9466 - Laserfiche WebLink

Home Browse Search

TABLE 1. -Population estimate for humpback chub in Black Rocks, 1998-2000 using Model Mo. Year N 95% Confidence Interval Coefficient of Variation Capture Probability 1998; all fish 948 603-1,573 0.25 0.07 1998; adults only' 764 512-1,206 0.23 0.08 1999 921 723-1,208 0.13 0.09 2000 ----------------------- --- 539 223-1497 0.54 0.04 Mean (all fish) - 803 ------------ --- Mean (adults only) 741 -------------------- ------------------ --------------- a Excludes three fish <200 mm TL. only three fish (none of which were recaptured) were removed from the analysis. Difference in point estimates for the other models were 25-50 individuals (Table A-1). Coefficient of variation (CV; standard error divided by estimate) is often used to measure precision of population estimates, with values less than 0.20 generally considered to be necessary for a robust estimate (K. Bestgen, personal communication). Values of CV for the Mo estimate ranged from 0.54 in 2000 to 0.13 in 1999 (Table A-1). The low value was apparently related to the fourth sampling effort in 1999 (compared with three rotations in 1998 and 2000). The high value in 2000 was mostly related to reduced catch rate (capture probability was about one-half of what it had been in the two previous years; Table 1) and the elimination of the fourth sampling rotation. Although a fourth rotation was planned for that year, it was not done because of concern for potential cumulative negative effects of sampling on the humpback chub population. It is not clear why catch rates were so low in 2000, but the detrimental effect of sampling was considered a possibility. Comparison With ISMP Data Relative Abundance -Catch rates of humpback chub in Black Rocks have been variable since ISMP sampling began in 1988 (Figure 5); however, there was a noticeable downward 9