I investigated the pollination ecology ofHesperidanthus suffrutescens, an endangered Utah shrub, as well as the effects that road dust and local plant density may have on successful reproduction. The breeding system study included four pollination treatments to determine the degree of self-compatibility. Flower visitors were collected and probable pollinators were identified. To determine the effects of road dust on reproduction, I measured dust deposition, plant size, and reproduction in plots of increasing distance from an unpaved road. I also measured dust effects on pollination success and stomatal conductance. I determined the number of individuals per plot and the relative isolation of individuals, and measured their reproductive success to determine any significant effects. Reproduction was measured as total fruit set per plant as well as estimates of total plant seeds and plant seed weight (mg). The results of the pollination treatments and the identity of pollinators are described in Chapter 2. Both autogamy and geitonogamy (selfing treatments) produced substantially and significantly fewer fruits, seeds, and seeds per fruit than xenogamy (outcrossing treatment). Additionally, outcrossed flowers produced significantly more fruits, seeds, and seeds per fruit than did open control flowers with no difference in mean seed weight. In total, 77 flower visitors were collected over the course of the study. Probable pollinators include severalAndrena,Halictus, andDialictusbees. The effects of road dust on the reproductive success ofH. suffrutescensare discussed in Chapter 3. Dust deposition decreased with increased distance from the road and was significantly correlated with decreased fruit set for plants of a given size. Other reproductive metrics showed the same negative patterns although not significantly. The results suggest that dust may disrupt pollination and affect the physiology of plants, resulting in decreased reproduction. Dust deposition did not decrease pollination success of dusted flowers but did negatively impact stomatal conductance of leaves. The main effects of patch size and relative isolation on reproductive success are discussed in Chapter 4. Both patch size and relative isolation had no significant impacts on reproductive success. Increased patch size did indicate negative patterns on reproductive metrics, although not significantly. Similarly, increased isolation indicated increased reproductive success, although not significantly.

Text-to-scene generation systems take input in the form of a natural language text and output a 3D scene illustrating the meaning of that text. A major benefit of text-to-scene generation is that it allows users to create custom 3D scenes without requiring them to have a background in 3D graphics or knowledge of specialized software packages. This contributes to making text-to-scene useful in scenarios from creative applications to education. The primary goal of this thesis is to explore how we can use text-to-scene generation in a new way: as a tool to facilitate the elicitation and formal documentation of language. In particular, we use text-to-scene generation (a) to assist field linguists studying endangered languages; (b) to provide a cross-linguistic framework for formally modeling spatial language; and (c) to collect language data using crowdsourcing. As a side effect of these goals, we also explore the problem of multilingual text-to-scene generation, that is, systems for generating 3D scenes from languages other than English. The contributions of this thesis are the following. First, we develop a novel tool suite (the WordsEye Linguistics Tools, or WELT) that uses the WordsEye text-to-scene system to assist field linguists with eliciting and documenting endangered languages. WELT allows linguists to create custom elicitation materials and to document semantics in a formal way. We test WELT with two endangered languages, Nahuatl and Arrernte. Second, we explore the question of how to learn a syntactic parser for WELT. We show that an incremental learning method using a small number of annotated dependency structures can produce reasonably accurate results. We demonstrate that using a parser trained in this way can significantly decrease the time it takes an annotator to label a new sentence with dependency information. Third, we develop a framework that generates 3D scenes from spatial and graphical semantic primitives. We incorporate this system into the WELT tools for creating custom elicitation materials, allowing users to directly manipulate the underlying semantics of a generated scene. Fourth, we introduce a deep semantic representation of spatial relations and use this to create a new resource, SpatialNet, which formally declares the lexical semantics of spatial relations for a language. We demonstrate how SpatialNet can be used to support multilingual text-to-scene generation. Finally, we show how WordsEye and the semantic resources it provides can be used to facilitate elicitation of language using crowdsourcing.

Abstract not available

This study arose from a conservation dilemma between the need to conserve a population of the endangered pink tailed legless lizard, Aprasia parapulchella and crucial upgrading of Canberra's sewerage treatment plant. The study focused on conservation biology and how it could be applied to the management of A. parapulchella. As little biological information was available on A. parapulchella, research was focused on obtaining data that would be beneficial in the conservation management of the species. This included the dietary and habitat requirements, population status and structure and population relocation and habitat restoration. A. parapulchella was found to occur in low relative densities at all sites (maximum 9.1 lizards per 1000 rocks turned, Mt Taylor). Populations of the species appear to be larger than has been considered previously, however numbers are still below 500 individuals per site. Sexual dimorphism is present in the species with females achieving longer snout-vent lengths than males. Sex ratios were equal in all sites, however average size of lizards was different at the four main study sites, indicating differences in population structure. The life history of A. parapulchella can be summarised as long-lived, late maturing (3rd or 4lh year of life) with a low reproductive rate (single clutch of two eggs per year). A. parapulchella is a dietary specialist feeding on about 11 different species of ants. Interestingly, the lizards also live with the same species that they feed upon. By far the most important species in the diet and homesite requirements of A. parapulchella is the ant Iridomyrmex rufoniger. A. parapulchella shows positive selection towards the species in homesite choice, is found most commonly in the nests of this ant, and selects its brood over the brood of other ant species in diet experiments. In addition this ant was represented in the diet of A. parapulchella more commonly than any other ant. The relationship between ants and A. parapulchella appears to be unique and the mechanisms of this specialisation require further investigation. The vegetation associations where A. parapulchella occur are different across the range of the species. Sites at Bendigo and West Wyalong support mallee communities, while in the ACT region, Bathurst and Tarcutta, A. parapulchella appears to be a habitat specialist, being found most commonly in sites dominated by grass species, both native and introduced. A. parapulchella lives beneath rocks in grassland sites. Rocks that the species uses for homesites are commonly between 100 and 150 mm wide, 120-220 mm long and 50 and 150 mm thick. There were significant differences between the sizes of rocks used at different sites and between different seasons. Rocks used by females were significantly larger than rocks used by juvenile lizards. Twelve other species of reptile were found to use rocks in grassland sites, with rocks partitioned within the community on the basis of size and shape. An experimental relocation of A. parapulchella was conducted to try to determine ideal habitat conditions for the survival of relocated individuals. Unfortunately of the 114 lizards released, only 15 were recaptured. Although recaptures were low, trends indicated that treatments of Themeda triandra and very high rock densities had the highest survival rates. Key habitat variables were reinstated at the Lower Molonglo Dam site however restoration is preliminary with more time needed for the habitat to be suitable for occupation by A. parapulchella. What started as a situation that was likely to have a deleterious impact on A. parapulchella has been turned into a conservation success story. Without the involvement of ACT Electricity and Water, much of the basic biology of A. parapulchella would still be unknown. The findings will now form a solid basis from which to effectively conserve A. parapulchella. The project provides an example of the conservation success that can be achieved between industry and conservation partnerships.

Habitat requirements and habitat use for Pseudophryne australis and Heleioporus australiacus were investigated to aid management of these threatened frogs around Sydney, Australia. Much of the work focussed on roads, commonly encountered features in the habitat of both species. The habitat requirements based on locality records of both frogs in the Sydney Basin were investigated at four spatial scales. Both species are habitat specialists. They showed a strong geological association with Hawkesbury Sandstone and occupy upper topographic areas with ephemeral watercourses of gentle gradients. Both frogs occur predominantly in areas of higher precipitation and milder temperature regimes compared to averages representative of the region. Leaf litter is an important feature of P. australis breeding sites, whereas H australiacus generally associate with crayfish burrows. Both species are dependent on natural vegetation with a complex structure. H australiacus have a relatively long larval period (3 - 12 months) and breed in ephemeral pools, exposing their tadpoles to the risk of dying due to early pond drying. In the laboratory, tadpoles responded to decreasing water levels by shortening their larval periods and metamorphosing earlier than siblings held at constant water level. Despite this plastic response, a number of pools in the field failed to produce metamorphs due to early drying, an observation also made on P. australis. Regular monitoring of breeding sites revealed increased reproductive success away from roads for both species probably because of relatively longer hydroperiods. Spatial distributions and associations with habitat features, and movement patterns of both frogs were further investigated using mark-recapture methods. Both species showed strong site fidelity. P. australis formed small aggregations and predominantly selected leaf litter piles despite their relatively low availability. Leaf litter piles in creeks moved over time and the animals moved with these piles. In contrast, H australiacus individuals formed no aggregations and showed no preference for any available structural vegetation type. Locations of individuals were independent of relative distances to creeks and artificial drains, but males appeared to be more common near culverts. However, individuals were randomly distributed in space and nearest-neighbour distances were high relative to individual movement distances, suggesting minimal overlap between relatively large home ranges. Radio-telemetry demonstrated that some H australiacus individuals burrow in the road environment. There they would be at risk of being dug up and possibly injured during road works. The results are discussed in relation to the spatial requirements of both species and the protection of utilised habitat features. Management options are suggested to mitigate the impacts of road works. Differences in spatial dynamics of both frogs with overlapping habitats highlighted in this study require species-specific management approaches.

Habitat requirements and habitat use for Pseudophryne australis and Heleioporus australiacus were investigated to aid management of these threatened frogs around Sydney, Australia. Much of the work focussed on roads, commonly encountered features in the habitat of both species. The habitat requirements based on locality records of both frogs in the Sydney Basin were investigated at four spatial scales. Both species are habitat specialists. They showed a strong geological association with Hawkesbury Sandstone and occupy upper topographic areas with ephemeral watercourses of gentle gradients. Both frogs occur predominantly in areas of higher precipitation and milder temperature regimes compared to averages representative of the region. Leaf litter is an important feature of P. australis breeding sites, whereas H australiacus generally associate with crayfish burrows. Both species are dependent on natural vegetation with a complex structure. H australiacus have a relatively long larval period (3 - 12 months) and breed in ephemeral pools, exposing their tadpoles to the risk of dying due to early pond drying. In the laboratory, tadpoles responded to decreasing water levels by shortening their larval periods and metamorphosing earlier than siblings held at constant water level. Despite this plastic response, a number of pools in the field failed to produce metamorphs due to early drying, an observation also made on P. australis. Regular monitoring of breeding sites revealed increased reproductive success away from roads for both species probably because of relatively longer hydroperiods. Spatial distributions and associations with habitat features, and movement patterns of both frogs were further investigated using mark-recapture methods. Both species showed strong site fidelity. P. australis formed small aggregations and predominantly selected leaf litter piles despite their relatively low availability. Leaf litter piles in creeks moved over time and the animals moved with these piles. In contrast, H australiacus individuals formed no aggregations and showed no preference for any available structural vegetation type. Locations of individuals were independent of relative distances to creeks and artificial drains, but males appeared to be more common near culverts. However, individuals were randomly distributed in space and nearest-neighbour distances were high relative to individual movement distances, suggesting minimal overlap between relatively large home ranges. Radio-telemetry demonstrated that some H australiacus individuals burrow in the road environment. There they would be at risk of being dug up and possibly injured during road works. The results are discussed in relation to the spatial requirements of both species and the protection of utilised habitat features. Management options are suggested to mitigate the impacts of road works. Differences in spatial dynamics of both frogs with overlapping habitats highlighted in this study require species-specific management approaches.

Bibliography: leaves 108-113. This dissertation givens and analytical overview of the research and recovery actions for the Orange-bellied Parrot and the Kangaroo Island Glossy Black-Cockatoo and evaluates the successful practices and those which have failed. This research shows that new recovery action initiatives can still be found and that not all recommendations in recovery plans are followed, which impairs the efficiency of the recovery efforts. It concludes that a successful recovery effort of these two bird species lies in continued research, a serious compliance with the recommendations made by recovery plans and by further environmental education in society.

The goal of this inductive study was to identify factors that facilitate and inhibit binational collaboration in the recovery of endangered species in the northern Mexico borderlands, focusing on the Mexican wolf (Canis lupus baileyi). A conceptual model was developed using qualitative techniques, providing the basis for design of a mail survey. The target population included participants with experience in recovery efforts for over a dozen species at risk in the region. Long interviews were recorded with 44 participants from Mexico and the United States. Thematic hierarchical analysis was used to develop a conceptual model of how interviewees talked about factors influencing binational collaboration. Issues were classified in five thematic clusters: project, organization, people, resources, culture/history. The survey was used to conduct a needs assessment, measuring respondents' attitudes about the relative priority of issues identified in the conceptual model. High priority needs were identified from each thematic cluster: (a) equitable participation in project design and implementation, (b) continuity of personnel, (c) coordination of federal, state and local efforts, (d) increased funding, managed with accountability, and (e) exchange visits to facilitate understanding of diverse perspectives. Responses to almost half the survey items indicated accord among the sample of respondents, providing a basis for shared common ground. The nature of discord was within the range of "manageable", with no clear polarization of attitudes measured. This exploratory data analysis suggested that the structure of the conceptual model developed from the Mexican wolf case study was generally a valid basis for future deductive analysis and reflection by practitioners. For 82% of 22 statements of need, priorities of participants in the Mexican wolf recovery efforts did not differ significantly from other respondents. Nationality (of respondents) significantly affected priority rankings for only 18% of the need statements. Significant effects of five demographic variables indicated that interactive effects should be examined in future multivariate analyses to determine how respondents' attitudes on issues related to priority rankings. Recommendations were provided for a more efficient and effective approach to collaborative problem-solving, engaging reflective practitioners from the private and public sectors in principled negotiation processes to better understand diverse perspectives.

The goal of my dissertation was to describe the dynamics of a group of Houston toads located at the Griffith League Ranch (GLR), Bastrop County, Texas. My research included using statistical modeling to predict activity and abundance, mark-recapture techniques to estimate survivorship, and simulation modeling to explore the impacts of the difference in age at first reproduction and to project the future dynamics of the population at the GLR. From 2001 – 2005, 225 individual Houston toads (199 M : 26 F) were captured using two methods: breeding pond surveys and drift fences. Houston toads were neither caught equally among capture methods, nor across years. Toad activity was mostly confined within their breeding season, and activity was not continuous. A logistic regression indicated activity depended on time of year, mean precipitation, mean minimum daily temperature, and mean percent lunation as well as two-way interactions with moon-phase and other variables. Abundance depended on time of year, current precipitation, minimum temperature, and two-way interactions between time of year and the other two variables. Twenty-one of the 199 males (10.5%) and no females were recaptured among years. The probability of male survival was estimated using program MARK. Eight of 16 candidate models were supported and all but one contained precipitation as a covariate, indicating precipitation is important for Houston toad survival. Survivorship estimates varied from 0.1 to 0.41. The sex ratio was significantly male-biased. The odds of catching females in traps were 3.5 greater than capturing females in a pond, while the odds of capturing males in a trap were 0.28 compared to ponds. Results from a simulation model indicated the sex ratio is biased because of the difference in maturation times between males and females, coupled with high juvenile mortality. Results from an individual-based, spatially-explicit, stochastic simulation model, indicated a relatively low probability (~ 0.013) of B. houstonensis going extinct at the GLR within the next 10 years. Emergent properties of the model were similar to results observed in the field or reported in the literature. The model also identified that dispersal of Houston toads should be a future research priority.

Genetic diversity has rarely been the focus of study in species at risk in Canada. <i>Tradescantia occidentalis</i> is one of 157 species listed by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC 2005a). This species is nationally threatened due to the limited number of populations, their geographic isolation from each other and from the main distribution in the United States of America, and habitat loss. The National Recovery Team for Plants at Risk in the Prairie Provinces and the Alberta Western Spiderwort Recovery Team have called for research into the habitat requirements, demography, and genetic diversity of <i>T. occidentalis</i> in Canada. As a result, this study was designed to address the following objectives: 1) to conduct an inventory of the Canadian populations, 2) to investigate intra- and interpopulation genetic diversity in <i>Tradescantia occidentalis</i>, and 3) to provide recommendations for the conservation management of this species. <p>Information on demography and plant communities in <i>Tradescantia occidentalis</i> habitats indicated that the numbers of individuals in the Saskatchewan and Manitoba populations were similar to previous surveys; however, the Alberta population was significantly larger in number than prior estimates, indicating population growth. Taxonomic lists were prepared for each province in habitats with and without <i>T. occidentalis</i>. Community types, as separated by RA analysis, differed by province and not by association with <i>T. occidentalis</i>. Euphorbia esula, an invasive species in Canada, was observed in the Saskatchewan and Manitoba populations but was absent in Alberta. <p> Using amplified fragment length polymorphisms (AFLPs), genetic diversity was assessed at the intra- and interpopulation levels. Relatively low levels of intrapopulation variation were observed in Saskatchewan and Alberta, while higher levels were found in Manitoba. Gene flow via pollen or propagule transfer may account for higher genetic diversity among the closely situated Manitoba populations. The lack of correlation between dendrogram topology and geographic distribution suggests panmixia in all populations. Levels of intrapopulation diversity were low to moderate depending on primer combination used, indicating that populations are isolated within each province. <p>Information on population demography and genetic diversity are important within a conservation context. The large number of individuals within each population and the perceived increase in some populations suggest that the existing populations of <i>Tradescantia occidentalis</i> are relatively stable. Although levels of genetic diversity are low in Saskatchewan and Alberta compared to Manitoba, it appears that all populations are adapted to their local environments based on their apparent size and stability. The most viable conservation strategy for this species is in situ protection. This should include controlling invasive plant species, monitoring grazing, and preventing further habitat fragmentation. Ex situ methods must also be explored. Transplantation of individuals from one population to the next may not be a successful conservation strategy due to the moderate level of population differentiation. Alternatively, it is recommended that a seed bank from each population be implemented in case of a drastic population decline.