Matching Items (7)
Filtering by
- All Subjects: Geography
- All Subjects: BART
- All Subjects: spatial statistics
- Creators: Brazel, Anthony
- Member of: ASU Electronic Theses and Dissertations
Description
This doctoral dissertation research aims to develop a comprehensive definition of urban open spaces and to determine the extent of environmental, social and economic impacts of open spaces on cities and the people living there. The approach I take to define urban open space is to apply fuzzy set theory to conceptualize the physical characteristics of open spaces. In addition, a 'W-green index' is developed to quantify the scope of greenness in urban open spaces. Finally, I characterize the environmental impact of open spaces' greenness on the surface temperature, explore the social benefits through observing recreation and relaxation, and identify the relationship between housing price and open space be creating a hedonic model on nearby housing to quantify the economic impact. Fuzzy open space mapping helps to investigate the landscape characteristics of existing-recognized open spaces as well as other areas that can serve as open spaces. Research findings indicated that two fuzzy open space values are effective to the variability in different land-use types and between arid and humid cities. W-Green index quantifies the greenness for various types of open spaces. Most parks in Tempe, Arizona are grass-dominant with higher W-Green index, while natural landscapes are shrub-dominant with lower index. W-Green index has the advantage to explain vegetation composition and structural characteristics in open spaces. The outputs of comprehensive analyses show that the different qualities and types of open spaces, including size, greenness, equipment (facility), and surrounding areas, have different patterns in the reduction of surface temperature and the number of physical activities. The variance in housing prices through the distance to park was, however, not clear in this research. This dissertation project provides better insight into how to describe, plan, and prioritize the functions and types of urban open spaces need for sustainable living. This project builds a comprehensive framework for analyzing urban open spaces in an arid city. This dissertation helps expand the view for urban environment and play a key role in establishing a strategy and finding decision-makings.
ContributorsKim, Won Kyung (Author) / Wentz, Elizabeth (Thesis advisor) / Myint, Soe W (Thesis advisor) / Brazel, Anthony (Committee member) / Guhathakurta, Subhrajit (Committee member) / Arizona State University (Publisher)
Created2011
Description
Urbanization, a direct consequence of land use and land cover change, is responsible for significant modification of local to regional scale climates. It is projected that the greatest urban growth of this century will occur in urban areas in the developing world. In addition, there is a significant research gap in emerging nations concerning this topic. Thus, this research focuses on the assessment of climate impacts related to urbanization on the largest metropolitan area in Latin America: Mexico City.
Numerical simulations using a state-of-the-science regional climate model are utilized to address a trio of scientifically relevant questions with wide global applicability. The importance of an accurate representation of land use and land cover is first demonstrated through comparison of numerical simulations against observations. Second, the simulated effect of anthropogenic heating is quantified. Lastly, numerical simulations are performed using pre-historic scenarios of land use and land cover to examine and quantify the impact of Mexico City's urban expansion and changes in surface water features on its regional climate.
Numerical simulations using a state-of-the-science regional climate model are utilized to address a trio of scientifically relevant questions with wide global applicability. The importance of an accurate representation of land use and land cover is first demonstrated through comparison of numerical simulations against observations. Second, the simulated effect of anthropogenic heating is quantified. Lastly, numerical simulations are performed using pre-historic scenarios of land use and land cover to examine and quantify the impact of Mexico City's urban expansion and changes in surface water features on its regional climate.
ContributorsBenson-Lira, Valeria (Author) / Georgescu, Matei (Thesis advisor) / Brazel, Anthony (Committee member) / Vivoni, Enrique (Committee member) / Arizona State University (Publisher)
Created2015
Description
ABSTRACT
Famine is the result of a complex set of environmental and social factors. Climate conditions are established as environmental factors contributing to famine occurrence, often through teleconnective patterns. This dissertation is designed to investigate the combined influence on world famine patterns of teleconnections, specifically the North Atlantic Oscillation (NAO), Southern Oscillation (SO), Pacific Decadal Oscillation (PDO), Atlantic Multidecadal Oscillation (AMO), or regional climate variations such as the South Asian Summer Monsoon (SASM). The investigation is three regional case studies of famine patterns specifically, Egypt, the British Isles, and India.
The first study (published in Holocene) employs the results of a Principal Component Analysis (PCA) yielding a SO-NAO eigenvector to predict major Egyptian famines between AD 1049-1921. The SO-NAO eigenvector (1) successfully discriminates between the 5-10 years preceding a famine and the other years, (2) predicts eight of ten major famines, and (3) correctly identifies fifty out of eighty events (63%) of food availability decline leading up to major famines.
The second study investigates the impact of the NAO, PDO, SO, and AMO on 63 British Isle famines between AD 1049 and 1914 attributed to climate causes in historical texts. Stepwise Regression Analysis demonstrates that the 5-year lagged NAO is the primary teleconnective influence on famine patterns; it successfully discriminates 73.8% of weather-related famines in the British Isles from 1049 to 1914.
The final study identifies the aggregated influence of the NAO, SO, PDO, and SASM on 70 Indian famines from AD 1049 to 1955. PCA results in a NAO-SOI vector and SASM vector that predicts famine conditions with a positive NAO and negative SO, distinct from the secondary SASM influence. The NAO-famine relationship is consistently the strongest; 181 of 220 (82%) of all famines occurred during positive NAO years.
Ultimately, the causes of famine are complex and involve many factors including societal and climatic. This dissertation demonstrates that climate teleconnections impact famine patterns and often the aggregates of multiple climate variables hold the most significant climatic impact. These results will increase the understanding of famine patterns and will help to better allocate resources to alleviate future famines.
Famine is the result of a complex set of environmental and social factors. Climate conditions are established as environmental factors contributing to famine occurrence, often through teleconnective patterns. This dissertation is designed to investigate the combined influence on world famine patterns of teleconnections, specifically the North Atlantic Oscillation (NAO), Southern Oscillation (SO), Pacific Decadal Oscillation (PDO), Atlantic Multidecadal Oscillation (AMO), or regional climate variations such as the South Asian Summer Monsoon (SASM). The investigation is three regional case studies of famine patterns specifically, Egypt, the British Isles, and India.
The first study (published in Holocene) employs the results of a Principal Component Analysis (PCA) yielding a SO-NAO eigenvector to predict major Egyptian famines between AD 1049-1921. The SO-NAO eigenvector (1) successfully discriminates between the 5-10 years preceding a famine and the other years, (2) predicts eight of ten major famines, and (3) correctly identifies fifty out of eighty events (63%) of food availability decline leading up to major famines.
The second study investigates the impact of the NAO, PDO, SO, and AMO on 63 British Isle famines between AD 1049 and 1914 attributed to climate causes in historical texts. Stepwise Regression Analysis demonstrates that the 5-year lagged NAO is the primary teleconnective influence on famine patterns; it successfully discriminates 73.8% of weather-related famines in the British Isles from 1049 to 1914.
The final study identifies the aggregated influence of the NAO, SO, PDO, and SASM on 70 Indian famines from AD 1049 to 1955. PCA results in a NAO-SOI vector and SASM vector that predicts famine conditions with a positive NAO and negative SO, distinct from the secondary SASM influence. The NAO-famine relationship is consistently the strongest; 181 of 220 (82%) of all famines occurred during positive NAO years.
Ultimately, the causes of famine are complex and involve many factors including societal and climatic. This dissertation demonstrates that climate teleconnections impact famine patterns and often the aggregates of multiple climate variables hold the most significant climatic impact. These results will increase the understanding of famine patterns and will help to better allocate resources to alleviate future famines.
ContributorsSantoro, Michael Melton (Author) / Cerveny, Randall S. (Thesis advisor) / McHugh, Kevin (Committee member) / Brazel, Anthony (Committee member) / Balling Jr., Robert C. (Committee member) / Arizona State University (Publisher)
Created2017
Description
While there are many elements to consider when determining one's risk of heat or cold stress, acclimation could prove to be an important factor to consider. Individuals who are participating in more strenuous activities, while being at a lower risk, will still feel the impacts of acclimation to an extreme climate. To evaluate acclimation in strenuous conditions, I collected finishing times from six different marathon races: the New York City Marathon (New York City, New York), Equinox Marathon (Fairbanks, Alaska), California International Marathon (Sacramento, California), LIVESTRONG Austin Marathon (Austin, Texas), Cincinnati Flying Pig Marathon (Cincinnati, Ohio), and the Ocala Marathon (Ocala, Florida). Additionally, I collected meteorological variables for each race day and the five days leading up to the race (baseline). I tested these values against the finishing times for the local runners, those from the race state, and visitors, those from other locations. Effects of local acclimation could be evaluated by comparing finishing times of local runners to the change between the race day and baseline weather conditions. Locals experienced a significant impact on finishing times for large changes between race day and the baseline conditions for humidity variables, dew point temperature, vapor pressure, relative humidity, and temperature based variables such as the heat index, temperature and the saturation vapor pressure. Wind speed and pressure values also marked a change in performance, however; pressure was determined to be a larger psychological factor than acclimation factor. The locals also demonstrated an acclimation effect as performance improved when conditions were similar on race day to baseline conditions for the three larger races. Humidity variables had the largest impact on runners when those values increased from training and acclimation values; however increased wind speed appeared to offset increased humidity values. These findings support previous acclimation research stating warm wet conditions are more difficult to acclimate to than warm dry conditions. This research while primarily pertaining to those participating physically demanding activities may also be applied to other large scale events such as festivals, fairs, or concerts.
ContributorsDeBiasse, Kimberly Michelle (Author) / Cerveny, Randall S. (Thesis advisor) / Brazel, Anthony (Committee member) / Selover, Nancy (Committee member) / Arizona State University (Publisher)
Created2011
Description
This dissertation centers on Bayesian Additive Regression Trees (BART) and Accelerated BART (XBART) and presents a series of models that tackle extrapolation, classification, and causal inference challenges. To improve extrapolation in tree-based models, I propose a method called local Gaussian Process (GP) that combines Gaussian process regression with trained BART trees. This allows for extrapolation based on the most relevant data points and covariate variables determined by the trees' structure. The local GP technique is extended to the Bayesian causal forest (BCF) models to address the positivity violation issue in causal inference. Additionally, I introduce the LongBet model to estimate time-varying, heterogeneous treatment effects in panel data. Furthermore, I present a Poisson-based model, with a modified likelihood for XBART for the multi-class classification problem.
ContributorsWang, Meijia (Author) / Hahn, Paul (Thesis advisor) / He, Jingyu (Committee member) / Lan, Shiwei (Committee member) / McCulloch, Robert (Committee member) / Zhou, Shuang (Committee member) / Arizona State University (Publisher)
Created2024
Description
Spatial regression is one of the central topics in spatial statistics. Based on the goals, interpretation or prediction, spatial regression models can be classified into two categories, linear mixed regression models and nonlinear regression models. This dissertation explored these models and their real world applications. New methods and models were proposed to overcome the challenges in practice. There are three major parts in the dissertation.
In the first part, nonlinear regression models were embedded into a multistage workflow to predict the spatial abundance of reef fish species in the Gulf of Mexico. There were two challenges, zero-inflated data and out of sample prediction. The methods and models in the workflow could effectively handle the zero-inflated sampling data without strong assumptions. Three strategies were proposed to solve the out of sample prediction problem. The results and discussions showed that the nonlinear prediction had the advantages of high accuracy, low bias and well-performed in multi-resolution.
In the second part, a two-stage spatial regression model was proposed for analyzing soil carbon stock (SOC) data. In the first stage, there was a spatial linear mixed model that captured the linear and stationary effects. In the second stage, a generalized additive model was used to explain the nonlinear and nonstationary effects. The results illustrated that the two-stage model had good interpretability in understanding the effect of covariates, meanwhile, it kept high prediction accuracy which is competitive to the popular machine learning models, like, random forest, xgboost and support vector machine.
A new nonlinear regression model, Gaussian process BART (Bayesian additive regression tree), was proposed in the third part. Combining advantages in both BART and Gaussian process, the model could capture the nonlinear effects of both observed and latent covariates. To develop the model, first, the traditional BART was generalized to accommodate correlated errors. Then, the failure of likelihood based Markov chain Monte Carlo (MCMC) in parameter estimating was discussed. Based on the idea of analysis of variation, back comparing and tuning range, were proposed to tackle this failure. Finally, effectiveness of the new model was examined by experiments on both simulation and real data.
In the first part, nonlinear regression models were embedded into a multistage workflow to predict the spatial abundance of reef fish species in the Gulf of Mexico. There were two challenges, zero-inflated data and out of sample prediction. The methods and models in the workflow could effectively handle the zero-inflated sampling data without strong assumptions. Three strategies were proposed to solve the out of sample prediction problem. The results and discussions showed that the nonlinear prediction had the advantages of high accuracy, low bias and well-performed in multi-resolution.
In the second part, a two-stage spatial regression model was proposed for analyzing soil carbon stock (SOC) data. In the first stage, there was a spatial linear mixed model that captured the linear and stationary effects. In the second stage, a generalized additive model was used to explain the nonlinear and nonstationary effects. The results illustrated that the two-stage model had good interpretability in understanding the effect of covariates, meanwhile, it kept high prediction accuracy which is competitive to the popular machine learning models, like, random forest, xgboost and support vector machine.
A new nonlinear regression model, Gaussian process BART (Bayesian additive regression tree), was proposed in the third part. Combining advantages in both BART and Gaussian process, the model could capture the nonlinear effects of both observed and latent covariates. To develop the model, first, the traditional BART was generalized to accommodate correlated errors. Then, the failure of likelihood based Markov chain Monte Carlo (MCMC) in parameter estimating was discussed. Based on the idea of analysis of variation, back comparing and tuning range, were proposed to tackle this failure. Finally, effectiveness of the new model was examined by experiments on both simulation and real data.
ContributorsLu, Xuetao (Author) / McCulloch, Robert (Thesis advisor) / Hahn, Paul (Committee member) / Lan, Shiwei (Committee member) / Zhou, Shuang (Committee member) / Saul, Steven (Committee member) / Arizona State University (Publisher)
Created2020
Description
Geographically Weighted Regression (GWR) has been broadly used in various fields to
model spatially non-stationary relationships. Classic GWR is considered as a single-scale model that is based on one bandwidth parameter which controls the amount of distance-decay in weighting neighboring data around each location. The single bandwidth in GWR assumes that processes (relationships between the response variable and the predictor variables) all operate at the same scale. However, this posits a limitation in modeling potentially multi-scale processes which are more often seen in the real world. For example, the measured ambient temperature of a location is affected by the built environment, regional weather and global warming, all of which operate at different scales. A recent advancement to GWR termed Multiscale GWR (MGWR) removes the single bandwidth assumption and allows the bandwidths for each covariate to vary. This results in each parameter surface being allowed to have a different degree of spatial variation, reflecting variation across covariate-specific processes. In this way, MGWR has the capability to differentiate local, regional and global processes by using varying bandwidths for covariates. Additionally, bandwidths in MGWR become explicit indicators of the scale at various processes operate. The proposed dissertation covers three perspectives centering on MGWR: Computation; Inference; and Application. The first component focuses on addressing computational issues in MGWR to allow MGWR models to be calibrated more efficiently and to be applied on large datasets. The second component aims to statistically differentiate the spatial scales at which different processes operate by quantifying the uncertainty associated with each bandwidth obtained from MGWR. In the third component, an empirical study will be conducted to model the changing relationships between county-level socio-economic factors and voter preferences in the 2008-2016 United States presidential elections using MGWR.
model spatially non-stationary relationships. Classic GWR is considered as a single-scale model that is based on one bandwidth parameter which controls the amount of distance-decay in weighting neighboring data around each location. The single bandwidth in GWR assumes that processes (relationships between the response variable and the predictor variables) all operate at the same scale. However, this posits a limitation in modeling potentially multi-scale processes which are more often seen in the real world. For example, the measured ambient temperature of a location is affected by the built environment, regional weather and global warming, all of which operate at different scales. A recent advancement to GWR termed Multiscale GWR (MGWR) removes the single bandwidth assumption and allows the bandwidths for each covariate to vary. This results in each parameter surface being allowed to have a different degree of spatial variation, reflecting variation across covariate-specific processes. In this way, MGWR has the capability to differentiate local, regional and global processes by using varying bandwidths for covariates. Additionally, bandwidths in MGWR become explicit indicators of the scale at various processes operate. The proposed dissertation covers three perspectives centering on MGWR: Computation; Inference; and Application. The first component focuses on addressing computational issues in MGWR to allow MGWR models to be calibrated more efficiently and to be applied on large datasets. The second component aims to statistically differentiate the spatial scales at which different processes operate by quantifying the uncertainty associated with each bandwidth obtained from MGWR. In the third component, an empirical study will be conducted to model the changing relationships between county-level socio-economic factors and voter preferences in the 2008-2016 United States presidential elections using MGWR.
ContributorsLi, Ziqi (Author) / Fotheringham, A. Stewart (Thesis advisor) / Goodchild, Michael F. (Committee member) / Li, Wenwen (Committee member) / Arizona State University (Publisher)
Created2020