worked well as there was no apparent build up of biological material inside the conductivity cells. Post-program calibrations reveal that the temperature drift on each of the four instruments was less than 0.001ºC. The maximum conductivity drift was less than 0.025 mS/cm for three instruments, but was 0.200 mS/cm for a fourth (MicroCat SN 0059). These values correspond to calculated salinity drifts on the order of 0.026 psu and 0.201 psu, respectively.