742 



Fishery Bulletin 100(4) 



125 OO'W 



44' 15'N. 



44°10'N. 



44'05'N. 



44'00'N 



43 55'N. 



Figure 3 



Backscatter imagery of Heceta Bank from Simrad EM300 

 multibeam survey. The lighter the shading the higher the 

 reflectivity of the substratum. 



videotape equipped with a timed data logger and audio 

 track. 



Direct and videotaped observations along the tran- 

 sects consisted of the number of species offish and mac- 

 roinvertebrates, and bottom-type characteristics. Fishes 

 along the transects were identified, counted, and lengths 

 were estimated to the nearest decimeter with a three- 

 decimeter fiberglass rod suspended within the observer's 

 view. Bottom type was categorized from videotapes by 

 using a two-code combination, the first letter indicating 

 the primary substratum (defined as covering at least 

 50% of the area viewed) and the second letter indicat- 

 ing the secondary substratum (defined as covering more 

 than 20% of the area viewed). If the field of view was a 

 single substratum, or the second most abundant substra- 

 tum covered less than 20% of the field, the same letter 

 was employed twice (e.g. MM for pure mud). The seven 

 possible categories in order of increasing particle size or 

 relief were mud (M), sand (S), pebble (P, diameter <6.5 

 cm), cobble (C, >6.5 and <25.5 cm), boulder (B, >25.5 cm), 

 continuous flat rock (F, low vertical relief), and diagonal 

 rock ridge (R, high vertical relief). The latitude and lon- 

 gitude positions of each transect were determined by 

 using Loran-C with a Trackpoint II ultrashort baseline 

 tracking system and by positioning the vessel directly 

 above the submersible every 10 to 15 minutes. At least 

 three position points were made per transect and the lo- 

 cations of bottom type and biological data were interpo- 

 lated between these points. The absolute accuracy of the 

 submersible's position, obtained by using Loran-C, was 

 within about 150 to 500 m (Melton, 1986). 



Multibeam sonar 



The acoustic survey of Heceta Bank was conducted in 

 May of 1998 with a Simrad EM300 (30 kHz) multibeam 

 sonar system on the RV Ocean Alert (Merle et al.*). This 

 survey provided a highly detailed, precisely navigated 

 seafloor map of bathymetry and seafloor texture ( Figs. 2 

 and 3). The data were processed with Swathed software 

 (Hughes-Clarke et al., 1996). The data processing steps 

 used in Swathed were the following: navigation and sound- 

 ing editing, roll bias correction, tide correction, refraction 

 correction, map sheet setup, gridding, and mosaicing 

 which resulted in a composite map made up of acoustic 

 backscatter imagery. The survey consisted of 47 overlap- 

 ping north-south swaths up to 45 km long, which provided 

 images of approximately 725 km^ of the seafloor and 

 nearly 100% coverage of high-resolution bathymetry and 

 backscatter amplitude. These data were displayed in grids 

 with a resolution of less than 5 meters on the shallowest 

 portions of the bank from depths of 70 to 150 meters, and of 

 about 5 to 10 meters at depths down to about 500 meters. 



Data integration and habitat assessment 



The sonar and submersible transect data were combined 

 by using ArcView and Arclnfo geographic informa- 

 tion system (GIS) software. The sonar data used were 



