科研之友
微信
新浪微博
Facebook
分享链接
摘要
Freshwater habitat quality is broadly recognized as fundamental to the viability of salmonid populations. Temporal trends in freshwater habitat have rarely been quantified, however, perhaps owing to a lack of methodical and rigorous time series data sets. We present an approach for evaluating change in freshwater habitat using data from a long-term program to monitor salmonid populations and their habitats in coastal drainages of Oregon. Our goals were to (1) evaluate the presence and magnitude of an underlying linear trend in freshwater habitat condition across coastal watersheds in Oregon and (2) determine the effectiveness of the current sampling design for meeting the monitoring objectives. Four features were selected to characterize freshwater habitat: percent of pool area, large wood volume, quantity of fine sediment, and stream size. We developed a statistical model to describe the trend in these features that incorporated an error structure to account for site, year, and site-by-year variability. Spatial variability accounted for most of the overall variation, and temporal variability was minimal. Trends were detected among several of the habitat metrics and these varied by geographic region. To evaluate the efficacy of the sampling design, we generated simulated data sets with hypothesized trends of 1-2% per year and estimated trend detection power under two different survey designs, variance structures, and monitoring durations. We conclude that the power to detect trends is sensitive to the duration of the monitoring program and the structure and magnitude of the variance. The monitoring program was effective in detecting subtle trends while providing a robust data set with which to address multiple monitoring objectives. Such a monitoring program is critical to assessing the viability of salmonid populations in the Pacific Northwest and tracking recovery and conservation efforts.
- 出版日期2011-1
