Hollow selection by vertebrate fauna in forests of southeastern Australia and implications for forest management

We examined the types of hollows, and types of hollow-bearing trees, occupi ed by vertebrate fauna in temperate eucalypt forests in southeastern Austra lia. Hollow-bearing trees are selected for retention in wood production for ests to mitigate the effects of logging on hole-nesting fauna. A total of 4 71 hollows was examined in 228 trees felled as part of routine logging oper ations. Fauna had occupied 43% of all hollows (greater than or equal to 2cm minimum entrance width; greater than or equal to 5cm depth). Hollows with small (2-5cm), medium (6-10cm) and large ( > 10cm) minimum entrance widths had occupancy rates of 29, 44 and 62%, respectively. The internal dimension s of hollows, especially hollow depth, were the best predictors of hollow o ccupancy, even when variables measured at the tree and site levels were con sidered. Fauna occupied 57% of all hollow-bearing trees. In a Poisson regre ssion model, the number of hollows in trees that contained evidence of occu pancy was positively associated with: (1) the total number of hollows visib le in the tree; (2) the proportion of the tree's crown that contained dead branches; and (3) tree diameter. The number of different vertebrate species that occupied trees was positively associated with the same explanatory va riables except tree diameter. Thus, our results suggest that trees with mul tiple hollows and dead branches in the crown should be preferentially selec ted for occupancy by hollow-using fauna. Our results suggest that trees wit h the largest diameter are not the most suitable for retention. Possibly be cause they contain proportionally fewer hollows with small entrances, which are favoured by some vertebrate species. (C) 2001 Elsevier Science Ltd. Al l rights reserved.