In recent years, the application of resampling methods to dependent data, such as time series or spatial data, has been a growing field in the study of statistics. In this dissertation, we discuss two such applications. In spatial statistics, the reliability of Kriging prediction methods relies on the observations coming from an underlying Gaussian process. When the observed data set is not from a multivariate Gaussian distribution, but rather is a transformation of Gaussian data, Kriging methods can produce biased predictions. Bootstrap resampling methods present a potential bias correction. We propose a parametric bootstrap methodology for the calculation of either a multiplicative or additive bias correction factor when dealing with Trans-Gaussian data. Furthermore, we investigate the asymptotic properties of the new bootstrap based predictors. Finally, we present the results for both simulated and real world data. In time series analysis, the estimation of covariance parameters is often of utmost importance. Furthermore, the understanding of the distributional behavior of parameter estimates, particularly the variance, is useful but often difficult. Block bootstrap methods have been particularly useful in such analyses. We introduce a new procedure for the estimation of covariance parameters for replicated time series data.

The purpose of this project is to design a computer program to help children understand a strategy for changing vantage points within an imagined space and gain knowledge of how spatial transformations work. The developed software, called Viewpoints, presents a virtual three-dimensional environment to be explored and modified by the user. Object and camera manipulations are illustrated through animation. Furthermore, the program was designed to have an intuitive interface and be easy to access. This should allow the software’s target audience of children to focus on the spatial orientation and spatial visualization aspects. A small study evaluated the software in terms of content, instructional design, technical quality, student use, and instructor use. The study provided valuable feedback on how to improve Viewpoints in the future. Information gathered suggests the issue of speed should be addressed and supplemental materials should be added.

Residential environments have a close relationship to everyday human life. They are not only significant places for human activity; they also play an important role in the physical and social contexts. Additionally, welcoming and attractive open spaces in residential neighbourhoods increase opportunities for informal interaction among residents, and create hospitable places for humans to live, build family, develop community,and interact with their natural environment. The intention of this practicum is to provide a lens through which to reconsider the definition of home, focusing on outdoor space as a crucial and key component to increasing overall satisfaction with one’s home. The study of Lilong housing in Shanghai plays a significant role in this practicum, through exploring the spatial-social relationship of Lilong housing, looking at how physical and spatial layout influence patterns of space use and movement, and examining impact on the social life (Hillier, 1987). These findings are derived from carefully investigating both the physical and community aspects of Lilong, and prove that the hierarchy of social structure is reflected and supported by a hierarchy of communal spaces (Gehl, 2011). Therefore, the most important spatial feature graduated, outdoor spaces with public, semi-public, semi-private and private areas - dominates the design decisions. The design is presented as a major means to redefine and reform outdoor spaces in residential Winnipeg, in relation to social interaction from a landscape architecture perspective. / February 2016

Scotland's tourist industry partly depends upon the quality of the Scottish landscape. However, despite demands for improved management of landscape resources, there is no standard method for the assessing landscape quality. This research takes a user-based approach to this problem and explores the use of Kelly's Personal Construct Theory (PCT) in eliciting underpinning attributes and dimensions of perception in a range of uses and across a range of Scottish landscapes. A novel aspect of the research is that it involves experimentation in mapping the resultant constructs through use of a Geographic Information System (GIS). Sixteen key constructs were gained from one to one. interviews and used in the preliminary mapping experiments. These showed that it was possible to express the constructs spatially. To evaluate between user group/landscape type responses a questionnaire was designed, piloted and applied. A total of 1286 responses were analyzed. Differences were found between landscape type but were stronger between user groups. The research has demonstrated that the application of PCT coupled to GIS is a valuable way of exploring landscape perception/landscape quality and their spatial expression.

This paper deals with evaluating 3D-bootstrap for the mining company New Boliden in an attempt to revise their current method of calculating average gold riches in different areas. The purpose is to find one-sided 3D-bootstrap confidence bound of the average gold riches. There lacks well-defined theory behind using 3D-bootstrap, in this paper the variogram is used as an estimate of dependencies between the observations, and the block length is chosen to be higher than this estimate. In aid of this, a simulated data material is conducted to check the validity of 3D-bootstrap in a controlled area where the theoretical value is known. The results are inconclusive, and further studies are needed.

In recent years, spatial data derived from cell phones has become increasingly recognized as a valuable data source for urban analysis. Using a sizeable dataset depicting the physical movements of several million Swedish and Danish cell phones during 24 hours, an analysis of aggregated individual mobility levels and mobility patterns is conducted. The analysis covers two measurements of mobility, total diurnal mobility and commuting mobility. Findings indicate that phone data may provide seminal insights on otherwise scarcely accessible information on how space is experienced and interacted with depending on the individual's residential location. The data is analysed using a broad set of spatial analysis techniques incorporating both statistical and visual representations of spatial mobility and spatial relationships.

Successful learning about physical systems is thought to depend on the development of a mental representation of the system's dynamic behavior, which constitutes a mental model, rather than only its static structure (e.g., Schnotz, 2005). Because dynamic mental models must be generated by learners from static diagrams, learning might be promoted by encouraging learners to visualize motion in static diagrams. However, mental models represent dynamic spatial information that might be difficult to construct for learners with lower spatial ability; they might benefit from instructional designs that support spatial reasoning, such as phase diagrams and depictive arrows. In Experiment 1, participants learned about air pumps, carburetors, and toilet tanks from single phase diagrams, multiphase diagrams, or multiphase diagrams followed by a prediction activity in which they predicted system behavior in novel situations. This prediction activity was expected to implicitly prompt mental visualization of motion. Learning in the latter condition (i.e., with the prediction activity) was significantly better than learning in the single phase condition. In the prediction condition, the enhancing effect of spatial ability on learning outcome was partially mediated by performance in the prediction activity. The mediation suggested that high spatial ability helped participants to accurately visualize the systems as they made predictions, which contributed to better performance on the learning assessment. Experiment 1 assessed visualizations during the prediction activity, whereas Experiment 2 assessed visualizations during the lessons. In two conditions in Experiment 2, participants were explicitly prompted to visualize motion in the system while viewing the lessons. Because learners with lower spatial ability were expected to have difficulty visualizing motion, arrows depicting motion were added in one condition. A baseline condition excluded the arrows and the prompt to visualize motion. In all three conditions, participants viewed multiphase diagrams followed by the prediction activity. Learning outcomes among the three conditions did not differ significantly: Depictive arrows and prompts to visualize motion were not significantly effective. Also, spatial ability did not interact with instructional condition. However, both spatial ability and subjective ratings of attempts to visualize motion were predictive of learning outcome. Overall, results from the two experiments indicated that participants with higher spatial ability were better able than participants with lower spatial ability to generate dynamic mental models from static instructions, particularly when they were implicitly prompted to reason about the system as they made predictions. Learners with lower spatial ability might need other forms of assistance for mental model generation, such as animated instructions.

In this thesis we investigate the performance of spatial reuse in wireless networks. More specifically, we investigate the performance of location-based spatial reuse built upon the identification of links in the network that form Collision-Free Sets (CFSs). We develop algorithms that efficiently find such CFSs. The main focus of this thesis is the application of our CFS-based algorithms to emerging WiMAX (Worldwide Interoperability for Microwave Access) networks. However, we will also apply our CFS-based algorithms to smaller scale networks, as well as to the issue of QoS routing. This thesis makes several contributions. We first show how CFSs can be found within polynomial time in small scale wireless networks. We then probe how CFSs can be used to enhance network efficiency when location is available. We also explore how CFS determination is affected by location error, determining what level of location error renders the use of CFSs ineffective. In this context, we discuss the density of access points required to ensure CFS-based spatial reuse remains effective. We then focus on the use of CFSs in emerging multihop WiMAX networks, showing in detail how enhanced spatial reuse is delivered. The spatial reuse gains are probed both in the presence of realistic channel conditions and realistic location errors. Within the context of the IEEE 802.16 standard we show how CFSs can enhance, by factors of two, the VoIP capacity of multihop WiMAX networks. We also discuss how our CFS algorithms can be applied to other ongoing efforts aimed at improving VoIP capacity in WiMAX networks. Finally, an application of our CFS algorithms in the context of QoS routing is studied. Specifically, we develop a two-hop QoS routing protocol that guarantees QoS specifications by securing higher bandwidth for the chosen routes.

In the past decade efforts to develop spatial data infrastructures (SDIs) have migrated from the initial “top-down” national approaches to “bottom-up” and cross jurisdictional efforts at the sub-national level. Although national SDI developments are fundamental to building the SDI culture and policy, it is sub-national and local SDI development that will deliver the immediate benefits to citizens and the community. In countries which have highly decentralised federations of states such as Australia, United States and Canada, the challenge is how to co-ordinate the literally thousands of often small local government jurisdictions which are important contributors to state and local SDIs. In recent years, a number of co-operative spatial data sharing partnerships between local and state government have emerged in various countries around the world. These partnerships are relatively new initiatives that have been established to facilitate more effective sharing of spatial data between organisations, but also as a mechanism to contribute to SDI development. To maximise the benefits from these partnerships it is essential to understand the factors that contribute to their successful operation and sustainability. Therefore, the focus of this research is to understand these collaborative arrangements so that future data sharing initiatives can be improved and sustained.

The development of a Regional Spatial Data Infrastructure (Regional SDI) is much more challenging than the development of a National SDI initiative within a nation. This is mainly because of the voluntary nature of cooperation at a multi-national level and participation in a Regional SDI initiative. As a result, despite considerable interest and activities, the development of an effective and comprehensive Asia-Pacific Regional Spatial Data Infrastructure (APSDI) is hampered by a lack of support from member nations which results in this initiative remaining only an innovative concept. Based on this situation, the aim of this research is to design an improved conceptual model for Regional SDI and an implementation strategy. It is proposed that this problem can be partly addressed by increasing the level of awareness about the nature and value of SDIs; improving the SDI conceptual model to better meet the needs of nations; and by identifying key factors that facilitate development by better understanding the complexity of the interaction between social, economic and political issues.