Air pollution simulation has several components, including data preparation, atmospheric circulation and air pollution dispersion modelling and computation, visualization of the model computation results, analysis and model evaluation. Of these components, only atmospheric circulation and air pollution dispersion modelling and computation are mature, while the other components are weak to a greater or lesser extent. To address these weaknesses, this thesis proposes to integrate the data, modelling and analysis into a multi-dimension, virtual geographically referenced environment. In addition, collaboration is used to solve the problem of multi-disciplinary knowledge requirements for conducting air pollution simulation. Based on this model, a concept of collaborative virtual geographic environments (CVGE) is proposed. / The air pollution that is associated with global economic growth is a global problem. Scientists, governmental officials and the public are focusing on improving understanding, accurately predicting and efficiently controlling levels of air pollution. Air pollution simulation is one method used to achieve these goals. This research will consider a computer supported simulation. / The contributions can be drawn from two aspects---CVGE and practice of air pollution simulation. Regarding CVGE, thesis 1 develops the conceptual framework of CVGE; 2 designs the architecture of a CVGE prototype in order to facilitate air pollution simulation; 3 proposes the concept of a "fuzzy boundary volume object", and designs a solution composed of a particle system wrapped in pollution boxes; and 4 examines the levels of geo-collaboration for air pollution simulation. For air pollution simulation, thesis 1 integrates air pollution sources, geo-data, an atmospheric circulation model, an air pollution dispersion model, geo-visualization and analysis into a collaborative virtual geographic environment, which is able to supply a new research methodology and platform for air pollution simulation; in 2, the new platform is scalable and able to free the restrictions of operations on visualization, which paves the way for further extension; 3 couples air pollution dispersion models with geo-information, opening up opportunities for cross studies between air pollution and other research areas, such as the economy, public health and urban planning. / The focus of this thesis is two-fold: one is on the development of a conceptual framework and prototype of CVGE from practice of air pollution simulation; the other is on applying this framework to facilitate air pollution simulation. The work of this thesis can be summarized as follows. (1) Defining the concept of CVGE, developing a conceptual framework for CVGE and discussing primary theories of CVGE. (2) Designing the architecture of a CVGE prototype to facilitate air pollution simulation. (3) Integration and computation of a complex atmospheric circulation model and an air pollution dispersion model based on high performance computation. (4) Geo-visualization of air pollution distribution and dispersion based on calculations using air pollution dispersion models. (5) Geo-collaboration for air pollution simulation. And finally (6) CVGE prototype based air pollution simulation. / The motivation for future research has two main aspects again---CVGE and practice of air pollution simulation. For the aspect of CVGE, possibilities for future research include: 1 more detailed research on the CVGE concept, primary theories and methodologies; 2 the efficient integration and management of heterogeneous geo-models with CVGE in standardization; and 3 the efficient rendering of a complex structured object in CVGE. Regarding practice, future research can be conducted into: 1 extending air pollution dispersion models; and 2 improving the efficiency of air pollutant rendering with a particle system wrapped in pollution boxes. / Xu, Bingli / Adviser: Hui Lin. / Source: Dissertation Abstracts International, Volume: 72-11, Section: A, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 251-264). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Microbial growth is a major cause of Indoor Air Quality (IAQ) problems. The implications of mold growth range from unacceptable musty smells and defacement of interior finishes, to structural damage and adverse health effects, not to mention lengthy litigation processes. Mold is likely to occur when a favorable combination of humidity, temperature, and substrate nutrient are maintained long enough. As many modern buildings use products that increase the likelihood of molds (e.g., paper and wood based products), reported cases have increased in recent years. Despite decades of intensive research efforts to prevent mold, modern buildings continue to suffer from mold infestation. The main reason is that current prescriptive regulations focus on the control of relative humidity only. However, recent research has shown that mold occurrences are influenced by a multitude of parameters with complex physical interactions. The set of relevant building parameters includes physical properties of building components, aspects of building usage, certain materials, occupant behavior, cleaning regime, HVAC system components and their operation, and other. Mold occurs mostly as the unexpected result of an unforeseen combination of the uncertain building parameters. Current deterministic mold assessment studies fail to give conclusive results. These simulations are based on idealizations of the building and its use, and therefore unable to capture the effect of the random, situational, and sometimes idiosyncratic nature of building use and operation. The presented research takes a radically different approach, based on the assessment of the uncertainties of all parameters and their propagation through a mixed set of simulations using a Monte Carlo technique. This approach generates a mold risk distribution that reveals the probability of mold occurrence in selected trouble spots in a building. The approach has been tested on three building cases located in Miami and Atlanta. In all cases the new approach was able to show the circumstances under which the mold risk could increase substantially, leading to a set of clear specifications for remediation and, in for new designs, to A/E procurement methods that will significantly reduce any mold risk.