Accuracy and Precision of Age Estimates for Northern Pikeminnow Using Scales and Opercles

Age-structure information is essential to many fisheries management programs and is used widely throughout the Northern Pikeminnow Management Program, which aims to reduce juvenile salmonid predation by removing large northern pikeminnow Ptychocheilus oregonensis from the Columbia and Snake rivers.  In order to evaluate the precision and accuracy of our ageing techniques we compared ages determined from scales and opercles between readers, used oxytetracycline (OTC) marking to validate opercular annuli, used tag recaptures to validate scale annuli, and compared ages derived from paired scale and opercle samples.  From 2000 to 2007 we evaluated over 2,300 scales and 1,100 opercles collected from 250 km of the Columbia and Snake rivers.  Complete precision on ages assigned by two readers ranged from 21% to 72% for scales and 19% to 41% for opercles.  The number of annuli identified past the OTC mark accurately reflected the time at large in 73% of the opercle samples and the ages assigned to tag recaptures accurately reflected the elapsed time in 38% of the scale samples.  Paired scale and opercle samples were aged identically in 29% of all cases, with the latter structure typically yielding older ages; scale ages ranged from 2 to 15 years whereas opercle ages ranged from 3 to 24 years.  Overall, the level of precision between readers was higher for scales, but the accuracy of annuli counts (i.e., accounting for the correct time at large) was higher for opercles.  Given these findings, we urge caution in using age data to quantify growth patterns for northern pikeminnow and encourage, when possible, using measured growth from mark and recapture studies.  Further, when age-structure data are needed, we suggest validation studies to provide a means for identifying and correcting for biased age assignments.