Estimating animal numbers in the wild is important for decisive conservation and management. However, it is difficult to perform, especially for rare or elusive species because it often requires assumption and high costs, and it is also labor intensive. The Asiatic black bear (Ursus thibetanus), distributed in southern and eastern Asia, is threatened in much of its native habitat and is designated as “vulnerable” according to the IUCN Red List (Garshelis and Steinmetz 2008). In Japan, the bear is divided into 19 local populations based on habitat continuity (Yoneda 2001). The population in Kyushu is probably extinct, and five populations located mainly in western Honshu are considered threatened (Ministry of the Environment 2002). On the other hand, conflicts with humans, such as crop damage, and injury to humans have recently increased in most local bear populations in Japan because of changes or expansion in bear distribution (Oi and Yamazaki 2006). These events have led to heavy culling to prevent nuisance behavior, but the effect of such operations on population viability is becoming a cause for concern. Therefore, accurate evaluation of the current status of each local population is essential for conservation and management. In recent years, the mark-recapture model has been applied using a DNA fingerprinting method (genetic tagging) based on genetic material obtained from collected bear hair (Woods et al. 1999; Mowat and Strobeck 2000). Development of this investigation method known as a “hair-trap” appears to be effective for estimating bear populations, which has been an almost impossible task in the past. However, it has been reported that the method may have a potential weakness associated with high genotyping error rates due to the low quantity and quality of DNA (Gagneux et al. 1997; Goossens et al. 1998). In addition, the high cost of DNA analysis limits extensive and consistent monitoring. Consequently, we paid attention to the “camera-trap” method, which is relatively simple, cost effective, and easy to perform. It also has great advantages in field studies for animal conservation and management. Recently, this method that identifies individuals based on natural markings such as coat patterns from photographs has been used frequently for estimating mainly felid populations (e.g., Foster and Harmsen 2011; O’Connell et al. 2011). The Asiatic black bear is well known to have a crescent-shaped chest mark which is considered to be unique to each bear (Higashide et al. 2012). In other words, if chest marks can be photographed by a cameratrap, this method will be applicable for estimating bear populations and for researching other ecological topics. However, the chest mark is difficult to photograph, because it is not very visible while the bear is walking, in contrast to many other animals which have natural marks over their whole body. Although Ngoprasert et al. (2010) managed to photograph some chest marks in their preliminary study, their results were unsuccessful. Consequently, it is necessary to develop a practical camera-trap technique for effectively photographing chest marks. In addition, a method must be devised for inducing a bear into a posture that enables their chest mark to be clearly photographed. We devised two types of camera-trap as part of a study to identify individual Asiatic black bears by natural markings as a prelude to estimating population size. This paper describes the design of the traps and assesses their utility. It also evaluates the photography success rate and photographic quality.
Read full abstract