abstract: This dissertation uses a comparative approach to investigate long-term human- environment interrelationships in times of climate change. It uses Geographical Information Systems and ecological models to reconstruct the Magdalenian (~20,000- 14,000 calibrated years ago) environments of the coastal mountainous zone of Cantabria (Northwest Spain) and the interior valleys of the Dordogne (Southwest France) to contextualize the social networks that could have formed during a time of high climate and resource variability. It simulates the formation of such networks in an agent-based model, which documents the processes underlying the formation of archaeological assemblages, and evaluates the potential impacts of climate-topography interactions on cultural transmission. This research then reconstructs the Magdalenian social networks visible through a multivariate statistical analysis of stylistic similarities among portable art objects. As these networks cannot be analyzed directly to infer social behavior, their characteristics are compared to the results of the agent-based model, which provide characteristics estimates of the Magdalenian latent social networks that most likely produced the empirical archaeological assemblage studied. This research contributes several new results, most of which point to the advantages of using an inter-disciplinary approach to the study of the archaeological record. It demonstrates the benefits of using an agent-based model to parse social data from long- term palimpsests. It shows that geographical and environmental contexts affect the structure of social networks, which in turn affects the transmission of ideas and goods that flow through it. This shows the presence of human-environment interactions that not only affected our ancestors’ reaction to resource insecurities, but also led them to innovate and improve the productivity of their own environment. However, it also suggests that such alterations may have reduced the populations’ resilience to strong climatic changes, and that the region with diverse resources provided a more stable and resilient environment than the region transformed to satisfy the immediate needs of its population. / Dissertation/Thesis / Appendix_D_Sites_Dates / Appendix_E_Flowchart_Biome_Reconstruction / Appendix_H_Flowchart_ABM / Appendix_I_Flowchart_Social_Network / Appendix_J_Portable_Art_Objects / Appendix_J_Art_Characteristics / Appendix_L_Poster_Summary / Appendix_A_Prehistoric_Fauna / Appendix_B_Modern_PFT_Distribution / Appendix_C_Prehistoric_PFT_Distribution / Doctoral Dissertation Anthropology 2017

Forecasting the potential range of invasive species is a critical component for risk assessment, monitoring, and management. However, many of these invasive species are not yet at equilibrium which can be problematic for many modelling approaches. Using the climate matching method, MaxEnt, a series of species distribution models (SDMs) and risk analysis maps were created for select apple snail species in Florida: Pomacea canaliculata, P. diffusa, and P. maculata. Apple snails, freshwater gastropods in the family Ampullariidae, are native to South America and were introduced to the United States via the pet trade approximately 40 years ago. These highly invasive species have already been introduced in ten states and established in at least seven. The models and risk analysis in this study show the majority of Florida was at least moderately suitable for all apple snails modeled, with P. maculata posing the greatest threat.

With exploding human populations and landscapes that are changing, an increasing number of wildlife species are brought to the brink of extinction. In Canada, the eastern hog-nosed snake, 'Heterodon platirhinos', is found in a limited portion of southern Ontario. Designated as threatened by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC), this reptile has been losing its habitat at an alarming rate. Due to the increase in development of southern Ontario, it is crucial to document what limits the snake’s habitat to direct conservation efforts better, for the long-term survival of this species. The goals of this study are: 1) to examine what environmental parameters are linked to the presence of the species at a landscape scale; 2) to predict where the snakes can be found in Ontario through GIS-based habitat suitability models (HSMs); and 3) to assess the role of biotic interactions in HSMs. Three models with high predictive power were employed: Maxent, Boosted Regression Trees (BRTs), and the Genetic Algorithm for Rule-set Production (GARP). Habitat suitability maps were constructed for the eastern hog-nosed snake for its entire Canadian distribution and models were validated with both threshold dependent and independent metrics. Maxent and BRT performed better than GARP and all models predict fewer areas of high suitability when landscape variables are used with current occurrences. Forest density and maximum temperature during the active season were the two variables that contributed the most to models predicting the current distribution of the species. Biotic variables increased the performance of models not by representing a limiting resource, but by representing the inequality of sampling and areas where forest remains. Although habitat suitability models rely on many assumptions, they remain useful in the fields of conservation and landscape management. In addition to help identify critical habitat, HSMs may be used as a tool to better manage land to allow for the survival of species at risk.

Despite an IUCN conservation status of critically endangered, Juliana’s golden mole (Neamblysomus julianae) has received no ecological research attention to date. The species urgently requires conservation attention, but a poor understanding its biology, ecology and distribution makes effective conservation planning almost impossible. In light of this, a thorough understanding of the habitat requirements for this habitat specific species was needed. Additionally, the cryptic lifestyle of Juliana’s golden mole has resulted in very few distribution records for the species. Little was known about the animals’ daily and seasonal activity patterns or movement patterns. This study revealed that Juliana’s golden mole is range restricted on account of sandy soils that have a uniform particle size distribution. This feature limits substrate compaction, making tunnelling far easier for these small mammals. In comparison to uninhabited areas, occupied habitat had greater vegetation cover provided by trees and shrubs, and this would be expected to provide a cooler and moister microenvironment. The ecological parameters associated with the species presence were then used, in addition to existing GIS data, to predict regions of potentially suitable habitat. This process revealed large potentially inhabitable areas in the northern parts of South Africa. Preliminary ground-truthing has shown the model to be relatively accurate with three potentially new populations having been identified. Contrary to what has been reported in the literature, Juliana’s golden mole does not appear to be strictly nocturnal. Instead, an adaptive pattern of activity was observed, governed by soil temperature. Body temperature was found to fluctuate to some degree with that of the soil temperature, but fluctuations were regulated by behavioural thermoregulation. Seasonal activity is highly correlated with rainfall. Rain moistens the soil making it significantly softer and far easier to tunnel through. Using this and other information acquired through the course of the study, the thesis culminates with an evaluation of conservation concerns and proposed conservation management actions. Copyright / Dissertation (MSc)--University of Pretoria, 2009. / Zoology and Entomology / unrestricted

This thesis aims to implement dynamic load balancing mechanism into the parallel simulation model of the heat distribution in a CPU cooler. The first part introduces theoretical foundations for dynamic load balancing, describing current solution approaches. The second part refers to the heat distribution model and related topics such as MPI communications library or HDF library for data storage. Then it proceeds to the implementation of simulation model with dynamic 2D decomposition of square model domain. Custom geometry based dynamic load balancing algorithm was introduced, which works with this decomposition. Important part of the implementation is Zoltan library, used especially for data migration. At the end, a set of experiments was presented, which demonstrates load balancing abilities of designed model together with conclusions and motivation for future research.

The dissertation builds a bridge from the social-philosophical works of Hobbes, Rousseau, and Smith to current game theoretical models explaining social order to two empirical applications in a rural region of Guinea. The work focuses on the central role of natural resources for human societies and uses two empirical studies to test general hypotheses on the drivers of the sustainable exploitation and the equitable distribution of natural resources. The work strictly follows a quantitative empirical approach. After discussing the methodological foundations of statistical causal analysis in a first step, the various approaches are evaluated in a simulation study. Finally, the most successful of these approaches finds its application in the analysis of the empirical data, which was collected during field research in Guinea. In the observational study, socioeconomic data are combined with environmental data to model the influence of humans on the occurrence of wild species that are exploited for economic purposes in the study area. In the experimental study, socioeconomic data is linked to experimental data from a resource distribution game to identify factors affecting the people’s behavior when sharing a common natural resource. Both the exploitation of wild species and the sharing of a natural resource show that the social context is crucial for the understanding of human behavior, as postulated by Hobbes, Rousseau, and Smith: reputation, market integration, inequality and homogeneity of the population, as well as the salience of moral norms and property rights are essential dimensions. / Die Dissertation widmet sich der zentralen Rolle von natürlichen Ressourcen für menschliche Gesellschaften. Anhand zweier empirischer Studien werden allgemeine Hypothesen zu den Determinanten der nachhaltigen Ausbeutung und gerechten Verteilung von natürlichen Ressourcen getestet. Dabei wird in der Arbeit eine Brücke von den Werken der Sozialphilosophen Hobbes, Rousseau und Smith, über aktuelle spieltheoretische Modelle zur Erklärung sozialer Ordnung, hin zu zwei empirischen Anwendungen in einer ländlichen Region Guineas geschlagen. Die Arbeit folgt strikt einem empirischen quantitativen Ansatz. Nach der Erörterung der methodischen Grundlagen der statistischen Kausalanalyse werden die verschiedenen Ansätze im Rahmen einer Simulationsstudie evaluiert. Schließlich findet der erfolgreichste dieser Ansätze in der Auswertung der Daten, die bei der Feldforschung in Guinea erhoben wurden, Anwendung. In der Beobachtungsstudie werden sozioökonomische Daten mit ökologischen Daten verknüpft, um den Einfluss der Menschen auf das Vorkommen von wilden Arten, die zu ökonomischen Zwecken ausgebeutet werden, im Untersuchungsgebiet zu modellieren. In der Experimentalstudie werden sozioökonomische Daten mit Experimentaldaten aus einem Ressourcenverteilungsspiel verknüpft, um Faktoren zu identifizieren, welche sich auf das Verhalten von Menschen beim Teilen einer gemeinsamen natürlichen Ressource auswirken. Sowohl für die Ausbeutung der wilden Arten, wie auch für das gemeinsame Teilen einer natürlichen Ressource gilt, dass der soziale Kontext für das Verhalten der Menschen von großer Relevanz ist, wie von Hobbes, Rousseau und Smith postuliert: Reputation, Marktintegration, Ungleichheit und Homogenität der Bevölkerung sowie die Salienz moralischer Normen und Eigentumsrechte sind entscheidende Dimensionen.

Large-scale land-use change across the tropics has led to the decline of animal populations and their habitat. With large investments into mining, hydropower dams and industrial agriculture this trend is likely to continue. Consequently, there is a need for systematic land-use planning to set aside areas for protection and allocate scarce conservation funding effectively. Even though primates are relatively well studied, data-driven systematic planning is still rarely implemented. The overall aim of this dissertation was to investigate population parameters needed for evidence-based conservation planning for the critically endangered western chimpanzee (Pan troglodytes verus) in West Africa. To this end, I compiled density datasets covering the entire geographic range of this taxon from the IUCN SSC A.P.E.S. database and modeled chimpanzee densities as a function of 20 social-ecological variables. I found that western chimpanzees seemingly persist within three social-ecological configurations: rainforests with a low degree of anthropogenic threats, steep areas that are less likely to be developed and are harder to access by humans, and areas with a high prevalence of cultural taboos against hunting chimpanzees. The third configuration of reduced hunting pressure is not yet reflected in commonly implemented conservation interventions, suggesting a need for designing new approaches aimed at reducing the threat of hunting. Based on the modeled density distribution, I estimated that 52,811 (95% CI 17,577-96,564) western chimpanzees remain in West Africa, and identified areas of high conservation value to which conservation interventions should be targeted. These results can be used to inform the expansion of the protected area network in West Africa, to quantify the impact of planned industrial projects on western chimpanzees, and to guide the systematic allocation of conservation funding. In addition, this thesis highlights the unique position of taxon-specific databases of providing access to high-resolution data at the scale needed for conservation planning. Data-driven conservation planning has the potential to enable conservationists to respond more proactively to current and emerging threats, and ultimately improve conservation outcomes.

No description available.

The genus Psilochalcis, of the Family Chalcididae, was originally described in 1905 by Kieffer. Mainly considered an Old-World taxon, the first North American Psilochalcis were not identified until 1981 by Grissell and Schauff. Little is known about the species distributions, biologies and ecological relationships of these parasitic wasps. This dissertation describes research conducted in central Utah setting arrays of Malaise traps in 4 different habitat types common to the Great Basin at three separate locations. A result of this sampling revealed a high abundance of multiple species of Psilochalcis wasps, particularly from one location and two habitat types. Chapter 1 describes three new species of Psilochalcis wasps namely; P. adenticulata Petersen, P. minuta Petersen, and P. quadratis Petersen. A review of all North American Psilochalcis species explains their distribution in Utah and throughout the surrounding western United States. A taxonomic key for all North American Psilochalcis species is given. Chapter 2 examines the seasonal abundance of P. minuta and P. quadratis and their associations with two common Great Basin habitat types. Both species show their highest abundance from late June through early August. Their peak abundance is shown to change dependent on the environmental conditions of temperature and precipitation. Psilochalcis minuta is significantly associated with pinyon/juniper (Pinus edulis or P. monophylla and Juniperus osteosperma) and P. quadratis is significantly associated with cheatgrass (Bromus tectorum). Chapter 3 describes the building of species distribution models for P. minuta and P. quadratis using a maximum entropy (Maxent) approach. Ten environmental variables were used to predict areas of optimal suitable habitat for each species. Multiple predicted sites were field sampled to test each model's effectiveness. Psilochalcis minuta occurred at nearly 90% of predicted sites, and P. quadratis occurred at 50% of predicted sites. Both species occurred at some non-predicted sites in other habitat types. Model analyses and field-testing results show the P. minuta model to be reliable in predicting areas of probable species occurrence, while the P. quadratis model is much less reliable in doing so. Aspect and fire disturbance show the highest percent contribution to both species' models. Slight differences in variable percent contribution between models suggest these species have sympatric distributions. Soil and slope are more important predictors of optimal suitable habitat for each species. Maintaining integrity between model predictions and field testing gave insights into other factors contributing to probable occurrence of Psilochalcis species.

Biodiversity monitoring and predictions of species occurrence are essential to develop outcome-oriented conservation management plans for endangered species and assess their success over time. To assess distribution and patterns of habitat use of large mammal assemblages in Thailand, with a focus on the endangered dhole (Cuon alpinus), I first implemented a long-term camera-trapping project carried out with park rangers from October 2003 through October 2007 in Khao Yai National Park. This project was extremely successful and may serve as a regional model for wildlife conservation. I found significantly lower relative abundance indices for carnivore species, and collectively for all mammals compared to data obtained in 1999-2000, suggesting population declines resulting from increased human activity. I integrated this data into maximum entropy modeling (Maxent) to further evaluate whether ranger stations reduced poaching activity and increased wildlife diversity and abundances. I then conducted a focused camera trap survey from January 2008 through February 2010 in Khao Ang Rue Nai Wildlife Sanctuary to gather critical baseline information on dholes, one of the predator species that seemed to have declined over time and that is exposed to continued pressure from humans. Additionally, I led a collaborative effort with other colleagues in the field to collate and integrate camera trap data from 15 protected areas to build a country-wide habitat suitability map for dholes, other predators, and their major prey species. The predicted presence probability for sambar (Rusa unicolor) and leopards (Panthera pardus) were the most important variables in predicting dhole presence countrywide. Based on my experience from these different field ecological surveys and endeavors, it became clear that local people's beliefs may have a strong influence on dhole management and conservation. Thus, I conducted villager interview surveys to identify local attitudes towards dholes, document the status of dholes in wildlife sanctuaries adjacent to Cambodia, and determine the best approach to improve local support for dhole conservation before proceeding with further field studies of the species in Thailand. A photograph of a dhole was correctly identified by only 20% of the respondents. My studies provide evidence that some protected areas in Thailand continue to support a diversity of carnivore speices of conservation concern, including clouded leopards (Neofelis nebulosa), dholes, and small felids. However, dholes' impact on prey populations may be increasing as tiger (Panthera tigris) and leopards are extripated from protected areas. The next step in dhole conservation is to estimate the size and stability of their fragmented populations and also focus on maintaining adequate prey bases that would support both large felids and dholes