The local extinction of habitat patches and dispersal between the patches are important processes structuring animal populations in heterogeneous environments. Understanding these two processes is crucial for the conservation of wildlife populations in landscapes impacted by human land-use. Approximately 50% of the subtropical rainforest in South East Queensland, Australia has been lost to deforestation over the last 100 years. While large areas of rainforest are reserved, little is known about the distribution and population status of rainforest birds within smaller remnants in the region. The overall research problem for this thesis was to understand how deforestation and fragmentation of subtropical rainforest affects the occurrence of rainforest birds and the effective dispersal of a rainforest-restricted species, the logrunner (Orthonyx temminckii). Understanding why some bird species are lost from habitat patches while others remain will lead to improved conservation of extinction prone species in fragmented landscapes. Although the mechanisms underlying local extinctions are well established in temperate systems, the relative importance of local and regional processes on species occurrence in subtropical and tropical rainforests is poorly understood. Chapter 2 investigated the relative effects of life history and scale of habitat modification on avian site occupancy using observational data collected at 46 rainforest sites in South East Queensland. A probabilistic model for the joint site occupancy of 29 bird species was used to evaluate hypotheses for the effects of avian life history traits on the occurrence of multiple species. The single-species occurrence models incorporated habitat effects on detection, which may be especially important in rainforests because dense vegetation and idiosyncratic occurrence of species can interfere with sampling. Occupancy rates for each species were modelled to determine the relative influence of process operating at the stand, landscape and patch scales. The life history analysis indicated taxonomic Family, body mass, migratory strategy and feeding strata had large effects on avian site occupancy, whereas abundance traits such as mean density and extent of occurrence showed little predictive ability. After accounting for correlated extinction risk attributed to life history, the degradation of stand structure at the local scale was more important for species richness than habitat modification at landscape or patch scales. While individual species showed various responses to the different scales of habitat modification, the distribution of many species was limited by vegetation structure at the landscape scale. Maintaining stand basal area and restoring degraded rainforests at the local scale will increase the probability of occupancy for members of the rainforest bird community. However, revegetation and retention of forest cover at the landscape scale may be necessary for the successful colonisation of many species. Chapter 3 introduced a predictive hypothesis-driven approach for quantifying the relative contribution of historic and contemporary processes to genetic connectivity. Current analytic frameworks in population genetics have difficulty evaluating meaningful hypotheses about spatial processes in dynamic landscapes. Confronting genetic data with models of historic and contemporary landscapes allowed the identification of dispersal processes operating in naturally heterogeneous and human-altered systems. Two measures of indirect gene flow were estimated from microsatellite polymorphism among 11 logrunner populations. Of particular interest was how much information in the genetic data was attributable to processes occurring in a reconstructed historic landscape and a contemporary human-modified landscape. A linear mixed model was used to estimate appropriate sampling variance from non-independent data and information-theoretic model selection provided strength of evidence for alternate hypotheses. The historic and contemporary landscapes explained an equal proportion of variation in genetic differentiation and there was considerable evidence for a temporal shift in dispersal pattern. Migration rates estimated from genealogical information were primarily influenced by contemporary landscape change. Landscape heterogeneity appeared to facilitate gene flow prior to European settlement, but contemporary deforestation is rapidly becoming the most important barrier to logrunner dispersal. Understanding asymmetric dispersal is becoming an important consideration for the conservation metapopulations. Populations acting as net exporters of dispersing animals may be able to rescue local populations from extinction and allow metapopulations to persist in degraded landscapes impacted by habitat loss. In Chapter 4, I estimated bidirectional migration rates from genetic data to infer dispersal among 11 logrunner populations. The first question posed was, does logrunner dispersal correspond to the source-sink or balanced model of dispersal? The second question involved determining the strength of evidence for two hypotheses about how landscape structure has affected asymmetric dispersal. Hypothesis one proposed that asymmetric dispersal was primarily influenced by naturally occurring habitat heterogeneity. Hypothesis two asserted that asymmetric dispersal was predominantly influenced by anthropogenic landscape change. The data were confronted with the alternate hypotheses using linear mixed models and landscape covariates extracted from digital maps. The results showed the direction of asymmetric dispersal was consistent with source-sink population structure. I also discovered that the asymmetry in dispersal was influenced more by anthropogenic landscape change than by naturally occurring habitat heterogeneity. Intact landscapes were net exporters of dispersing logrunners while landscapes heavily impacted by rainforest clearing were net importers of individuals. Elevated immigration rates into landscapes impacted by rainforest clearing appeared to arrest population declines in accordance with the rescue effect. The primary conclusion emerging from the study of patch occupancy and dispersal was that logrunner populations in South East Queensland conformed to a mainland-island metapopulation. Asymmetric dispersal from the largest expanse of upland rainforest appeared to prevent fragmented rainforests in close proximity from going locally extinct. While the distribution of logrunners was limited by the spatial configuration of rainforest patches, other rainforest birds exhibited variable responses to scale of habitat modification. The most consistent pattern was several species dropping-out of the community in degraded stands affected by selective timber harvest. Deforestation at the landscape scale also played a role in the extremely low patch occupancy rates of Albert’s lyrebirds (Menura alberti) and green catbirds (Ailuroedus crassirostris).

This thesis fills a research gap in our knowledge of kokako song by looking at how song evolves in multi-dialect areas. Kokako only exist in small remnant populations separated by large tracts of unsuitable land. Kokako are very poor flyers, only able to fly for approximately 100 metres at a time. In order to prevent inbreeding and a loss of genetic diversity in the remaining kokako populations, managers are carrying out translocations to establish new populations, maintain gene flow and prevent inbreeding. However, these translocations have the potential to be unsuccessful because kokako exhibit macrogeographic variation in their dialects, and tend not to breed with individuals who do not share the same dialect as them. If the purpose of the translocation is to enhance genetic diversity by having kokako from different areas breed then song is an important factor that must be dealt with. Song is extremely important to kokako for a number of reasons. Chiefly, it is hypothesised that song is their primary means of territory defence. A kokako gains all its resources from its territory, so it is imperative that they successfully defend it. Their duet song functions in territory defence, but also acts to form and maintain pair bonds. In order to try and address the problem that song causes in translocations I studied if and how song evolves in a multiple dialect area. I conducted research at Pukaha Mount Bruce, where there is a population of kokako originating from two different source populations, and thus two different dialects, Northern Mapara and Mangatutu. I looked at the Northern Mapara dialect, and recorded kokako belonging to three groups; kokako currently living in the source population, kokako that were translocated to Pukaha and kokako that were born at Pukaha. In order to determine how the song may be changing I looked at element repertoires, the levels of sharing between groups, the number of unique elements in repertoires and the syntactical and temporal characteristics of phrases. This study shows that translocation into multiple dialect areas can affect kokako song. Translocation did not affect the size of the kokako repertoire, but it seemed to affect the amount of sharing within and between different groups of kokako. Currently the level of sharing within the groups at Pukaha is lower than the level of sharing within the source Mapara population, indicating that the song may be diverging. There also seem to be more unique elements found at Pukaha, which in part explains the lack of sharing. There appears to be microgeographic variation at Pukaha, with birds clustered around the second (Mangatutu) dialect sharing less with the source population than do those kokako whose territories are lower down in the reserve. The phrases which are used are also evolving, with only one phrase truly shared among all groups. There are other phrases which show additions or deletions of elements, and so are evolving. The main change found in the phrases is the timing between elements, with six out of seven phrases examined showing changes. These results have repercussions for future kokako translocations, and the future of the Pukaha kokako. A low amount of phrase and element type sharing combined with changes of intra-phrase timing could lead to the Pukaha kokako’s inability to successfully defend their territories. This research shows how kokako song can give conservation managers information on the status of their populations in regards to interbreeding and raises questions which can be answered by further research, both at Pukaha and in other mixeddialect kokako populations.

This thesis fills a research gap in our knowledge of kokako song by looking at how song evolves in multi-dialect areas. Kokako only exist in small remnant populations separated by large tracts of unsuitable land. Kokako are very poor flyers, only able to fly for approximately 100 metres at a time. In order to prevent inbreeding and a loss of genetic diversity in the remaining kokako populations, managers are carrying out translocations to establish new populations, maintain gene flow and prevent inbreeding. However, these translocations have the potential to be unsuccessful because kokako exhibit macrogeographic variation in their dialects, and tend not to breed with individuals who do not share the same dialect as them. If the purpose of the translocation is to enhance genetic diversity by having kokako from different areas breed then song is an important factor that must be dealt with. Song is extremely important to kokako for a number of reasons. Chiefly, it is hypothesised that song is their primary means of territory defence. A kokako gains all its resources from its territory, so it is imperative that they successfully defend it. Their duet song functions in territory defence, but also acts to form and maintain pair bonds. In order to try and address the problem that song causes in translocations I studied if and how song evolves in a multiple dialect area. I conducted research at Pukaha Mount Bruce, where there is a population of kokako originating from two different source populations, and thus two different dialects, Northern Mapara and Mangatutu. I looked at the Northern Mapara dialect, and recorded kokako belonging to three groups; kokako currently living in the source population, kokako that were translocated to Pukaha and kokako that were born at Pukaha. In order to determine how the song may be changing I looked at element repertoires, the levels of sharing between groups, the number of unique elements in repertoires and the syntactical and temporal characteristics of phrases. This study shows that translocation into multiple dialect areas can affect kokako song. Translocation did not affect the size of the kokako repertoire, but it seemed to affect the amount of sharing within and between different groups of kokako. Currently the level of sharing within the groups at Pukaha is lower than the level of sharing within the source Mapara population, indicating that the song may be diverging. There also seem to be more unique elements found at Pukaha, which in part explains the lack of sharing. There appears to be microgeographic variation at Pukaha, with birds clustered around the second (Mangatutu) dialect sharing less with the source population than do those kokako whose territories are lower down in the reserve. The phrases which are used are also evolving, with only one phrase truly shared among all groups. There are other phrases which show additions or deletions of elements, and so are evolving. The main change found in the phrases is the timing between elements, with six out of seven phrases examined showing changes. These results have repercussions for future kokako translocations, and the future of the Pukaha kokako. A low amount of phrase and element type sharing combined with changes of intra-phrase timing could lead to the Pukaha kokako’s inability to successfully defend their territories. This research shows how kokako song can give conservation managers information on the status of their populations in regards to interbreeding and raises questions which can be answered by further research, both at Pukaha and in other mixeddialect kokako populations.

This thesis fills a research gap in our knowledge of kokako song by looking at how song evolves in multi-dialect areas. Kokako only exist in small remnant populations separated by large tracts of unsuitable land. Kokako are very poor flyers, only able to fly for approximately 100 metres at a time. In order to prevent inbreeding and a loss of genetic diversity in the remaining kokako populations, managers are carrying out translocations to establish new populations, maintain gene flow and prevent inbreeding. However, these translocations have the potential to be unsuccessful because kokako exhibit macrogeographic variation in their dialects, and tend not to breed with individuals who do not share the same dialect as them. If the purpose of the translocation is to enhance genetic diversity by having kokako from different areas breed then song is an important factor that must be dealt with. Song is extremely important to kokako for a number of reasons. Chiefly, it is hypothesised that song is their primary means of territory defence. A kokako gains all its resources from its territory, so it is imperative that they successfully defend it. Their duet song functions in territory defence, but also acts to form and maintain pair bonds. In order to try and address the problem that song causes in translocations I studied if and how song evolves in a multiple dialect area. I conducted research at Pukaha Mount Bruce, where there is a population of kokako originating from two different source populations, and thus two different dialects, Northern Mapara and Mangatutu. I looked at the Northern Mapara dialect, and recorded kokako belonging to three groups; kokako currently living in the source population, kokako that were translocated to Pukaha and kokako that were born at Pukaha. In order to determine how the song may be changing I looked at element repertoires, the levels of sharing between groups, the number of unique elements in repertoires and the syntactical and temporal characteristics of phrases. This study shows that translocation into multiple dialect areas can affect kokako song. Translocation did not affect the size of the kokako repertoire, but it seemed to affect the amount of sharing within and between different groups of kokako. Currently the level of sharing within the groups at Pukaha is lower than the level of sharing within the source Mapara population, indicating that the song may be diverging. There also seem to be more unique elements found at Pukaha, which in part explains the lack of sharing. There appears to be microgeographic variation at Pukaha, with birds clustered around the second (Mangatutu) dialect sharing less with the source population than do those kokako whose territories are lower down in the reserve. The phrases which are used are also evolving, with only one phrase truly shared among all groups. There are other phrases which show additions or deletions of elements, and so are evolving. The main change found in the phrases is the timing between elements, with six out of seven phrases examined showing changes. These results have repercussions for future kokako translocations, and the future of the Pukaha kokako. A low amount of phrase and element type sharing combined with changes of intra-phrase timing could lead to the Pukaha kokako’s inability to successfully defend their territories. This research shows how kokako song can give conservation managers information on the status of their populations in regards to interbreeding and raises questions which can be answered by further research, both at Pukaha and in other mixeddialect kokako populations.

The local extinction of habitat patches and dispersal between the patches are important processes structuring animal populations in heterogeneous environments. Understanding these two processes is crucial for the conservation of wildlife populations in landscapes impacted by human land-use. Approximately 50% of the subtropical rainforest in South East Queensland, Australia has been lost to deforestation over the last 100 years. While large areas of rainforest are reserved, little is known about the distribution and population status of rainforest birds within smaller remnants in the region. The overall research problem for this thesis was to understand how deforestation and fragmentation of subtropical rainforest affects the occurrence of rainforest birds and the effective dispersal of a rainforest-restricted species, the logrunner (Orthonyx temminckii). Understanding why some bird species are lost from habitat patches while others remain will lead to improved conservation of extinction prone species in fragmented landscapes. Although the mechanisms underlying local extinctions are well established in temperate systems, the relative importance of local and regional processes on species occurrence in subtropical and tropical rainforests is poorly understood. Chapter 2 investigated the relative effects of life history and scale of habitat modification on avian site occupancy using observational data collected at 46 rainforest sites in South East Queensland. A probabilistic model for the joint site occupancy of 29 bird species was used to evaluate hypotheses for the effects of avian life history traits on the occurrence of multiple species. The single-species occurrence models incorporated habitat effects on detection, which may be especially important in rainforests because dense vegetation and idiosyncratic occurrence of species can interfere with sampling. Occupancy rates for each species were modelled to determine the relative influence of process operating at the stand, landscape and patch scales. The life history analysis indicated taxonomic Family, body mass, migratory strategy and feeding strata had large effects on avian site occupancy, whereas abundance traits such as mean density and extent of occurrence showed little predictive ability. After accounting for correlated extinction risk attributed to life history, the degradation of stand structure at the local scale was more important for species richness than habitat modification at landscape or patch scales. While individual species showed various responses to the different scales of habitat modification, the distribution of many species was limited by vegetation structure at the landscape scale. Maintaining stand basal area and restoring degraded rainforests at the local scale will increase the probability of occupancy for members of the rainforest bird community. However, revegetation and retention of forest cover at the landscape scale may be necessary for the successful colonisation of many species. Chapter 3 introduced a predictive hypothesis-driven approach for quantifying the relative contribution of historic and contemporary processes to genetic connectivity. Current analytic frameworks in population genetics have difficulty evaluating meaningful hypotheses about spatial processes in dynamic landscapes. Confronting genetic data with models of historic and contemporary landscapes allowed the identification of dispersal processes operating in naturally heterogeneous and human-altered systems. Two measures of indirect gene flow were estimated from microsatellite polymorphism among 11 logrunner populations. Of particular interest was how much information in the genetic data was attributable to processes occurring in a reconstructed historic landscape and a contemporary human-modified landscape. A linear mixed model was used to estimate appropriate sampling variance from non-independent data and information-theoretic model selection provided strength of evidence for alternate hypotheses. The historic and contemporary landscapes explained an equal proportion of variation in genetic differentiation and there was considerable evidence for a temporal shift in dispersal pattern. Migration rates estimated from genealogical information were primarily influenced by contemporary landscape change. Landscape heterogeneity appeared to facilitate gene flow prior to European settlement, but contemporary deforestation is rapidly becoming the most important barrier to logrunner dispersal. Understanding asymmetric dispersal is becoming an important consideration for the conservation metapopulations. Populations acting as net exporters of dispersing animals may be able to rescue local populations from extinction and allow metapopulations to persist in degraded landscapes impacted by habitat loss. In Chapter 4, I estimated bidirectional migration rates from genetic data to infer dispersal among 11 logrunner populations. The first question posed was, does logrunner dispersal correspond to the source-sink or balanced model of dispersal? The second question involved determining the strength of evidence for two hypotheses about how landscape structure has affected asymmetric dispersal. Hypothesis one proposed that asymmetric dispersal was primarily influenced by naturally occurring habitat heterogeneity. Hypothesis two asserted that asymmetric dispersal was predominantly influenced by anthropogenic landscape change. The data were confronted with the alternate hypotheses using linear mixed models and landscape covariates extracted from digital maps. The results showed the direction of asymmetric dispersal was consistent with source-sink population structure. I also discovered that the asymmetry in dispersal was influenced more by anthropogenic landscape change than by naturally occurring habitat heterogeneity. Intact landscapes were net exporters of dispersing logrunners while landscapes heavily impacted by rainforest clearing were net importers of individuals. Elevated immigration rates into landscapes impacted by rainforest clearing appeared to arrest population declines in accordance with the rescue effect. The primary conclusion emerging from the study of patch occupancy and dispersal was that logrunner populations in South East Queensland conformed to a mainland-island metapopulation. Asymmetric dispersal from the largest expanse of upland rainforest appeared to prevent fragmented rainforests in close proximity from going locally extinct. While the distribution of logrunners was limited by the spatial configuration of rainforest patches, other rainforest birds exhibited variable responses to scale of habitat modification. The most consistent pattern was several species dropping-out of the community in degraded stands affected by selective timber harvest. Deforestation at the landscape scale also played a role in the extremely low patch occupancy rates of Albert’s lyrebirds (Menura alberti) and green catbirds (Ailuroedus crassirostris).

The local extinction of habitat patches and dispersal between the patches are important processes structuring animal populations in heterogeneous environments. Understanding these two processes is crucial for the conservation of wildlife populations in landscapes impacted by human land-use. Approximately 50% of the subtropical rainforest in South East Queensland, Australia has been lost to deforestation over the last 100 years. While large areas of rainforest are reserved, little is known about the distribution and population status of rainforest birds within smaller remnants in the region. The overall research problem for this thesis was to understand how deforestation and fragmentation of subtropical rainforest affects the occurrence of rainforest birds and the effective dispersal of a rainforest-restricted species, the logrunner (Orthonyx temminckii). Understanding why some bird species are lost from habitat patches while others remain will lead to improved conservation of extinction prone species in fragmented landscapes. Although the mechanisms underlying local extinctions are well established in temperate systems, the relative importance of local and regional processes on species occurrence in subtropical and tropical rainforests is poorly understood. Chapter 2 investigated the relative effects of life history and scale of habitat modification on avian site occupancy using observational data collected at 46 rainforest sites in South East Queensland. A probabilistic model for the joint site occupancy of 29 bird species was used to evaluate hypotheses for the effects of avian life history traits on the occurrence of multiple species. The single-species occurrence models incorporated habitat effects on detection, which may be especially important in rainforests because dense vegetation and idiosyncratic occurrence of species can interfere with sampling. Occupancy rates for each species were modelled to determine the relative influence of process operating at the stand, landscape and patch scales. The life history analysis indicated taxonomic Family, body mass, migratory strategy and feeding strata had large effects on avian site occupancy, whereas abundance traits such as mean density and extent of occurrence showed little predictive ability. After accounting for correlated extinction risk attributed to life history, the degradation of stand structure at the local scale was more important for species richness than habitat modification at landscape or patch scales. While individual species showed various responses to the different scales of habitat modification, the distribution of many species was limited by vegetation structure at the landscape scale. Maintaining stand basal area and restoring degraded rainforests at the local scale will increase the probability of occupancy for members of the rainforest bird community. However, revegetation and retention of forest cover at the landscape scale may be necessary for the successful colonisation of many species. Chapter 3 introduced a predictive hypothesis-driven approach for quantifying the relative contribution of historic and contemporary processes to genetic connectivity. Current analytic frameworks in population genetics have difficulty evaluating meaningful hypotheses about spatial processes in dynamic landscapes. Confronting genetic data with models of historic and contemporary landscapes allowed the identification of dispersal processes operating in naturally heterogeneous and human-altered systems. Two measures of indirect gene flow were estimated from microsatellite polymorphism among 11 logrunner populations. Of particular interest was how much information in the genetic data was attributable to processes occurring in a reconstructed historic landscape and a contemporary human-modified landscape. A linear mixed model was used to estimate appropriate sampling variance from non-independent data and information-theoretic model selection provided strength of evidence for alternate hypotheses. The historic and contemporary landscapes explained an equal proportion of variation in genetic differentiation and there was considerable evidence for a temporal shift in dispersal pattern. Migration rates estimated from genealogical information were primarily influenced by contemporary landscape change. Landscape heterogeneity appeared to facilitate gene flow prior to European settlement, but contemporary deforestation is rapidly becoming the most important barrier to logrunner dispersal. Understanding asymmetric dispersal is becoming an important consideration for the conservation metapopulations. Populations acting as net exporters of dispersing animals may be able to rescue local populations from extinction and allow metapopulations to persist in degraded landscapes impacted by habitat loss. In Chapter 4, I estimated bidirectional migration rates from genetic data to infer dispersal among 11 logrunner populations. The first question posed was, does logrunner dispersal correspond to the source-sink or balanced model of dispersal? The second question involved determining the strength of evidence for two hypotheses about how landscape structure has affected asymmetric dispersal. Hypothesis one proposed that asymmetric dispersal was primarily influenced by naturally occurring habitat heterogeneity. Hypothesis two asserted that asymmetric dispersal was predominantly influenced by anthropogenic landscape change. The data were confronted with the alternate hypotheses using linear mixed models and landscape covariates extracted from digital maps. The results showed the direction of asymmetric dispersal was consistent with source-sink population structure. I also discovered that the asymmetry in dispersal was influenced more by anthropogenic landscape change than by naturally occurring habitat heterogeneity. Intact landscapes were net exporters of dispersing logrunners while landscapes heavily impacted by rainforest clearing were net importers of individuals. Elevated immigration rates into landscapes impacted by rainforest clearing appeared to arrest population declines in accordance with the rescue effect. The primary conclusion emerging from the study of patch occupancy and dispersal was that logrunner populations in South East Queensland conformed to a mainland-island metapopulation. Asymmetric dispersal from the largest expanse of upland rainforest appeared to prevent fragmented rainforests in close proximity from going locally extinct. While the distribution of logrunners was limited by the spatial configuration of rainforest patches, other rainforest birds exhibited variable responses to scale of habitat modification. The most consistent pattern was several species dropping-out of the community in degraded stands affected by selective timber harvest. Deforestation at the landscape scale also played a role in the extremely low patch occupancy rates of Albert’s lyrebirds (Menura alberti) and green catbirds (Ailuroedus crassirostris).

This thesis fills a research gap in our knowledge of kokako song by looking at how song evolves in multi-dialect areas. Kokako only exist in small remnant populations separated by large tracts of unsuitable land. Kokako are very poor flyers, only able to fly for approximately 100 metres at a time. In order to prevent inbreeding and a loss of genetic diversity in the remaining kokako populations, managers are carrying out translocations to establish new populations, maintain gene flow and prevent inbreeding. However, these translocations have the potential to be unsuccessful because kokako exhibit macrogeographic variation in their dialects, and tend not to breed with individuals who do not share the same dialect as them. If the purpose of the translocation is to enhance genetic diversity by having kokako from different areas breed then song is an important factor that must be dealt with. Song is extremely important to kokako for a number of reasons. Chiefly, it is hypothesised that song is their primary means of territory defence. A kokako gains all its resources from its territory, so it is imperative that they successfully defend it. Their duet song functions in territory defence, but also acts to form and maintain pair bonds. In order to try and address the problem that song causes in translocations I studied if and how song evolves in a multiple dialect area. I conducted research at Pukaha Mount Bruce, where there is a population of kokako originating from two different source populations, and thus two different dialects, Northern Mapara and Mangatutu. I looked at the Northern Mapara dialect, and recorded kokako belonging to three groups; kokako currently living in the source population, kokako that were translocated to Pukaha and kokako that were born at Pukaha. In order to determine how the song may be changing I looked at element repertoires, the levels of sharing between groups, the number of unique elements in repertoires and the syntactical and temporal characteristics of phrases. This study shows that translocation into multiple dialect areas can affect kokako song. Translocation did not affect the size of the kokako repertoire, but it seemed to affect the amount of sharing within and between different groups of kokako. Currently the level of sharing within the groups at Pukaha is lower than the level of sharing within the source Mapara population, indicating that the song may be diverging. There also seem to be more unique elements found at Pukaha, which in part explains the lack of sharing. There appears to be microgeographic variation at Pukaha, with birds clustered around the second (Mangatutu) dialect sharing less with the source population than do those kokako whose territories are lower down in the reserve. The phrases which are used are also evolving, with only one phrase truly shared among all groups. There are other phrases which show additions or deletions of elements, and so are evolving. The main change found in the phrases is the timing between elements, with six out of seven phrases examined showing changes. These results have repercussions for future kokako translocations, and the future of the Pukaha kokako. A low amount of phrase and element type sharing combined with changes of intra-phrase timing could lead to the Pukaha kokako’s inability to successfully defend their territories. This research shows how kokako song can give conservation managers information on the status of their populations in regards to interbreeding and raises questions which can be answered by further research, both at Pukaha and in other mixeddialect kokako populations.

Conservation in New Zealand has begun to focus heavily on the restoration of degraded mainland ecosystems and the reintroduction of native species that have become locally extinct. In many cases the individuals that are selected for reintroduction are harvested from ‘mammal-free’ offshore islands. This thesis examines the effects of isolation from mammalian predators on the predator avoidance behaviours and predator recognition abilities of New Zealand birds using the North Island robin as a model. It also investigates whether any effects of isolation from mammalian predators has a lasting impact on mainland populations founded by individuals from offshore islands. Nest site selection behaviours were compared across three populations that are exposed to different suites of predators and have differing translocation histories; Benneydale, Tiritiri Matangi and Wenderholm. Point height intercept and point-centred quarter surveys were used to compare habitat availability between the sites and to compare nest sites with the available habitat. Eight nest characteristic variables were also compared across the three sites using a principle component analysis. Benneydale nests were located higher in the trees and were more concealed than nests at the other two sites. Nests on Tiritiri Matangi were supported by large numbers of thin branches and were located toward the periphery of the nest tree. Unfortunately these differences are very difficult to interpret due to a high degree of variation in the habitat types present at the three sites. The anti-predator behaviours initiated in response to a model stoat, model morepork and control were used to test the ability of nesting robins to recognise the threat that each of these treatments might pose to nest success. Behavioural variables were compared between Benneydale, Tiritiri Matangi and Wenderholm using a response intensity scoring system and a principle component analysis. The results indicated that isolation from mammalian predators on Tiritiri Matangi has suppressed the ability of robins on the island to recognise the predatory threat posed by a stoat. They also suggest that the intense mammal control carried out at Wenderholm may have inhibited the ability of local robins to produce strong anti-predator responses when faced with a stoat.

Conservation in New Zealand has begun to focus heavily on the restoration of degraded mainland ecosystems and the reintroduction of native species that have become locally extinct. In many cases the individuals that are selected for reintroduction are harvested from ‘mammal-free’ offshore islands. This thesis examines the effects of isolation from mammalian predators on the predator avoidance behaviours and predator recognition abilities of New Zealand birds using the North Island robin as a model. It also investigates whether any effects of isolation from mammalian predators has a lasting impact on mainland populations founded by individuals from offshore islands. Nest site selection behaviours were compared across three populations that are exposed to different suites of predators and have differing translocation histories; Benneydale, Tiritiri Matangi and Wenderholm. Point height intercept and point-centred quarter surveys were used to compare habitat availability between the sites and to compare nest sites with the available habitat. Eight nest characteristic variables were also compared across the three sites using a principle component analysis. Benneydale nests were located higher in the trees and were more concealed than nests at the other two sites. Nests on Tiritiri Matangi were supported by large numbers of thin branches and were located toward the periphery of the nest tree. Unfortunately these differences are very difficult to interpret due to a high degree of variation in the habitat types present at the three sites. The anti-predator behaviours initiated in response to a model stoat, model morepork and control were used to test the ability of nesting robins to recognise the threat that each of these treatments might pose to nest success. Behavioural variables were compared between Benneydale, Tiritiri Matangi and Wenderholm using a response intensity scoring system and a principle component analysis. The results indicated that isolation from mammalian predators on Tiritiri Matangi has suppressed the ability of robins on the island to recognise the predatory threat posed by a stoat. They also suggest that the intense mammal control carried out at Wenderholm may have inhibited the ability of local robins to produce strong anti-predator responses when faced with a stoat.

The local extinction of habitat patches and dispersal between the patches are important processes structuring animal populations in heterogeneous environments. Understanding these two processes is crucial for the conservation of wildlife populations in landscapes impacted by human land-use. Approximately 50% of the subtropical rainforest in South East Queensland, Australia has been lost to deforestation over the last 100 years. While large areas of rainforest are reserved, little is known about the distribution and population status of rainforest birds within smaller remnants in the region. The overall research problem for this thesis was to understand how deforestation and fragmentation of subtropical rainforest affects the occurrence of rainforest birds and the effective dispersal of a rainforest-restricted species, the logrunner (Orthonyx temminckii). Understanding why some bird species are lost from habitat patches while others remain will lead to improved conservation of extinction prone species in fragmented landscapes. Although the mechanisms underlying local extinctions are well established in temperate systems, the relative importance of local and regional processes on species occurrence in subtropical and tropical rainforests is poorly understood. Chapter 2 investigated the relative effects of life history and scale of habitat modification on avian site occupancy using observational data collected at 46 rainforest sites in South East Queensland. A probabilistic model for the joint site occupancy of 29 bird species was used to evaluate hypotheses for the effects of avian life history traits on the occurrence of multiple species. The single-species occurrence models incorporated habitat effects on detection, which may be especially important in rainforests because dense vegetation and idiosyncratic occurrence of species can interfere with sampling. Occupancy rates for each species were modelled to determine the relative influence of process operating at the stand, landscape and patch scales. The life history analysis indicated taxonomic Family, body mass, migratory strategy and feeding strata had large effects on avian site occupancy, whereas abundance traits such as mean density and extent of occurrence showed little predictive ability. After accounting for correlated extinction risk attributed to life history, the degradation of stand structure at the local scale was more important for species richness than habitat modification at landscape or patch scales. While individual species showed various responses to the different scales of habitat modification, the distribution of many species was limited by vegetation structure at the landscape scale. Maintaining stand basal area and restoring degraded rainforests at the local scale will increase the probability of occupancy for members of the rainforest bird community. However, revegetation and retention of forest cover at the landscape scale may be necessary for the successful colonisation of many species. Chapter 3 introduced a predictive hypothesis-driven approach for quantifying the relative contribution of historic and contemporary processes to genetic connectivity. Current analytic frameworks in population genetics have difficulty evaluating meaningful hypotheses about spatial processes in dynamic landscapes. Confronting genetic data with models of historic and contemporary landscapes allowed the identification of dispersal processes operating in naturally heterogeneous and human-altered systems. Two measures of indirect gene flow were estimated from microsatellite polymorphism among 11 logrunner populations. Of particular interest was how much information in the genetic data was attributable to processes occurring in a reconstructed historic landscape and a contemporary human-modified landscape. A linear mixed model was used to estimate appropriate sampling variance from non-independent data and information-theoretic model selection provided strength of evidence for alternate hypotheses. The historic and contemporary landscapes explained an equal proportion of variation in genetic differentiation and there was considerable evidence for a temporal shift in dispersal pattern. Migration rates estimated from genealogical information were primarily influenced by contemporary landscape change. Landscape heterogeneity appeared to facilitate gene flow prior to European settlement, but contemporary deforestation is rapidly becoming the most important barrier to logrunner dispersal. Understanding asymmetric dispersal is becoming an important consideration for the conservation metapopulations. Populations acting as net exporters of dispersing animals may be able to rescue local populations from extinction and allow metapopulations to persist in degraded landscapes impacted by habitat loss. In Chapter 4, I estimated bidirectional migration rates from genetic data to infer dispersal among 11 logrunner populations. The first question posed was, does logrunner dispersal correspond to the source-sink or balanced model of dispersal? The second question involved determining the strength of evidence for two hypotheses about how landscape structure has affected asymmetric dispersal. Hypothesis one proposed that asymmetric dispersal was primarily influenced by naturally occurring habitat heterogeneity. Hypothesis two asserted that asymmetric dispersal was predominantly influenced by anthropogenic landscape change. The data were confronted with the alternate hypotheses using linear mixed models and landscape covariates extracted from digital maps. The results showed the direction of asymmetric dispersal was consistent with source-sink population structure. I also discovered that the asymmetry in dispersal was influenced more by anthropogenic landscape change than by naturally occurring habitat heterogeneity. Intact landscapes were net exporters of dispersing logrunners while landscapes heavily impacted by rainforest clearing were net importers of individuals. Elevated immigration rates into landscapes impacted by rainforest clearing appeared to arrest population declines in accordance with the rescue effect. The primary conclusion emerging from the study of patch occupancy and dispersal was that logrunner populations in South East Queensland conformed to a mainland-island metapopulation. Asymmetric dispersal from the largest expanse of upland rainforest appeared to prevent fragmented rainforests in close proximity from going locally extinct. While the distribution of logrunners was limited by the spatial configuration of rainforest patches, other rainforest birds exhibited variable responses to scale of habitat modification. The most consistent pattern was several species dropping-out of the community in degraded stands affected by selective timber harvest. Deforestation at the landscape scale also played a role in the extremely low patch occupancy rates of Albert’s lyrebirds (Menura alberti) and green catbirds (Ailuroedus crassirostris).