馬來西亞婆羅洲沙巴低地雨林馬來熊（Helarctos malayanus）之棲地利用

Abstract

The Malayan sun bear (Helarctos malayanus),the least studied animal of the Ursidae family, is maily distributed in the tropical rainforests of Southeast Asia. Studies of its habitat use in the wild are particularly rare and with inconsistent results. The objective of this study was to ascertain the pattern of habitat use of the Malayan sun bear in the Malaysian state of Sabah. This investigation of sun bear habitat use was conducted through strip transect between May 2006 and January 2007 in lowland rainforests of Borneo, around the Danum Valley Field Center, in the Ulu Segama Forest Reserve. The 200 km2 study area was divided into 100 sampling units of 1 km-wide and 2 km-long grids. Within each sampling unit, a 10 m-wide and 400 m-long strip transect for a sign survey was then selected. This research relied on an actual sample size of 94 strip transects for its data analysis on the presence (including numerical data) or absence of bear. It analyzed 15 factors, i.e. slope, elevation, distance to the river, distance to active logging road, distance to inactive road, density of active logging road, density of inactive road, index of trail, canopy cover, ground vegetation cover, large tree density, forest type, dead wood density, termite nest density, density of the genus of Ficus. Bear presence was noted in 75 strip transects, i.e. 80% of all strip transects; while bear absence characterized the remaining 19 strip transects, i.e. 20% of all strip transects. Of bear claw marks, 29% occurred on trees in the primary forest, for an average density of claw mark on tree of 4.94 trees/ha; 71% occurred on trees in the logged forest, with an average density of 8.77 trees/ha. For 60% of all transects, claw marks were concentrated on one to three trees. By differentiating between absence (n = 19) and presence (n = 75) of bear claw marks in transects for the above-mentioned 15 factors, it revealed that the transects for which there was presence of claw marks on trees were the ones with fewer trails (Mann-Whitney U Test, p = 0.028), a lower large tree density (p = 0.029), and with less primary forest (p = 0.008). The Modified Ivlev’s Electivity Index revealed that the Malayan sun bear avoided trails, had a preference for lower densities of large trees and for logged forest. I then proceeded with a logistic regression model, with presence or absence of Malayan sun bear, and the Omnibus test of model coefficients was significant (p = 0.008), yet only forest type could be considered as significant explanatory factor (p = 0.011). In a generalized linear model, with numerical data of Malayan sun bear, the Likelihood Ratio Chi-square test of model coefficients was not really significant (p＝0.059), except for the forest type variable which could be considered as a significant explanatory factor (p = 0.016). The results indicat that the Malayan sun bear is not really a forest-interior species. I suggest that, in addition to addressing the importance of primary rainforests for conservation, with well designed logging practices for quality secondary succession and with a limited impact from human activity, logged forest could become an important habitat for wildlife.