Fault detection and diagnosis is quite important in engineering systems, and deserves further attention in view of the increasing complexity of modern machinery. Traditional single-sensor methods of fault monitoring and diagnosis may find it difficult to meet modern industrial requirements because there is usually no direct way to measure and accurately correlate a machine fault to a single sensor output. Fusion of information from multiple sensors can overcome this shortcoming. In this thesis, a neural-fuzzy approach of multi-sensor fusion is developed for a network-enabled remote fault diagnosis system. The approach is validated by applying it to an industrial machine called the Iron Butcher, which is a machine used in the fish processing industry for the removal of the head in fish prior to further processing for canning. An important characteristic of the fault diagnosis approach developed in this thesis is to make an accurate decision of the machine condition by fusing information from different sensors. First, sound, vibration and vision signals are acquired from the machine using a microphone, an accelerometer and a digital CCD camera, respectively. Second, the sound and vibration signals are transformed into the frequency domain using fast Fourier transformation (FFT). A feature vector from the FFT frequency spectra is defined and extracted from the acquired information. Also, a feature based vision tracking approach—the Scale Invariant Feature Transform (SIFT)—is applied to the vision data to track the object of interest (fish) in a robust manner. Third, Sound, vibration and vision feature vectors are provided as inputs to a neuro-fuzzy network for fault detection and diagnosis. A four-layer neural network including a fuzzy hidden layer is developed in the thesis to analyze and diagnose existing faults. By training the neural network with sample data for typical faults, faults of five crucial components in the fish cutting machine are detected with high reliability and robustness. Alarms to warn about impending faults may be generated as well during the machine operation. A network-based remote monitoring architecture is developed as well in the thesis, which will facilitate engineers to monitor the machine condition in a more flexible manner from a remote site. Developed multi-sensor approaches are validated using computer simulations and physical experimentation with the industrial machine, and compared with a single-sensor approach.

Research Doctorate - Doctor of Philosophy (PhD) / Environmental sustainability through the successful management of the impacts of forestry related activities on native wildlife in New South Wales (NSW) is a core requirement for Forests NSW, the government authority overseeing timber harvesting in the state. Populations of all native species must be protected adequately to allow their survival into the foreseeable future. A major proportion of the timber harvesting in NSW occurs along the coast and adjacent ranges of central and northern NSW. Timber is extracted using selective logging (not clearfelling) and the impacts of this activity on wildlife ameliorated through the use of strategies such as the placement of buffers around riparian sites, and the retention of hollow bearing trees, downed woody debris and undisturbed corridors of vegetation. This region, encompassing Latitudes 30o to 34o South, is dominated by native temperate wet and dry sclerophyll forests distributed along a gradient from coastal lowlands to montane areas over 1000 m. There are two families of anurans present in this region, the hylidae (approximately 25 species) and myobatrachidae (approximately 20 species), and are considered to have evolved in the region over at least 60 million years. Understanding of the effectiveness of protective measures requires a confident understanding of the habitat requirements of the frogs present in the forests so that they are used in appropriate locations. Presenting scientifically based information on the response of frogs to disturbances with these measures in place is also of primary interest to provide indications as to the effectiveness of these measures. Studies of pond breeding frogs in this region were undertaken to provide information on their habitat requirements and so direct the appropriate placement of the required protective measures. Tests were also undertaken to assess if there is an evident negative relationship between forestry disturbances and the use of ponds by frogs, indicating if current protective measures are effective. Current monitoring data was also investigated to provide insights into successful monitoring strategies for frogs which will provide the opportunity to more effectively demonstrate ecologically sustainable management of frogs into the future. Finally, records available on the distributions of frogs were used to study relationships between frog records and the current protected area system to assess its value as a means of protecting the frog species in the study region and throughout Australia. The majority of the research work undertaken was based on surveys of anurans at 93 semi-permanent and permanent ponds found in four areas: The Watagan Mountains, Bulahdelah, Wauchope and Dorrigo. Each site was surveyed on multiple occasions between 2001 and 2008, using aural and visual nocturnal searches conducted over different seasons to provide long-term information on the species using the sites. The presence and abundance of each species at each pond on each occasion was recorded as was micrometeorological data during the surveys. Each site was also visited in the day-time to collect data on variables associated with both the breeding pond and with the habitat within an 80 m radius of the site. A basic understanding of the extent and types of habitats individual frogs use is an important first step in understanding the extent of area requiring management for ongoing conservation. Published studies of frog movements were collated and reviewed to provide data on the likely habitat requirements and patterns of habitat use of frogs. The data, obtained from a 68 studies covering 50 species world-wide, indicating that in most cases (39 of 57 studies) adults have both specific breeding and non-breeding habitats. Mean distances moved from breeding to non-breeding habitats ranged from 38-1810 m and means for different anuran families falling around 300 m. Individuals in non-breeding habitat tended to restrict their activity to core areas (means from 6.3 m2 to 5099m2). This review indicated the strong need to consider both the breeding ponds and the surrounding habitat in the management of frogs. The calling seasons assigned by field guides to anuran species present within the study region are both highly variable and often vague. Therefore an objective assessment of the calling seasons of the frogs in eastern mesic NSW was developed to more accurately determine the months during which each species is most likely to call – the core calling period – and so be best surveyed. This was defined for each species as the months that covered > 90% of calling records. The 17,461 records covering 67 species provided enough information to reasonably determine the core calling periods for 46 species. Of these, 43 had clearly definable core calling seasons centred on the spring-summer months, two called year-round and one had an uncertain calling season, but probably calls year-round. Increasing latitude usually, but not always, led to a small reduction in the core calling period. This information confirmed the appropriate timing of surveys used in this study and should improve the outcomes of future research, management, and conservation. Micrometeorlogical variables such as temperature and rainfall are well recognised for their potentially significant influence on the calling activity of anurans, but this effect has been little tested in Australia. The calling records of frogs obtained through the study were compared with the micrometeorological variables temperature, humidity and (previous) 24 and 72 hour rainfall to assess how these related to calling activity. Models obtained for nine species all found calling activity to relate to at least two of these four variables. Calling showed a positive relationship with 72 hour rainfall in eight species, but no relationship for one species. Calling showed a positive relationship with 24 hour rainfall for six species, a negative relationship for two species and no relationship for one species. Temperature and humidity were found to have a mixture of positive, negative or no relationships with calling being evident. Rainfall over the previous 72 hours appears to be the best predictor consider when planning surveys. Other factors such as social environment and circannual rhythms are likely to be interacting with micrometeorology; hence the complex results. Further work is required to understand them fully. The results do indicate that surveys undertaken for this study were suitably timed to provide good results. Understanding if there are specific features of available habitats that correlate strongly with the presence/absence of frogs is important for managing frogs. This information may identify which breeding sites specific frogs are likely to use (or use in greater numbers) and habitat elements that may be critical for their survival. It may also provide indications of how human induced modifications may be affecting the frogs in an area. The presence/absence and abundance of anurans was recorded at 45 ponds in the Watagan Mountains, 100 km north of Sydney, and compared with 24 habitat variables. Only nine species occupied > 33% of the ponds. Canonical Correspondence Analysis (CCA) indicated there was no strong community structure. Species richness at a site was related significantly to depth and % emergent vegetation, but neither provided great explanatory power, based on their parameter estimates. Total anuran abundance related significantly to 12 variables that accounted for substantial variation in the data, but all contributed small amounts. Models for the four commonest species found many variables with statistically significant relationships to the frog counts, but neither pond attributes nor surrounding habitat attributes explained large amounts of variance in the data. There was little commonality in the habitat variables significant for the four species. This study indicated that no single variable could be said to influence general anuran presence and abundance and that habitat may not play a large role in determining the presence or abundance of anurans in this system. The initial study was confined to a relatively small geographic area to minimise differences resulting from variations in altitude, climate and geology. This was followed with a generalised linear model analysis that covered the 93 sites in the four forest blocks. The comparison with the recorded anurans and habitat also considered variation in altitude, geology and habitat types and included GIS derived variables covering more landscape habitat features around the ponds (eg, moisture indices and solar radiation indices). Species richness increased significantly with increasing emergent vegetation and the presence of sandstone, and decreased as the Prescott Index (a measure of ground moisture), elevation and latitude increased. Total frog abundance increased with increasing emergent vegetation and pond area and decreased with elevation, pond density and the Prescott Index. No consistent patterns were evident between the anuran counts and the significant habitat variables. Pond shading was the variable most commonly related to the presence/abundance of individual species (appearing in models for eight species), followed by tree height (six species). Models explained up to 64% of the deviance for presence-absence and 48% for abundance, but usually explained < 35% deviance. Local scale variables account for 8/17 significantly related variables in presence models and 26/42 in abundance models. There was no pattern evident based on phylogeny and human disturbances showed little relationship with anuran counts, suggesting their influence was minimal. Managing forest frogs requires consideration of multiple scales and multiple features of the environment. No one pond fits all frogs. Not all breeding species that occur in the study area are considered common and widespread. Conservation of rare and threatened species is usually considered to be of most importance and relies on identifying their specific habitat requirements or specialisations. The heath frog, Litoria littlejohni, is a rarely observed threatened species confined to a limited area of the Watagan Mountains. It is protected by placing broad buffer zones around known breeding sites with little knowledge of how this might protect individuals. The distribution pattern of this anuran in the Watagan Mountains was assessed by comparing 12 habitat variables associated with 10 presence and 36 absence sites from the region. Logistic Regression Modelling indicated only that the species was more prevalent at ponds in forests with little grass cover (P = 0.0479, Beta = -0.1759). A logistic regression comparing GIS derived habitat features within 500 m of 61 presence and 20,000 randomly allocated absence sites located within the Sydney region indicated that the presence of the heath frog was positively associated with an increasing Prescott (Moisture) Index (Z = 4.22; P < 0.001) and negatively associated with increasing roughness of the surrounding terrain (Z = -3.62; P < 0.001). Using a regression tree to predict the presence of this species within an area, I classified presence sites (94%) based on a combination of the Prescott Index and Solar Radiation. However, these features provide only a very general ability to identify potential sites. The continued presence of the species at sites logged and burnt multiple times provides confidence that this species is able to cope with some degree of disturbance. The capacity to effectively monitor populations of anurans to detect early population changes is paramount in their ongoing management. A central requirement is to regularly assess the data being obtained to determine if the program has the statistical power to detect significant changes in population size. The power of the current monitoring program for the threatened northern corroboree frog (Pseudophryne pengilleyi) was assessed to determine its sensitivity to detect both increases and declines in the numbers of calling males being recorded at 14 sites. Based on the data collected over nine years, the power analysis indicated that only an annual decline of 7% or greater would be statistically detectable after 10 years (cumulative > 52% decline). Being able to successfully detect a 3% annual decline over ten years (25% total decline) in this population will require sixty sites to be monitored. This increase in sites could be achieved by combining data from other monitoring programs being undertaken in the region. The northern corroboree frog has consistent and easily monitored calling behaviour, yet detecting changes in numbers of calling males is difficult due to inter-year variance in counts at sites. Developing standard effective monitoring programs for “typical” species with inconsistent calling spread over long seasons will be difficult and likely require considerable resources. In addition to the application of wildlife management practices to maintain populations of amphibians in production forests, there is a need to ensure that species are adequately represented in protected areas (PAs) that can provide populations of animals that migrate into production forests to maintain metapopulations and/or genetic flows. The relative reservation status of anuran species in Australia was determined using the Australian Natural Heritage Tool Database (ANHAT), a very large database with multiple sources of anuran site records. The number of records for each species inside and outside protected areas was noted, as well the number of reserves from which each species was recorded. Given approximately 10.5% of Australia is covered by PAs, a random allocation of records would result in species having a mean 10% of records falling in PAs. The proportion of records coming from reserves was greater than expected with a mean 30% of sites per species occurring within PAs. Thirty six species had > 45% of record sites in PAs and 18 species < 10% within PAs. The majority of the latter group of species are found in < 10 reserves. Relatively poorly “reserved” species occur within northern Australia or highly productive and poorly protected agricultural belts in eastern and western Australia. All forest dwelling species from eastern NSW are relatively well recorded from the PAs. Species dependent on native vegetation prevalent in agricultural areas are the most pressing conservation issue. Forestry management practices include the use of controlled burns to reduce the incidence of high intensity and uncontrolled wildfires that threaten forest assets. The effect of these controlled burns on wildlife is widely debated. The impact of a controlled burn on anurans in the Dorrigo area was examined by comparing the trends in counts of anurans obtained at four burnt “experimental” ponds with those obtained at four unburnt “control” ponds. The analysis indicated that the trends in counts at burnt ponds and unburnt ponds did not differ. The fauna and flora of this region has evolved in an environment subjected to regular fires and the anurans present may have developed strategies to cope with low intensity fires. Current protection of anurans in forests in NSW are based around the retention of areas of undisturbed habitat, with an emphasis on protecting ponds and the adjacent 20-30 m band of vegetation. Linkage corridors of undisturbed vegetation are also required for migration. For threatened species, larger protection zones are usually set in place, often specifying the exclusion of fire as an important element in conserving species. Hydrology research indicates that 20-30 m buffer zones around breeding sites will protect water quality and such buffers will also protect individuals engaged in reproductive activities. Such a band will not provide significant protection to the surrounding forest that is used as non-breeding habitat and, where there is genuine cause for concern in regard to negative impacts of forestry activities on a species, this buffer should be expanded to a minimum 300 m. However, the studies on habitat relationships indicate that anurans in the region do not have specialised habitat requirements that may become unusable through disturbance and there were no obvious negative relationships between logging, fire or grazing and the presence/absence or abundance of anurans. In general, anurans appear to be relatively robust to the types of forest habitat modification resulting from selective logging and buffer zones greater than 20 m are not likely to be required to maintain populations in such an environment. Species with narrow habitat requirements, such as rainforest/wet forest specialists, may be the exception to this and require consideration in management plans.

This thesis addresses the hypothesis that unobtrusive monitoring of daily living could be used to evaluate changes in the health status of frail elderly people living alone at home. Low cost motion sensors can be used for monitoring the long term trends in occupant?s well-being in terms of physical, mental, social and environmental factors. The monitored data can be used to quantitatively measure parameters that can provide insight into the level of activity and functional ability of the subject. Any deviations in these parameters can provide information on the changing health status of the subject. This thesis attempts for the first time to establish a mathematical and statistical framework for the monitoring of functional health status in the home using a network of wireless sensors to monitor occupancy in each room of the house. A low power and low cost, unobtrusive occupancy monitoring system using ZigBee wireless technology and passive infrared sensors has been developed by the Biomedical Systems Laboratory at the University of New South Wales. The essence of the occupancy monitoring system is to detect variations in the activities of daily living (ADL) of elderly people living alone at home. The finite state, discrete parameter, time homogeneous Markov chain represents a theoretical framework for an unobtrusive occupancy monitoring system. An implementation of this framework for monitoring occupancy pattern is presented in real time use. The system was evaluated in a series of field studies in laboratory and home environment, in supervised and unsupervised settings, using cohorts of healthy elderly subjects living alone in their homes in community dwelling setting. A pilot trial was conducted in which four healthy elderly subjects living alone had PIR motion sensors installed in their homes at strategic points for a period of up to 13 weeks. The functionality of the system was evaluated over a domain of basic daily activities. A profile of the activities, in real time environment, for different times and days was stored as transition probability matrices. Automatic techniques for interpreting the test data captured by the system in terms of human movements were evaluated and compared with the wellness profile of the subject. Trial results exhibited that clinically significant model parameters were able to detect longitudinal deviations in the functional health status of elderly people.

During the recessional flow period of 2004, benthic diatoms were sampled from four catchments in the tropics of the Northern Territory of Australia. Each of the catchments showed evidence of acid mine drainage. Diatoms were used in conjunction with physical and chemical water parameters to provide a biomonitoring approach with the ultimate goal to assess the water quality of the mine impacted catchments. A total of 267 species of diatoms from 45 genera were recorded from 50 sites in the Northern Territory. One of the primary aims of this study was to assess diatom classification techniques, morphological and genetic, for the accurate identification of the morphologically variable taxon, Nitzschia palea. A second primary aim was to determine the degree to which diatom taxa and community structures can be effectively used as biological indicators and monitors of heavy metal pollution in tropical aquatic environments. To address the lack of diatom floristic studies from the Northern Territory, an iconograph of the most abundant taxa was created. Forty eight Nitzschia palea-like cells were cultured and then grouped by hierarchical cluster analysis (Ward’s method). The reliability of the groups was tested with discriminant analysis. The morphological groups were compared to the phylogenetic grouping of 20 of the cultures which were each genetically sequenced using the nuclear-encoded small subunit rDNA. For monitoring studies, multivariate statistical techniques were used to produce models to infer environmental variables from diatom distributions. The three datasets used varied in terms of the level of taxonomic identification and the inclusion of bioavailable or total metal concentrations. The program AquaRisk was employed to determine the bioavailable concentrations of the heavy metals. Each dataset was related to water quality parameters using canonical correspondence analysis. The Simpson’s index and the Shannon-Wiener diversity index were statistically related, together with species richness, to the variables pH and copper through bivariate regression analysis. To determine the usefulness of individual diatom taxa as indicators, species were selected based on their occurrence, weighted average optima and tolerance values, and species response curves. Additionally, teratogenic frustules were statistically correlated with environmental variables by bivariate and linear regression. Although the Nitzschia palea-like cultures displayed morphological variation, which could be used separate the cultures into five distinct groups, the cultures were not found to be genetically variable. However, one new species, Nitzschia sp. 39, was defined genetically and described morphologically. As with other analyses of mine impacts, canonical correspondence analysis identified pH and aluminium as the principal environmental factors structuring the diatom communities. The calibration set models generated to infer pH and heavy metals had high predicative capabilities. Overall, the species dataset, rather than the datasets utilising genus level identification or bioavailable metal fractions, provided the statistically strongest results. Monitoring using diversity indices was less successful for indicating pollution. Contrary to findings from other studies, species richness tended to be higher in polluted waters than control sites. Unlike other studies of acidic environments, the species Chamaepinnularia mediocris, Naviculadicta subtilissima, Nitzschia vasta and Pinnularia schoenfelderi were found to be the best indicators. Nitzschia vasta was the only teratogenic diatom within the dataset. However, as with other analyses, the relationship between teratogenic valves and mine impact was not significant. This study demonstrates the continued need for research in the tropics in order to better understand biological responses and enable study comparisons. Although at least one new xii species was identified, the genetic research indicated that taxonomic texts produced for northern hemisphere regions can be applied to tropical diatom taxa as long as the risk of taxonomic force fitting is avoided. Additionally, the benefit of diatoms as biological monitors varies depending on the method utilised. In these highly impacted systems, transfer functions produced the strongest results. In contrast, it is evident that the abundance and distribution of teratological forms of diatoms are poor indicators of impact. Further work culturing diatoms and combining this technique with ecotoxicological work will help verify autecologies of taxa and their responses to co-varying pollutants. This will strengthen use of diatom taxa as indicator species. Continued monitoring of these sites can add much to our ecological understanding of these highly impacted systems which, in turn, will lead to better management of the systems for both sustainable resource development and conservation. / Thesis (Ph.D.) -- University of Adelaide, School of Social Sciences, 2008

Research Doctorate - Doctor of Philosophy (PhD) / Environmental sustainability through the successful management of the impacts of forestry related activities on native wildlife in New South Wales (NSW) is a core requirement for Forests NSW, the government authority overseeing timber harvesting in the state. Populations of all native species must be protected adequately to allow their survival into the foreseeable future. A major proportion of the timber harvesting in NSW occurs along the coast and adjacent ranges of central and northern NSW. Timber is extracted using selective logging (not clearfelling) and the impacts of this activity on wildlife ameliorated through the use of strategies such as the placement of buffers around riparian sites, and the retention of hollow bearing trees, downed woody debris and undisturbed corridors of vegetation. This region, encompassing Latitudes 30o to 34o South, is dominated by native temperate wet and dry sclerophyll forests distributed along a gradient from coastal lowlands to montane areas over 1000 m. There are two families of anurans present in this region, the hylidae (approximately 25 species) and myobatrachidae (approximately 20 species), and are considered to have evolved in the region over at least 60 million years. Understanding of the effectiveness of protective measures requires a confident understanding of the habitat requirements of the frogs present in the forests so that they are used in appropriate locations. Presenting scientifically based information on the response of frogs to disturbances with these measures in place is also of primary interest to provide indications as to the effectiveness of these measures. Studies of pond breeding frogs in this region were undertaken to provide information on their habitat requirements and so direct the appropriate placement of the required protective measures. Tests were also undertaken to assess if there is an evident negative relationship between forestry disturbances and the use of ponds by frogs, indicating if current protective measures are effective. Current monitoring data was also investigated to provide insights into successful monitoring strategies for frogs which will provide the opportunity to more effectively demonstrate ecologically sustainable management of frogs into the future. Finally, records available on the distributions of frogs were used to study relationships between frog records and the current protected area system to assess its value as a means of protecting the frog species in the study region and throughout Australia. The majority of the research work undertaken was based on surveys of anurans at 93 semi-permanent and permanent ponds found in four areas: The Watagan Mountains, Bulahdelah, Wauchope and Dorrigo. Each site was surveyed on multiple occasions between 2001 and 2008, using aural and visual nocturnal searches conducted over different seasons to provide long-term information on the species using the sites. The presence and abundance of each species at each pond on each occasion was recorded as was micrometeorological data during the surveys. Each site was also visited in the day-time to collect data on variables associated with both the breeding pond and with the habitat within an 80 m radius of the site. A basic understanding of the extent and types of habitats individual frogs use is an important first step in understanding the extent of area requiring management for ongoing conservation. Published studies of frog movements were collated and reviewed to provide data on the likely habitat requirements and patterns of habitat use of frogs. The data, obtained from a 68 studies covering 50 species world-wide, indicating that in most cases (39 of 57 studies) adults have both specific breeding and non-breeding habitats. Mean distances moved from breeding to non-breeding habitats ranged from 38-1810 m and means for different anuran families falling around 300 m. Individuals in non-breeding habitat tended to restrict their activity to core areas (means from 6.3 m2 to 5099m2). This review indicated the strong need to consider both the breeding ponds and the surrounding habitat in the management of frogs. The calling seasons assigned by field guides to anuran species present within the study region are both highly variable and often vague. Therefore an objective assessment of the calling seasons of the frogs in eastern mesic NSW was developed to more accurately determine the months during which each species is most likely to call – the core calling period – and so be best surveyed. This was defined for each species as the months that covered > 90% of calling records. The 17,461 records covering 67 species provided enough information to reasonably determine the core calling periods for 46 species. Of these, 43 had clearly definable core calling seasons centred on the spring-summer months, two called year-round and one had an uncertain calling season, but probably calls year-round. Increasing latitude usually, but not always, led to a small reduction in the core calling period. This information confirmed the appropriate timing of surveys used in this study and should improve the outcomes of future research, management, and conservation. Micrometeorlogical variables such as temperature and rainfall are well recognised for their potentially significant influence on the calling activity of anurans, but this effect has been little tested in Australia. The calling records of frogs obtained through the study were compared with the micrometeorological variables temperature, humidity and (previous) 24 and 72 hour rainfall to assess how these related to calling activity. Models obtained for nine species all found calling activity to relate to at least two of these four variables. Calling showed a positive relationship with 72 hour rainfall in eight species, but no relationship for one species. Calling showed a positive relationship with 24 hour rainfall for six species, a negative relationship for two species and no relationship for one species. Temperature and humidity were found to have a mixture of positive, negative or no relationships with calling being evident. Rainfall over the previous 72 hours appears to be the best predictor consider when planning surveys. Other factors such as social environment and circannual rhythms are likely to be interacting with micrometeorology; hence the complex results. Further work is required to understand them fully. The results do indicate that surveys undertaken for this study were suitably timed to provide good results. Understanding if there are specific features of available habitats that correlate strongly with the presence/absence of frogs is important for managing frogs. This information may identify which breeding sites specific frogs are likely to use (or use in greater numbers) and habitat elements that may be critical for their survival. It may also provide indications of how human induced modifications may be affecting the frogs in an area. The presence/absence and abundance of anurans was recorded at 45 ponds in the Watagan Mountains, 100 km north of Sydney, and compared with 24 habitat variables. Only nine species occupied > 33% of the ponds. Canonical Correspondence Analysis (CCA) indicated there was no strong community structure. Species richness at a site was related significantly to depth and % emergent vegetation, but neither provided great explanatory power, based on their parameter estimates. Total anuran abundance related significantly to 12 variables that accounted for substantial variation in the data, but all contributed small amounts. Models for the four commonest species found many variables with statistically significant relationships to the frog counts, but neither pond attributes nor surrounding habitat attributes explained large amounts of variance in the data. There was little commonality in the habitat variables significant for the four species. This study indicated that no single variable could be said to influence general anuran presence and abundance and that habitat may not play a large role in determining the presence or abundance of anurans in this system. The initial study was confined to a relatively small geographic area to minimise differences resulting from variations in altitude, climate and geology. This was followed with a generalised linear model analysis that covered the 93 sites in the four forest blocks. The comparison with the recorded anurans and habitat also considered variation in altitude, geology and habitat types and included GIS derived variables covering more landscape habitat features around the ponds (eg, moisture indices and solar radiation indices). Species richness increased significantly with increasing emergent vegetation and the presence of sandstone, and decreased as the Prescott Index (a measure of ground moisture), elevation and latitude increased. Total frog abundance increased with increasing emergent vegetation and pond area and decreased with elevation, pond density and the Prescott Index. No consistent patterns were evident between the anuran counts and the significant habitat variables. Pond shading was the variable most commonly related to the presence/abundance of individual species (appearing in models for eight species), followed by tree height (six species). Models explained up to 64% of the deviance for presence-absence and 48% for abundance, but usually explained < 35% deviance. Local scale variables account for 8/17 significantly related variables in presence models and 26/42 in abundance models. There was no pattern evident based on phylogeny and human disturbances showed little relationship with anuran counts, suggesting their influence was minimal. Managing forest frogs requires consideration of multiple scales and multiple features of the environment. No one pond fits all frogs. Not all breeding species that occur in the study area are considered common and widespread. Conservation of rare and threatened species is usually considered to be of most importance and relies on identifying their specific habitat requirements or specialisations. The heath frog, Litoria littlejohni, is a rarely observed threatened species confined to a limited area of the Watagan Mountains. It is protected by placing broad buffer zones around known breeding sites with little knowledge of how this might protect individuals. The distribution pattern of this anuran in the Watagan Mountains was assessed by comparing 12 habitat variables associated with 10 presence and 36 absence sites from the region. Logistic Regression Modelling indicated only that the species was more prevalent at ponds in forests with little grass cover (P = 0.0479, Beta = -0.1759). A logistic regression comparing GIS derived habitat features within 500 m of 61 presence and 20,000 randomly allocated absence sites located within the Sydney region indicated that the presence of the heath frog was positively associated with an increasing Prescott (Moisture) Index (Z = 4.22; P < 0.001) and negatively associated with increasing roughness of the surrounding terrain (Z = -3.62; P < 0.001). Using a regression tree to predict the presence of this species within an area, I classified presence sites (94%) based on a combination of the Prescott Index and Solar Radiation. However, these features provide only a very general ability to identify potential sites. The continued presence of the species at sites logged and burnt multiple times provides confidence that this species is able to cope with some degree of disturbance. The capacity to effectively monitor populations of anurans to detect early population changes is paramount in their ongoing management. A central requirement is to regularly assess the data being obtained to determine if the program has the statistical power to detect significant changes in population size. The power of the current monitoring program for the threatened northern corroboree frog (Pseudophryne pengilleyi) was assessed to determine its sensitivity to detect both increases and declines in the numbers of calling males being recorded at 14 sites. Based on the data collected over nine years, the power analysis indicated that only an annual decline of 7% or greater would be statistically detectable after 10 years (cumulative > 52% decline). Being able to successfully detect a 3% annual decline over ten years (25% total decline) in this population will require sixty sites to be monitored. This increase in sites could be achieved by combining data from other monitoring programs being undertaken in the region. The northern corroboree frog has consistent and easily monitored calling behaviour, yet detecting changes in numbers of calling males is difficult due to inter-year variance in counts at sites. Developing standard effective monitoring programs for “typical” species with inconsistent calling spread over long seasons will be difficult and likely require considerable resources. In addition to the application of wildlife management practices to maintain populations of amphibians in production forests, there is a need to ensure that species are adequately represented in protected areas (PAs) that can provide populations of animals that migrate into production forests to maintain metapopulations and/or genetic flows. The relative reservation status of anuran species in Australia was determined using the Australian Natural Heritage Tool Database (ANHAT), a very large database with multiple sources of anuran site records. The number of records for each species inside and outside protected areas was noted, as well the number of reserves from which each species was recorded. Given approximately 10.5% of Australia is covered by PAs, a random allocation of records would result in species having a mean 10% of records falling in PAs. The proportion of records coming from reserves was greater than expected with a mean 30% of sites per species occurring within PAs. Thirty six species had > 45% of record sites in PAs and 18 species < 10% within PAs. The majority of the latter group of species are found in < 10 reserves. Relatively poorly “reserved” species occur within northern Australia or highly productive and poorly protected agricultural belts in eastern and western Australia. All forest dwelling species from eastern NSW are relatively well recorded from the PAs. Species dependent on native vegetation prevalent in agricultural areas are the most pressing conservation issue. Forestry management practices include the use of controlled burns to reduce the incidence of high intensity and uncontrolled wildfires that threaten forest assets. The effect of these controlled burns on wildlife is widely debated. The impact of a controlled burn on anurans in the Dorrigo area was examined by comparing the trends in counts of anurans obtained at four burnt “experimental” ponds with those obtained at four unburnt “control” ponds. The analysis indicated that the trends in counts at burnt ponds and unburnt ponds did not differ. The fauna and flora of this region has evolved in an environment subjected to regular fires and the anurans present may have developed strategies to cope with low intensity fires. Current protection of anurans in forests in NSW are based around the retention of areas of undisturbed habitat, with an emphasis on protecting ponds and the adjacent 20-30 m band of vegetation. Linkage corridors of undisturbed vegetation are also required for migration. For threatened species, larger protection zones are usually set in place, often specifying the exclusion of fire as an important element in conserving species. Hydrology research indicates that 20-30 m buffer zones around breeding sites will protect water quality and such buffers will also protect individuals engaged in reproductive activities. Such a band will not provide significant protection to the surrounding forest that is used as non-breeding habitat and, where there is genuine cause for concern in regard to negative impacts of forestry activities on a species, this buffer should be expanded to a minimum 300 m. However, the studies on habitat relationships indicate that anurans in the region do not have specialised habitat requirements that may become unusable through disturbance and there were no obvious negative relationships between logging, fire or grazing and the presence/absence or abundance of anurans. In general, anurans appear to be relatively robust to the types of forest habitat modification resulting from selective logging and buffer zones greater than 20 m are not likely to be required to maintain populations in such an environment. Species with narrow habitat requirements, such as rainforest/wet forest specialists, may be the exception to this and require consideration in management plans.

The term “blue-sky eruption” (BSE) can be used to describe eruptions which are unexpected or have no detected precursory activity. Case study analyses indicate that they have a diverse range of characteristics and magnitudes, providing both direct and indirect hazards and occur in both under-developed and developed countries. BSEs can be a result of physical triggers (e.g. the lack of physically detectable precursors or a lack of understanding of the eruption model of the volcano), social triggers (such as an inadequate monitoring network), or a combination of the two. As the science of eruption forecasting is still relatively young, and the variations between individual volcanoes and individual eruptions are so great, there is no effective general model and none should be applied in the absence of a site-specific model. Similarly, as methods vary between monitoring agencies, there are no monitoring benchmarks for effective BSE forecasting. However a combination of seismic and gas emission monitoring may be the most effective. The United States began a hazard and monitoring review of their volcanoes in 2005. While the general principles of their review would be beneficial in a monitoring review of New Zealand’s volcanoes, differences in styles of volcanism, geographic setting and activity levels mean changes would need to be review to fully appreciate the risk posed by New Zealand’s volcanoes. Similarly, the monitoring benchmarks provided in the U.S. review may not be fully applicable in New Zealand. While advances in technology may ultimately allow the effective forecasting of some BSEs, the immediate threat posed by unexpected eruptions means that effective management and mitigation measures may be the only tools currently at our disposal to reduce the risks from BSEs.

Chlorpyrifos, an organophosphate pesticide moderately toxic to humans via inhalation and dermal absorption (LD50 oral, rat = 226 mg kg-1, LD50 skin, rabbits = 1265 mg kg-1), is widely used to eradicate termites in Australia. A series of 28 surveys totaling 32 separate assessments, or 10% of all professional users in Perth, Western Australia, comprised biological monitoring, exposure assessment techniques, a health symptoms and work practices questionnaire. Chlorpyrifos metabolite 3,5,6-trichloro-2-pyridinol and alkyl phosphates were extracted from urine, and serum cholinesterase (SChE) and erythrocyte acetylcholinesterase from blood. Chlorpyrifos was extracted from 24 patches removed from a supplied cotton overall, cotton gloves worn under protective gloves, 7 absorbent patches placed on the skin and an organic vapour collection tube. Surface wipes were collected in the workers vehicle and on the workers forehead. Chlorpyrifos was applied in either 0.5% (n=2) or 1% (n=26) concentration of active ingredient in water solution. Surveys took place at pre-construction sites (n=5) where pesticide was sprayed onto a prepared site, existing buildings with concrete foundations (n=17) where pesticide was injected under pressure around the perimeter of the building and existing buildings with suspended floors requiring the worker to spray under floor (n=6). Combined left and right glove deposition was 9 mg hour- 1 (SD = 18 mg.hour-1). Mean deposition on overalls was 14 mg.hour-1 (SD = 12 mg.hour-1), on skin patches was 0.2 ƒÝg.cm-1.hour-1, on vehicle gear-stick was 3 ƒÝg (SD = 8 ƒÝg) and, on steering wheels¡¦ was 3 ƒÝg (SD = 3 ƒÝg). The mean protection 4 factor of overalls, a ratio of outer layer and inner levels, was 75 (SD = 411). Mean air concentration of chlorpyrifos during an application was 30 ƒÝg m-3, and 17 ƒÝg m-3 8 hour TWA (SD = 40 ƒÝg m-3 8 hour TWA), and in one group of 17 workers correlated (p<0.05) with ambient air temperature (15 to 38 oC). Urinary metabolites and SChE activity were effective indicators of exposure. The health symptoms questionnaire did not highlight significant health effects. A discrepancy between operators¡¦ perception of risk and their actual exposure requires addressing, for example the measured high deposition rate to hands was ineffectively controlled, as 48% or workers wore inappropriate or no gloves and only 26% washed their hands after completing their tasks. All workers indicated in the questionnaire they would wash their hands after completing their tasks. The questionnaire also highlighted a high incidence of poor work practices, 58% spilt the concentrate at least once a week, 74% had recently spilt/splashed diluted chlorpyrifos in their eyes and 90% on their boots, and 52% believed they would benefit from more education concerning chlorpyrifos. Observations concluded that workers unnecessarily increased their exposure by poor work practice. Recommendations include modification to pesticide worker education, licencing and health surveillance systems; an improvement in the understanding of the benefits of a health and safety management systems for employers, and pesticide suppliers taking a stewardship role in the usage of their products.

Thesis (M.S.)--University of Arizona, 1990. / Includes bibliographical references (pp. 197-199).

Thesis (M.S.)--Northwest Missouri State University, 2007. / The full text of the thesis is included in the pdf file. Title from title screen of full text.pdf file (viewed on January 25, 2008) Includes bibliographical references.

Proefschrift Universiteit van Amsterdam. / Lit. opg.: p. 157-163. - Met samenvatting in het Nederlands.