This qualitative research study investigated the expression of critical thinking in environmental subjects at university level in Australia. It also explored the strategies used by lecturers to encourage student critical thinking. Initially an open-ended questionnaire was sent to lecturers and tutors in universities across Australia and the responses from thirty participants were analysed thematically. From these participants semi-structured interviews were arranged with six lecturers and eight students from three different universities. The third phase of the research process involved analysing subject outlines and student assignments. Findings were triangulated to create a picture of the teaching and learning practices of critical thinking in this field. The three cases of Environmental Science, Environmental Engineering and Environmental Law were identified as having some common yet distinctively different themes in relation to critical thinking. In each discipline area three broad approaches to the expression of critical thinking were identified: scientific/technical, paradigmatic and reflective. The scientific/technical approach in science was most concerned with critique of experiments and studies, in engineering with problem solving and in law with the technical aspects of language and application of the law. The paradigmatic approach included critical analysis of the different theories, frameworks and paradigms of each subject. The reflective approach to critical thinking included students’ self evaluation of their learning within a specific subject. Notable differences between the three cases were the types of strategies used by lecturers to encourage student critical thinking. The Environmental Science lecturers were more inclined to use a wide variety of strategies, including provision of written material about critical thinking with critical thinking identified in marking criteria, discussion in class and guided exercises. The Environmental Engineering lecturers placed more emphasis on discussion in class than on providing written materials. The Environmental Law lecturers predominantly focused on the strategy of debating in class for the development of student critical thinking. The findings of this study are significant in providing support for the argument that development of critical thinking is subject specific rather than generic. This calls into question the use of generic skills testing for university graduates as a reliable measure of their ability to think critically in their field. The findings indicate that it is necessary for subject lecturers to be explicit with students about their expectations regarding critical thinking, particularly in relation to assessment. Findings suggest guidelines for lecturers including strategies and attitudes that encourage or hinder student critical thinking. Findings also suggest that the nature of the field of environmental and sustainability subjects lends itself to developing critical thinking, which is considered to be important.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / This research provides insight into large-scale spatial variation of ozone (O3) distribution in the Auckland region and was designed to increase our understanding of O3 behaviour and distribution in a coastal-urban situation. The research was also carried out with a view to assisting regulatory agencies optimize future monitoring networks, and to help identify locations where human health and natural resources could be at risk in the future. Although the research was limited to one region, the results are valuable for improving the conceptual understanding of formation of high ozone concentrations in a more general sense in the New Zealand and Southern Hemisphere. The work reported in this thesis is aimed at studying O3 concentrations and the influence of the most relevant meteorological variables on an average coastal New Zealand city where precursor emissions are mainly due to traffic exhaust. It deals with the use of Principal Component Analysis method for determining O3 concentrations as a function of meteorological parameters. The study region includes the entire Auckland isthmus, and extends from Whangaparaoa in the North to Pukekohe in the South. Surface O3 data from four sites (Whangaparaoa, Musick Point, the Sky Tower and Pukekohe) for a 4-year period(October 1997 to October 2001) for the Auckland region were examined. Ambient concentration of O3 was characterized in terms of diurnal, weekday/weekend, seasonal and spatial variations in concentration using O3 measurements from the four air quality sites. The monthly average ambient background O3 concentrations at the monitoring sites during this study ranged from 16-30 ppb, much lower than those found in the Northern Hemisphere. The measured seasonal O3 record in the Auckland region, in common with many other remote sites in the Southern Hemisphere, exhibited a summer minimum and a winter maximum. Background concentrations of O3 (as seen in air of marine origin) made a significant contribution to the observed ambient concentrations. A unique feature of Auckland's air quality was the dilution of polluted city air due to the mixing of east coast air into the cleaner west coast circulation leading to overall lower average O3 concentrations in summer. The magnitude, frequency and spatial extent of maximum O3 concentrations were identified, and the observed patterns linked to the prevailing meteorological, topographic, and emission characteristics of the region. However, at no time at any site or season did the O3 concentration exceed the l-hour New Zealand Ministry for the Environment guideline of 75 ppb. O3 depletion was observed to occur at the urban sites, with O3 scavenging by nitric oxide believed to be the dominant depletion mechanism. The seasonal cycle was characterized by elevated O3 concentrations in the winter (nighttime level >24 ppb) and low mixing ratios in the summer (nighttime levels in the range 14-20 ppb). The afternoon O3 maxima found at the three low elevation sites under the impact of "Auckland city" plume were on average, 1.5 to 1.7 times higher than those associated with the "marine sector". A state of the art diagnostic meteorological model, namely CALMET, was used to generate wind fields for the Auckland region. These wind fields were then used to construct backward trajectories on days when high O3 concentrations (episodes up to 6l ppb) were observed. The un-even distribution of the meteorological monitoring sites provided justification for running the diagnostic model CALMET and exploring the utility of using such a model when topography/land use prohibits monitoring sites in certain sections of the domain. It was found that the high O3 events recorded at the monitoring sites during the study period coincided with transport episodes originating from Auckland's urban and industrial areas. For the occurrence of high O3 concentrations at downwind sites, it was found that not only wind direction from the urban areas of Auckland, but that high solar radiation was important as well. On four out of the five cases when photochemical production of O3 was found to occur, high O3 concentrations were associated with a particular type of diurnal evolution of wind direction wind fields associated with the sea breezes. In addition to the analysis of ambient o3 concentrations and O3 episodes, field measurements were carried out in an attempt to detect and subsequently understand the interaction between particulate matter and O3 in the Auckland region, a multivariate statistical analysis approach was utilized. Particulate matter in the size range 2.75-6.25 μm accounted for over 70% of the total aerosol concentration at all sampling sites. Surface area of particulate matter variable (especially in the size class 2.75-4.25 μm) was statistically significant in explaining variation in O3 concentration. However, the net change in the adjusted R2 indicated that the effect of adding particulate matter in the multiple regression model for the present dataset was relatively minor except at Musick Point.

Each chapter in this thesis has been written as a separate paper intended for later publication. Rather than write one summary, an abstract for each chapter has been given in the order they are presented in the thesis. Chapter I. The prey species of domestic cats (Felis catus) in two suburbs of Auckland city, New Zealand. The prey brought in by 80 cats (Felis catus), over a year was monitored in two suburbs of Auckland New Zealand. The survey technique followed that of Churcher & Lawton (1987) where cat owners were asked to record (and if possible keep) the prey their cats brought in. The results for the year indicated that there were distinct differences in the type of prey taken by cats in each area. Rodents were the most important prey brought in by domestic cats in an urban / forest fringe habitat and invertebrates were the main prey brought in by domestic cats in a fully urban habitat. Birds and lizards were the second and third most important prey groups in both study areas. Rat and mouse snap trap indexes which were run at each study location, did not detect rodents in the urban habitat. The results from the urban/forest fringe concurred more with studies of feral cat diet in New Zealand whilst those from the urban area compared more to studies of domestic and stray cat diet in urban areas overseas. There were seasonal trends in the prey captured in each area and in both areas cats hunted less over the winter months. Chapter II. Diets of coexisting alien mammalian carnivores in Northern New Zealand. 120 feral cats, 85 stoats (Mustela erminea), 28 weasels (M. nivalis) and 16 ferrets (M. furo) were caught over three years, in Northland, New Zealand. The gut contents of these animals were examined and their food habits described. The main prey groups of cats were (in order of importance by weight), lagomorphs, rodents (Rattus spp & Mus musculus), other mammals, birds and invertebrates. The main prey of stoats were rodents, birds, lagomorphs, skinks (Cyclodina spp) and invertebrates. Skinks, followed by mice, were the main prey of weasels. Rabbits (Oryctolagus cuniculus) were the most commonly occurring prey item in the ferret guts. The food habits for cats and stoats were compared between different habitats within the Northland region where these animals were collected from. There was little difference between habitats, but invertebrates occurred more frequently and lagomorphs less frequently in the diet of animals from forest habitats compared to those from forest / pasture and coastal habitats. No differences were found in the prey of male and female cats, but invertebrates occurred more frequently in the guts of sub-adult cats compared to adult cats. Skinks occurred more frequently in the guts of female stoats, which tended to take smaller prey items than males. Chapter III. Home range of introduced mammalian carnivores, feral cats (Felis catus), stoats (Mustela erminea) and a ferret (M. furo) at Trounson Kauri Park, Northland, New Zealand. The minimum home ranges of eleven feral cats, four stoats and one male ferret were examined by radio telemetry at Trounson Kauri Park in Northland New Zealand. The average minimum home range of male feral cats was 305 ± 74ha. This was not significantly larger than the 122 ± 35 ha minimum average home range of female cats. The minimum average home range of three male stoats was 108 ± 19 ha and the minimum home range of one female stoat was 50 ha. The male ferret had a minimum home range of 179 ha. There was no overlap spatially, in the home range of three adult male cats. There was also very little spatial overlap in the home range of four female cats. There was substantial home range overlap with four sub-adult male cats. The home ranges of these sub-adult male cats also overlapped spatially with those of the adjacent females and adult males. The home ranges of two male stoats overlapped spatially to some degree and the home range of another male stoat overlapped that of the female stoat substantially. The home range of the male ferret overlapped the home ranges of all of the cats occupying similar areas of the park. The home ranges of two male stoats overlapped the home ranges of the adjacent cats but not into the "core" areas of those cats ranges. Chapter IV. Secondary poisoning of introduced mammalian carnivores during possum and rodent control operations at Trounson Kauri Park, Northland New Zealand. Predatory mammals were monitored by radio telemetry through a 1080 then brodifacoum poison baiting operation at Trounson Kauri Park in Northland, New Zealand to target possums (Trichosurus vulpecula) and rodents (Rattus rattus, Rattus norvegicus & Mus musculus). All six feral cats (Felis catus), and the single stoat (Mustela erminea) and ferret (Mustela furo) being monitored at the beginning of the operation died of secondary poisoning following the 1080 operation. A further two cats and four stoats were radio tagged and monitored through the ongoing poisoning campaign using brodifacoum in a continuous baiting regime. None of these radio tagged carnivores died of secondary poisoning. However, tissue analysis of additional carnivores trapped at Trounson found that cats, weasels (Mustela nivalis) and to a lesser extent stoats did contain brodifacoum residues. The duration that the radio tagged predators were alive in and around Trounson Kauri Park suggested that the secondary poisoning effect was much reduced under the continuous baiting strategy, compared to the initial 1080 poison operation. Chapter V. Managing alien mammals in mainland New Zealand: The implications for predator/prey interactions in complex communities. Recent advancements in alien pest mammal control technologies have enabled conservation managers to attempt long term, large scale, predator and browser control programmes at mainland sites in New Zealand. At Trounson Kauri Park (in Northland, New Zealand) feral cats, ferrets, stoats and weasels plus rodents and possums were controlled for two and a half years. Early results from Trounson suggested that some native birds and plants responded positively and quickly to the control of these alien pest mammals. However, the management also affected the predator-prey and predator-predator interactions, within the mammalian community, with changes in the predator guild and some prey species increasing in abundance. The long term effects of these responses within the mammalian community to control efforts are unclear. A review of the international literature served only to confirm the view that managing alien mammals where re-invasion is constant will be a complex task. Conservation management at mainland sites like Trounson Kauri Park, may in effect be acting as crude mammalian predator removal experiments. Monitoring changes in predator-prey interactions and predator-predator interactions at this and other intensive management sites over sufficient time frames could provide scientists with the critical ecological data required to construct useful predictive models. The ability of managers to target specific pests at critical times would represent a significant advancement in controlling predatory mammals on the New Zealand mainland.

A new air sampler, the Mist-UV sampling system, has been developed to trap and detect metals and metal compounds present in atmospheric samples, as non-filterable metal (NFM). The Mist-UV system is based on the combination of two ideas, mixing the filtrated air sample with nitric acid mist and simultaneous ultraviolet irradiation of the mist. Atmospheric NFM in the introduced air samples, which were first passed through a 0.2 µm filter, were oxidised by UV irradiation and dissolved in the nitric acid mist. The trapped metals resulting from the NFM in nitric acid were quantitatively determined using ICP-MS. Using the Mist-UV system, measurements of atmospheric NFM concentrations were carried out to obtain their temporal and spatial trends in Auckland, New Zealand. Five metals (Mg, Al, Cu, Ba and Sr) were detected as NFM in ambient air. Elevated atmospheric concentrations of Mg, Cu, and Ba were detected in midsummer and in daytime samples. The UV oxidisable fraction dominated the NFM and high enrichment factors (EFcrust) were found for these metals. This suggested that natural emissions of organometals or reduced metals were responsible for the elevated atmospheric concentrations. Meteorological factors, particularly air temperature, light irradiation and rainfall, promoted biogenic activity and resultant NFM emissions. Remarkably high concentrations of Al and Sr as NFM were discovered in several air samples and were dominated by the non-oxidisable compounds. This implies that the sources of Al and Sr were mainly from inorganic or oxidised metals associated with fine particulates. Wet and dry soils appeared to be the sources of Mg+Ba and Al+Sr, respectively. Plant emissions are recognised as important sources of Mg, Cu and Ba as NFM in ambient air, whereas the suspension of fine particulates derived from road dust and soil enriched with non-oxidisable compounds are seen as the major sources of Al and Sr. The emissions from automobiles and seawater samples were found to be unimportant NFM sources. The NFM concentrations were found to be significantly higher than concentrations of particulate metals, therefore the atmospheric importance and role of NFM cannot be ignored in the estimation of global metal circulation and budgets.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / This research provides insight into large-scale spatial variation of ozone (O3) distribution in the Auckland region and was designed to increase our understanding of O3 behaviour and distribution in a coastal-urban situation. The research was also carried out with a view to assisting regulatory agencies optimize future monitoring networks, and to help identify locations where human health and natural resources could be at risk in the future. Although the research was limited to one region, the results are valuable for improving the conceptual understanding of formation of high ozone concentrations in a more general sense in the New Zealand and Southern Hemisphere. The work reported in this thesis is aimed at studying O3 concentrations and the influence of the most relevant meteorological variables on an average coastal New Zealand city where precursor emissions are mainly due to traffic exhaust. It deals with the use of Principal Component Analysis method for determining O3 concentrations as a function of meteorological parameters. The study region includes the entire Auckland isthmus, and extends from Whangaparaoa in the North to Pukekohe in the South. Surface O3 data from four sites (Whangaparaoa, Musick Point, the Sky Tower and Pukekohe) for a 4-year period(October 1997 to October 2001) for the Auckland region were examined. Ambient concentration of O3 was characterized in terms of diurnal, weekday/weekend, seasonal and spatial variations in concentration using O3 measurements from the four air quality sites. The monthly average ambient background O3 concentrations at the monitoring sites during this study ranged from 16-30 ppb, much lower than those found in the Northern Hemisphere. The measured seasonal O3 record in the Auckland region, in common with many other remote sites in the Southern Hemisphere, exhibited a summer minimum and a winter maximum. Background concentrations of O3 (as seen in air of marine origin) made a significant contribution to the observed ambient concentrations. A unique feature of Auckland's air quality was the dilution of polluted city air due to the mixing of east coast air into the cleaner west coast circulation leading to overall lower average O3 concentrations in summer. The magnitude, frequency and spatial extent of maximum O3 concentrations were identified, and the observed patterns linked to the prevailing meteorological, topographic, and emission characteristics of the region. However, at no time at any site or season did the O3 concentration exceed the l-hour New Zealand Ministry for the Environment guideline of 75 ppb. O3 depletion was observed to occur at the urban sites, with O3 scavenging by nitric oxide believed to be the dominant depletion mechanism. The seasonal cycle was characterized by elevated O3 concentrations in the winter (nighttime level >24 ppb) and low mixing ratios in the summer (nighttime levels in the range 14-20 ppb). The afternoon O3 maxima found at the three low elevation sites under the impact of "Auckland city" plume were on average, 1.5 to 1.7 times higher than those associated with the "marine sector". A state of the art diagnostic meteorological model, namely CALMET, was used to generate wind fields for the Auckland region. These wind fields were then used to construct backward trajectories on days when high O3 concentrations (episodes up to 6l ppb) were observed. The un-even distribution of the meteorological monitoring sites provided justification for running the diagnostic model CALMET and exploring the utility of using such a model when topography/land use prohibits monitoring sites in certain sections of the domain. It was found that the high O3 events recorded at the monitoring sites during the study period coincided with transport episodes originating from Auckland's urban and industrial areas. For the occurrence of high O3 concentrations at downwind sites, it was found that not only wind direction from the urban areas of Auckland, but that high solar radiation was important as well. On four out of the five cases when photochemical production of O3 was found to occur, high O3 concentrations were associated with a particular type of diurnal evolution of wind direction wind fields associated with the sea breezes. In addition to the analysis of ambient o3 concentrations and O3 episodes, field measurements were carried out in an attempt to detect and subsequently understand the interaction between particulate matter and O3 in the Auckland region, a multivariate statistical analysis approach was utilized. Particulate matter in the size range 2.75-6.25 μm accounted for over 70% of the total aerosol concentration at all sampling sites. Surface area of particulate matter variable (especially in the size class 2.75-4.25 μm) was statistically significant in explaining variation in O3 concentration. However, the net change in the adjusted R2 indicated that the effect of adding particulate matter in the multiple regression model for the present dataset was relatively minor except at Musick Point.

1. The aim of this research was to determine appropriate microbial indicator systems for use in monitoring the occurrence and source of faecal contamination in aquatic environments. Particular emphasis has been placed on the identification of indicators which might enable faecal contamination of human origin to be distinguished from that of non-human derivation. Two lines of investigation have been pursued; the use of alternative microbial indicator systems for water quality monitoring, and the development of diagnostic molecular methods for identifying the origin of faecal organisms. 2. Field and treatment plant studies were undertaken to investigate the utility of three indicator systems for monitoring water quality impacts in the coastal area surrounding the discharge from Auckland's North Shore City wastewater treatment plant. F-specific bacteriophage (F-phage) and ratios of F-phage to enterococci and faecal coliforms were examined over a twelve month period in which indicator levels at various stages of the wastewater treatment process were measured. These indicators and levels of Clostridium perfringens spores in sediment were measured in four outfall impact studies. 3. F-phage levels in effluent from the treatment plant (median=2.5 x 102 cfu/l00 ml) were too low to enable their use as source-specific tracers of effluent dispersion. However, higher F-phage levels in raw sewage (median=6.2 x l05 cfu/l00 ml) suggested that phage may be an appropriate indicator of raw sewage contamination in the receiving environment. F-phage was detected in coastal waters remote from the sewage outfall, providing presumptive evidence of sewage contamination entering the sea from sources other than the outfall discharge itself. 4. F-phage/enterococci and F-phage/faecal coliform ratios that were determined for sewage and treated effluent derived from the North Shore plant proved to be highly variable and were therefore unreliable for determining the source faecal contamination. 5. The concentration of Clostridium perfringens spores in sediments did not provide a useful means for tracing the distribution of faecal contaminants in the coastal system because the spore levels were found to be similar throughout the study area. This widespread occurrence of C. perfringens spores prevented identification of defined impact zones for any of the potential sources of faecal contaminants examined. 6. Randomly Amplified Polymorphic DNA (RAPD) analysis was used to investigate whether genetic markers could be identified for Escherichia coli isolates derived from different sources. RAPD profiling was found to be a highly discriminative method able to subtype E. coli strains present in both sewage and in the wastewater stream of a mechanical wood-pulping operation. In the pulp mill effluent, RAPD analysis indicated the presence of a restricted range of E. coli strains compared to the range found in the water intake and sewage effluent discharges. The results obtained provide presumptive evidence for the multiplication of E. coli strains within the wastewater stream. 7. Comparison of RAPD profiles derived from a range of human and environmental E. coli isolates led to the identification of a 1.6-kb marker. present in a subset of human isolates, but found only infrequently in isolates from animal origin. Primers based on sequence information obtained from this region of the genome yielded a direct PCR assay for the 'human' marker. The reliability and utility of the assay was improved when fluorescently-labelled primers and an automated DNA detection system was employed for the analysis of the amplified PCR products. 8. The diagnostic polymorphism was localised to the region immediately downstream of the gene encoding the glycine decarboxylase P-protein (gcvP). Comparison of DNA sequences from this region in 'marker-positive' and 'marker-negative' E. coli isolates revealed an abrupt loss of homology immediately downstream of the transcription termination point of the gcvP gene. The region of non-homology extends for at least 130 base pairs beyond the gcvP transcription terminator. Sequences spanning this region in marker-negative isolates exhibited a high level homology with that determined for E. coli K-12. The origin of the sequence for the same region in marker-positive isolates was unable to be identified. 9. The utility of the marker for environmental monitoring was investigated in a field study of a rural stream receiving bacterial contamination of both human and animal origin. The marker was consistently detected both in isolates derived from sewage effluents and in those found immediately downstream from sewage inputs-despite the presence of a high background of E. coli of animal origin. However the utility of the marker as a sewage-specific tracer was limited by the relatively low numbers of marker-positive d. coli in sewage effluents (approximately 8%). 10. In summary, the work described in this thesis has confirmed the need for multiple indicator systems in water quality monitoring and has identified appropriate areas for further research. The most promising area for future research is in the development of molecular tools which might provide both a better understanding of genetic variation in indicator organisms and enable methods to be developed for the detection of host-specific strains. Location of other polymorphisms likely to be present in the E. coli genome should enable further host-specific markers to be identified which could complement the 'human' marker identified in this study. / Whole document restricted, but available by request, use the feedback form to request access.

Each chapter in this thesis has been written as a separate paper intended for later publication. Rather than write one summary, an abstract for each chapter has been given in the order they are presented in the thesis. Chapter I. The prey species of domestic cats (Felis catus) in two suburbs of Auckland city, New Zealand. The prey brought in by 80 cats (Felis catus), over a year was monitored in two suburbs of Auckland New Zealand. The survey technique followed that of Churcher & Lawton (1987) where cat owners were asked to record (and if possible keep) the prey their cats brought in. The results for the year indicated that there were distinct differences in the type of prey taken by cats in each area. Rodents were the most important prey brought in by domestic cats in an urban / forest fringe habitat and invertebrates were the main prey brought in by domestic cats in a fully urban habitat. Birds and lizards were the second and third most important prey groups in both study areas. Rat and mouse snap trap indexes which were run at each study location, did not detect rodents in the urban habitat. The results from the urban/forest fringe concurred more with studies of feral cat diet in New Zealand whilst those from the urban area compared more to studies of domestic and stray cat diet in urban areas overseas. There were seasonal trends in the prey captured in each area and in both areas cats hunted less over the winter months. Chapter II. Diets of coexisting alien mammalian carnivores in Northern New Zealand. 120 feral cats, 85 stoats (Mustela erminea), 28 weasels (M. nivalis) and 16 ferrets (M. furo) were caught over three years, in Northland, New Zealand. The gut contents of these animals were examined and their food habits described. The main prey groups of cats were (in order of importance by weight), lagomorphs, rodents (Rattus spp & Mus musculus), other mammals, birds and invertebrates. The main prey of stoats were rodents, birds, lagomorphs, skinks (Cyclodina spp) and invertebrates. Skinks, followed by mice, were the main prey of weasels. Rabbits (Oryctolagus cuniculus) were the most commonly occurring prey item in the ferret guts. The food habits for cats and stoats were compared between different habitats within the Northland region where these animals were collected from. There was little difference between habitats, but invertebrates occurred more frequently and lagomorphs less frequently in the diet of animals from forest habitats compared to those from forest / pasture and coastal habitats. No differences were found in the prey of male and female cats, but invertebrates occurred more frequently in the guts of sub-adult cats compared to adult cats. Skinks occurred more frequently in the guts of female stoats, which tended to take smaller prey items than males. Chapter III. Home range of introduced mammalian carnivores, feral cats (Felis catus), stoats (Mustela erminea) and a ferret (M. furo) at Trounson Kauri Park, Northland, New Zealand. The minimum home ranges of eleven feral cats, four stoats and one male ferret were examined by radio telemetry at Trounson Kauri Park in Northland New Zealand. The average minimum home range of male feral cats was 305 ± 74ha. This was not significantly larger than the 122 ± 35 ha minimum average home range of female cats. The minimum average home range of three male stoats was 108 ± 19 ha and the minimum home range of one female stoat was 50 ha. The male ferret had a minimum home range of 179 ha. There was no overlap spatially, in the home range of three adult male cats. There was also very little spatial overlap in the home range of four female cats. There was substantial home range overlap with four sub-adult male cats. The home ranges of these sub-adult male cats also overlapped spatially with those of the adjacent females and adult males. The home ranges of two male stoats overlapped spatially to some degree and the home range of another male stoat overlapped that of the female stoat substantially. The home range of the male ferret overlapped the home ranges of all of the cats occupying similar areas of the park. The home ranges of two male stoats overlapped the home ranges of the adjacent cats but not into the "core" areas of those cats ranges. Chapter IV. Secondary poisoning of introduced mammalian carnivores during possum and rodent control operations at Trounson Kauri Park, Northland New Zealand. Predatory mammals were monitored by radio telemetry through a 1080 then brodifacoum poison baiting operation at Trounson Kauri Park in Northland, New Zealand to target possums (Trichosurus vulpecula) and rodents (Rattus rattus, Rattus norvegicus & Mus musculus). All six feral cats (Felis catus), and the single stoat (Mustela erminea) and ferret (Mustela furo) being monitored at the beginning of the operation died of secondary poisoning following the 1080 operation. A further two cats and four stoats were radio tagged and monitored through the ongoing poisoning campaign using brodifacoum in a continuous baiting regime. None of these radio tagged carnivores died of secondary poisoning. However, tissue analysis of additional carnivores trapped at Trounson found that cats, weasels (Mustela nivalis) and to a lesser extent stoats did contain brodifacoum residues. The duration that the radio tagged predators were alive in and around Trounson Kauri Park suggested that the secondary poisoning effect was much reduced under the continuous baiting strategy, compared to the initial 1080 poison operation. Chapter V. Managing alien mammals in mainland New Zealand: The implications for predator/prey interactions in complex communities. Recent advancements in alien pest mammal control technologies have enabled conservation managers to attempt long term, large scale, predator and browser control programmes at mainland sites in New Zealand. At Trounson Kauri Park (in Northland, New Zealand) feral cats, ferrets, stoats and weasels plus rodents and possums were controlled for two and a half years. Early results from Trounson suggested that some native birds and plants responded positively and quickly to the control of these alien pest mammals. However, the management also affected the predator-prey and predator-predator interactions, within the mammalian community, with changes in the predator guild and some prey species increasing in abundance. The long term effects of these responses within the mammalian community to control efforts are unclear. A review of the international literature served only to confirm the view that managing alien mammals where re-invasion is constant will be a complex task. Conservation management at mainland sites like Trounson Kauri Park, may in effect be acting as crude mammalian predator removal experiments. Monitoring changes in predator-prey interactions and predator-predator interactions at this and other intensive management sites over sufficient time frames could provide scientists with the critical ecological data required to construct useful predictive models. The ability of managers to target specific pests at critical times would represent a significant advancement in controlling predatory mammals on the New Zealand mainland.

A new air sampler, the Mist-UV sampling system, has been developed to trap and detect metals and metal compounds present in atmospheric samples, as non-filterable metal (NFM). The Mist-UV system is based on the combination of two ideas, mixing the filtrated air sample with nitric acid mist and simultaneous ultraviolet irradiation of the mist. Atmospheric NFM in the introduced air samples, which were first passed through a 0.2 µm filter, were oxidised by UV irradiation and dissolved in the nitric acid mist. The trapped metals resulting from the NFM in nitric acid were quantitatively determined using ICP-MS. Using the Mist-UV system, measurements of atmospheric NFM concentrations were carried out to obtain their temporal and spatial trends in Auckland, New Zealand. Five metals (Mg, Al, Cu, Ba and Sr) were detected as NFM in ambient air. Elevated atmospheric concentrations of Mg, Cu, and Ba were detected in midsummer and in daytime samples. The UV oxidisable fraction dominated the NFM and high enrichment factors (EFcrust) were found for these metals. This suggested that natural emissions of organometals or reduced metals were responsible for the elevated atmospheric concentrations. Meteorological factors, particularly air temperature, light irradiation and rainfall, promoted biogenic activity and resultant NFM emissions. Remarkably high concentrations of Al and Sr as NFM were discovered in several air samples and were dominated by the non-oxidisable compounds. This implies that the sources of Al and Sr were mainly from inorganic or oxidised metals associated with fine particulates. Wet and dry soils appeared to be the sources of Mg+Ba and Al+Sr, respectively. Plant emissions are recognised as important sources of Mg, Cu and Ba as NFM in ambient air, whereas the suspension of fine particulates derived from road dust and soil enriched with non-oxidisable compounds are seen as the major sources of Al and Sr. The emissions from automobiles and seawater samples were found to be unimportant NFM sources. The NFM concentrations were found to be significantly higher than concentrations of particulate metals, therefore the atmospheric importance and role of NFM cannot be ignored in the estimation of global metal circulation and budgets.

1. The aim of this research was to determine appropriate microbial indicator systems for use in monitoring the occurrence and source of faecal contamination in aquatic environments. Particular emphasis has been placed on the identification of indicators which might enable faecal contamination of human origin to be distinguished from that of non-human derivation. Two lines of investigation have been pursued; the use of alternative microbial indicator systems for water quality monitoring, and the development of diagnostic molecular methods for identifying the origin of faecal organisms. 2. Field and treatment plant studies were undertaken to investigate the utility of three indicator systems for monitoring water quality impacts in the coastal area surrounding the discharge from Auckland's North Shore City wastewater treatment plant. F-specific bacteriophage (F-phage) and ratios of F-phage to enterococci and faecal coliforms were examined over a twelve month period in which indicator levels at various stages of the wastewater treatment process were measured. These indicators and levels of Clostridium perfringens spores in sediment were measured in four outfall impact studies. 3. F-phage levels in effluent from the treatment plant (median=2.5 x 102 cfu/l00 ml) were too low to enable their use as source-specific tracers of effluent dispersion. However, higher F-phage levels in raw sewage (median=6.2 x l05 cfu/l00 ml) suggested that phage may be an appropriate indicator of raw sewage contamination in the receiving environment. F-phage was detected in coastal waters remote from the sewage outfall, providing presumptive evidence of sewage contamination entering the sea from sources other than the outfall discharge itself. 4. F-phage/enterococci and F-phage/faecal coliform ratios that were determined for sewage and treated effluent derived from the North Shore plant proved to be highly variable and were therefore unreliable for determining the source faecal contamination. 5. The concentration of Clostridium perfringens spores in sediments did not provide a useful means for tracing the distribution of faecal contaminants in the coastal system because the spore levels were found to be similar throughout the study area. This widespread occurrence of C. perfringens spores prevented identification of defined impact zones for any of the potential sources of faecal contaminants examined. 6. Randomly Amplified Polymorphic DNA (RAPD) analysis was used to investigate whether genetic markers could be identified for Escherichia coli isolates derived from different sources. RAPD profiling was found to be a highly discriminative method able to subtype E. coli strains present in both sewage and in the wastewater stream of a mechanical wood-pulping operation. In the pulp mill effluent, RAPD analysis indicated the presence of a restricted range of E. coli strains compared to the range found in the water intake and sewage effluent discharges. The results obtained provide presumptive evidence for the multiplication of E. coli strains within the wastewater stream. 7. Comparison of RAPD profiles derived from a range of human and environmental E. coli isolates led to the identification of a 1.6-kb marker. present in a subset of human isolates, but found only infrequently in isolates from animal origin. Primers based on sequence information obtained from this region of the genome yielded a direct PCR assay for the 'human' marker. The reliability and utility of the assay was improved when fluorescently-labelled primers and an automated DNA detection system was employed for the analysis of the amplified PCR products. 8. The diagnostic polymorphism was localised to the region immediately downstream of the gene encoding the glycine decarboxylase P-protein (gcvP). Comparison of DNA sequences from this region in 'marker-positive' and 'marker-negative' E. coli isolates revealed an abrupt loss of homology immediately downstream of the transcription termination point of the gcvP gene. The region of non-homology extends for at least 130 base pairs beyond the gcvP transcription terminator. Sequences spanning this region in marker-negative isolates exhibited a high level homology with that determined for E. coli K-12. The origin of the sequence for the same region in marker-positive isolates was unable to be identified. 9. The utility of the marker for environmental monitoring was investigated in a field study of a rural stream receiving bacterial contamination of both human and animal origin. The marker was consistently detected both in isolates derived from sewage effluents and in those found immediately downstream from sewage inputs-despite the presence of a high background of E. coli of animal origin. However the utility of the marker as a sewage-specific tracer was limited by the relatively low numbers of marker-positive d. coli in sewage effluents (approximately 8%). 10. In summary, the work described in this thesis has confirmed the need for multiple indicator systems in water quality monitoring and has identified appropriate areas for further research. The most promising area for future research is in the development of molecular tools which might provide both a better understanding of genetic variation in indicator organisms and enable methods to be developed for the detection of host-specific strains. Location of other polymorphisms likely to be present in the E. coli genome should enable further host-specific markers to be identified which could complement the 'human' marker identified in this study. / Whole document restricted, but available by request, use the feedback form to request access.

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / This research provides insight into large-scale spatial variation of ozone (O3) distribution in the Auckland region and was designed to increase our understanding of O3 behaviour and distribution in a coastal-urban situation. The research was also carried out with a view to assisting regulatory agencies optimize future monitoring networks, and to help identify locations where human health and natural resources could be at risk in the future. Although the research was limited to one region, the results are valuable for improving the conceptual understanding of formation of high ozone concentrations in a more general sense in the New Zealand and Southern Hemisphere. The work reported in this thesis is aimed at studying O3 concentrations and the influence of the most relevant meteorological variables on an average coastal New Zealand city where precursor emissions are mainly due to traffic exhaust. It deals with the use of Principal Component Analysis method for determining O3 concentrations as a function of meteorological parameters. The study region includes the entire Auckland isthmus, and extends from Whangaparaoa in the North to Pukekohe in the South. Surface O3 data from four sites (Whangaparaoa, Musick Point, the Sky Tower and Pukekohe) for a 4-year period(October 1997 to October 2001) for the Auckland region were examined. Ambient concentration of O3 was characterized in terms of diurnal, weekday/weekend, seasonal and spatial variations in concentration using O3 measurements from the four air quality sites. The monthly average ambient background O3 concentrations at the monitoring sites during this study ranged from 16-30 ppb, much lower than those found in the Northern Hemisphere. The measured seasonal O3 record in the Auckland region, in common with many other remote sites in the Southern Hemisphere, exhibited a summer minimum and a winter maximum. Background concentrations of O3 (as seen in air of marine origin) made a significant contribution to the observed ambient concentrations. A unique feature of Auckland's air quality was the dilution of polluted city air due to the mixing of east coast air into the cleaner west coast circulation leading to overall lower average O3 concentrations in summer. The magnitude, frequency and spatial extent of maximum O3 concentrations were identified, and the observed patterns linked to the prevailing meteorological, topographic, and emission characteristics of the region. However, at no time at any site or season did the O3 concentration exceed the l-hour New Zealand Ministry for the Environment guideline of 75 ppb. O3 depletion was observed to occur at the urban sites, with O3 scavenging by nitric oxide believed to be the dominant depletion mechanism. The seasonal cycle was characterized by elevated O3 concentrations in the winter (nighttime level >24 ppb) and low mixing ratios in the summer (nighttime levels in the range 14-20 ppb). The afternoon O3 maxima found at the three low elevation sites under the impact of "Auckland city" plume were on average, 1.5 to 1.7 times higher than those associated with the "marine sector". A state of the art diagnostic meteorological model, namely CALMET, was used to generate wind fields for the Auckland region. These wind fields were then used to construct backward trajectories on days when high O3 concentrations (episodes up to 6l ppb) were observed. The un-even distribution of the meteorological monitoring sites provided justification for running the diagnostic model CALMET and exploring the utility of using such a model when topography/land use prohibits monitoring sites in certain sections of the domain. It was found that the high O3 events recorded at the monitoring sites during the study period coincided with transport episodes originating from Auckland's urban and industrial areas. For the occurrence of high O3 concentrations at downwind sites, it was found that not only wind direction from the urban areas of Auckland, but that high solar radiation was important as well. On four out of the five cases when photochemical production of O3 was found to occur, high O3 concentrations were associated with a particular type of diurnal evolution of wind direction wind fields associated with the sea breezes. In addition to the analysis of ambient o3 concentrations and O3 episodes, field measurements were carried out in an attempt to detect and subsequently understand the interaction between particulate matter and O3 in the Auckland region, a multivariate statistical analysis approach was utilized. Particulate matter in the size range 2.75-6.25 μm accounted for over 70% of the total aerosol concentration at all sampling sites. Surface area of particulate matter variable (especially in the size class 2.75-4.25 μm) was statistically significant in explaining variation in O3 concentration. However, the net change in the adjusted R2 indicated that the effect of adding particulate matter in the multiple regression model for the present dataset was relatively minor except at Musick Point.