In situations of mass disaster (1), which include airline crashes (2), terrorist attacks (2), large fires (3), and mass homicide (4), the human remains are often damaged beyond recognition (5). In these cases, bones and teeth are potentially the only acceptable source of deoxyribonucleic acid, or DNA (6). Previous studies have evaluated a plethora of techniques to purify DNA from hard tissue, but there is no consensus on the optimal process by which to extract and purify DNA from these samples. Not only are hard tissue samples difficult to process, in many cases the samples are subjected to extreme environmental conditions, such as high temperature. Thus, there is interest in obtaining information on DNA quality from samples exposed to high temperatures (7). This work hopes to fill the gap by: 1) optimizing a DNA extraction protocol from hard tissue; and 2) measuring the degree to which the DNA is degraded in an effort to link the quantity and quality of the DNA recovered to the outer appearance of the tooth. To accomplish this, individual teeth were burned in a furnace at 100 °C, 200 °C, 300 °C, 400 °C, 500 °C, and 600 °C for 10, 20, and 30 minutes. The optimal extraction procedure utilized Amicon® Ultra-4 Centrifugal 30K filter devices and the QIAGEN MinElute Polymerase Chain Reaction (PCR) Purification Kit. Samples were quantified using the Quantifiler® Trio quantification kit to obtain the quantity and quality metrics. After heat exposure, each tooth was photographed and subsequently given a color designation or value: light yellow to beige teeth were assigned a value of 1; dark yellow to orange were assigned a value of 2; brown was assigned a value of 3; shiny black was assigned a value of 4; and black to light gray teeth were assigned a value of 5. Both carbonization and the early stages of calcination were observed. The mass of DNA per mass of tooth was determined by examining quantitative PCR (qPCR) results for both a large and small autosomal fragment. The degradation index, or DI, was also calculated from qPCR measurements. The results demonstrate a strong correlation between the quantity of DNA recovered, the quality of the DNA obtained, and the designated color value. The highest recovery rates were obtained from teeth assigned a color value of 1 (unaltered beige) or 2 (yellow to orange). These teeth were exposed to either room temperature, 100 °C or 200 °C. At temperatures exceeding 300 °C, the amount of DNA recovered drastically decreased and was inconsistent. Some of the samples subjected to temperatures at and above 300 °C resulted in no quantifiable DNA. In contrast, the DI results suggested that when the teeth were subjected to temperatures ≤ 100 °C, the quality of the DNA was good, wherein the DI value was approximately 1. At 200 °C, the temperature began to impact the DI value, which increased with time to the point where a DI was no longer able to be calculated because the large autosomal fragment could not be detected. In conclusion, the current work compares five different methods of DNA extraction to establish a best practice extraction procedure for these difficult samples. Furthermore, this work suggests that examination of the tooth’s appearance can be used to deduce whether successful DNA recovery is likely. In summary, the results demonstrate that when the tooth sample was assigned a color value of 1, the quantity and quality of the DNA obtained was high. Once the color value of the sample rose to 2, the quantity and quality of DNA varied greatly and the probative value of the sample was diminished. Samples that exhibited large color changes or had begun the calcination process resulted in no recoverable DNA.

Design and analysis of computer experiments is an area focusing on efficient data collection (e.g., space-filling designs), surrogate modeling (e.g., Gaussian process models), and uncertainty quantification. Survival analysis focuses on modeling the period of time until a certain event happens. Data collection, prediction, and uncertainty quantification are also fundamental in survival models. In this dissertation, the proposed methods are motivated by a wide range of real world applications, including high-performance computing (HPC) variability data, jet engine reliability data, Titan GPU lifetime data, and pine tree survival data. This dissertation is to explore interfaces on computer experiments and survival analysis with the above applications. Chapter 1 provides a general introduction to computer experiments and survival analysis. Chapter 2 focuses on the HPC variability management application. We investigate the applicability of space-filling designs and statistical surrogates in the HPC variability management setting, in terms of design efficiency, prediction accuracy, and scalability. A comprehensive comparison of the design strategies and predictive methods is conducted to study the combinations' performance in prediction accuracy. Chapter 3 focuses on the reliability prediction application. With the availability of multi-channel sensor data, a single degradation index is needed to be compatible with most existing models. We propose a flexible framework with multi-sensory data to model the nonlinear relationship between sensors and the degradation process. We also involve the automatic variable selection to exclude sensors that have no effect on the underlying degradation process. Chapter 4 investigates inference approaches for spatial survival analysis under the Bayesian framework. The Markov chain Monte Carlo (MCMC) approaches and variational inferences performance are studied for two survival models, the cumulative exposure model and the proportional hazard (PH) model. The Titan GPU data and pine tree survival data are used to illustrate the capability of variational inference on spatial survival models. Chapter 5 provides some general conclusions. / Doctor of Philosophy / This dissertation focus on three projects related to computer experiments and survival analysis. Design and analysis of the computer experiment is an area focusing on efficient data collection, building predictive models, and uncertainty quantification. Survival analysis focuses on modeling the period of time until a certain event happens. Data collection, prediction, and uncertainty quantification are also fundamental in survival models. Thus, this dissertation aims to explore interfaces between computer experiments and survival analysis with real world applications. High performance computing systems aggregate a large number of computers to achieve high computing speed. The first project investigates the applicability of space-filling designs and statistical predictive models in the HPC variability management setting, in terms of design efficiency, prediction accuracy, and scalability. A comprehensive comparison of the design strategies and predictive methods is conducted to study the combinations' performance in prediction accuracy. The second project focuses on building a degradation index that describes the product's underlying degradation process. With the availability of multi-channel sensor data, a single degradation index is needed to be compatible with most existing models. We propose a flexible framework with multi-sensory data to model the nonlinear relationship between sensors and the degradation process. We also involve the automatic variable selection to exclude sensors that have no effect on the underlying degradation process. The spatial survival data are often observed when the survival data are collected over a spatial region. The third project studies inference approaches for spatial survival analysis under the Bayesian framework. The commonly used inference method, Markov chain Monte Carlo (MCMC) approach and the approximate inference approach, variational inference's performance are studied for two survival models. The Titan GPU data and pine tree survival data are used to illustrate the capability of variational inference on spatial survival models.

This thesis presents the results of a series of undrained cyclic triaxial tests carried out on four lime-cement stabilized specimens and clay specimen. The shear resistance degradation rate of lime-cement column subjected to cyclic loading simulated from heavy truck was investigated based on stress-controlled test. The influence of lime and cement on the degradation rate was investigated by comparing the behavior of stabilized kaolin and unstabilized kaolin with similar initial condition. The results indicate an increase in degree of degradation as the number of loading cycles and cyclic strain increase. It is observed that the degradation index has approximately a parabolic relationship with the number of cycles. Generally adding lime and cement to the clay will increase the degradation index which means lower degree of degradation. The degradation parameter, t has a hyperbolic relationship with shear strain, but it loses its hyperbolic shape as the soil getting stronger. On the other hand, for unstabilized clay an approximate linear relationship between degradation index and number of cycles was observed and the degradation parameter has a hyperbolic shape with the increase number of cycles. It was also observed that the stronger the material was, the lesser pore pressure developed in the lime-cement stabilized clay.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Environmental degradation presents itself as a major challenge for a variety of countries. This fact is due to the increased knowledge obtained regarding the changes that the global environment has suffered. This phenomenon can be understood as the destruction, damage or wearing that is generated in the environment by human activities and inherent to nature. In this sense, many of the changes originated from environmental degradation are consequences of agricultural activities, and in the Brazilian scenario, this issue also arises from livestock exploration, given the importance of this type of activity for the country. Environmental degradation can be measured by the creation of an Agricultural Environmental Degradation Index (IDAA). Due to the impacts of environmental degradation caused by livestock activities, there are changes occurring regarding the rural development of the localities, because the rural environment is in a reality closer to the occurrence of this phenomenon for this type of activity. Within the context of Rio Grande do Sul, there is a lack of studies that measure environmental degradation. In order to characterize the agricultural environmental degradation in Rio Grande do Sul and also fill this gap in the literature regarding the relationship between this phenomenon and other aspects of development of the regions, the research problem of this study can be summarized as follows: what is the pattern of agricultural environmental degradation of the cities from Rio Grande do Sul and how this phenomenon is impacted by factors of rural development in two distinct periods of time. Therefore, this study aimed to analyze the pattern of agricultural environmental degradation of the cities in the state of Rio Grande do Sul and see how this pattern is affected by the factors of rural development in these same cities in two different periods of time. In this sense, the methodology used was the Agricultural Environmental Degradation Index (IDAA) as a proxy for agricultural environmental degradation and the technique of factor analysis was used to find the determinants of rural development of cities in the state. In order to study the impact of these factors on agricultural environmental degradation of Rio Grande do Sul, a regression model with panel data was estimated by the method of Fixed Effects. The values of agricultural environmental degradation for the mesoregions of Rio Grande do Sul shown to be high for the two years studied, and the central-eastern mesoregion presented the greatest degradation averages, with IDAA values of 84.58% in 1996 and 85.16% in 2006.Referring to its scale, this mesoregion also showed the worst results, with 47.62% of its cities with degradation patterns with scales of degree above the average value in 1996 and 2006. Regarding the variation of agricultural environmental degradation in the two years surveyed, there is a small variation in the value of the index from one year to another, ie, only 0.02%.The factors of rural development found in Rio Grande do Sul were Conditions of Housing and Rural Education (F1), Structure and Performance of the Agricultural Sector (F2); Leverage and Correction of Rural Production (F3), Agricultural Production Area (F4); Rural Electricity (F5) and Economic and Financial Rural (F6).In the relationship between IDAA and the factors, all coefficients were statistically significant. F1, F3 and F5 presented a positive relationship with degradation, demonstrating that the more developed the regions are, in relation to these aspects, the greater the levels of degradation, which are justified, respectively, by the environmental dilemma of Rio Grande do Sul, because of the overuse of these practices and the irrational use of electrical resources. In the contrary, F2, F4 and F6 showed a negative relationship, justified, respectively, by agro-ecological assumptions, by concerns regarding environmental issues and the inverse relationship between degradation and the economic aspect. / A degradação ambiental apresenta-se como um grande desafio para uma diversidade de países. Isso é decorrente do maior conhecimento que se obtém sobre as transformações que o meio ambiente mundial vem sofrendo. Este fenômeno pode ser entendido como a destruição, deterioração ou desgaste que é gerado no meio ambiente a partir das atividades humanas e inerentes à natureza. Neste sentido, muitas das mudanças oriundas da degradação ambiental são consequências da atividade agropecuária, e, no cenário brasileiro, essa questão tem também grande parcela decorrente da exploração da agropecuária, em vista da importância desse tipo de atividade para o país. A degradação ambiental pode ser mensurada por meio da criação de um Índice de Degradação Ambiental Agropecuária (IDAA). Devido aos impactos da degradação ambiental oriundos das atividades agropecuárias, ocorrem alterações no desenvolvimento rural das localidades, pois o ambiente rural está em uma realidade mais próxima da ocorrência desse fenômeno por esse tipo de atividade. Dentro do contexto do Rio Grande do Sul, verifica-se a escassez de trabalhos que mensurem a degradação ambiental. A fim de caracterizar a degradação ambiental agropecuária no Rio Grande do Sul e ainda preencher a lacuna na literatura quanto à relação entre esse fenômeno e outros aspectos do desenvolvimento das regiões, o problema de pesquisa deste trabalho pode ser sintetizado da seguinte forma: qual o padrão de degradação ambiental agropecuário dos municípios gaúchos e como esse fenômeno é impactado pelos fatores de desenvolvimento rural em dois períodos distintos de tempo. Portanto, o objetivo geral deste trabalho foi o de analisar o padrão de degradação ambiental agropecuário dos municípios gaúchos e verificar como esse padrão é impactado pelos fatores de desenvolvimento rural desses mesmos municípios em dois períodos distintos de tempo. Neste sentido, foi utilizada a metodologia do Índice de Degradação Ambiental Agropecuária (IDAA) como proxy para a degradação ambiental agropecuária e a técnica de análise fatorial para encontrar os fatores determinantes do desenvolvimento rural dos municípios gaúchos. A fim de estudar o impacto desses fatores na degradação ambiental agropecuária do Rio Grande do Sul, foi estimado um modelo de regressão com dados em painel por meio do método de Efeitos Fixos. Os valores de degradação ambiental agropecuária para as mesorregiões gaúchas mostraram-se elevados para os dois anos estudados, e a mesorregião Centro Oriental apresentou as maiores médias de degradação, com valores de IDAA de 84,58% em 1996 e 85,16% em 2006. Referente à sua escala, essa mesorregião também apresentou os piores resultados, com 47,62% dos seus municípios com padrão de degradação com escalas de grau acima do valor médio em 1996 e em 2006. Com relação à variação da degradação ambiental agropecuária nos dois anospesquisados, nota-se uma pequena variação no valor do índice de um ano para o outro, isto é, de apenas 0,02%. Os fatores de desenvolvimento rural encontrados para o Rio Grande do Sul foram Condições de Moradia e Educação Rurais (F1); Estrutura e Desempenho do Setor Agropecuário (F2); Alavancagem e Corretivos da Produção Rural (F3); Área de Produção da Agropecuária (F4); Energia Elétrica Rural (F5); e Econômico e Financeiro Rural (F6). Nas relações entre o IDAA e os fatores, todos os coeficientes demonstraram significância estatística. Os fatores F1, F3 e F5 apresentaram relação positiva com a degradação, demonstrando que, quanto mais desenvolvidas as regiões nesses aspectos, maiores serão os níveis de degradação, justificados, respectivamente, pelo dilema ambiental do Rio Grande do Sul, pelo uso excessivo dessas práticas e pelo uso irracional dos recursos elétricos. De maneira contrária, os fatores F2, F4 e F6 apresentaram relação negativa, justificada, respectivamente, pelo pressuposto agroecológico, pela preocupação com as questões ambientais e pela relação inversa entre degradação e o aspecto econômico.