440 



Effect of towing speed on retention of 

 zooplankton in bongo nets 



Joseph Kane (contact author) 

 Jacquelyn L. Anderson 



Email address for J. Kane: |kane(5iwhsunl. wh.whoi.edu 



National Oceanic and Atmospheric Administration 



National Manne Fisheries Service 



Northeast Fisheries Science Center 



28 Tarzwell Drive 



Narragansett, Rhode Island 02882-1152 



Long-term time series of zooplank- 

 ton data provide invaluable informa- 

 tion about the fluctuations of species 

 abundance and the stability of marine 

 community structure. These data 

 have demonstrated that environmen- 

 tal variability have a profound effect 

 on zooplankton communities across 

 the Atlantic basin (Beaugrand et al., 

 2002; Frank et al., 2005; Pershing 

 et al., 20051. The value of these time 

 series increases as they lengthen, but 

 so does the likelihood of changes in 

 sampling or processing methods. Sam- 

 pling zooplankton with nylon nets is 

 highly selective and biased because of 

 mesh selectivity, net avoidance, and 

 damage to fragile organisms. One 

 sampling parameter that must be 

 standardized and closely monitored 

 is the speed of the net through the 

 water column. Tow speed should be as 

 fast as possible to minimize net avoid- 

 ance by the organisms, but not so fast 

 as to damage soft bodied zooplankters 

 or extrude them through the mesh 

 (Tranter et al., 1968; Anderson and 

 Warren, 1991). 



The bongo plankton sampler (Pos- 

 gay and Marak, 1980) has been 

 used by many investigators to moni- 

 tor plankton populations throughout 

 the world's oceans (e.g., Lough et al., 

 1985; Kane, 1993; Licandro et al., 

 2001). Tow speed on bongo surveys 

 has ranged from approximately 1 

 to 4 knots, depending on the target 

 population of the survey program. 

 With the growing interest in long- 

 term time series, the purpose of our 

 study was to evaluate differences in 

 catches with bongo nets towed at two 



of the common speeds: 1.5 and 3.5 

 knots. If the differences can be quan- 

 tified, zooplankton data collected at 

 the two towing speeds could be used 

 to compare or extend time series at 

 the regional level. We made replicate 

 collections at the two different tow 

 speeds and compared species com- 

 position, community structure, and 

 overall biomass. 



Materials and methods 



Plankton samples were collected 

 aboard the RV Delaware II from 

 March 29 to April 9 1999 in north- 

 eastern U.S. continental shelf waters 

 (Fig. 1). Initially a cross-shelf tran- 

 sect of 10 stations was surveyed, 

 extending from the island of Mar- 

 tha's Vineyard out to just beyond the 

 200-m isobath. For the remainder of 

 the cruise, samples were collected at 

 29 stations in a saw tooth pattern 

 throughout the Georges Bank region, 

 most within the 200-m isobath. 



At each sampling location two con- 

 secutive hauls were conducted with 

 a 61-cm bongo frame fitted with a 

 white colored 0.333-mm mesh net 

 towed obliquely to a maximum depth 

 of 200 m or to within 5 m from the 

 bottom and back to the surface. Wire 

 was payed out at 50 m/min and re- 

 trieved at 20 m/min during both tows 

 and a conductivity-temperature-depth 

 (CTD) probe was attached above the 

 bongo to measure sampling depth. A 

 flowmeter was suspended in the cen- 

 ter of the bongo frame to measure the 

 volume of water filtered during each 



tow. The first tow was performed 

 at a ship speed of 1.5 knots, which 

 varyied slightly at times to maintain 

 an approximately 45° wire angle. The 

 sampler was depressed with a simple 

 45-kg lead ball. After the first tow 

 had been completed, the vessel re- 

 turned to the exact location where 

 the tow was initiated. The gear was 

 redeployed with a wedge-shaped 1.2- 

 m 108-kg vfin depressor and towed at 

 3.5 knots without regard to the wire 

 angle. Specimens from all samples 

 were preserved in 5% formalin. 



In the laboratory, biomass was 

 measured by displacement volume. 

 The plankton sample, with preserv- 

 ing liquid, was measured in a gradu- 

 ated cylinder, poured through a mesh 

 cone into a second cylinder, and 

 drained until the interval between 

 drops was 15 seconds. The liquid in 

 the second cylinder was measured 

 and the displacement volume of the 

 sample was the difference between 

 readings. Samples were then reduced 

 to approximately 500 organisms by 

 subsampling with a modified box 

 splitter. Zooplankton were sorted, 

 counted, and identified to the low- 

 est possible taxa. Abundance and 

 biomass were expressed as number/ 

 100 m^ and cubic centimeters/100 m^, 

 respectively. 



Abundance of dominant taxa, de- 

 fined as those contributing on av- 

 erage >!% to the total abundance, 

 were compared between tow speeds. 

 Survey mean abundance values were 

 tested for significant differences by 

 using the conventional paired ^-test, 

 and its nonparametric counterpart, 

 the Wilcoxon test, was used to com- 

 pare median abundance levels. The 

 stage distribution of the five most 

 abundant copepods was also exam- 

 ined for significant differences by us- 

 ing the same statistical methods. 



Zooplankton community structures 

 from the two tow speeds were com- 

 pared by using multivariate statisti- 



Manuscript submitted 12 Deeembber 2006 

 to the Scientific Editor's Office. 



Manuscript approved for publication 

 14 March 2007 by the Scientific Editor. 



Fish. Bull. 105:440-444 (2007). 



