Laserfiche WebLink
<br />242 <br /> <br />COGGINS ET AL. <br /> <br />8,000 <br />.~7,000 <br />t 6,000 <br />~ 5,000 <br />'84,000 <br />.... <br />~3,OOO <br />S 2,000 <br />::s <br />Z 1,000 <br />o <br /> <br />--+-- Data through 1996 <br />:::~ Data through 1997 <br />---<>-- Data through 1998 <br />_Data through 1999 <br />. .. .... Data through 2000 <br />- . - . - Data through 200 I <br />- Data through 2002 <br /> <br /> <br />~ <br />~ <br /> <br />9::,0, <br />~ <br /> <br />~'\. <br />~ <br /> <br />cu 16,000 <br />.!::! 14000 <br />Vl ' <br />g 12,000 <br />.~ 10,000 <br />'[ 8,000 <br />&. 6,000 <br />'3 4,000 <br />~ 2,000 <br />o <br /> <br /> <br />9::,0, <br />'\.0, <br /> <br />~'\. <br />~ <br /> <br />~,." <br />~ <br /> <br />~,." <br />'\.~ <br /> <br />~" <br />'\.~ <br /> <br />~'\ <br />~ <br /> <br />~o, <br />'\.~ <br /> <br />Brood Year <br /> <br />--+-- Data through 1996 <br />- Data through 1997 <br />---<>-- Data through I 998 <br />_ Data through 1999 <br />-- h ... Data through 2000 <br />-.-.- Data through 2001 <br />- Data through 2002 <br /> <br /> <br />~" <br />.~ <br /> <br />~'\ <br />~ <br /> <br />~o, <br />~ <br /> <br />~'\. <br />');~ <br /> <br />Year <br /> <br />FIGURE 6.-Retrospective analysis of age-2 humpback chub recruinnent by brood year (top panel) and adult (age-4 and older) <br />humpback chub abundance (bottom panel) using the first of the three annual age-structured, open-population capture-recapture <br />models (ASMR 1). The trend lines reflect analyses using annually truncated data sets beginning in 1989 and ending as specified <br />in the legend. <br /> <br />based on different total numbers of years of sample <br />data. This retrospective analysis (Figure 6) indicates <br />that the recruitment estimates for the early years (brood <br />years 1987-1992) of the sampling program are quite <br />stable (i.e., are not altered by including more years of <br />data), but that the estimates for the later sampling years <br />are more variable. Additionally, the adult abundance <br />estimates display more variability in both trend and <br />magnitude and generally suggest lower initial abun- <br />dance (1989) and higher terminal abundance (2001) <br />when more years of data are considered. This trend in <br />adult abundance is a result of changes in sampling <br />intensity over time that essentially cause the model to <br />trade off higher mortality estimates with lower capture <br />probability during the years of low sampling intensity <br />(1996-1999). As the analysis is supplied with more <br />recapture data during the 2000--2002 time period, <br />mortality estimates fall and the trend in adult <br />abundance becomes less severe. Although the retro- <br /> <br />spective analysis does not suggest any severe structural <br />problem in model formulation, as there is a general <br />convergence in recruitment and abundance estimates <br />during the early years of data collection, it does further <br />reinforce the need to minimize the variability in <br />sampling effort over time to minimize assessment <br />errors resulting from parameter confounding. <br />All of the evidence suggests that the LCR humpback <br />chub population suffered a major decline in adult <br />abundance between the late 1980s and the present. All <br />of the evidence also points to major declines in <br />recruitment sometime in the 1980s. Though it is <br />difficult to believe that the adult population size <br />remained stable over the 1990s despite these re- <br />cruitment declines, changes in sampling intensity <br />between the early 1990s and early 2000s make it <br />impossible to categorically deny that the adult <br />population size has been stable since 1990. Efforts to <br />improve monitoring since 2000 (e.g., baited hoop nets) <br />