9711 - Laserfiche WebLink

Home Browse Search

on growth has been focused on these smaller individuals. The analyses described in this report address the effect of tempera- ture on growth rate ofHBC and attempt to estimate the length at which fish transition from primarily LCR occupancy to primarily mainstem occupancy. The general implication from the tindings reported herein is that growth rate will increase substantially with a tempemtureincrease from HfC to 20"C, as is indicated by the values of Q< = 4.6 and Qm = 2.0. These coef- ficients suggest that anabolic processes will more than double relative to catabolic processes across this temperature range. As an additional evaluation of the results of the analy- sis. monthly growth rates from the TDGM were compared to laboratory observations of juvenile HBC growth. Clarkson and Childs (2000) conducted laboratory experiments to evaluate the growth rate of larval HBC at WOC, 140C, and 200C. They report monthly growth rates of I mm/month, 13 mm/month, and 17 mm/month for these temperatures, respectively. Con- sidering the estimated monthly growth rates from the TDGM in tigure 24, the TDGM tends to overestimate the growth rates reported by Clarkson and Childs (2000) at WOC and underes- timate the growth rate at 20"C. However, the results reported herein are in overall agreement with the laboratory study. The TDGM and related age-length function should be of considerable use to researchers studying HBC throughout the Colorado River Basin. Additionally, this case history should also be useful to anyone wishing to recover temperature- dependent bioenergetic parameters for fish using capture- recapture data, or to estimate the relationship and a<ssociated uncertainty between fish age and length using non-lethal techniques. This technique shows considerable promise in extracting useful information on fish growth from field data, rather than from laboratory studies, where such information is typically obtained. Incorporation of Ageing Error in ASMR Assessments A major criticism of the ASMR technique, as previously applied. is that it does not explicitly account for uncertainty in the assignment of age to individual tish (Kitchell and oth- ers. 20(3). As a result, abundance, recruitment, and mortality estimates may contain excessive bias. Additionally, estimates of precision are likely overstated by not incorporating this important source of uncertainty. The analyses presented in this report attempt to address these concems by constructing a more rigorous model to predict length as a function of age and to incorporate uncertainty fi'om age assignments into estimates of abundance and recruitment. Coggins and others (2006b) conducted sensitivity analyses on the effect of random ageing en-or and found little systematic bias in reconstructed recmit- ment trends. However, the cUlTent analysis is a more rigorous treatment of the problem and has two major implications. First, model results of estimated adult abundance are still very precise even when uncertainty in the assignment of age is explicitly accounted for in the assessment. Following reviews 15 by Kitchell and others (2003) and Otis and Wickham (U.S. Geological Survey, written commun., 2(06), this assessment lends additional credibility to results frol1l ASMR, indicating that it provides a rigorous measure of the state of the adult portion of the LCR HBC population. It is recommended that this assessment be considered "best available science" for use in contemplating management decisions both within the Glen Canyon Dam Adaptive M.magement Program and the U.S. Fish and Wildlife Service. Second, this analysis points out the difficulty that open population models generally have in the precise estimation of recruitment (Williams and others, 200 I; Pine and others, 20(3). Because many of the most critical management ques- tions for HBC center around how best to improve recruitment, pal1icularly considering improved rearing conditions in the mainstem Colorado River, it will be difficult for ASMR to detect statistically significant changes in recruitment, unless those changes are quite large. As a result, experimental adaptive management actions designed to increa<;e recruit- ment should consider first and foremost how to achieve large changes in recruitment. Small-scale experimental treatments of short duration, or so-called "mini-experiments," should be summarily discounted recognizing that the monitoring pro- gram is unlikely to detect small recruitment change even if it occurs. Additionally, multiyear experiments should be strongly favored in order to help offset not only unexpected and uncon- trollable effects, but the low precision in recruitment estimates. Acknowledgements For their thorough review of the GCMRC assessment methodology, I would like to thank members of the Hump- back Chub Assessment Review Panel: James F. Kitchell, Churchill Grimes, Steven T. Lindley, David Otis, and Carl Schwarz. Steve Martell is acknowledged for his help and council in developing ADMB programs and conceptualizing how to incorporate ageing elTor into the cun'ent assessment program. Carl Walters, Bill Pine, Mike Allen, and Tom Frazer are deserving of thanks for early reviews and suggestions that much improved this manuscript. I would also like to thank Tina Kister and Lara Schmit for publication assistance. Finally. and perhaps most deservedly, my thanks to field crews past and present whose tireless efforts provide the data upon which this assessment relies completely. References Akaike, H.x., 1973, Information theory as an extension of the maximum likelihood principle, in Petrov, B.N., and Csaki, F., eds., Intemational symposium on information theory, 2nd, Akademiai Kiado, Budapest, 1973 [Proceedings], p. 267-281.