FISHERY BULLETIN: VOL. 84, NO. 2 



The results of echo integration, line intercept, and 

 line transect analyses were compared using data col- 

 lected during the 1980 and 1981 FRI cruises 

 (Gunderson et al. fn. 7, 8). Large differences were 

 seen between echo integration and line transect 

 estimates in a situation where schools were relatively 

 small and scarce (1980 transect data, Table 6). The 

 principal reason for this is that the threshold echo 

 voltage required to trigger the sonar CRT display 

 was higher than that needed to detect a school on 

 the echo integration system, so many of the sparser 

 schools detected by the echo sounder were not 

 detected with the sonar. In situations where schools 

 were larger and more plentiful (1980 nonrandom 

 runs and 1981 transects) all three methods produced 

 similar estimates. The precision of abundance 

 estimates generated by line transect and line inter- 

 cept methods is usually comparable to that of con- 

 ventional echo integration methods and can exceed 

 it in some cases (Gunderson et al. fn. 7). The major 

 factors which led us to concentrate our efforts on 

 line transect surveys were the ability to cover large 

 areas rapidly and the ability to expand the number 

 of schools sighted by a detection function, yielding 

 more accurate estimates of school abundance. 



APPLICATION OF 

 ASSESSMENT METHODOLOGY 



By 1982 the aforementioned studies had provided 

 a foundation of information on which to expand 

 developmental research. The behavioral observations 

 suggested that widow rockfish aggregations were 



most stable and susceptible to assessment during the 

 night. Line transect estimation of school abundance 

 through the use of sonar and echo integration equip- 

 ment was found to be the most effective of the tech- 

 niques compared, especially when school abundance 

 was likely to be low. The next step in the project was 

 to evaluate the feasibility of applying the line 

 transect survey method in a comprehensive survey 

 to assess and monitor widow rockfish stocks. 



Methods 



The trawler Ocean Leader was chartered to survey 

 five areas off Oregon (Fig. 1) where widow rockfish 

 had been caught consistently between 1980 and 

 1982. Specifications of the vessel, fishing gear, and 

 hydroacoustic equipment used appear in Tables 2-4. 

 The proximity of alternative grounds was important 

 for the success of the survey, should widow rockfish 

 not be found in one or more of the areas. At each 

 of the grounds the survey procedure was as follows: 



1) The ground was systematically surveyed with 

 hydroacoustic equipment during the night to 

 determine whether fish schools were in the area. 

 The locations of schools suspected to be composed 

 of widow rockfish, or species likely to be confused 

 with widow rockfish, were noted. The final bound- 

 aries of the study area were then delineated. 



2) The study area was surveyed at night along 

 parallel tracklines about 1 km apart using the line 

 transect survey technique The tracklines were 

 replicated as many times as practical throughout 



Table 6.— Summary of estimates of school abundance (D), mean school biomass (MSB), 

 and total biomass (6) for widow rockfish. Coefficients of variation (CV) are given for each 

 estimate. 1 



'Gunderson, D. R., G. L. Thomas, P. Cullenberg, and R. E. Thorne. 1981. Rockfish investigations 

 off the coast of Washington and Oregon. Final report. Unpubl. manuscr, 45 p. Univ. Wash., Fish. 

 Res. Inst. FRI-UW-8125. 



302 



