9588 - Laserfiche WebLink

Home Browse Search

that low capture probabilities at the extremes of the length ranges observed here are biased low because the length terms in the quadratic equation attempt to fit the few data at the extremes. The result is steep downward trajectories of the tails of the distribution of probability of capture values that are forced to near zero, where estimates are also likely imprecise. It may be reasonable to truncate those distributions so marginal values do not approach zero, but criteria to do such were lacking. While it is conceivable that more small fish exist in the system than abundance estimates suggested, it is unrealistic to speculate that large numbers of very large and old individuals are going undetected by our sampling efforts. Thus, bias of abundance estimates for adult Colorado pikeminnow caused by low capture probabilities for large individuals is assumed negligible. Transition rates.--Transition probabilities (0) reflected a general movement pattern of Colorado pikeminnow from the lower Green River and the Desolation-Gray Canyon reach upstream. This was not surprising given the abundance of relatively small Colorado pikeminnow in those reaches, including recruits, compared to other reaches. This pattern was also consistent with the Colorado River, where relatively small-bodied Colorado pikeminnow reared in lower sections of the river and eventually moved upstream as sub-adults or adults (Osmundson et al. 1998). Prevailing wisdom suggests that those mostly adult-sized fish would establish and maintain stable home ranges in most areas of the Green River Basin (e.g. Irving and Modde 2000). Perhaps drought or other conditions in some years, particularly 2002, forced fish to move from those relatively small streams such as the Yampa River to larger ones such as the middle Green River, a reach where net Ori was positive. However, we recommend against overinterpreting O's presented here, given that most are rather imprecise and likely represent only a few fish. 57