Repatriated razorback suckers (Xyrauchen texanus) in Lake Mohave have been monitored for more than 20 years, but low recapture rates have inhibited evaluation of factors contributing to highly variable post-stocking survival. In 2010, deployment of remote passive integrated transponder (PIT) scanners able to detect 134.2 kilohertz (kHz) PIT tags was initiated to increase the number of encounters with marked fish. The program was expanded in 2012 and 2013, while traditional capture methods (i.e., trammel nets) continued to be employed to collect comparable long-term monitoring data and estimate abundance of all repatriated and wild razorback suckers marked with either 400 or 134.2 kHz PIT tags.
Twenty-one razorback suckers were handled by Marsh & Associates (M&A) during FY18; eight fish on November 28-29, 2017 with assistance from Arizona Fish and Wildlife Conservation Office (AZFWCO), and 13 fish during March 1216, 2018 multi-agency monitoring activities. PIT tags were undetected in three of the 21 captures and their histories were recorded as unknown in the database. These three unknown fish plus one other PIT tagged capture with no rearing history were omitted from further consideration, leaving 17 fish for analysis. Sex was determined at both events, and captures included 17 females and four males. Based on monitoring data from March 2017 and 2018, there is no effective wild razorback sucker population remaining in Lake Mohave. The repatriated razorback sucker population estimate in 2017, based on March 2017 and 2018 capture data, was 841 (95% confidence interval [CI] from 694 to 4,487).
Total deployment time for remote PIT scanners from October 1, 2017 through August 31, 2018 was 37,903 scan hours resulting in 131,131 PIT tag contacts, representing 3,835 unique PIT tags for which 3,652 had a razorback sucker marking record (i.e., implanted with a PIT tag and associated data recorded) in the Native Fish Database (as of August 31, 2018). Among fish with a marking record, 3,615 were repatriates, nine were wild, and 28 were of unknown origin. Based on 2017 and 2018 remote PIT scanning, the 134.2 kHz tagged repatriate population in 2017 was 3,471 (95% CI from 3,365 to 3,576). Basin and River subpopulation estimates based on zone specific scanning in 2017 and 2018 also were calculated. The Basin subpopulation was estimated at 1,872 (95% CI from 1,804 to 1,940) and River at 2,093 (95% CI from 1,966 to 2,220). The subpopulation in Liberty zone was not estimated because there were no recaptures there. Too few wild fish were contacted to estimate Basin and River subpopulations separately (six and three contacts, respectively). The lake-wide estimate of the wild population based on PIT scanning in 2017 and 2018 was nine fish (95% CI from 4 to 23).
A robust mark-recapture model was applied to a subset of razorback suckers contacted by remote PIT scanning. This population was known to be at large during a six year period (FY12 through FY18) allowing survival estimation to be removed from the analysis. This analysis was used to assess temporary emigration and to determine if capture parameters could be accurately assessed from PIT scanning data within the robust model framework. Temporary emigration was estimated at up to 8.1% of the known population (sample years 2014 to 2015) and estimates of razorback sucker returning to availability peaked at 54.9% for the last estimable period (2015 to 2016 and 2016 to 2017 combined). Temporary emigration could represent “skip” spawning or the existence of additional spawning areas that are not currently covered by remote PIT scanning deployments.
Stocking displacement was examined to determine distance traveled from stocking locations and to identify movement between zones. In River and Basin zones, most fish were contacted within their zone of release and this result was consistent across years. Razorback suckers stocked in Liberty zone were contacted in River or downstream in Basin and fish stocked in Katherine zone all were contacted upstream of their release locations. Results are congruent with 2016 and 2017 cohort analyses (Wisenall et al. 2016; Leavitt et al. 2017) but provide a more spatially explicit illustration of movement patterns among years.