A Description Of A Discrete Depth Plankton Sampler With Some Notes

Transcript

A DESCRIPTION OF A DISCRETE DEPTH PLANKTON SAMPLER WITH SOME NOTES ON THE TOWING BEHAVIOR OF A B-FOOT ISAACS-KIDD MID-WATER TRAWL AND A ONE-METER RING NET William Aron, Newell Raxter, Roy Noel, and William Andrews GM Defense Research Laboratories, General Motors Corporation, Santa Barbara, California ABSTBACI’ This paper describes an electrically operated sampling system developed for capturing pelagic animals at a known depth while also providing a continuous shipboard record of the depth of the net and the temperature of the water being sampled. It includes data on the towing behavior of a 6-ft Isaacs-Kidd mid-water trawl and a l-m ring net that were obtained with the new sampIer. INTRODUCI’ION THE DISCRETE DEPTH PLANKTON SYSTEM SAMPLER The need for adequate sampling equipment for studying the distribution of The electrically operated discrete depth plankton was considered in the 1961 Interplankton sampler ( DDPS ) system ( schenational Council for the Exploration of the matically illustrated in Fig, 1) evolved Sea, Symposium on Zooplankton Producfrom an acoustical-mechanical system that tion, particularly in the contributions by could be used without the necessity of Aron ( 1962), Currie ( 1962)) Laevastu electrical cable or winch modifications. ( 1962), Motoda ( 1962)) and Yentsch, One such unit (the Mark I) was conGrice, and Hart (1962). These authors structed and field tested. Ship noise, parreviewed existing techniques and generticularly at towing speeds of over 6 knots ally agreed that there was still considcra( 11 km/hr ), made detection of the acousble room for improvement of the existing tic signals difficult, and the reliability of gear, especially in regard to reliability, the simple pressure switch over long periease and speed of operation, and the abilods was questionable. The precision of ity to provide simultaneous environmental the depth determination was also, doubtdata. ful. None of these difficulties was insurThe above inadequacies are particularly however, it was considered limiting to detailed studies of the deep mountable; scattering layer. The equipment described that their solutions would not merit the in this paper allows the capture of animals Because of additional effort involved. from known depths and temperatures. It these problems, the acoustical-mechanical was developed by the GM Defense Re- system was abandoned in favor of an elccsearch Laboratories in support of their bitrically controlled system that also permitological and acoustic studies off the Calited the addition of electronic depth and fornia coast. temperature sensors. The paper also considers some of the The DDPS system consists of four comresults obtained with the new sampler, ponents : 1) a cod-end sampler, usable particularly in regard to the towing charwith various nets; 2) underwater electronic acteristics of a 6-ft Isaacs-Kidd mid-water devices that include the depth and temperand a l-m ring net. It is trawl (IKMT) ature sensors and the stepping switch for shown that changes in towing speeds prooperating the cod-end sampler; 3) single duce a significant change in the sampling depth and that previous assumptions re- conductor towing cable; 4) the shipboard garding the towpath of a net during sam- gear, which includes the controls, readout equipment, and winch with slip rings. pling may be in considerable error. 324 325 FIG. 3a. The Mark III cod-end sampler. diffrrcnt from 1%cm sampler.) (Dimensions in parentheses are for 15.cm sampler when three configurations have been extensively and successfully used. In using the Mark II sampler, however, several problems were encountered. The filtering area provided in the walls of the tube appeared inadequate and restricted water flow. When the tandem arrangement was employed, this resulted in contamination of the forward chambers of the sampler with animals that were caught during the first part of the sampling period hut were not washed into the last of the three chambers. Also, when the samplers were mounted in tandem, it became necessary to disassemble the unit to remove the catch from each of the sections (in the Mark II the catch was removed through the doors at the ends of the sampler). It was also suggested that the 10. cm diameter of the sampler was insufficient if the equipment were to he used with larger sampling gear, such as a lo-ft IKMT, or used for long tows in areas of high plankton concentrations. The sampler was redesigned to provide 8 greater filtration area (preliminary evidence indicates that the increased filtration area has almost completely eliminated the contamination problem mentioned above) and side-opening panels for sample removal. Two models (the Mark III) were built; one 10 cm in diameter (the same as the original), the other 15 cm in diameter. These units have four sets of doors and take four samples during each tow-three from depth and a fourth from the last sampling depth to the surface (Figs. 3a and 3b). The total length of the new unit is about the same as of the earlier device when used in the tandem arrangement. A feature that has been retained in the new models is the “dead space,” created by allowing an unscreened distance equal to at least one-half the diameter of the tube between the after edge of the filtering screw and the doors. Because of this space, the condition of animals taken by the sampler, even during tows at 8.5 knots (16 km/hr), has been remarkably good as compared with catches obtained by cod-end buckets with terminal straining areas. The underwater electronic equipment 1s enclosed in a pressure housing and mounted in the spreader bar of the IKMT (Fig. 4). When used with a l-m ring net, the spreader bar is attached as a link between the cable termination and the towing point of the net. The underwater unit includes the selector switch for operating the doors of the cod-end sampler and the temperature and depth sensors. Temperature and depth are obtained with a modified electionic bathythermom- eter (Hytech Corporation Model 472). A new pressure housing was built to accommodate the added door-control circuitry. The pressure transducer was replaced with a unit (Rahm Model PT-143.1000X-lOhI) that has a conductive plastic potentiometer rather than the wire-wound potentiometer OS the original equipment. This eliminated most of the stepwise motion initially encountered in the readout. Tho towing cable that was developed for the program by the Packard Electric Division of General Motors is a single-conductor armored cable 11.18 mm in diameter (Table 1, Fig. 5). Shipboard equipment The winch used is equipped with slip rings, designed and manufactured by the Pacific Scirntifir Company. to permit con- FIG. 4. The spreader bar assembly showing the pressure housing and underwater ture and depth information. The X-Y recorder of the original equipment was replaced by a two-channel strip-chart recorder (Leeds and Northrup Speedomax G, XI -X2) on which the haul is plotted directly as a function of time. The sampler control section consists of a power source, control switches, and a tinuous recording of the depth and temperature information. electronic equipment The shipboard consists of three sections: BT deck unit, recorder, and sampler control. The BT deck unit contains the d-c power supply, the band-pass amplifiers, and the frequency-to-d-c discriminators for the temperaTABLE 1. 227 454 908 1,362 1,816 2,270 2,724 3,178 3,632 4,086 4,540 Physical (500) (1,000) characteristics 0.069 0.173 0.449 0.725 1.00 1.28 1.48 1.86 2.07 2.31 2.73 electronics. of armowd electrical iZi ii::;; iii i :::;ii i:i:;:i (19.75) cable used in DDPS 0.762 2.54 5.59 8.89 12.7 18.3 24.1 30.7 37.8 44.7 55.4 :iEi system* 5 10 30 i::::i ii:::1 ;;:::i ii:;:; (2.18) ii 100 125 155 180 225 325 DISCRETE DEPTH PLANKTON SAMPLER 329 shipboard indicator for the door selector in the underwater unit. CALIBRATION OF THE ELECTRIC BATHYTHERMOGRAPII Laboratory and field calibrations of the bathythermograph over a period of two years have indicated that the accuracy of the depth channel is better than 41.4% of full scale (600 m), and the tcmpcraturc channel has an accuracy of *l.OC. The equipment is checked in the field by attaching a mechanical BT to the spreader bar and has also been checked against hydrographic casts with standard reversing thermometers. The laboratory calibration procedures are as follows. Depth channel The depth transducer is calibrated by an Amthor dead weight testor, Model 452 (?O.l% of test point). The pressure is varied in 3.5 kg/cm2 (50 psi) steps to a pressure equivalent to a depth of 600 m. The frequency output and the rccordcr d-c output are recorded as a function of pressure for the average of increasing and decreasing pressure points. From this is obtained a calibration curve. This procedure is performed before and after each field operation to maintain a current check on the calibration of the instrument. Temperature channel The temperature transducer is calibrated in a temperature bath having a mercury thermometer calibrated to ?O.lC. The bath is then varied in steps of about 2C from 0 to 3OC, and the data are recorded as in the depth calibration. USE OF TIIE DISCRETE SAMPLER DEPTH PLANKTON SYSTEM The basic system was considered opcrational by October 1962. Since that time, about 300 tows have been made in the Santa Barbara Channel and California offshore area from the RV Suxn. A number of changes have been made, mainly to improve the mechanical design and increase ru2. 5. Cross section of single-conductor Packard Electric Division armored towing cable. the accuracy of temperature and depth determinations. The unit has been reliable, with only seven failures since operational status was attained. Two of the failures were associated with the electronic system; the remaining failures were mechanical and caused either preclosure of the doors or failure of the doors to close. In each case, repairs were made in the field, and in no case did the repair time cxcced 1 hr. The sampler is operated from the control panel ( Fig. 6). An a-c signal ( 115 v, 60 cycles) is rectified and applied to the line, operating the door selector in the underwater unit and the indicator on the control panel. The door selector stepping switch also removes either temperature or depth information, or both, from the line, giving an indication on the recorder that a particular door has closed. This allows a cheek of the selector operation but gives no assurance that the door actually closed. For example, if the animals are large the doors may jam on them; similarly, any damage to the doors may cause such a condition. After operation of the selector switch is indicated, the selector is stepped again, reconnecting the sensor outputs to the cable and recorder. During the actual operation of the door selector, the BT portion of the electronic system is bypassed to prevent damage by the comparatively large door-closure signals. The equipment has been used at speeds ranging from 1.5 to 8.5 knots (2.8-16 km/ Preliminary evidence, however, indicates that this is not a serious problem. RESULTS id FIG. 6. Control pxlel for DDPS showing mig. X-Y rccordtr. hr). During tows at speeds of more than 7.0 knots (13 km/hr), prior to the use of an electrical swivel between the cable termiuatiorl and the net, the armored cable showed evidence of kinking. The exact cause of this is unknown, but it is believed that the unrelieved torque in the towing cable at the higher towing speeds was responsible. With the addition of a swivel, this problem appears to be solved. No kinking has occurred at the lower towing speeds, with or without a swivel in the system. It is anticipated that the useful life of this cable will be the sane as nonelectrical galvanized cable of similar diameter. It is assumed that all animals entering the net pass through the open cod end, thus preventing contamination of the catch during the lowering process. In fact, some contamination probably occurs; animals are sometimes trapped in the meshes while the net is lowered and then wash into the cod end after the rear doors are closed. The use of the tandem configuration and the newer four-door samplers has permitted the start of investigations of vertical plankton and nekton stratification. A study that has been accomplished with the DDPS system is the examination of the towing behavior of nets relative to changes in wire length and ship speed. The following data are from 35 tows using a 6.ft IKMT and 5 tows made with a l-m ring net. Fig. 7 presents a family of caves for the mid-water trawl taken with 100, 200, 400, 800, and 1,600 m of wire. In each of these tows, the ship was brought to towing speed (5.0-5.75 knots, 9.3-10.7 km/hr) before the cable was released. When all of the cable was paid out, the net was allowed to stabilize (reach a constant depth), after which the ship was slowed (to 2.25-3.0 knots, 4.2-5.6 km/hr) to permit the recovery of the trawl. The tows in this series show several common characterisncs. In each case, the net is at only about two-thirds of its stabilized towing depth at the moment when all of the wire has been paid out. The time for the net to stabilize was longer with increasing length of wire out, for example, FIG. 7. Towing characteristics of a 6.ft IsaacsKidd mid-water trawl. First arrow on each curve indicates the point when all wire had been paid out. Second arrow indicates when the ship was slowed and wire recovery commenced. DISCRETE DEPTII PLANKTON 13 min when 1,600 m of wire had been used. When the ship was slowed to recover the net (wire recovery began as soon as the engine speed was reduced ), the net went to greater depth and would remain for about half of the recovery period below the stabilized towing depth. Most of the above features appear in the family of curves obtained in an cxpcriment to determine the effect of speed on towing depth. In this series (Fig. S), the trawl was lowered with the ship traveling at the highest speed to be used during the test. Once the net stabilized, the towing speed was reduced about 2 knots (3.7 km/ hr ) , and the net was allowed to stabilize at this new speed, after which time the speed of the ship was again reduced about 2 knots. After the net stabilized this time, it was recovered without Eurthcr reduction in speed. As in the previous series, the net was at only about two-thirds of its initial stabilized depth at the moment all of the wire was out, and the time for the net to stabilize increased with increasing lengths of wire. The depth of the net increased with decreasing towing speed in all cases, but speed changes at the lower speeds produced a greater effect than at the higher speeds. For cxamplc, with 100 m of wire out, a change from 5.2 to 2.6 knots (9.6-4.8 km/hr) resulted in a 9-m increase in towing depth, while a reduction in speed from 2.6 to 1.6 knots (4.8-3.0 km/hr) produced an increase of 11 m in towing depth. With 200 m of wire out, a change from 7.3 to 5.4 knots (13.6-10.0 km/hr) altered the depth I~~~~~~~~;,~~ LlllCII 01wisrSLOW SLOW Irco”,a” 400 100 :: :: ,600 30 II FIG. 8. Towing characteristics Kidd mid-water trawl relative to ing speed. First arrow on each the point when all wire had been arrows indicate points when ship reduced. of a 6-ft Isaacschanges in towcurve indicates paid out. Other speed had been 331 SAMPLER !!i 300 s Ly 250 200 iF 150 ‘:;$*<~ ,,,,,,,I 0 100 200 ELECTRONIC 300 400 BT DEPTH FIG. 8. Depth dctermincd by plotted against depth detcrmincd mcasurcmcnt for 35 tows. 500 (METERS) clcctronic l3T by wire-angle of fishing only 2 m, whereas the change from 5.4 to 3.6 knots (10.0-6.7 km/hr) produced a 12-m increase in towing depth. All of the tows represented in the above two series were repeated at least twice. The additional data, except for minor variations in detail, are consistent with the information in Figs. 7 and 8. During the above series, the wire-angle depth was also determined. Fig. 9 comparts the equilibrium depth determined by the wire-angle measurement and that obtained by the DDPS system. The close agreement between the two measurements is somewhat surprising; however, the wire lengths used in the tests were relativcly short and all of the work was done in the Santa Barbara Charmel during ideal weather conditions. The winds never excecded 15 knots (28 km/hr) and generally were below 5 knots (9 km/hr). During a cruise made in cooperation with Dr. E. II. Ahlstrom of the U.S. Fish and Wildlife Service, La Jolla, California, five tows using a l-m ring net were taken. The net was equipped with a NO-lb (45 kg) weight suspended from the towing cable at the point of attachment of the net bridle. During the experiment, wind speeds ranged between 18 and 30 knots 332 WILLIAM ARON, NEWELL RAXTER, ROY NOEL, AND WILLIAM ANDREW!3 during the deep portion of the second tow is not understood, but it may be the result of irregularities in the wind and surface currents. It should be noted also that for hauls 4 and 5, which are repeats of hauls 2 and 3 but in the opposite towing direction, greater lengths of wire were needed to attain the depths reached in the earlier pair of tows. DISCUSSION FIG. 10. Towing characteristics of a 1-m ring net. Arrows indicate times when cod-end sampler doors were closed. (33 and 56 km/hr); however, the sea state did not interfere with the program. The paths followed by the net in each of the tows are shown in Fig. 10. During all tows, the ship speed was about 2.5 knots ( 4.6 km/hr ) . Unfortunately, neither wire angle nor information on the path of the net during the lowering process was recorded. On Station 1, the net was lowered to sampling depth and then recovered at a constant winch speed in an effort to obtain an oblique sample. On this tow, the gear was lowered with the rear doors of the cod-end sampler closed. On attaining sampling depth, the second set of doors was closed and the recovery of the net began. Doors 3 and 4 were closed as the net was being recovered. On the remaining stations in the series, the net was lowered with the cod-end sampler completely open. On attaining sampling depth, the rear ,doors were closed, After towing at this depth for about 5 min, the second set of doors was closed and the net was raised to the next sampling depth. On attaining the new depth, no. 3 doors were closed and the net was towed briefly prior to closing no. 4 doors and bringing the net back to the surface. Examination of Fig. 10 indicates that, as in the cast of the mid-water trawl, there is frequently a lag between the start of recovery and the time the net begins to rise. The reason for the sinusoidal path Our findings on the towing characteristics of nets strengthen the argument in favor of devices for continuously monitoring fishing depth and for obtaining samples from specific depths. If the time required for the net to attain sampling depth after deployment is added to sampling time during net recovery, it becomes apparent that the catches will be strongly biased unless such devices are employed. The recent samplers developed by B6 ( 1962) and by Foxton ( 1963) appear to provide a significant step toward solving problems regarding the vertical distribution of plankton. The use of Foxton’s equipment in association with the National Institute of Oceanography Depth Telemeter is especially valuable for detecting unexpected perturbations in the towing path such as the sinusoidal track shown in Fig. 10. Both 136’s and Foxton’s devices, however, sample a wide vertical column, and although they are useful for many survey operations they appear to have limited value for studies of the thermocline or other discontinuities where vertical gradients are sharp and large changes in plankton may occur in short vertical distances. The need for better data requires the use of more sophisticated instrumentation. The equipment described in this paper permits the capture of animals from known depths while continuously monitoring biologically important physical parameters. It is perhaps of greater importance to note that the use of electrical cable permits the addition of other sensors to the system in addition to depth and temperature. Besides a new bathythermometer to replace the existing unit, a bathyphotometer and a flowmeter are currently under DISCREm DEPTII PLANKTON development at our laboratories for incorporation in the system. Information from these sensors wifi be telemetered up the single conductor cable. Additional sensors can be added without increasing the numbcr of conductors. This increasing ability to capture animals and record cnvi?onmeG tal data simultaneously will permit a more comprehensive undektanding of animal distribution in the sea. REFERENCES ARON, W. 1962. Some aspects of sampling the macroplankton. Rappt. Proccs-Vcrbaux Reunions, Conscil Pcrm. Intern. Exploration Mcr, 153: 29-38. BJ$ A. W. H. 1962. Quantitative multiple opening and closing plankton samplers. Deep-Sea Res., 9: 144-151. SAMPLER 333 CURRIE, R. I. 1962. Net closing gear. Rappt. Proccs-Vcrbaux Reunions, Conseil Perm. Intern. Exploration Mcr, 153: 48-54. FOXTON, P. 1963. An automatic opcning-closing device for large plankton nets and midwater trawls. J. M arine Biol. Assoc. U.K., 43 : 295-308. LAEVASTU, T. 1962. The adequacy of plankton sampling. Rappt. Proces-Verbaux Reunions, Conscil Pcrm. Intern. Exploration Mer, 153: 66-73. MOTODA, S. 1962. Plankton sampler for collccting uncontaminated materials from scvcral different zones by a single vertical haul. Rappt. Proces-Verbaux Reunions, Con&l Pcrm. Intern. Exploration Mer, 153: 55-58. YENTSCII, C. S., G. D. GRICE, AND A. D. HART. 1962. Some opening-closing devices for plankton nets operated by pressure, electrical and mechanical action. Rappt, Proccs-Vcrbaux Reunions, Conseil Perm. Intern. Exploration Mcr, 153: 59-65.