17-8 Composite priors for survey bottom trawl catchability using inshore longfin squid as an example

Tuesday, September 14, 2010: 11:00 AM
402 (Convention Center)
Larry Jacobson, PhD , Population Dynamics branch, NOAA, NEFSC, Woods Hole, MA
Lisa Hendrickson , Northeast Fisheries Science Center, Woods Hole, MA
Tim Miller, PhD , Northeast Fisheries Science Center, Woods Hole, MA
James R. Bence, PhD , Department of Fisheries and Wildlife, Quantitative Fisheries Center, Michigan State University, East Lansing, MI
Many invertebrate and finfish species are difficult to assess using traditional stock assessment models. We use a direct approach to set bounds and estimate uncertainty for survey catchability, stock biomass, and fishing mortality.  Bounds and uncertainty for inshore longfin squid are based on a vessel calibration study, trawl sensor data, biological information, expert judgment, and capture efficiency. Prior distributions for each factor can be used to calculate a composite prior distribution for survey catchability. This composite prior for squid is asymmetric with a broad flat region implying a range of values that are almost equally probable.  There are interesting mathematical and policy aspects because it is easier to make precise assumptions about the upper bound of survey catchability (lower bound for biomass) than the lower bound for survey catchability (upper bound for biomass).  Our method could be applied to other stocks, and is especially useful when information about area swept by a research survey trawl is available.