Wf. Laurance et Gn. Harrington, ECOLOGICAL ASSOCIATIONS OF FEEDING SITES OF FERAL PIGS IN THE QUEENSLAND WET TROPICS, Wildlife research, 24(5), 1997, pp. 579-590
Data from 152 plots (0.8 ha) and 659 small quadrats (0.04 ha) were use
d to assess rooting activity by feral pigs in forest communities in no
rth Queensland. Study sites spanned the rainforest-sclerophyll-forest
gradient along the western margin of the wet tropics region. Detailed
floristic, physiognomic and edaphic data were recorded for each plot a
nd used to develop a predictive model of pig activity in these habitat
s. The most striking result was that rooting activity varied markedly
among different forest types. Wet sclerophyll forests consistently had
the greatest area disturbed, followed by mesic and dry sclerophyll fo
rests. Both rainforest and rainforest-invaded sclerophyll forests had
relatively low activity levels. There were some differences in rooting
activity among different geographic regions, but few effects of local
topography, soil type or proximity to water. A mathematical model was
developed to predict the ecological associations of pig rooting activ
ity, using generalised linear modeling. Pig rooting was associated wit
h certain attributes of wet sclerophyll forests and with slopes and ri
dge tops, but the model had limited effectiveness, with fitted values
explaining 16% of the actual variation in rooting activity. This may h
ave resulted because microhabitat preferences of pigs varied among dif
ferent forest types and seasons. We suggest that pigs could be consumi
ng fungal fruit-bodies in sclerophyll forests, and if so they may comp
ete for food with some native, mycophagous mammals.