Evaluation of the recovery period in mark-recapture population estimates of rainbow trout in small streams

We compared the backpack electrofishing capture efficiencies and Petersen-type mark-recapture abundance estimates of resident rainbow trout Oncorhynchus mykiss that had recovered for 24 h versus 3 h after electrofishing, handling, marking, and release in thirteen 100-m sites in four Yakima River basin tributary streams in central Washington State. Our results indicate that the catchability of rainbow trout was not significantly different between the two recovery periods (P = 0.27). Similarly, Petersen-type mark-recapture abundance estimates did not differ between the two recovery periods (P = 0.20). Despite vigilant effort at installing and maintaining block nets during the 24-h period, we detected fish movement out of 75% of our sites. In addition, our block nets collapsed or were destroyed by small animals in 36% of sites used for a 24-h recovery period: therefore, valid estimates could not be calculated. In contrast, no movement or net failure was detected during the 3-h recovery period. Some of the advantages of a 3-h recovery period between mark-recapture backpack electrofishing events include (1) increased probability of generating a population estimate because of a low threat of block-net failure: (2) lower probability of violating the movement assumption associated with the Petersen-type mark-recapture estimator: and (3) completion of field sampling within a single site visit on a single day. We believe that these advantages should be considered when designing sampling protocols for enumerating stream fish populations.