No abstract

Rangers collect law enforcement monitoring (LEM) data during their patrols in protected areas. These data are increasingly used to interpret patrolling effectiveness and to predict poaching activity. However, LEM data can contain biases that may weaken the conclusions derived from such analyses. Research data, collected from 82 km of desnaring transects and interviews with 31 rangers, were compared with recorded LEM data. The latter included a logbook in which poacher sightings were documented, desnaring reports containing the locations of known snaring hotspots, and patrol strategies, containing the allocation of ranger patrols within the conservancy. Our findings suggest that the poaching prevalence reported through the LEM data is likely to constitute an underestimation of the true prevalence. Patrolling strategies were found to be predictable, allowing poachers to evade detection. One‐third of the interviewed rangers admitted to not reporting sighted poachers. We conclude that the use of LEM data for analysis or poaching prediction by managers of protected areas or researchers requires careful consideration of patrol predictability, possible displacement of poaching activity, and ranger culture and morale in order to avoid underestimation of true poaching prevalence.