9565 - Laserfiche WebLink

Home Browse Search

1360 sampling passes (usually 7-11 d each) through each reach during the same period each year and repeated that design in other years. Sampling passes were conducted after ice-off and prior to or during spring runoff before Colorado pikeminnow spawning migra- tions began (Tyus 1990). Approximately 7-10 d elapsed between sampling to allow for sufficient mixing of marked and unmarked fish. Pulsed-DC Coffelt or Smith-Root electrofishing units were the primary sampling gear, used from flat-bottomed boats in alluvial reaches and from inflatable rafts in canyon reaches. Each craft had one or two anode spheres suspended forward from booms; cathodes were either stainless steel dropper cables or the aluminum boat itself. Two boats or rafts were typically used on each sampling pass, one along each shore, and sampling was in a downstream direction. The exception was the Yampa River in 2000, where a single electrofishing boat sampled portions of each shoreline on each pass and effort was supplemented with passive sampling gears such as fyke nets and trammel nets in low- velocity areas. Colorado pikeminnow were measured (TL [mm]), weighed (g), and scanned for the presence of a passive integrated transponder (PIT) tag; unmarked fish (> 150 mm) received a PIT tag inserted into the body cavity just posterior to the pelvic girdle. Capture and release location was determined from river maps and a Global Positioning System unit, and all Colorado pikeminnow were released within 0.2 rkm of their capture location. We used Recovery Goal criteria to define the lengths of adult (2':450-mm) and recruit (40~49-mm) Colorado pikeminnow, recognizing that some Colorado pike- minnow reach sexual maturity at larger sizes (Tyus 1990; U.S. Fish and Wildlife Service 2002; Osmund- son 2006). Additional data utilized in this study were from an Interagency Standardized Monitoring Program (ISMP; McAda 2002). The ISMP sampled adult and subadult Colorado pikeminnow from 10 reaches (8-25 kIn long) in the Green River basin, 5 in the Green River (rkm 539.4-526.5, 513.6-483.0,418.6-394.5, 185.2-154.6, and 90.0-64.0), 3 in the Yampa River (rkm 167.4- 153.0,128.8-112.7, and 86.9-78.9), and 2 in the White River (rkm 169.1-153.0 and 33.8-0). The reaches totaled about 23% of the critical habitat of Colorado pikeminnow in the Green River basin and were chosen because they were accessible by a flat-bottom electro- fishing boat and represented reaches known to support Colorado pikeminnow. Sampling was conducted during spring each year from 1986 to 2000. A single electro fishing sampling pass was made down each shoreline with one boat, stunned Colorado pikeminnow were captured with dip nets, and electrofishing time BESTGEN ET AL. was recorded. Captured fish were tagged, measured, and released following a protocol similar to that described above for abundance estimation sampling. Electrofishing effort was stratified among four to eight subreaches of each river reach, and catch per unit effort (CPUE) statistics were calculated for each subreach sample. Variances and standard errors for CPUE indices were calculated based on those samples. Prior to 1991, Colorado pikeminnow were tagged with Carlin dangler tags, which were presumed to be subject to relatively high tag loss that would confound analyses based on recaptures. Therefore, data from 1986 to 1990 were not included in any analyses presented here. After 1991, all Colorado pikeminnow captured were PIT- tagged, so tag loss was considered negligible. To allow comparisons with the ISMP data collected from 1991 to 2000, we designated a sampling pass during 2001- 2003 that conformed in space and time to the ISMP sampling. To compare the abundance estimates of Colorado pikeminnow with the CPUE indices, we averaged capture rates for ISMP reaches that were within the main river reaches used for abundance estimation sampling. Those annual composite ISMP CPUE indices for the Yampa River, White River, middle Green River, and lower Green River were then statistically correlated with the abundance estimates for the same reaches. No ISMP sampling was conducted in the Desolation-Gray Canyon reach of the Green River, so no comparisons are available for that reach. We also compared the size structure of adult and recruit-size Colorado pikeminnow that were captured only in ISMP reaches in 1991-1999 and 2000-2003 to determine whether there were differences between the two periods. Robust design for capture-recapture studies.- Robust-design sampling and analysis capitalizes on the strengths of the closed and open population models that are used to estimate demographic parameters (Pollock 1982; Pollock et al. 1990). Following this approach, we made sampling passes at closely spaced intervals (e.g., about 1 week) to estimate annual population size. In some reaches and years, we conducted a fourth, relatively late sampling pass. We combined the fourth-pass data (when available) with the third-pass data because equal numbers of passes created consistency in the capture history formats for modeling. Pooling was done only after analyses showed that only a few fish (much less than 1%) moved between sampling reaches between passes within a year. Statistical modeling.-The combined robust-design (Kendall 1999; Kendall et al. 1995, 1997), multi- stratum (Brownie et al. 1993; Hestbeck et al. 1991) model in Program MARK (White and Burnham 1999)