7867 - Laserfiche WebLink

Home Browse Search

.. 1. MARKING TECHNIQUES FOR COLORADO SQUA WFISH 911 ! proved survival to cooling water temperatures plus a modification of the marking procedure that in- volved holding the fish in a 0.5% salt solution from capture to release. When the elastic polymer was used in the field study, three people sampling for 6 d produced a three-pass mark-recapture population estimate of 5,595 fish (::'::38%) and 2,523 fish (::'::38%) for the autumn and spring, respectively. These estimates are within the precision criteria (1.5N > N> 0.5N) proposed by Robson and Regier (1964) to address management level objectives. An important assumption of the M(t) population estimator (closed population, capture probabilities vary among sampling passes) is that on a given sampling pass, every fish has the same probability of capture. However, we suspect that heterogeneity of capture probabilities may result in bias up to 18%. Our sampling observations suggest that one way to overcome this problem is to confine seining to warm afternoons when most of the catch of young Colorado squawfish comes from one or two seine hauls at the warm, shallow ends of both deep and shallow backwaters. However, this is often difficult to achieve. During the 3 weeks of autumn and spring sampling, the weather can be extremely variable, and cool temperatures result in reduced catches, particularly in deep backwaters. The best procedure is to begin during warm weather and complete sampling as quickly as possible. Our winter survival estimate of 0.45 lacked de- sired precision (::':: 54%). Poor estimate precision was probably caused by the movement of un- marked fish into the study reach, the movement of marked fish out of the study area, or a combination of the two. The short distances traveled between autumn and spring by the five fish marked in the autumn and recaptured the following spring and the lack of recaptures 16 km downstream from the study reach suggest limited movement. However, the 5 recaptures in the spring (of 297 fish marked in the autumn) is far less than the 22.7 recaptures expected under the 0.45 winter survival estimate, the assumption that there was no movement out of the study reach, and estimated probabilities of cap- ture of 0.04, 0.03, and 0.10 for three sampling passes. Either fish lost their marks, marked fish died at a higher rate than unmarked fish, or more likely, some marked fish moved out of the study reach. The simulation results suggest that increas- ing the length of the study reach and thereby in- creasing numbers of captures would result in sub- stantial reductions in bias and Cv. The question of winter fish movement, however, needs further study. Survival estimates based on mark-recapture procedures had nearly twice the precision of those based on CPUE, even when seining effort was equal. When survival was calculated from CPUE for single passes through the study reach in autumn and the following spring, the estimates were more variable because seining efficiency varied greatly from pass to pass due to differences in river con- ditions and temperature. This resulted in an un- reliable index of population change. Survival es- timates derived from mark-recapture, on the other hand, are unaffected by changes in seining effi- ciency. The usefulness of CPUE increases when multiple years of data are available or when pop- ulation indices for extensive reaches of river are needed and mark-recapture is unfeasible. How- ever, for short-term studies in limited reaches of river, mark-recapture is a viable method because it enables increased precision with only moder- ately greater expenditure of resources. Our study found that marking young individuals is a satisfactory method for estimating population size, winter survival, and movement of age-O and age-l Colorado squawfish. Acknowledgments We thank Steve Severson and Ron Nichols of the Ouray Fish Hatchery, Colorado River Fish Pro- ject (CRFP), U.S. Fish and Wildlife Service (USFWS), for assistance with the laboratory stud- ies. We thank Tom Hatch and the fishery techni- cians of the CRFP for assistance with the field studies. Kevin Bestgen, Colorado State University, Chuck McAda and Bob Burdick of USFWS, Grand Junction, Colorado, and Dick Wydoski, USFWS, Denver, Colorado, reviewed an earlier draft of the manuscript. Three anonymous reviewers provided many helpful comments that improved the manu- script. References Arnold, D. E. 1966. Use of the jaw injection technique for marking warmwater fish. Transactions of the American Fisheries Society 95:432-433. David, C. S. 1955. Injection of latex solution as a fish marking technique. Investigations of Indiana Lakes and Streams 4:111-116. Hart, P. J. B., and T. J. Pitcher. 1969. Field trials offish marking using a jet inoculator. Journal of Fish Bi- ology 1:383-385. Hilborn, R. 1990. Determination of fish movement pat- terns from tag recoveries using maximum likelihood