Spatial distribution of browsing damage and mammalian herbivores in Tasmanian eucalypt plantations

The Bennett's wallaby (Macropus rufogriseus), Tasmanian pademelon (Thylogale billardierii), common brushtail possum (Trichosurus vulpecula) and European rabbit (Oryctolagus cuniculus) can damage eucalypt seedlings growing in forestry plantations. The browsing damage caused by these herbivores is spatially variable, both within and between plantations. This study tests the hypothesis that the severity of browsing damage and the relative abundance of herbivores within a plantation, declines with increasing distance from adjacent cover habitat (remnant bushland or established plantation) and windrows. Randomly located damage plots (five seedlings) and scat plots were established at 15 plantations at the time of planting. Browsing damage was assessed at six and 12 months after planting. Scat counts were carried out from planting to 320 days. Correlation analyses indicated that browsing damage levels were related to distance from cover habitat in some, but not all, plantations. Damage scores measured at six months were negatively correlated with distance from cover habitat at four plantations (r=0.373). Damage scores measured at 12 months were negatively correlated with distance from cover habitat at five plantations (r=0.456). Damage was never correlated with distance from windrows. There was very little evidence to suggest that herbivores used areas close to cover habitat or windrows in preference to areas that were far from these habitat features. Significant correlation between the number of scats that were found in a plot and the distance from cover habitat was observed at only one plantation (pademelon: r=0.603). There was no correlation between scat numbers and distance from windrow. These results suggest that there will often be little scope for reducing the economic impact of browsing damage, using methods weighted towards reducing levels of damage at the perimeter of the plantation. However, it may occasionally be possible to target edges in certain plantations, provided that the site characteristics conducive to edge effects are identified.