Robust estimation of snare prevalence within a tropical forest context using N-mixture models

Abstract

Hunting with snares is indiscriminate and wasteful, and this practice is currently one of the gravest threats to terrestrial vertebrates in the tropics. However, as snares are difficult to detect and often dispersed widely across large, inaccessible areas it is problematic to reliably estimate their prevalence and no standard survey methods exist. Conservation managers need reliable, timely, information on the spatio-temporal patterns of hunting and on responses to interventions, and we present an innovative sampling and analysis framework that allows for the rigorous estimation of snare detectability and 'abundance', but which can be feasibly implemented in challenging field contexts. This new approach was used to undertake a large-scale systematic snare survey in Keo Seima Wildlife Sanctuary, in Eastern Cambodia, and the resulting data were analysed using a novel application of N-mixture models. A range of environmental and management factors were examined as potential determinants of snare abundance and detectability, and proximity to the Vietnamese border was shown to be overwhelmingly the most influential factor. Snares were more common in the wet season rather than the dry season, and the detection probability of snares was shown to be low (~ 0.33), as predicted. No clear relationships between snaring levels, anti-poaching patrol effort and ungulates densities were evident from these data. There was clear evidence that certain factors, such as the percentage of dense forest cover, will exert confounding effects on both detectability and abundance, highlighting the critical need to take account of the imperfect detection when designing threat monitoring systems.

Ex situ management as insurance against extinction of mammalian megafauna in an uncertain world

“…2018; O'Kelly et al. 2018). Likewise, armed conflicts often lead to withdrawal of conservation funding by international development programs (Hanson et al.…”

Malignant Snare Traps Threaten an Irreplaceable Megafauna Community

“…Unlike gun-hunting, snares remain a threat long after the hunter has departed the forest and they consistently evade detection by even the most experienced antipoaching teams (Gray et al, 2018;Moore et al, 2018). Among Cambodian patrol teams, for example, snare detection probabilities were 0.28-0.36, which suggests that the magnitude of this problem is considerably greater than observed (O'Kelly et al, 2018). Thus, we further recommend implementing patrols and subsequent analyses that account for imperfect detection to avoid the risk of biased interpretations resulting from sites where snares were present, but undetected.…”

Conservation status, ex situ priorities and emerging threats to small carnivores

“…The overall declines of traded wildlife in faunally impoverished countries, such as Vietnam, are causing increases in hunting in other regions: in border areas adjacent to demand countries (e.g. O'Kelly et al ., 2018; Gray & Gauntlett, 2019) and as far as Latin America (Reuter et al ., 2018; Verheij, 2019). There are Chinese‐ and Vietnamese‐led criminal networks that are illegally trafficking high‐value wildlife and plants from Africa, and more recently from Latin America, to markets in Asia.…”