Temporal Variation in Trout Populations: Implications for Monitoring and Trend Detection

—We summarized the temporal variation in populations of inland trout Salmo, Salvelinus, and Oncorhynchus spp. from streams in North America and determined the statistical power to detect trends over time. The coefficients of variation in abundance averaged 49% (SD1⁄4 27%; range1⁄4 15–108%) over time for all ages of trout. Temporal variation was lower when more age-classes were monitored, but whether abundance or biomass was more variable differed among populations. Detecting population trends was difficult when using the traditional a1⁄4 0.05 criterion. For example, detecting a 5% annual decline with good power (1 – b 0.80) would require about 20 years if only one site were monitored. Even when a was relaxed to 0.20, 15 years were required to detect a 5% annual decline when the variation was average. Using a network of sites improved the ability to detect changes: a 5% annual decline at a1⁄4 0.05 could be detected in 10 years when 30 sites were monitored. For high-value populations, it may require relaxing a to ensure that declines are detected, even if this increases the risk of claiming change when none has occurred and thus undertaking unnecessary management action. For example, a 5% annual decline could be detected with good power ( 0.80) in 8 years when a network of 30 sites is monitored at a1⁄40.20. Thus, biologists should monitor the least-variable component of a population, monitor a network of sites, and increase a for species of concern to ensure that real population trends are detected. Estimates of trend parameters (and their uncertainty) should be considered in addition to whether or not a statistical test for trend is significant. A pilot study or existing data can help estimate the variation that is typical of the population(s) to be monitored, determine whether trends can be reliably detected, and identify how much risk needs to be incurred to detect trends. Fisheries biologists often implement monitoring programs to detect trends in fish populations and habitat conditions (Firman and Jacobs 2001; Larsen et al. 2004; Wagner et al. 2007). Monitoring can be conducted at a single location to examine local populations or at multiple locations to assess regional trends (Larsen et al. 2001). For example, Lyons et al. (1996) monitored one site on each of several Wisconsin rivers to evaluate the effect of a minimum length limit on the abundance and size structure of smallmouth bass Micropterus dolomieu in each river. The Oregon Department of Fish and Wildlife monitors stream habitat, spawning salmon, and juvenile salmon at a network of sites to make inferences regarding regional trends in stream habitat and salmon populations (Urquhart and Kincaid 1999; Firman and Jacobs 2001). The temporal variation in fish population abundance is higher than that in many plant and animal populations that represent a variety of life history strategies (Gibbs et al. 1998). Natural population variation, sampling error, and sampling variation determine the amount of temporal variation observed. Natural variation results from environmental and demographic stochasticity. Although natural variation cannot be controlled, efforts should be made to reduce sampling error and sampling variation (Sulkava et al. 2007). Sampling error results from having to estimate the number or biomass of fish in a stream reach with methods, such as electrofishing, that are not 100% efficient (Link and Nichols 1994). Sampling error can be reduced by using a more efficient sampling gear, sampling more intensively, or using a more precise abundance estimator (Seber 1982; Riley and Fausch 1992; Bailey 2005). Sampling variation is caused by sampling a population at a different time each year or, for populations that are at a different point in an annual cycle when they are sampled (such as spawning, the timing of which can change from year to year), even by sampling at the same time each year. Sampling variation can be reduced by shortening the time window during which populations are sampled each year or by scheduling annual sampling to match an annual population cycle (Larsen et al. 2001). Temporal variation that occurs even when habitat * Corresponding author: [email protected] 1 Present address: Trout Unlimited, 910 Main Street, Suite 342, Boise, Idaho 83702, USA. Received July 20, 2007; accepted August 25, 2008 Published online February 5, 2009 38 Transactions of the American Fisheries Society 138:38–51, 2009 Copyright by the American Fisheries Society 2009 DOI: 10.1577/T07-154.1 [Article]