Spatiotemporal Stability Patterns of Brook Trout Abundance and Implications for Stream Research and Monitoring

High spatial and temporal variation in many stream resident trout populations complicates basic sampling questions. We used 4 years of census data from 3 km of stream to examine (1) spatial and temporal stability in Brook Trout Salvelinus fontinalis populations and the implications for representative reaches, (2) stream lengths necessary to estimate abundance and biomass, and (3) comparative performance of site selection methods. Sections of highest relative abundance and biomass were consistently high across years and had significantly lower coefficients of variation. Greater pool area and depth were associated with higher temporal stability, which suggests that sections with the best habitats can serve as effective representative reaches for trend monitoring. However, extrapolating from reference reaches grossly overestimated the census size. Randomizing site sections generally required less stream to estimate population size at a desired precision, but the required stream lengths were sensitive to the census size; less stream was required in years of higher abundances. Regardless of the year, sampling at least 10% of the available stream length was required under most scenarios, with some requiring substantially more. Our results highlight trade-offs between the information needs versus the effort required and the importance of tailoring sampling strategy to the question.