Patterns of genetic diversity and differentiation.
We found that each of the disturbance processes (natural vs linear transport infrastructure) resulted in a significant decrease in allelic richness (\({}^{0}D\)) but saw no changes in Ho, He, \({}^{1}D\) and\({}^{2}D\) (see Table 1). The population decline caused by ‘natural’ mortality events resulted in a significant loss in \({}^{0}D\) (Wilcoxon test: \({}^{0}D\_dataset1\)vs \({}^{0}{D\_}dataset2\), p-value < 0.0001). This was also the case for the subsequent population decline caused by the linear transport infrastructure project (Wilcoxon test;\({}^{0}D\_dataset2\)vs \({}^{0}{D\_}dataset3\_above\), p-value < 0.0001;\({}^{0}D\_dataset2\) vs \({}^{0}{D\_}dataset3\_below\), p-value < 0.0001). Importantly, the subdivision caused by the linear transport infrastructure resulted in the population of koalas located above the linear transport infrastructure having a significantly smaller\({}^{0}D\) than its neighbouring population of koalas located below the linear transport infrastructure (Wilcoxon test:\({}^{0}D\_dataset3\_above\) vs \({}^{0}{D\_}dataset3\_below\), p-value < 0.0001). It is also important to note that the percentage of monomorphic loci increased at each stage of the disturbance process: from 1.07% to 2.65% because of the population decline caused by ‘natural’ mortality events, and from 2.65% to 8.07% (above) and 4.27% (below) because of the subsequent population decline caused by the linear transport infrastructure project (i.e. translocations and population subdivision, Table 1). The extent of genetic differentiation between those genetic datasets, however, was negligible (Appendix 1, Table S1).