122 



Discussion 



The naCure eind limitations of the study should be recognized. 

 First of all, the time series of each locality is called a survey, 

 not a popuiation or stock (Simon and Larkin, 1972), because the 

 authors presently have no basis to make this judgement. While the 

 authors are optimistic that most of the surveys described as "wild 

 Chinook" and "hatchery Chinook" will ultimately turn out to be 

 indicators of self-sustaining breeding populations, or stocks, some 

 may not. Questions of how to aggregate the individual groups of 

 spawners described in these surveys (i.e., Table l) into self- 

 sustaining stocks have not been answered here, however such 

 recomm en dations axe the subject of much current research, including 

 the authors. Secondly, the integrity of the survey data has been 

 assumed without direct discussion with the agencies involved. The 

 assuu^tion of integrity means that the locality surveyed and the 

 survey methods have remained the same throughout the time span of 

 the surveys. 



The statistical methods were chosen in the belief that annual 

 observations within a locality were conq^arable from year to year. 

 Further, while differences in the type of data among localities 

 make con^jarisons among localities difficult or in5)ossible, the 

 methods permit the slopes of regressions to be compared. Surveys 

 reporting in units of redd counts do not produce measures of 

 abundance that are necessaxily proportional to total population 

 size in the drainage basin, since the standard reach probably 

 contains an inconstant portion of the total spawning population 

 each year. However, if the same reach is counted each year, then 

 the annual redd counts do accurately reflect the population of that 

 reach, and that reach alone. Since the standard index reaches are 

 usually the best spawning groxinds accessible to the surveyors, 

 trends in redd coimts should be excellent measures of the status of 

 wild Chinook populations in general. So, while it is recognized 

 that the combined redd count data are not necessarily proportional 

 to total Chinook eibundance in the Columbia River Basin as a whole, 

 the redd counts are considered by the authors to be directly 

 prop<3Ttional to Chinook spawning abundance in the reaches surveyed. 



It is recognized that escapement data may not fully reflect 

 fluctuations in total population size because of una ccounted 

 removals by fisheries, and other sources of human- induced mortality 

 (see Symons and Waldichuk 1984; Schwartzberg and Roger 1986) . From 

 the standpoint of measuring stock status of very small chinook 

 populations, the authors view this as an academic concern. The 

 purpose of this effort is to identify siit^jle measures that may be 

 related to stock status, and to inform on the status of these 

 measures by species and locality, not to identify the causes of 

 population declines. If numbers of spawners sufficient to sustain 

 the populations axe not being recruited annually, in the end it 

 makes very little difference why the populations have disappeared. 

 The authors hope that this type of work can help to point out where 



