The main metric used to manage harvest is changes in abundance. The various data used are complicated by a number of factors including weather during the harvest period, etc. Can DNA results replace the current modeling to reach a usable conclusion?
Dieter, the use of DNA results to estimate fish population abundance is in its infancy. I am unaware of a definitive answer yet to your question yet developed. The Ecosystem Science and Management Working Group (ESMWG) of the NOAA Science Advisory Board has received information recently that addresses the question. You might want to contact Dave Fluharty for more details about presentations made to the ESMWG.
I'm a co-convenor on an ICES Annual Science Meeting symposium on the subject. It will be called " Modernizing Fisheries Stock Assessment and Monitoring with Genetic Methods". I believe that the call for abstracts will be in January. We identified close kin mark recapture (in addition to Edward Farrell's point above see also https://projecteuclid.org/download/pdfview_1/euclid.ss/1464105042 on close kin), gene tagging( http://www.molecularfisherieslaboratory.com.au/portofolio/genetic-mark-recapture/ for other very interesting work) and eDNA as among the key areas where this is moving forward. Gary Matlock's point still stands that the work is in its infancy but the potential is very exciting!