Identifying drivers of population dynamics for a stream breeding amphibian using time series of egg mass counts

Abstract/Summary

The decline of amphibian populations is one of the starkest examples of the biodiversity crisis. For stream-breeding amphibians, alteration of natural flow regimes by dams, water diversions, and climate change have been implicated in declines and extirpations. Identifying drivers of amphibian declines requires long time series of abundance data because amphibian populations can exhibit high natural variability. Multiple population viability analysis (MPVA) models integrate abundance data and share information from different populations to estimate how environmental factors influence population growth. Flow alteration has been linked to declines and extirpations in the Foothill Yellow-Legged Frog (Rana boylii), a stream-breeding amphibian native to California and Oregon. To date, no study has jointly analyzed abundance data from populations throughout the range of R. boylii in an MPVA model. We compiled time series of egg mass counts (an index of adult female abundance) from R. boylii populations in 36 focal streams and fit an MPVA model to quantify how streamflow metrics, stream temperature, and surrounding land cover affect population growth. We found population growth was positively related to stream temperature and was higher in the years following a wet year with high total annual streamflow. Density-dependence was weakest (i.e., carrying capacity was highest) for streams with high seasonality of streamflow and intermediate rates of change in streamflow during the spring. Our results highlight how altered streamflow can further increase the risk of decline for R. boylii populations. Managing stream conditions to better match natural flow and thermal regimes would benefit the conservation of R. boylii populations.