AN APPROACH FOR LONG-TERM MONITORING OF RECOVERING POPULATIONS OF NEARCTIC RIVER OTTERS (LONTRA CANADENSIS) IN THE SAN FRANCISCO BAY AREA, CALIFORNIA

Abstract

We present results from the first 5 y of a long-term monitoring study of recovering populations of the North American River Otter, Lontra canadensis, in Marin County, California. Historically present, but extirpated due to trapping and habitat loss, this apex aquatic predator is recolonizing coastal, lentic, and riverine habitat areas around San Francisco Bay. Using camera traps and a community science effort, we estimated an annual minimum population size at each of 14 focal study sites (FSS). Based on those estimates over 5 y, we developed a new Bayesian-statistics model to estimate population parameters for each FSS, including initial population abundance, annual rate of change in abundance, and probability that abundance is in decline. Our results show significantly different changes in abundance among the various FSS, with annual rates ranging from a high of 0.86 to a low of –0.44. Using a Random Forest framework, we then investigated the relative value of select spatial, environmental, and anthropogenic variables as predictors for each FSS population parameter. In our analysis spatial factors, and specifically latitude, were the best predictor of differences in FSS population parameters. Higher latitudes correlated with higher initial population abundance, greater annual increase in abundance, and lower probability that abundance is in decline. Our results provide new information about the rate and pattern of natural River Otter recolonization of areas from which they have been absent for decades. The results also serve as a demonstration of an approach to long-term monitoring, with the goal of increased understanding of the ecological function of River Otters in the San Francisco Bay Area.