Laserfiche WebLink
against point estimates for those years. In most cases there was poor agreement between the <br />predictions and point estimates. Discrepancies suggested that the method, or the abundance <br />point estimates against which the results were compared, is not reliable enough to forecast <br />trends. Imprecision of the point estimates or inaccurate representation of subadults in length <br />frequencies of sampled fish are likely causes for the method failure. <br />Continuation of the current capture-recapture methodology is recommended for <br />monitoring this population. The three-year-on, three-year-off sampling regimen should also <br />be continued. Back-to-back years of sampling maximize life-history information and allow <br />averaging of annual population estimates; three years of rest allow the population to complete <br />its life functions with minimal disruption. It is strongly recommended that in years of <br />sampling the number of passes and the per-pass effort should be increased to reduce potential <br />bias of abundance estimates and tighten confidence interval width. Concurrent sampling by <br />four crews will be required to accomplish four annual sampling passes in each reach during <br />the eight weeks of spring runoff. <br />x