BARNETT ET AL.: DISTRIBUTION OF ICHTHYOPLANKTON OFF SAN ONOFRE, CALIF 



1981) as our neustonic sampler. This net had an 88 

 cm wide mouth and fished to a depth of 16 cm. 

 Fiberglass-covered styrofoam floats kept the top of 

 the net out of water, and a 3 m spar and asymmetrical 

 bridle kept the gear outboard of the bow wave. A 

 weight suspended from the end of the wire held the 

 bridle well below the surface, out of the path of the 

 net. The sampler was launched and recovered off the 

 quarter by means of a tag line. Both a Tsurumi-Seiki 

 (TSK) flowmeter and a GO flowmeter were mounted 

 in the mouth of the net. The GO meter served as a 

 back-up for the TSK, which sometimes fouled with 

 kelp and eelgrass. 



The Auriga net, 7 used to sample the epibenthic 

 layer, consisted of a rectangular net frame (0.5 m high 

 X 2 m wide) attached to a chassis equipped with a 

 pair of side-mounted, 2 m diameter wheels. The 

 device rolled on the bottom so that the mouth of the 

 net was 10 cm (original design) or 17 cm (later ver- 

 sions) above the bottom of the wheels. A series of 12 

 cm diameter plastic rollers below the mouth of the 

 net helped prevent the sampler from digging into the 

 bottom and presumably minimized escapement 

 below the net. Both GO and TSK flowmeters were 

 mounted within the mouth of the Auriga net. The 

 Auriga net was towed off the stern. Divers have 

 observed (M. Sowby 8 ) that the mouth of the Auriga 

 assumes a horizontal attitude when the wheels are off 

 the bottom. We therefore believe that contamination 

 of the epibenthic samples by midwater plankton was 

 minimal during launch and recovery, when the main 

 component of (relative) water movement was across, 

 rather than through, the mouth. Any contamination 

 that did occur should have been a function of depth, 

 which was always <209r of the length of an 

 epibenthic tow (this potential source of error has 

 been ignored in the density calculations). 



Although serious clogging was not apparent in the 

 preliminary study, denser plankton concentrations 

 at other times of the year might clog the nets before 

 400 m 1 of water could be filtered. Clogging would be 

 most serious for oblique bongo tows, because it 

 would result in undersampling of the upper part of 

 the water column. In anticipation of this possibility, 

 the area of mesh in all nets was increased according to 

 the criteria suggested by Smith et al. (1968, equation 

 5) in order to sample 500 m 3 (bongo), 400 m 3 (Auriga), 

 and 200 m 3 (Manta) for "green" coastal waters. The 

 filtering ratios (R = mesh pore area/net mouth area) 

 of bongo, Auriga, and Manta nets were increased to 



7.8, 6.6, and 1 0.7, respectively, by adding mesh cylin- 

 ders ahead of the conical portions of the nets. Exter- 

 nal flowmeters were not used in the subsequent 

 surveys, but tows were carefully timed. Internal flow- 

 meter readings were checked upon recovery, and 

 samples were repeated if the readings differed by 

 more than 20% from expected values. 



Except for the limited study of daily vertical migra- 

 tion, all sampling was done at night. The deck lights 

 were always off during the neuston tows. All samplers 

 were launched, towed, and recovered with the vessel 

 underway at about 1 m/s. For bongo tows, wire was 

 paid out (scope about 2:1) until the weight, located 

 1.5 below the center of the net frame, bumped the 

 bottom. Then the nets were opened, and a stepped 

 oblique tow was made consisting of 18 30-s steps. 

 The Auriga sampler was towed with a scope of 3:1 

 and recovered after 6.5 min on the bottom. With the 

 small-mouthed Manta net, the volume of 400 m 3 was 

 achieved by towing two nets simultaneously, off port 

 and starboard, for 20 min (about 1.4 km). 



Samples were preserved in 5-10% seawater- 

 Formalin. 



Sampling Locations and Frequency 



Since we eventually wanted to assess the effects of a 

 power plant cooling system, it was necessary to con- 

 centrate much of our sampling effort within the depth 

 contours encompassing the cooling structures. At the 

 same time, in order to estimate the abundance of 

 nearshore species, we needed to sample far enough 

 from shore to delimit their centers of abundance. We 

 decided upon a stratified random sampling design 

 (Snedecor and Cochran 1967) wherein, on each sam- 

 pling date, the neustonic, midwater, and epibenthic 

 layers were sampled along a randomly chosen depth 

 contour in each of five blocks (Figs. 1, 2). The five 

 blocks were defined by depth contours: A) 6-9 m, cor- 

 responding to cooling water intake locations; B) 9- 1 2 

 m and C) 12-22 m, both corresponding to future dif- 

 fuser discharge locations; D) 22-45 m, corresponding 

 to a faunal break between inshore and coast- 

 al zooplankton assemblages (Barnett and Sertic 9 ); 

 and E) 45-75 m, chosen a priori as the likely offshore 

 limit of most nearshore larval fishes. 



The sampling transect thus consisted of 15 strata: 

 Three depth layers in each of five blocks (Fig. 2). To 



"Marine Biological Consultants, Inc., 947 Newhall Street, Costa 

 Mesa, CA 92627. 



"M. L. Sowby. Marine Biological Consultants, Inc., 947 Newhall 

 Street, Costa Mesa, CA 92627, pers. commun. 1979. 



'Barnett, A.M., and P. D. Sertic. 1979. Spatial and temporal pat- 

 terns of temperature, nutrients, seston, chlorophyll-a and plankton 

 off San Onofre from August 1976 - September 1978, and the 

 relationships of these patterns to the SONGS cooling system. In 

 Marine Review Committee Document 79-01, p. vii through 9- 

 89. Marine Review Committee of the California Coastal Commis- 

 sion, 631 Howard Street, San Francisco, CA 94105. 



99 



