Conservation Genetics of Alpine Rock Wren (Xenicus gilviventris)

Many species occur in naturally subdivided populations due to spatial heterogeneity of the landscape. Such a pattern is especially evident in alpine species, where naturally fragmented habitat forms an ‘alpine archipelago’. High altitude habitat patches and the species they harbour can serve as effective models for monitoring global change processes in sensitive ecosystems. The rock wren (Xenicus gilviventris) is a threatened alpine passerine belonging to the endemic New Zealand wren family (Acanthisittidae). This ancient family was once represented by at least seven species, however due to the impacts of introduced mammalian predators, only two species remain. Conservation management of rock wren has only recently commenced via the translocation of individuals to offshore islands, but genetic considerations are not currently a part of management practices. In this thesis, I investigated the role of genetic factors in the conservation management of rock wren and applied my findings to improve understanding of the species’ ecology and better inform future management efforts. I sampled rock wren (n=221) from throughout their range and using 14 microsatellite markers combined with nuclear and mitochondrial DNA sequence data, describe significant differences in genetic variation and differentiation between rock wren populations across the South Island. A deep North–South genetic divergence was evident (3.7 ± 0.5% at cytochrome b), consistent with the ‘biotic gap’ hypothesis whereby Northern and Southern populations became restricted in ice-free refugia during the Pleistocene era of extensive glaciation c. 2 mya. There is some evidence for a larger refugium within the south of the rock wren’s range, as estimates of genetic variation and long-term effective population size are consistently larger for the Southern lineage. Although this finding may also be indicative of more optimal habitat in the south of the species’ range; supporting a higher density of rock wren long-term. Designation of Northern and Southern rock wren lineages as separate evolutionarily significant units (ESUs) is proposed. Estimates of the long-term effective population size of rock wren are dramatically larger relative to contemporary estimates, indicating that in the past, rock wren sustained a much higher abundance than today. Whilst a genetic signature linking population decline within the Northern lineage to a timeframe of anthropogenic disturbance (i.e. the past c. 100 years) was not detected, there is some evidence for a recent population bottleneck within this timeframe in the South. This suggests that although natural historical climate fluctuations have clearly played an important role driving patterns of rock wren abundance in the past, these impacts are now being compounded by much more recent anthropogenic impacts, most likely, predation by introduced mammalian predators. Significant fine-scale spatial genetic structure in rock wren was also detected, and a strong pattern of isolation by distance whereby genetic relatedness among neighbouring individuals is significantly greater than that among more distant or randomly located individuals. This pattern of gene flow is indicative of a stepping stone model of dispersal. A potential sex-bias in dispersal, suggestive of male natal philopatry, was also detected which may have further contributed to the strong pattern of fine-scale structuring. The spatial scale of positive genetic structure or ‘genetic patch size’ (i.e. the distance over which individuals were not genetically independent) was unexpectedly large (c.70 km) given the rock wren’s limited flight ability. Asymmetrical gene flow is also evident among populations within the Southern lineage, indicative that source-sink dynamics are operating. The Murchison Mountains appear to be a particularly important source of migrants for other populations. Therefore, management efforts, such as predator control, to ensure this population is conserved should be prioritised. Conversely, the Upper Hollyford and Lake Roe populations appear to be functioning as sink populations, with migration occurring into, but not out of these areas. By improving habitat quality in these areas (e.g. by controlling invasive species) there is potential that they may be converted from sinks into new source populations