Calculation of untrawlable areas within the boundaries of a bottom trawl survey

A geographic information system (GIS) analysis of 5039 bottom trawl events from U.S. west coast bottom trawl surveys (1977-1998) estimated that the survey area was about 77% trawlable but five of the 30 strata were less than 50% trawlable. Untrawlable areas, by definition, cannot be surveyed with the bottom trawl; however, there has never been a means of identifying and excluding these areas from relative abundance estimates, which are calculated only from hauls completed in the trawlable portions of each stratum. Unknown amounts of untrawlable habitat are a problem for relative abundance estimation in many bottom trawl surveys. This manuscript describes one method of using the bottom trawling events of a survey, such as ripped-up hauls and abandoned stations, to calculate the amount of area that is untrawlable. A comparison of catch rates between undamaged tows and a limited number of damaged tows, which are normally discarded as faulty samples, showed that Sebastes catch rates were generally higher in damaged tows. Thus untrawlable areas may have substantial importance on relative abundance estimates of Sebastes, the original target species group for this survey.