Attrition of bush trees after suburban development in Hobart, Australia

The patterns and correlates of native tree loss after suburbs are built in forest and woodland are poorly known yet can be important for nature conservation and scenic amenity. We use Hobart, Tasmania, Australia, to determine the rate of such tree loss and the factors that relate to this. Twenty suburbs in Hobart that had transitioned from native vegetation to suburbs between 1946 and 2021 were randomly sampled. Ten cadastral parcels that were native forest or woodland in 1946 were randomly selected from each suburb. Data on tree and housing density for years 1946, 1971, 1996 and 2021 were obtained from stereoscopic aerial photographs. Correlation and multiple regression were used to analyse the relationships between tree loss rates and house numbers, mean block size, biophysical variables, and sociodemographic variables. The mean annual percentage loss of trees between 1946 and 2021 was 1.03%, while data on tree decline after houses were built indicated a mean annual percentage loss of 0.18%. Loss between 1946 and 2021 at the suburb scale was predicted by a multiple regression model that constituted number of houses in 1996, mean household income and mean number of people per household. This model reflected the surge of tree removal with house building, the tendency of people with high income to have trees in their gardens and the tendency for new suburbs to be occupied by young families to a greater degree than old suburbs. The rate of original tree loss after house construction is low, but points to a need to provide a steady stream of substitute trees.