Understanding the Significance of Redd Counts: a Comparison between Two Methods for Estimating the Abundance of and Monitoring Bull Trout Populations

—For salmonids that exhibit multiple life history forms within a single population, it may be necessary to evaluate the inconsistencies associated with population monitoring techniques. We compared mark– resight population estimates with those based on annual redd counts for bull trout Salvelinus confluentus in eastern Oregon. Our data suggest that across years, the trends in population estimates based on expanded redd count data and those based on the mark–resight method may be similar within basins. Across basins, however, there appear to be inconsistencies between mark–resight population estimates for different size-classes of bull trout and the expanded redd count data. In some systems, only the larger, potentially migratory fish are represented in redd counts, whereas in others some combination of small resident and large, potentially migratory fish is represented. The disparity between redd counts and population estimates for the reproductive population suggests that caution be invoked when choosing the monitoring techniques used to set recovery or monitoring goals for bull trout populations. The trend in abundance of the reproductive portion of a population is often one of the most important characteristics in the recovery and conservation of a species, yet the estimation of this trend can be a deceivingly difficult task. This task is especially complicated for species exhibiting multiple life history forms that coexist within a single population unit and that are behaviorally cryptic (e.g., bull trout Salvelinus confluentus; Maxell 1999). In addition to species-specific challenges, managers are often resource limited and must rely on monitoring techniques that (1) are cost and time effective, (2) are specifically focused on the demographic portion of the population of * Corresponding author: [email protected]. Received January 13, 2005; accepted June 13, 2005 Published online November 4, 2005 interest (e.g., reproductive adults), and (3) have not been adequately validated with alternative approaches. Given these limitations, the accuracy and precision of population monitoring techniques must be critically evaluated, as these factors can affect our ability to assess the impacts of implemented management and recovery actions on longterm species persistence. A considerable number of salmonid populations have experienced significant declines throughout their native ranges (Frissell 1993; Thurow et al. 1997). Bull trout have been listed as threatened in the United States since 1998 and have been designated as a species of special concern since 1995 in Canada. Bull trout require cold water temperatures (Selong et al. 2001; Dunham et al. 2003) and are often associated with complex habitats (Rieman and McIntyre 1993; Watson and Hillman 1997), which can limit the accuracy and precision of population-level monitoring. Like many other salmonid species (e.g., coastal cutthroat trout Oncorhynchus clarkii clarkii), bull trout are known to exhibit multiple life history forms, including resident and migratory, that can coexist within a single population (Rieman and McIntyre 1993, 1996; Nelson et al. 2002). Resident bull trout remain in their headwater natal streams for their entire life cycle and are thought to remain relatively small (,300 mm; Goetz 1989; Buchanan et al. 1997). Migratory bull trout remain in their natal streams for 1–3 years before migrating to a larger stream or lake or to the ocean and ultimately returning to their natal streams to spawn (Rieman and McIntyre