WILKINS: ABUNDANCE OF WIDOW ROCKFISH 



havior and stock size should be used to evaluate 

 survey methodology. The groundfish management 

 team of the Pacific Fisheries Management Council 

 (see footnotes 2 and 3) used stock reduction and 

 cohort analyses to estimate the abundance of this 

 species. In an area comparable to our 1982 survey 

 area, the widow rockfish biomass was estimated to 

 be 21,664 t at the begining of 1982. This estimate 

 is based in part on commercial landing information 

 and, consequently, the definition of the grounds to 

 which it applies is somewhat vague. The fishery- 

 based estimates are much higher than those derived 

 from the 1982 survey data (about 830 1). The relative- 

 ly low sensitivity of the sonar systems used would 

 result in underestimating biomass and is undoubted- 

 ly responsible for much of this difference The 

 discrepancy is also partly due to the fact that our 

 survey methods only estimate the portion of the 

 stock present as detectable schools and are therefore 

 a measure of relative, rather then absolute, abun- 

 dance. This is true to some extent for most types 

 of surveys. 



Innovations are also needed to resolve the techni- 

 cal problems related to data collection, identification 

 of school species composition, and survey design. 

 Some suggestions include 



1) a two-vessel survey to improve the efficiency of 

 data collection— such a technique would separate the 

 chore of delineating areas of widow rockfish ag- 

 gregations, estimating school abundance, and test 

 fishing from that of estimating mean school biomass 

 (Gunderson et al. fn. 7); 



2) a means of recording a time base on both the 

 audio and video tape records of the echo sounder 

 and sonar to simplify finding the same school on 

 each system for school dimension measurements; 

 and 



3) a method of estimating all school dimensions 

 and the density within the school from a single data 

 collection system— this would entail development of 

 a sophisticated, quantitative sonar-integration sys- 

 tem with the capability of recording the output onto 

 videotape (Ehrenberg 1979). 



Such refinements could probably be implemented 

 with relative ease. The methodology should be re- 

 evaluated when these technological and sampling 

 improvements have been made. Widow rockfish 

 management could have been significantly improved 

 with the knowledge of stock size from an effective 

 resource assessment survey. There are also other 

 species which exhibit similar behavior and which, 

 although presently unexploited, need to be assessed 



(eg, shortbelly, redstripe, and black rockfish). This 

 methodology could probably be easily adapted for 

 surveying these resources. 



CONCLUSIONS 



Based on the results of research conducted dur- 

 ing this project, the line transect survey method 

 using a sector scanning sonar and a quantitative 

 echo sounder appears to be the best means of assess- 

 ing widow rockfish abundance with research surveys. 

 A weakness of this method is that it only measures 

 the portion of the population existent as distinguish- 

 able schools and that portion may be quite variable 

 It also relies heavily on subjective experience for 

 identifying the species composition of schools. Its 

 strengths are that large areas can be covered quickly 

 and it is not necessary that all schools within sighting 

 range be detected in order to estimate school abun- 

 dance It appears that this could be a useful assess- 

 ment method for widow rockfish and for several 

 other Pacific coast groundfish species which are not 

 yet being seriously exploited. The effectiveness of 

 the technique could be enhanced by employing or 

 developing more sensitive and specialized quan- 

 titative sonars and by improving the methods of data 

 collection. The technological and survey design prob- 

 lems encountered should be relatively easy, though 

 somewhat costly, to resolve The method should then 

 be reevaluated to determine its utility. As the tech- 

 nique is used, scientists will gain a better under- 

 standing of the behavior and habits of the target 

 species. 



ACKNOWLEDGMENTS 



The work described in the developmental section 

 was ably conducted by Donald R. Gunderson and 

 Gary L. Thomas and their associates at the Fisheries 

 Research Institute, University of Washington Col- 

 lege of Fisheries, Seattle, WA, under contract to the 

 National Marine Fisheries Service (contract no. 79- 

 ABC-00203). Much of the information presented in 

 that section is extracted from their contract reports. 

 The hydroacoustic expertise during the 1982 survey 

 work was provided by the Pelagic Resources Assess- 

 ment Task of the Resource Assessment and Conser- 

 vation Engineering (RACE) Division, Northwest and 

 Alaska Fisheries Center, NMFS; in particular, Ed- 

 mund Nunnallee, Jimmie J. Traynor, and John Gar- 

 rison. I am expecially grateful to Nunnallee for 

 advice and guidance during the analysis of echo 

 sounder and sonar data. I also wish to thank Thomas 

 A. Dark, RACE Division, and Nunnallee and Traynor 



309 



