Despite the rapidly growing interest in Bayesian inference, recent publications reveal common issues that suggest even experienced researchers across disciplines, including geospatial research, may not always follow the proper procedure in conducting Bayesian analysis and report results. This study aims to promote Bayesian inference and the best practice guidelines of Bayesian analysis, targeting beginners in the geospatial community. We selected the Bayesian multilevel model (BMLM) to demonstrate proper Bayesian analysis through a case study examining the effect of neighbourhood-level social deprivation on birthweight in Fulton County, Georgia, USA. We constructed both frequentist multilevel models (MLMs) and BMLMs, and these were two-level varying intercept models. Following the Avoid the Misuse of Bayesian Statistics (WAMBS) checklist, we illustrated the BMLM workflow and highlighted key steps in fitting, reporting, and interpreting BMLM. Our case study shows several advantages of the Bayesian approach over the frequentist method, including incorporating prior information, better handling of uncertainty, more intuitive interpretation, and greater transparency in reporting. However, some benefits, such as avoiding multiple comparisons and generating robust estimations, are challenging to explicitly illustrate. Other benefits such as handling small sample sizes and complex models cannot be showcased with our simple models. Notably, BMLM is more computationally intensive compared to MLMs. Adhering to established guidelines can enhance the quality, transparency, and reproducibility of Bayesian analysis, allowing us to fully harness the potential of Bayesian inference in advancing geospatial knowledge.

A good understanding of the associations of COVID-19 infection and mortality with contextual factors when vaccines were not widely available is necessary for human societies to be better prepared for future outbreaks of infectious diseases. This retrospective ecological study aimed to explore the spatially varying associations of COVID-19 incidence, death, and case fatality rates with contextual socioeconomic, health, and environmental factors during the period of partial population coverage of vaccination at county level in the state of Georgia, USA. The associations of COVID-19 rates and contextual factors were analyzed using geographically weighted regression (GWR), compared with ordinary least squares regression (OLS) analysis. OLS results showed that most factors were significantly associated with COVID-19 death rate and case fatality rate, but not incidence rate. GWR results demonstrated that the associations of all three COVID-19 rates with factors varied across space: A factor might have a significant positive, significant negative, or nonsignificant association with each rate in certain counties. Most factors for poor health outcomes were significantly associated with higher risks of COVID-19 infection and mortality in more counties compared to non-significant or inverse associations. The spatially varying associations for some contextual factors were related to the socioeconomic and urbanization characteristics of counties. Some factors also affected COVID-19 infection and mortality differently. For example, persons aged 65 and older percentage was not a significant risk factor of COVID-19 infection in most counties, but it was the most spatially consistent risk factor of COVID-19 death in Georgia; fully vaccinated percentage was a more significant indicator of reducing COVID-19 infection in rural counties compared to urban and suburban areas. This study provides useful information for public health agencies and professionals to make and implement more specific and targeted local health policies.

Roads and traffic are important elements of urbanization, but their spatial associations with surface water quality in watersheds have been seldom studied. In this study, the spatially varying associations of three urbanization indicators, including road density, traffic density, and percentages of urban land, with twenty water quality indicators, including dissolved oxygen (DO), specific conductance (SC), dissolved solids (DS), suspended solids (SS), biochemical oxygen demand (BOD), dissolved nutrients, dissolved ions, heavy metals, and coliform bacteria, across the watersheds in the northern part of the state of Georgia, USA, have been examined by a conventional statistical method, ordinary least squares regression (OLS), and a spatial statistical method, geographically weighted regression (GWR). The results from OLS show that the urbanization indicators all have significant positive associations with the majority of the studied water pollutants, indicating that water pollution is significantly contributed by human activities related to urbanization in northern Georgia. In contrast, GWR results show that the associations vary across the watersheds affected by their urbanization levels. Significant positive associations are found between each urbanization indicator and each of the studied water pollutants, but not in all watersheds. The associations of suspended solids, nitrogen nutrients, and coliform bacteria with all three urbanization indicators are more significant in less-urbanized watersheds, while the associations of dissolved ions, BOD, and orthophosphate (PO4) with road density and traffic density are more significant than those with urban land in more-urbanized watersheds, indicating that those water pollutants are more contributed by human activities associated with roads and traffic than other activities in more-urbanized areas. As a pilot study to explore how and why the associations of surface water quality with roads and traffic change across watersheds with different urbanization levels, its findings suggest that the policies of watershed management, land-use planning, and transportation planning should be tailored in local areas based on the locally important water pollutants and their associated urbanization indicators.

The COVID-19 disease caused by the SARS-CoV-2 virus first identified in December 2019 has resulted in millions of deaths so far around the world. Controlling the spread of the disease requires a good understanding of the factors (e.g. air pollutants) that influence virus transmission and the conditions under which it spreads. This study analyzed the relationships between COVID-19 cases and both short-term (6-month) and long-term (60-month) exposures to eight air pollutants (NO, NO2, NOx, CO, SO2, O3, PM2.5 and PM10) in Tehran city, Iran, by integrating geostatistical interpolation models, regression analysis, and an innovated COVID-19 incidence rate calculation (Q-index) that considered the spatial distributions of both population and air pollution. The results show that the higher COVID-19 incidence rate was significantly associated with the exposure to higher concentrations of CO, NO, and NOx during the short-term period; the higher COVID-19 incidence rate was significantly related to the exposure to higher concentrations of PM2.5 during the long-term period; while COVID-19 incidence rate was not significantly associated with the concentrations of O3, SO2, PM10 and NO2 in either period. This study indicates that exposure to air pollutants can effect an increase in the number of infected people by transmitting the virus through the air or by predisposing people to the disease over time. The Q-index calculation method developed in this study can be also used by other studies to calculate more accurate disease rates that consider the spatial distribution of both population and air pollution.

Preterm birth (PTB) is an important cause of infant morbidity and mortality around the world. A good understanding of its associations with relevant factors is essential for an effective reduction in PTB risk. This study explores the spatially varying relationships between PTB and demographic, socioeconomic, and behavioral factors in the state of Georgia, USA using Geographically Weighted Logistic Regression (GWLR). The results show the relationships between PTB and factors vary over space, and the spatial patterns in the varying relationships are related to the socioeconomic and urbanization characteristics of the communities where the births were located. For example, PTB has significant positive relationships with black mothers and single mothers, and significant negative relationships with maternal education in the majority of Georgia, but such relationships are insignificant in small portions of rural Georgia with low socioeconomic status (SES). PTB has significant positive relationships with maternal age and parity, and significant negative relationships with female births in small portions of Georgia, particularly in the Atlanta Metropolitan Area, while in larger portions of rural Georgia with overall low SES, those relationships are not significant. Thus, health prevention and intervention policies should be tailored to address the locally important factors to reduce PTB risk.

Preterm birth (PTB) is a major cause of infant mortality and morbidity. The relationships between PTB and ambient air pollution have been examined by many previous studies worldwide, but the results vary among different studies, and no general conclusion could be drawn about the relationships. We analyzed 116,112 live and singleton births in year 2000 in Georgia, USA in this cross-sectional study. A spatial statistical method, Geographically Weighted Logistic Regression (GWLR), was employed to model the relationships between PTB and two ambient air pollutants, ozone (O3) and fine particulate matter (PM2.5), with nine individual-level birth and maternal demographic, socioeconomic, behavioral, and lifestyle factors, and three community socioeconomic status (SES) and urbanization variables as covariates. Different from the results calculated from global logistic regression, the results obtained from the GWLR model show that the relationships between PTB and the two pollutants vary over space. Positively significant (a higher risk of PTB is associated with a higher concentration of air pollutant), negatively significant (a lower risk of PTB is associated with a higher concentration of air pollutant), and non-significant relationships between PTB and air pollutants are all discovered in different regions of Georgia, and the varying relationships are strongly related to the varying SES and urbanization level of the communities of the births. PTB is not significantly associated with either O3 or PM2.5 in most of the state, especially in urban communities. The positively significant relationship between PTB and O3 or PM2.5 that indicates either air pollutant might be a significant PTB risk factor is primarily located in rural communities with low SES. These findings suggest that in order to more successfully reduce PTB risk, it is necessary to consider the varying relationships between PTB and air pollution across the communities with different levels of urbanization and SES for making and implementation of local public health policies.

Birth weight is an important indicator of overall infant health and a strong predictor of infant morbidity and mortality, and low birth weight (LBW) is a leading cause of infant mortality in the United States. Numerous studies have examined the associations of birth weight with ambient air pollution, but the results were inconsistent. In this study, a spatial statistical technique, geographically weighted regression (GWR) is applied to explore the spatial variations in the associations of birth weight with concentrations of ozone (O3) and fine particulate matter (PM2.5) in the State of Georgia, USA adjusted for gestational age, parity, and six other socioeconomic, behavioral, and land use factors. The results show considerable spatial variations in the associations of birth weight with both pollutants. Significant positive, non-significant, and significant negative relationships between birth weight and concentrations of each air pollutant are all found in different parts of the study area, and the different types of the relationships are affected by the socioeconomic and urban characteristics of the communities where the births are located. The significant negative relationships between birth weight and O3 indicate that O3 is a significant risk factor of LBW and these associations are primarily located in less-urbanized communities. On the other hand, PM2.5 is a significant risk factor of LBW in the more-urbanized communities with higher family income and education attainment. These findings suggest that environmental and health policies should be adjusted to address the different effects of air pollutants on birth outcomes across different types of communities to more effectively and efficiently improve birth outcomes.

This study estimates the neighborhood socioeconomic status (SES) effect on the risk of preterm birth (PTB) using multilevel regression (MLR) models. Birth data retrieved from year 2000 and 2010 Georgia Vital Records were linked to their respective census tracts. Principle component analysis (PCA) was performed on nine selected census variables and the first two principal components (Fac1 and Fac2) were used to represent the neighborhood-level SES in the MLR models. Two-level random intercept MLR models were specified using 122,744 and 112,578 live and singleton births at the individual level and 1613 and 1952 census tracts at the neighborhood level, for 2000 and 2010, respectively. After adjustment for individual level factors, Fac1, which represents disadvantaged SES, respectively generated an Odds Ratio of 1.056 (95% CI: 1.031-1.081) and 1.080 (95% CI: 1.056-1.105) for these two years, showing a modest but statistically significant effect on PTB. After adjusting for individual level factors and the census tract level factors, Intra-class correlation (ICC) was 1.2% and 1.4%, for year 2000 and 2010, respectively. The two IOR-80% intervals, 0.73-1.52 (year 2000) and 0.73-1.59 (year 2010) suggest large unexplained between census tract variation. The Median Odds Ratio (MOR) value of 1.21(year 2000) and 1.23 (year 2010) revealed that the un-modeled neighborhood effect was smaller than two individual-level predictor variables, race, and tobacco use but larger than the fixed effect of census tract-level predicting variable, Fac1 and all the other individual level factors. Overall, better census tract level SES was found to have a modest protective effect for PTB risk and the effects of the two examined years were similar. Large unexplained between census tract heterogeneity warrants more sophisticated MLR models to further investigate the PTB risk factors and their interactions at both individual and neighborhood levels.

This study investigates the contextual effect of neighbourhood socio-economic status (SES) on the risk of preterm birth (PTB) using multilevel models. Birth data retrieved from 2000 Georgia Vital Records were geocoded and joined to their respective census tracts. The census tract level Index of Deprivation (IoD) was calculated using nine 2000 Census variables based on a previously proposed ‘standard’ index. Two-level random intercept regression models were developed using 117,329 live and singleton births at the individual level and 1618 census tracts at the neighbourhood level. After adjustment for individual-level factors, IoD generated an odds ratio of 1.006 (95% CI 1.00–1.01), showing a modest but significant effect on PTB. Intra-class correlation (ICC) was 0.83% after adjusting for individual-level factors and the census tract level IoD. A wide IOR-80% interval (0.74–1.36) suggests large unexplained residual in between census tract variation remained. The median odds ratio (MOR) value of 1.17 revealed that the unmodelled neighbourhood effect was stronger than the fixed effect of census tract-level predicting variable, IoD, but weaker than the effects of several individual-level predictor variables, including race, tobacco use, prenatal care, foetal death history and marital status. Overall, better census tract-level SES would have a modest protective effect for PTB risk. The full strength of multilevel models should be exploited further to help our understanding of PTB aetiology.

A spatial statistical technique, Geographically Weighted Regression (GWR) is applied to study the spatial variations in the relationships between four land use indicators, including percentages of urban land, forest, agricultural land, and wetland, and eight water quality indicators including specific conductance (SC), dissolved oxygen, dissolved nutrients, and dissolved organic carbon, in the watersheds of northern Georgia, USA. The results show that GWR has better model performance than ordinary least squares regression (OLS) to analyze the relationships between land use and water quality. There are great spatial variations in the relationships affected by the urbanization level of watersheds. The relationships between urban land and SC are stronger in less-urbanized watersheds, while those between urban land and dissolved nutrients are stronger in highly-urbanized watersheds. Percentage of forest is an indicator of good water quality. Agricultural land is usually associated with good water quality in highly-urbanized watersheds, but might be related to water pollution in less-urbanized watersheds. This study confirms the results obtained from a similar study in eastern Massachusetts, and so suggest that GWR technique is a very useful tool in water environmental research and also has the potential to be applied to other fields of environmental studies and management in other regions.

The impact of land use changes caused by urban sprawl on water quality since the 1970s in northern Georgia, USA is studied by examining the spatial and temporal relationships between land use and water quality indicators in 43 watersheds through geographic information system (GIS) and statistical analyses. GIS analyses are used to delineate watersheds using Digital Elevation Models for water sampling sites and to derive land use indicators such as urban land, forest, agricultural land, and wetland for each watershed. Statistical analyses are used to examine, quantify, and compare the relationships between water quality and land use indicators and to find good predictors of water quality changes in response to the spatial and temporal variations of land uses. Significant spatial relationships are found between all the 14 studied water quality indicators, including dissolved oxygen, specific conductance, dissolved nutrients, dissolved ions, dissolved solids, and suspended sediment, and two land use indicators (percentages of urban land and forest). The more urbanized watersheds with higher percentage of urban land and lower percentage of forest tend to have higher concentrations of water pollutants. However, no significant temporal relationships are found between water quality and land use indicators. To fully evaluate, explain, and predict water quality long-term change, it is necessary to consider more natural and anthropogenic factors, such as water pollution control technology, policy, and climate change.

The impact of urbanization on water quality might vary over space because watershed characteristics, pollution sources, and land use patterns are not the same in different places. However, the spatially varying impact is usually not considered using conventional statistical methods, such as ordinary least squares regression (OLS) and Spearman's rank correlation analysis. This study applies a geospatial statistical technique, geographically weighted regression (GWR), to analyze the relationships between urbanization and water quality indicators across watersheds with varied urbanization levels in eastern Massachusetts, USA. The study finds that the relationships between water quality and urbanization indicators vary across the urbanization gradient in the studied watersheds. Percentage of developed land and population density are more strongly related to concentrations of:water pollutants in less-urbanized areas than in highly-urbanized areas. The adverse impact of urbanization on water quality is more substantial in less-urbanized suburban areas than highly-urbanized central cities, which is associated with the dominant pattern of urbanization in the study area: urban sprawl. The study suggests that GWR is a useful geospatial technology for policy makers, regional and local agencies, and researchers to unveil the local pollution causes, to improve the understanding of local pollution status, and to adopt appropriate environmental and land use planning policies suitable to the local watershed conservation and management.

This study developed new procedures to loosely integrate an air dispersion model, AERMOD, and a geographic information system (GIS) package, ArcGIS, to simulate air dispersion from stationary sources in the Bronx, New York City, for five pollutants: PM(10), PM(2.5), NO(x), CO, and SO(2). Plume buffers created from the model results were used as proxies of human exposure to the pollution from the sources and they modified the commonly used fixed-distance proximity buffers by considering the realities of air dispersion. The application of the plume buffers confirmed that the higher asthma hospitalization rates were associated with the higher potential exposure to local air pollution. The air dispersion modeling exhibited advantages over proximity analysis and geostatistical methods for environmental health research. The loose integration provides a relatively simple and feasible method for health scientists to take advantage of both air dispersion modeling and GIS by avoiding the need for intensive programming and substantial GIS expertise.

A study of water quality, land use, and population variations over the past three decades was conducted in eastern Massachusetts to examine the impact of urban sprawl on water quality using geographic information system and statistical analyses. Since 1970, eastern Massachusetts has experienced pronounced urban sprawl, which has a substantial impact on water quality. High spatial correlations are found between water quality indicators (especially specific conductance, dissolved ions, including Ca, Mg, Na, and Cl, and dissolved solid) and urban sprawl indicators. Urbanized watersheds with high population density, high percentage of developed land use, and low per capita developed land use tended to have high concentrations of water pollutants. The impact of urban sprawl also shows clear spatial difference between suburban areas and central cities: The central cities experienced lower increases over time in specific conductance concentration, compared to suburban and rural areas. The impact of urban sprawl on water quality is attributed to the combined effects of population and land-use change. Per capita developed land use is a very important indicator for studying the impact of urban sprawl and improving land use and watershed management, because inclusion of this indicator can better explain the temporal and spatial variations of more water quality parameters than using individual land use or/and population density.

Continuous measurements of ozone and its precursors including NO, NO2, and CO at an urban site (32°03′N, 118°44′E) in Nanjing, China during the period from January 2000 to February 2003 are presented. The effects of local meteorological conditions and distant transports associated with seasonal changed Asian monsoons on the temporal variations of O3 and its precursors are studied by statistical, backward trajectory, and episode analyses. The diurnal variation in O3 shows high concentrations during daytime and low concentrations during late night and early morning, while the precursors show high concentrations during night and early morning and low concentrations during daytime. The diurnal variations in air pollutants are closely related to those in local meteorological conditions. Both temperature and wind speed have significant positive correlations with O3 and significant negative correlations with the precursors. Relative humidity has a significant negative correlation with O3 and significant positive correlations with the precursors. The seasonal variation in O3 shows low concentrations in late autumn and winter and high concentrations in late spring and early summer, while the precursors show high concentrations in late autumn and winter and low concentrations in summer. Local mobile and stationary sources make a great contribution to the precursors, but distant transports also play a very important role in the seasonal variations of the air pollutants. The distant transport associated with the southeastern maritime monsoon contributes substantially to the O3 because the originally clean maritime air mass is polluted when passing over the highly industrialized and urbanized areas in the Yangtze River Delta. The high frequency of this type of air mass in summer causes the fact that a common seasonal characteristic of surface O3 in East Asia, summer minimum, is not observed at this site. The distant transports associated with the northern continental monsoons that dominate in autumn and winter are related to the high concentrations of the precursors in these two seasons. This study can contribute to a better understanding of the O3 pollution in vast inland of China affected by meteorological conditions and the rapid urbanization and industrialization.

A study of water quality, land-use patterns, and population data is conducted at several watersheds in eastern Massachusetts to examine the impact of urban sprawl on water quality at three different spatial scales: watersheds, buffered streams, and buffered water sampling sites. GIS analysis is used to delineate the drainage areas using digital elevation models for the water flow passing the corresponding water sampling sites; to generate buffers of different distances for the streams and the sampling sites; and to derive the indicators of suburbanization such as developed land and population density for different scales. Statistical analyses are used to examine and quantify the relationships between water quality parameters and the indicators of suburbanization. Results from this study will contribute to a better understanding of not only the impact of urban sprawl on water quality but also the appropriate scale for effective management of watersheds.