Integrating encounter theory with decision analysis to evaluate collision risk and determine optimal protection zones for wildlife

摘要

1.Better understanding human–wildlife interactions and their links with management can help improve the design of wildlife protection zones.One example is the problem of wildlife collisions with vehicles or human-built structures (e.g.,power lines,wind farms).In fact,collisions between marine wildlife and watercraft are among the major threats faced by several endangered species of marine mammals.Natural resource managers are therefore interested in finding cost-effective solutions to mitigate these threats.2.We combined abundance estimators with encounter rate theory to estimate relative lethal collision risk of the Florida manatee (Trichechus manatus latirostris) from watercraft.We first modelled seasonal abundance of watercraft and manatees using a Bayesian analysis of aerial survey count data.We then modelled relative lethal collision risk in space and across seasons.Finally,we applied decision analysis and Linear Integer Programming to determine the optimal design of speed zones in terms of relative risk to manatees and costs to waterway users.We used a Pareto efficient frontier approach to evaluate the performance of alternative zones,which included additional practical considerations (e.g.,spatial aggregation of speed zones) in relation to the optimal zone configurations.3.Under the various relationships for probability of death given strike speed that we considered,the current speed zones reduced the relative lethal collision risk by an average of 51.5% to 70.0% compared to the scenario in which all speed regulations were removed (i.e.,the no-protection scenario).We identified optimal zones and near-optimal zones with additional management considerations that improved upon the current zones in terms of cost or relative risk.4.Policy implications.Our analytical framework combines encounter rate theory and decision analysis to quantify the effectiveness of speed zones in protecting manatees while accounting for uncertainty.Our approach can be used to optimize