Chang: Effects of sewage sludge clumping on fishery resources 



597 



species, and of all these species combined. The general 

 hypothesis was subdivided into three more specific hy- 

 potheses, each of which was likewise tested: 1 ) there 

 are no temporal differences in species abundance be- 

 tween the period prior to sewage sludge dumping at 

 the 106-MDS and the period after dumping resumed; 

 2) there are no spatial differences in species abun- 

 dance between the area north of the 106-MDS and the 

 area south of the dump site, the latter more likely to 

 be influenced by the sewage sludge that supposedly 

 moved through the dump site toward the west-south- 

 west shallow outer shelf (PAI; Fig. 1); and 3) there are 

 no seasonal differences for individual and all species 

 for the north and south areas within the pre- and 

 post-dumping periods. 



Methods 



Data from NOAA's Northeast Fisheries Science Center 

 (NEFSC) bottom trawl surveys were used for analysis. 

 From 1963 to the present, NEFSC has conducted bot- 

 tom trawl surveys on the Northwest Atlantic Shelf 

 and upper slope, from the Scotian Shoals to Cape 

 Hatteras including the areas adjacent to the 106-MDS 

 (Fig. 1). The entire survey area has been stratified by 

 depth down to 365 m, and surveys have been based on 

 a stratified random design that provides statistically 

 valid samples for estimating indices of species popula- 

 tion abundance. All fishes and invertebrates are sorted, 

 counted, and weighed by species to the nearest 0.1kg. 

 Large catches are subsampled by weight and volume 

 and expanded to estimate the entire catch. Detailed 

 information pertinent to the rationale, descriptions of 

 trawl gears and sampling schemes, as well as history 

 of the NEFSC bottom trawl survey are provided by 

 Grosslein (1974) and more recently Azarovitz (1981). 

 Grosslein et al. (1979) summarized research, assess- 

 ment, and management of the northwest Atlantic eco- 

 system using data from NEFSC bottom trawl surveys. 

 Data from the spring 1982 through spring 1986 

 NEFSC surveys were evaluated as the pre-dumping 

 period, and data from autumn 1986 through spring of 

 1990 as the post-dumping period. Eleven economically 

 important species were selected for analysis: silver hake 

 (Merluecius bilinearis); red hake (Urophycis ehuss); 

 summer flounder (Paralichthys dentatus); goosefish 

 {Lophius americanus); black sea bass (Centropristes 

 stria.tus); scup (Stenotomus chrysops); butterfish 

 (Peprilus triacantus); longfin squid (Loligo pealei); 

 American lobster (Homarus americanus); sea scallop 

 iPlacopecten magellanicus); and spiny dogfish (Squalus 

 acanthias). Species were selected on the basis of their 

 commercial values and landings, and the known avail- 

 ability of data for the entire 106-MDS studv area 



(Chang, 1990). In addition, all species, a category com- 

 prising the total catch for each cruise, was also used. 



The multivariate rank sum test (Puri and Sen, 1971) 

 was employed for temporal, spatial, and seasonal dif- 

 ferences in the population abundance analyses by us- 

 ing the catch per unit of effort (CPUE) as a measure of 

 population abundance: 1) the average CPUE values 

 for each bottom trawl survey stratum for all cruises 

 occurring in the pre-dumping period cruises (spring 

 1982-spring 1986), and for all cruises occurring in the 

 post-dumping period cruises (autumn 1986-spring 

 1990), and 2) the average CPUE data from the area 

 defined as south of the 106-MDS, including strata 61- 

 76 and data from the area defined north of the site 

 including strata 1-12 (Fig. 1). Test statistics for CPUE 

 were calculated and compared with critical values from 

 the chi-square table to determine the following: 



1. there are no temporal differences in CPUE of pre- 

 and post-dumping periods for spring and autumn 

 cruises collectively in the regions north and south 

 of the 106-MDS (cf. Table 1), 



2. there are no spatial differences in CPUE north and 

 south of the 106-MDS for pre- and post-dumping 

 period cruises collectively in spring and autumn sea- 

 sons (cf. Table 2), 



3. there are no seasonal differences in CPUE of spring 

 and autumn surveys for pre- and post-dumping pe- 

 riod cruises collectively in the regions of north and 

 south of the 106-MDS (cf. Table 3). 



Test statistics were also computed and compared with 

 the critical values for testing CPUE for spatial differ- 

 ences for individual species and for all species taken 

 on each cruise in the area north and south of the 106- 

 MDS. Test statistics and abundance indices of indi- 

 vidual species in tabulated form are not included among 

 tables given here. 



The values of test statistics for temporal, spatial, 

 and seasonal differences are readily comparable (Tables 

 1-3 ). When examining values of the test statistics for 

 temporal differences, secondary differences for species, 

 regions, and seasons were also assessed. Test statis- 

 tics were similarly examined for spatial and seasonal 

 differences among species, regions, and seasons. 



Changes in significance or non-significance of the 

 test statistics between pre- and post-dumping period 

 surveys may be interpreted as indicative of temporal 

 differences in CPUE. Significant changes of test sta- 

 tistics between regions north and south of the 106- 

 MDS indicate spatial differences. Changes between 

 spring and autumn surveys indicate seasonal differ- 

 ences. Significant differences in CPUE in time, space, 

 and season may thus be interpreted as some shift in 

 species abundance. Negative differences are taken here 

 as a response of the population to some adverse 



