Advanced Sampling Technologies Research Group
Introduction to Operational Methodology
The Northeast Fisheries Science Center's (NEFSC) Advanced Sampling Technologies Research Group was established to implement state-of-the-art acoustic and optic technologies, in conjunction with midwater trawling, to improve fisheries-independent stock assessments for pelagic species. Pilot and experimental studies were conducted during 1995-97 to develop the NEFSC's fisheries acoustic capabilities. Annual NEFSC Atlantic Herring Hydroacoustic Surveys began in 1998. NEFSC acoustic capabilities include echo-integration, omni-directional sonar, Acoustic Doppler Current Profiler (ADCP), and trawl monitoring systems (Figure 2.0.1). Midwater trawling, plankton sampling, and underwater video operations provide biological samples and target verification. CTD (conductivity-temperature-depth) vertical profiles are also obtained during survey operations.
Figure 2.0.1. The Northeast Fisheries Science Center's fisheries acoustic survey operations aboard the NOAA FR/V Delaware II.
The Advanced Fisheries Tow Vehicle (AFTV) is a horizontally stable towfish platform with Ethernet-based sensor integration which is used to implement advanced sampling technologies in support of NMFS strategic missions (SAIP, EFH, ICOM). The AFTV system consists of a DTMarine Portable Winch Model No. 1015-EHLWR, 2,000 meters of 0.322 inch armored fiberoptic cable, an AFTV control panel at the surface which provides power to the tow vehicle, and the AFTV towbody itself.
The winch is a self contained unit consisting of a 15 HP, 230/460 VAC, 3 Phase, 60 Hz motor which drives a hydraulic pump. The gross weight of the winch is 5,000 pounds with the 2,000 meters of armored cable. The fiberoptic cable is manufactured by Rochester Corportation, and consists of 3 copper conducters (20 AWG) rated at 1,200 Volts with 8.3/125/245 µm single-mode optical fiber embedded in each conductor.
The AFTV control panel was designed and developed by Deep Sea Systems, Inc. This control panel converts the ship's 240 VAC to 1,100 VAC which is then sent through the 2,000 meters of cable to the AFTV tow vehicle. The 1,100 VAC is applied to a step down transformer in the towbody which provides 115 VAC to power the vehicle.
The AFTV tow vehicle, also designed by Deep Sea Systems, Inc., was delivered to NEFSC in August, 2006. The towbody consists of an 2,000 meter rated electronics enclosure which houses all of the electronics that interface the sensors with the surface. Communications within the electronics housing, and to the surface as well, consists of a 10/100-BaseT local area network (LAN). Currently, the AFTV is fitted with a SIMRAD deepwater EK60 38kHz transducer, a SIMRAD EK60 38kHz general purpose transceiver (GPT), a MicroCAT SBE 37-SMP CTD profiler, two DeepSea Power and Light SSC-5000 Super Sea Cameras for stereo video applications, a Remote Ocean Systems PT-10 pan and tilt unit, and two OceanTools OceanLED Underwater lights with variable intensity. In September/October 2008, the Didson High-Definition Sonar was deployed on the AFTV towbody to test the DIDSONs capabilities.
Deployment of the AFTV tow vehicle is performed off the aft end of the research vessel. A 2-point deployment is required. First, the AFTV vehicle is lowered into the water using the research vessel's crane. Once the vehicle is in the water, a 100 pound weight is attached to the armored cable approximately 30 meters from the AFTV towbody. This weight decouples the affects of the ship's heave and roll from the AFTV tow vehicle, thus providing a stable sensor platform.
The Simrad EK500 Scientific Echo sounder is designed for fisheries acoustic research and survey operations. The EK500 provides low self-noise, high transmit power, instantaneous dynamic range of 160 dB, unlimited range compensation (TVG), efficient transducers, and a wide frequency range (12 to 200 kHz) for scientific-grade detection capabilities that can be applied to fisheries research and assessment.
The EK500 (v. 5.30) is the primary scientific echo sounder used during NEFSC cruises since 1995 for conducting fisheries acoustic surveys and research. The EK500 was installed with hull-mounted 38 and 120 kHz split-beam transducers aboard the FR/V Albatross in 1996 and a third hull-mounted transducer (single-beam 12 kHz) was installed on the FR/V Albatross in 1997. In January 2002, the 12 kHz single-beam transducer was replaced with an 18 kHz split-beam transducer. From 1995-2000, EK500 data were logged to a SUN Sparc workstation with the BI500 software (Simrad, Inc.) which logged, processed, and archived EK500 data as binary files and in the UNIX-based INGRES relational database. SonarData software replaced the BI500 software for EK500 data logging and post-processing during 2001, after preliminary comparisons between the two packages were conducted during 1999-2000. EK500 data were logged simultaneously to the SonarData and BI500 software during the 1999-2000. Beginning with the 2001 Atlantic Herring Hydroacoustic Survey, EK500 data were logged solely to the SonarData personal computer post-processor.
EK500 operations during the NEFSC cruises involve continuous data logging from the EK500's three frequencies (12 or 18, 38, and 120 kHz), along the cruise track. Vessel speed is maintained at a consistent 10 2 knots during transect operations and is variable during other operations. Data telegrams outputs by the EK500 are: depth, navigation, vessel log, echo integration echograms, and echo traces (individual target information). The EK500 echo-integration echograms consist of volume backscatter (Sv, dB) values at a vertical resolution of 500 data points. During 1997-1999, echo integration data were collected to 250 m (0.5 m resolution) and during 2000-2001, data were collected to 500 m (1 m resolution). Volume backscatter estimates were vertically integrated to obtain areal backscatter (sA in units of m2/nm2) as a relative index of abundance along the cruise track. Individual target strength (TS) measurements were also collected by the EK500.
Routine calibrations for the 12, 18, 38, and 120 kHz on the FR/V Delaware and the 38 and 120 kHz on the FR/V Albatross have been conducted to ensure proper operation of the EK500 system and high precision of the acoustical measurements. The EK500 test/transceiver menu is also routinely monitored before, during and at the end of each cruise to determine if there were system problems (i.e., transceiver card, transducer connections).
The EK500 is calibrated using the standard sphere calibration procedure (Foote et al. 1987). For each frequency, a calibration sphere of known target strength is suspended under each transducer. Each sphere is moved throughout the acoustic beam using three remotely controlled downriggers positioned around the vessel. EK500 calibrations have been conducted at sea while drifting during calm weather conditions. The 38 and 120 kHz transducers have also been calibrated alongside the Woods Hole Oceanographic Institution's pier where bottom depth is about 21 m. The Simrad Lobe program is used during calibration of the 38 and 120 kHz split-beam transducers to obtain TS and Sv gain settings and angle offset parameters. 45, 63, 60, and 23 mm copper spheres are used to calibrate the 12, 18, 38, and 120 kHz systems respectively.
EK500 data acquisition and post-processing was accomplished using the Simrad BI500 Bergen Integrator software until 2001 on a Sun workstation and then using SonarData's EchoView software on a PC. Scientists post-process the acoustic data in near real time to review the EK500 data. Unwanted noise is filtered, a Sv threshold set, bottom detection lines are modified to remove bottom echoes from the water column, and volume backscatter is partitioned to species.
The FR/V Delaware's Furuno CSH-5 omni-directional sonar (user selectable frequencies: 55 or 64 kHz) is used intermittently during survey operations for locating fish aggregations and documenting the horizontal spatial patterns along the transects. The CSH-5 sonar simultaneously scans a full 360 degrees with a cone-shaped beam. The acoustic beam can be tilted at various angles from the surface to eliminate surface noise. The vertical width of the receiving beam results in a horizontal search radius of 800 m in waters with bottom depths of about 200 m. The search radius on the display is typically set to 400 m during survey operations, but can be adjusted to maintain bottom on the display to locate fish schools near bottom. A major disadvantage with the Furuno CSH-5 is the lack of digital output. A video capture board was installed to capture the sonar analog images and these graphic images have been stored at a rate of every 30 seconds. The sonar images can be merged with navigational data and archived. To date, these omni-directional sonar data have not been processed for quantitative analysis because of the difficulty with working with analog output and the limitations of using side-looking sonar for quantitative estimates (e.g., high variability from lateral aspect of fish). The omni-directional sonar, however, has provided useful operational and biological information when conducting field experiments. For example, the omni-directional sonar reduces search time when locating aggregations and shows horizontal movements and spatial structure of schooling fish during in-situ experiments. Femto Sonar Logging and Post-processing Software is available to for editing these archived sonar images for future analysis.
The minimum requirements for the Northeast Fisheries Science Center's (NEFSC) pelagic trawling operations were a trawl that could be readily deployed from the FR/V Delaware, minimal drag for high sustainable tow speeds (up to 5 knots) by 1200 hp vessels such as the FR/V Delaware, net mouth opening of no less than 300 m2 to minimize net avoidance, and dependable technical support and mending supplies from the vender. The requirement for fishing the pelagic trawl at high speeds was in anticipation that pelagic surveys would be targeting fast swimming fish and squid in the near future. A High Speed Midwater Rope Trawl (Gourock HSMRT design R202825A) was selected as the preferred pelagic trawl design for the Northeast Fisheries Science Center's fisheries acoustic surveys.
The HSMRT is the primary sampling gear used to verify fish backscatter identified by the EK500. The HSMRT is a commercial midwater trawl designed for high tow speeds with minimal drag. Its symmetrical four-seam box design had 53.1 m (174') footrope, headrope, and breastlines. The mouth opening of the HSMRT is approximately 13 3 m vertical and 27 5 m horizontal. The trawl is rigged with four 54.8 m (179'9") bridles to 1.8 m2 US Jet double-foiled Suberkrub-type doors with double door weights. Optimum tow configuration was determined from trawl performance tests conducted during 1997-1998. For each side, the total (forward and aft sections) setback was adjusted to 2.5 m (8') with 273 kg (600 lb) tom-weights (refer to Cruise Results DE9809 for further details).
A series of trawl performance studies were conducted during 1994 - 1998 to test and evaluate various types of pelagic trawls. The objective of these studies was to identify a pelagic trawl that could be used for routine pelagic fish and squid surveys along the Northwest Atlantic continental shelf region. Four-bridle and six-bridle semi-pelagic Shuman trawls were tested during 1994-95, and two Gourock 42350S Quantum trawls were tested during 1997. Minimized drag for high tow speeds was the most difficult requirement to satisfy. The performance of the High Speed Midwater Rope Trawl (HSMRT) was optimized during a 1998 study by testing various door attachments, tomweights, and setback adjustments at various speeds (Fig. 2.5.1).
HSRMT trawl deployments are targeted on selected aggregations observed in the EK500 echograms. The duration of a HSMRT tow generally varies from 5 to 60 minutes. Tow duration is set by the watch chief who monitors acoustical targets going into the trawl from the FS903 display. Tow duration is defined as the time between setting the doors and when the doors are hauled out of the water. Trawl duration, tow depths, and tow speeds were not standardized or consistent between trawls and catch data have not been used for separate abundance estimates. We are currently utilizing trawl catch information to verify acoustic backscatter, thus midwater trawling is often targeted at specific aggregations or layers in the water column observed from the EK500 and FS903 display. The HSMRT was towed at speeds ranging from 4.0 - 4.5 knots. The trawl is often fished obliquely by incrementally increasing the wire-out until the trawl is close to the bottom (no closer than 5-10 m from the bottom).
Trawl performance and mensuration is monitored using Simrad FS903 and ITI trawl monitoring systems. The Simrad FS903 Trawl Monitoring System is a third-wire device that provides real-time sonar images of the trawl opening and performance. The Simrad ITI wireless trawl sensors provide point measurements of the trawl depth, horizontal and vertical opening, and door spread. Vemco depth-temperature probes are also attached to the trawl headrope and footrope to provide continuous depth-temperature profile data for each deployment. These are small (~2x6 cm) probes that record temperature and depth at regular time intervals. A sampling rate of 2 to 10 seconds has been our standard. The FS903 provides more reliable response time than the Simrad ITI system, while the ITI provides supplemental trawl measurements. The FS903 third-wire system is also used during trawl deployments to keep the headrope sensors out of the trawl meshes.
Trawl mensuration data (trawl depth, vertical and horizontal mouth opening, door spread, wing spread, and in situ temperature) are recorded onto log sheets, then and entered into computer forms for auditing and archive. In the future, these data will be imported to an Oracle database table at the NEFSC.
A Static Underwater Stereo Video System (SUSVS) was designed for a 1997 field study to directly verify the in-situ acoustic measurements of Atlantic herring in the Gulf of Maine. The SUSVS was deployed mid-ship while the vessel drifted over fish aggregations. A pair of underwater video cameras (DSPL Micro-SeaCam models 1050 and 2050) was mounted on the SUSVS to obtain stereo imagery of fish schools. These video cameras had a low light (0.05 lux) auto-adjusting iris with a 770 horizontal and 590 vertical view fields. DSPL SeaLasers 110-15 were mounted in parallel (5.4 cm off center) for measuring target size. Two DSPL Multi-SeaLites provided illumination that were dimmed remotely using an isolated 0-140V variable transformer. Real-time depth, temperature, compass bearing, and three-dimensional orientation was obtained every 10 seconds from a JASCO Underwater Attitude Measurement Sensor (model UWINSTRU, rated to 200 m). The dual video and environmental data were transmitted in real-time through a 100 m multi-conductor cable to a PC computer and SVHS video recorders. Each video frame was time-stamped with a time-code generator (Horita TG-50 and WG-50) to synchronize the stereo video recordings. The SUSVS video operations were upgraded during 1998-2000 with a portable video winch, 310 m multi-conductor dual-video cable, and CCD cameras with improved low-light and depth capabilities.
A Multi-purpose Underwater Video System (MUVS) was designed in 2001 to provide a more adaptable platform for equipment upgrades and additional underwater video requirements of the NEFSC. The MUV tow-body has adjustable mounting brackets and tow-points that enable various configurations for observing acoustical backscatter, habitat assessment, fish and invertebrate behavior, and gear selectivity and performance. The MUVS was intermittently deployed to directly verify backscatter and obtain behavioral observations during the 2001 Atlantic Herring Hydroacoustic Survey. A pair of low light CCD video cameras (DSPL Super SeaCam SSC-5000-T) was mounted to obtain stereo imagery of herring schools. These CCD video cameras can operate at low light sensitivity (0.001 lux @ f0.80) with 570 lines horizontal resolution. The view field in water is 117o (D) x 97o (H) x 77o (V) and working distance of 60 cm to infinity. The titanium housing allows for a 6000 m depth rating. DSPL Multi-SeaLites dimmed by an isolated 0-140V variable regulator provides illumination. A JASCO Underwater Attitude Sensor (UWINSTRU-D, rated to 1500 m) provides real-time depth, temperature, compass bearing, and pitch-roll of the MUVS platform during deployments. The three dimensional orientation of the observed herring and other biological targets can be determined from the JASCO data. During the 2001 survey, the MUVS was deployed mid-ship (alongside the EK500's acoustical beam) while the vessel drifted over selected backscattering aggregations.
Underwater video operations were intermittently successful during the 1997-2000 surveys due to avoidance by the white (unfiltered) lights of the video system. This was apparent by the disappearance of backscatter that often occurred when lights were turned on or brightened. During the 2001 Atlantic Herring Hydroacoustic Survey, various light filters were tested in an effort to reduce the avoidance of herring and other biological organisms to the MUVS's lights. Stainless steel collars were installed on a pair of DSPL Multi-SeaLites for easy placement of the light filters. Infrared, red, green, and blue filters were tested, and herring appeared to react less to the green filters. Although herring are believed to not react to infrared and red wavelengths, these light sources resulted in limited visual range during underwater video observations. Greater light penetration can be achieved with blue and green light, although herring are known to detect the blue-green wavelengths. Presently, we can only speculate from a few in-situ video observations that the green filtered light appeared less intrusive to herring behavior in comparison to the unfiltered white and blue light. We did not have time to test an ultraviolet filter. Although fish are known to not react to ultraviolet light, ultraviolet light tends to emphasize particulate material resulting in less range in turbid waters. Further tests in the laboratory and field will be conducted to determine the optimal wavelengths to reduce in-situ avoidance reactions of herring during video operations.
The Scientific Computer System (SCS) has been used throughout the 1995-2001 cruises to continuously collect navigational, oceanographic, and meteorological data. The SCS system continuously samples data streams from shipboard instrumentation at 2-second intervals throughout the cruise. Approximately 150 data variables are collected, including date and time (GMT), multiple latitude and longitude positions (PCODE, differential, LORAN), water and air temperatures, salinity, fluorometry, wind speed, pitch and roll, and bottom depths and vessel log values from the EK500.
An electronic event log is maintained to link EK500 data with other scientific operations. The event log is routinely used throughout all cruises to document chronological events of the acoustic sampling, deployments, and other operational details that are important for data processing and management. The implementation of the event log has been upgraded each year from using hand-written logsheets to various versions of the SCS Event Log program from 1999 to present. The generic SCS event logger was developed by a team of scientists and programmers at NOAA headquarters and NEFSC, and can be customized for specific applications. We presently use a SCS Event Log window that was developed for the NEFSC Atlantic Herring Aacoustic Surveys which closely links acoustical, SCS, and deployment (e.g., trawl, video, CTD) data by time and the EK500 vessel log. This event log was constructed to register the start and stop of transects, gear deployments, sites and transect series and associated data such as date-time stamps, geographic location, EK500 vessel logs, and comments during survey operations. The event log was hand-written and entered at sea into a computer spreadsheet during the 1997-98 cruises, and directly into the SCS event log program from 1999 to present. The event log has served as an important data management tool for linking EK500 data with deployment (e.g., trawling, video, CTD) and SCS data. All computers, instrumentation, acoustic data collection, and data recording were synchronized to the SCS master clock.
The Fisheries Scientific Computer System (FSCS) was implemented in 2001 to improve on-board entry of catch data from trawl operations. The Window's-based menu system provides effective and efficient recording of biological data with near-real-time auditing routines. Biological catch data are audited on shore and linked to the acoustic and event log data in an Oracle database.
Trawl catches are sorted by species, and the total weight of each species catch and up to 100 individual lengths are measured (Fork Length (FL) to the nearest cm) according to standard NEFSC procedures. For Atlantic herring, approximately 150 Atlantic herring are processed for individual lengths (FL in mm) and weights (to nearest 0.1 g). In addition, detailed biological sampling (sex, maturity, and stomach contents) of herring is taken for one fish per centimeter interval below 25 cm and three fish per centimeter interval above 25 cm. We collect the additional herring above 25 cm to increase the sample size of older fish. During 1998-99 cruises, fork lengths (FL) and total lengths (TL) were recorded from Atlantic herring to provide FL-TL conversions. The FL-TL conversion was needed when comparing our FL data with the acoustic literature that typically refers to TL in regard to relationships between acoustic measurements and fish size.
Age data for Atlantic herring are obtained from otolith (sagittal) samples from the Atlantic Herring Acoustic Surveys, and from spring and autumn Bottom Trawl Surveys. From each trawl station, a length stratified selection of herring samples (one from each 1 cm length interval per station) is frozen whole, and their otoliths are removed in the laboratory. The herring otoliths are mounted in black resin and their annuli read with a binocular scope (Fig 2.11.1). Atlantic herring in the Gulf of Maine and Georges Bank region spawn primarily during the autumn, resulting in the formation of the otolith's nucleus occurring during period of slowest growth in the fall and winter. The outer edge of this first hyaline zone is interpreted as the first annulus. The age is determined from the annuli from the outer (less opaque) edge of each hyaline zone. False annuli (check) occurrs as thin or discontinuous hyaline zones due to anomalies in growth patterns. Age validation is accomplished using back-calculation and length-frequency analyses. Recent herring growth appear to be slower when compared to the 1980's when herring biomass was lower, possibly due to density-dependent effects.
A Conductivity-Temperature-Depth (CTD) Profiler is deployed at the beginning and ending of each transect, at the beginning of each deployment, and at selected locations to define the hydrographic conditions. Water bottle casts are also done to collect salinity samples for CTD calibration.