Impacts of road mortality on the Western Rattlesnake (Crotalus oreganus) in British Columbia

Direct mortality due to wildlife-vehicle collisions has emerged as a major and worldwide conservation threat to wildlife. This source of mortality may be particularly adverse for populations persisting at the periphery of their range, where existing natural constraints already limit population growth and vigour. As a result, conservation assessment and planning for many peripheral species-at-risk will benefit from a fundamental understanding of the impacts of road mortality, yet these can be difficult to isolate due to the interaction of a number of factors. Using population viability analysis (PVA) I evaluated the persistence of a Western Rattlesnake (Crotalus oreganus) population threatened by road mortality in the dry interior of British Columbia, Canada. From 2015-2016 I quantified road mortality through methodical road surveys and coincidental assessments of scavenging rates and observer detection probability using planted snake carcasses. Additionally, I conducted intensive mark-recapture and radio-telemetry to estimate population density, size, and the range of the study population. After accounting for sources of error, my modelling showed that the estimated number of rattlesnake deaths was 2.7× the number of carcasses detected through unadjusted surveys and incidental observations. Overall, an estimated 6.6% of the population was killed on the road annually under traffic conditions that amounted to a maximum of only 350 vehicles per day. The PVA indicated that the population still was likely to persist for the next 100 years, but with a continual decline under the current, observed road mortality rate. With the loss of 6.6% of the population/year, the projected probability of extinction was <0.01 in 100 years and 0.0 in 50 years. At simulated road mortality rates of ≤6% there was zero probability of extinction for this population of rattlesnakes within the next 100 years. However, at the extinction threshold of road mortality of 6%, the stochastic growth rate was -0.032, and the mean population size was estimated to decrease by 96% in 100 years. Simulations with road mortality rates >6% consistently put the population at risk of extinction over 100 years. In comparison, the growth rate in the absence of road mortality was 0.0047 and the population was projected to increase (60% increase over 100 years). My results also suggest that in theory, improving adult female survival as well as overall longevity of rattlesnakes would significantly increase the population growth rate. My method of estimating population size for the area impacted by road deaths likely presents an overestimate, suggesting that the actual risk to this population is greater than what the models have implied. The detailed PVA using refined road mortality estimates provides strong evidence that road mortality is and will be a significant contributor to population decline, and adds to the growing body of evidence that large populations of long-lived species will face extirpation under low levels of road mortality, even in the absence of other sources of disturbance. Conservation priorities should focus on reducing road mortality and improving habitat availability away from roads.