Fnedlander et al : Sidescan-sonar mapping of benthic trawl marks off Eureka, California 



793 



was particularly true in the deeper blocks 

 where the ratio of area trawled to block areas 

 ranged from 127% to 339%. 



The mean trawl depth and mean duration 

 of hauls for bottom trawls off Eureka has in- 

 creased steadily since the early 1980s (Fig. 

 12A). Along with this change in fishing depth 

 has come a shift in the composition of species 

 landed. Thornyheads are deeper-water species 

 and their catch contribution has increased 

 from -4% of the total catch in the late 1970s 

 to -14% in the 1990s. Conversely, the English 

 sole is a shallow-water flatfish whose landings 

 have declined from nearly 77c of the total catch 

 in the late 1970s to 2-3% in the 1990s. 



Discussion 



124°40' 

 4r20' 



124°30' 



124.°20' 



124°10' 



Utility of sidescan sonar 



41''10' 



40°50' 



Acoustics are useful in fisheries for stock as- 

 sessment (Karp, 1990), for remotely sensing 

 characteristics of bottom habitat (Able et al., 

 1987; Greenstreet et al., 1997; Mayer et al., 

 1997; Yoklavich et al., 1997; Auster et al., 

 1998a, 1998b), and for assessing impacts to 

 fisheries habitat (Auster et al., 1996; Collie et 

 al., 1997). Our results demonstrate that the 

 use of acoustic remote sensing also presents a 

 promising independent approach for evaluat- 

 ing fishing effort on a spatial scale consistent 

 with commercial fishing activities. Sidescan- 

 sonar data are useful for identifying the loca- 

 tion of trawling activity, as well as for evalu- 

 ating the amount of disturbance to the sea- 

 bed, and may be used in developing indices of 

 habitat disturbance caused by fishing activi- 

 ties. Despite minimal sidescan coverage in 

 some reporting blocks and biases in the log- 

 book data associated with misreporting, there 

 was a significant positive correlation between 

 the density of trawl marks per reporting block from 

 sidescan-sonar data and the mean annual number 

 of trawl hours. A quadratic relationship with depth 

 was observed for the number of trawl marks and fish- 

 ing effort per reporting block with maximum values 

 at -400 m and -500 m. respectively. This relation- 

 ship, together with the density contour plot of trawl 

 marks and the vector plot of trawl orientation, shows 

 that fishing effort was concentrated on the upper to 

 mid slope and along bathymetric contours. 



The entire study area off Eureka is trawled -1.5 

 times its total area on an annual basis. Because of 

 narrow bands of rocky outcrops and anticlines. 



124° 

 41°20' 



41°10' 



41° 



40°50' 



40°40' 



40°40' 



40°30' 



124°40' 



124°30' 



124°20' 



124°10' 



124° 



40°30' 



Trawl marks/sq. km 

 0-10 



5 10 nautical miles 



5 10 15 20 kilometers 



Figure 5 



Contour plot of trawl mark densities (number/km-) de- 

 rived from 10-minute intervals along sidescan survey track 

 lines. Actual locations of data points are in Fig. 6. 



trawlable fishing area within the mid slope report- 

 ing blocks is less than the total area of the blocks. 

 The area swept by trawls is therefore an under-esti- 

 mate of the effective fishing area that is impacted by 

 trawl gear in these blocks. Gislason ( 1994) estimated 

 that trawling swept the entire area of the North Sea 

 annually, with some fishing grounds in the southern 

 North Sea likely being swept more than three times 

 a year by beam trawls. This type of chronic impact 

 on the seafloor can have a long-term effect on the 

 local benthic epifauna (Rogers et al., 1998). For ex- 

 ample, there is abundant evidence of rich ecological 

 communities associated with cold-seeps and exten- 



