8162 - Laserfiche WebLink

Home Browse Search

DISPERSAL PATTERNS OF COLORADO SQUAWFISH TABLE l.-Locations of study strata and subreaches sampled for potential Colorado squawfish food fishes. Lo- cations are described in river kilometers (rkm) upstream of the confluence of the Green River. Stratum 4 was not sampled (see text). Mean Strata Subreach gradient location location Stratum (m/km) (rkm) Subreach (rkm) 1 0.21 0.0-112.6 1 5.2-6.0 2 12.2-13.0 3 20.9-21.7 4 30.1-30.9 5 61.6-62.4 6 100.6-101.4 7 103.5-104.3 2 1.33 112.6-140.6 8 121.3-122.1 9 129.0-129.8 3 0.61 140.6-181.0 10 148.5-149.3 11 156.6-157.4 12 172.2-173.0 4 2.42 181.0-200.0 5 0.91 200.0-245.5 13 212.9-213.7 14 228.0-228.8 15 234.6-235.4 6 1.27 245.5-275.1 16 252.9-253.7 17 270.0-270.8 7 1.70 275.1-298.1 18 287.2-288.0 19 293.8-294.6 19-km section dominated by large rapids and tur- bulent eddies, was excluded for logistical reasons and because few Colorado squawfish were found there in past studies (McAda and Kaeding 1991). The river was divided into seven strata based on major changes in average channel gradient (Table 1): three downstream (strata 1-3) and three up- stream (strata 5-7) of Westwater Canyon (stratum 4). Distribution and movement.-We documented distribution of Colorado squawfish by systemati- cally sampling the study area from mid-April through mid-June for 1991-1994. Trammel nets were set in backwaters, flooded ponds, and flooded canyon mouths because Colorado squawfish are attracted to calm habitats during the spring runoff period (Osmundson and Kaeding 1989). Capture techniques are described in detail in Osmundson and Burnham (1998). Three sampling passes (7- 9 d each) through the upper reach were made each spring, and every zero- or low-velocity habitat large enough to contain adult Colorado squawfish was netted. Similar preliminary data collected dur- ing spring 1990 were also used. In the lower reach, one pass was made in 1991 and two passes were made each spring during 1992-1994. In some por- tions of both reaches where zero-velocity habitats were rare, shorelines were electrofished with a 5- m, electrofishing boat equipped with a Coffelt 945 VVP-15 electrofisher (Coffelt Manufacturing, Flagstaff, Arizona) that produced pulsed DC. Captured Colorado squawfish were anesthetized with tricaine methanesulfonate and scanned for the presence of a passive integrated transponder (PIT) tag (Biomark, Inc., Boise, Idaho); if no tag was detected, one was implanted (see Osmundson and Burnham 1998). Fish were measured (nearest 1 mm) for maximum total length (TL; Anderson and Gutreuter 1983), weighed with an electronic bal- ance (nearest 1 g), and released after recovery from the anesthetic. Netting catch per unit effort (CPUE) was used to assess river-wide differences in relative abun- dance of adult Colorado squawfish. However, be- cause differences in fish density among river strata may have been influenced by the number of back- waters available for use by fish (higher concen- trations could occur in strata with few backwaters), we pooled netting data by reach for between-reach comparisons rather than attempt to compare den- sities among river strata. To measure movement, we compared locations of consecutive captures of marked individuals. Movement patterns were compared among 100- mm length-classes and between fish initially cap- tured in the lower reach and fish initially captured in the upper reach. Localized movements were considered to be those less than 10 km and were not analyzed. For assessing movement, we con- sidered only consecutive captures at least 1 year apart to allow fish adequate time to disperse be- tween captures. Also, captures during spawning periods and during pre- and postspawning migra- tory periods were excluded so that temporary spawning movements would not be confused with dispersal movements. Since concurrent radiote- lemetry studies indicated that annual spawning movements did not commence until late June or later, only samples through 16 June were consid- ered (except in 1994, when spawning movements began early and all June samples were excluded). Additional capture data collected during 1991- 1995 by the Colorado Division of Wildlife (from stratum 5) and the Utah Division of Wildlife Re- sources (from portions of strata 1 and 3) were also used in the movement analysis (both agencies con- duct annual spring electrofishing surveys). Re- cords from a 1994-1995 U.S. Fish and Wildlife Service (USFWS) survey of the lower 3.5 km of the Gunnison River and a 1995 survey of flooded ponds near rkm 262 were also used. Body condition.-Relative body condition (Le Cren 1951) was calculated for Colorado squawfish