Geographically Weighted Regression Model Research Articles

Integrating the Budyko framework with the emerging hot spot analysis in local land use planning for regulating surface evapotranspiration ratio

Land use planning regulates surface hydrological processes by adjusting land properties with varied evapotranspiration ratios. However, a dearth of empirical spatial information hampers the regulation of place-specific hydrological processes. Therefore, this study proposed a Local Land Use Planning framework for EvapoTranspiration Ratio regulations (ETR-LLUP), which was tested for the developments of spatially-varied land use strategies in the Dongjiang River Basin (DRB) in Southern China. With the first attempt at integrating the Emerging Hot Spots Analysis (EHSA) with the Budyko framework, the spatiotemporal trends of evapotranspiration ratios based on evaporative index and dryness index, from 1992 to 2018, were illustrated. Then, representative land-cover types in each sub-basin were defined using Geographically Weighted Principal Component Analysis, in two wet years (1998 and 2016) and three dry years (2004, 2009, and 2018), which in turn were identified using the Standard Precipitation Index. Finally, Geographically Weighted Regressions (GWRs) were used to detect spatially-varied relationships between land-cover proportions and evaporative index in both dry and wet climates. Results showed that the DRB was consistently a water-limited region from 1992 to 2018, and the situation was getting worse. We also identified the upper DRB as hotspots for hydrological management. Forests and croplands experienced increasingly water stress compared to other vegetation types. More importantly, the spatial results of GWR models enabled us to adjust basin land use by 1) expanding and contracting a combination of ‘mosaic natural vegetation’ and ‘broadleaved deciduous trees’ in the western and eastern parts of the basin, respectively; and 2) increasing ‘broadleaved evergreen trees’ in the upstream parts of the basin. These spatially-varied land use strategies based on the ETR-LLUP framework allow for place-specific hydrological management during both dry and wet climates.

Spatiotemporal estimates of daily PM2.5 concentrations based on 1-km resolution MAIAC AOD in the Beijing–Tianjin–Hebei, China

The accurate prediction of PM2.5 concentrations is essential for health risk assessment and the formulation of air pollution prevention and control strategies. In this study, we selected a two-stage statistical regression model, which combined the linear mixed effect (LME) and geographically weighted regression (GWR) model, and applied aerosol optical depth (AOD) with 1-km spatial resolution, meteorological variables and land use parameters as predictors, to predict the daily near-ground PM2.5 concentrations (ρ(PM2.5)) from 2013 to 2017 in the Beijing–Tianjin–Hebei (BTH) region. The model was fitted annually, and the coefficient of determination (R2), mean prediction error (MPE), root-mean-squared prediction errors (RMSPE), and relative prediction error (RPE) ranged of the model's cross-validation results were 0.85–0.94, 8.24−17.78 µg/m3, 13.14−29.90 µg/m3 and 21.42−32.71%, respectively. The LME + GWR model significantly outperformed the LME model, but with a slight increase in overfitting. The ρ(PM2.5) in the BTH displayed obvious temporal (decrease per year, highest in winter and lowest in summer) and spatial (high in southern plains and low in northwestern mountain areas) characteristics. During the investigated period, the annual mean ρ(PM2.5) dropped by 35.5%, and the areas with high concentrations shrank significantly. Moreover, a sharply decline occurred from 2015 to 2017 due to the strict implementation of the government's pollution-preventing policies on pollutant emissions. This study can provide a scientific basis and support for the understanding and prevention of air pollution in the BTH region and beyond.

Estimating 1 km gridded daily air temperature using a spatially varying coefficient model with sign preservation

Near-surface air temperature (Ta) is one of the key variables in a variety of studies such as hydrological modeling, assessment of heat waves, and energy modeling. Among existing methods, statistical algorithms are suitable for integrating auxiliary spatial data with station-based Ta data to produce gridded Ta over large areas. However, existing statistical algorithms (e.g., Geographically Weighted Regression (GWR)) cannot always correctly capture and preserve relationships between Ta and explanatory variables, which may increase uncertainties of relevant applications based on the estimated Ta with abnormal spatial patterns. This issue is mainly caused by the lack of enough observations due to the limited spatial coverage of weather stations, leading to abnormal relationships between Ta and explanatory variables. In order to address this issue, in this study, we introduced a new method named the Spatially Varying Coefficient Models with Sign Preservation (SVCM-SP) to estimate gridded Ta using gridded land surface temperature (LST) and elevation as explanatory variables with presetting positive and negative signs for coefficients, respectively. Using this method, first, we calculated the preset parameters of the bivariate spline surface. Second, we used the input data at weather stations and constrained least squares regression to obtain the coefficient surface for both the explanatory variables (i.e., elevation and LST) and the intercept. Third, we calculated the gridded Ta using the 1 km gridded LST and elevation data, and the estimated spatially varying coefficient surfaces. We evaluated the model performance for estimating 1 km gridded daily maximum and minimum Ta (i.e., Tmax and Tmin) data in mainland China from 2003 to 2016 using 10-fold cross-validation and compared its performance with the GWR model. The average root mean square error (RMSE) and mean absolute error (MAE) based on the SVCM-SP are 1.75 °C and 1.22 °C for Tmax, and 1.82 °C and 1.30 °C for Tmin, respectively. The SVCM-SP method showed better performance than the GWR in terms of accuracy, computing efficiency, and has more interpretable coefficients for explanatory variables to get more realistic spatial pattern of gridded Ta. More important, the sign preservation of the SVCM-SP method can mitigate the issue of abnormal relationships between Ta and explanatory variables in the traditional methods such as GWR, and therefore will contribute to future studies in developing better gridded air temperature or relevant data products.

Multi-scale spatio-temporal analysis of soil conservation service based on MGWR model: A case of Beijing-Tianjin-Hebei, China

To improve regional ecosystem services and ecological management feedback, insights should be gained into the driving mechanisms of ecological service. As reported by existing studies in ecosystem drivers, ecological and socioeconomic factors at different spatial scales internally can drive the structure, process and service of the ecosystem, whereas the intrinsic multi-scale mechanism of the drivers above remain unclear. This study attempts to clarify the drivers of soil conservation service and investigate the multi-scale driving mechanism by employing an optimized Multiscale Geographically Weighted Regression model (MGWR) in Beijing-Tianjin-Hebei region. As indicated by the result, (1) the analysis of the bandwidth size and spatial characteristics of the driving factors revealed that the dominant drivers of soil conservation service exerted a wide-changed scale effect, in which four scales were presented (i.e., region, city clusters, urban and county scale) from an administrative spatial perspective. (2) At a spatial scale, the contribution of natural factors exceeded landscape index and socioeconomic factors at the region spatial scale in accordance with the significance test value. However, with the decrease in the spatial scale range, the driving force of landscape index and socioeconomic factors tended to increase. (3) At a time scale, the drivers of soil conservation service are changed over time as impacted by regional land use, economic and social, ecological restoration project, etc. As suggested by the analytical results, coordinated regional eco-socio-economic management and regional sustainable development with ecosystem service enhancement as the goal orientation can be scientifically and effectively guided through the assessment of multi-scale spatial and temporal variation patterns of ecosystem service drivers.

Spatial analysis of echinococcosis in pastoral area of Qinghai province, 2019

Objective: To understand the spatial characteristics of echinococcosis and associated factors in the pastoral area of Qinghai province, and provide evidence for the effective prevention and control of echinococcosis. Methods: The number of echinococcosis cases in the pastoral areas of Qinghai in 2019 was collected to perform spatial epidemiological analysis. The thematic map of the distribution of echinococcosis cases was generated with software ArcGIS 10.8 for visual analysis and spatial autocorrelation analysis. The spatial autocorrelation and spatial scanning analysis were performed to estimate the clustering of echinococcosis with software SaTScan 9.5. Software GeoDa 1.14 and ArcGIS 10.8 were used to establish spatial lag model and geographical weighted regression model to analyze the related factors of echinococcosis epidemic. Results: In 2019, the echinococcosis surveillance covered 64 741 people in the pastoral area of Qinghai, and 829 echinococcosis cases were found, with a prevalence rate of 1.28%. The distribution of the cases had spatial correlation (Moran's I=0.41, P<0.001). The most possible clustering areas indicated by spatial scanning analysis included Banma, Jiuzhi, Dari and Gande counties of Guoluo Tibetan Autonomous Prefecture (LLR=460.77, RR=9.20, P<0.001). The prevalence of echinococcosis in the pastoral areas was positively associated with the total annual precipitation (β=0.13, P=0.036), and negatively associated with population density (β=-1.36, P=0.019) and doctors/nurse ratio (β=-25.60, P=0.026). Conclusions: The distribution of echinococcosis cases in the pastoral areas of Qinghai in 2019 had spatial correlation, and the prevalence was affected by total annual precipitation, population density, and doctors/nurse ratio.

Spatial non‐stationary effects of explanatory variables on soil bulk density in the critical zone of the Chinese Loess Plateau

AbstractBulk density (BD) is one of the most important soil physical properties in the hydrological processes of Earth's Critical Zone (CZ). The heterogeneity and related controlling factors of surface BD have received extensive research attention. However, it is unclear whether the effects of controlling factors on BD are homogeneous or vary spatially. This study quantified topsoil BD with a high‐density sampling strategy (10 m × 10 m; n = 1134) across the ShuangChaGou watershed in the critical zone of the Chinese Loess Plateau (CLP‐CZ). Stepwise multiple linear regression (SMLR), regression kriging (RK), random forest (RF) and geographically weighted regression (GWR) were established as modelling approaches to (i) compare the performance of these four models in explaining BD variation, (ii) assess the main factors driving BD, and (iii) evaluate the spatial non‐stationary relationship between BD and its explanatory variables. GWR explained the BD variation to the same extent as the RK model (35%), outperforming both the SMLR (15%) and RF models (19%). The variation in topsoil BD depended mainly on: soil organic carbon (SOC) content, normalised difference vegetation index (NDVI), plan curvature, and slope length. The spatial distribution of the local R2, regression coefficient magnitude, significance and sign calculated by GWR revealed that the effects of variables accounting for variation in BD varied spatially. The explanatory power of these variables in terms of BD variation was strongest in northern and northeastern areas of the watershed and weakest in southern areas of the watershed. Both significant negative and positive correlations between BD and silt content, NDVI, and plan curvature were found. This study enriched traditional research linked to the complicated relationships between BD and controlling factors by incorporating spatial non‐stationarity, which is useful for improving the accuracy of hydrological modelling in the CLP‐CZ and in similar CZs worldwide.Highlights Non‐stationary relationships between BD (N = 1134) and its explanatory variables were captured at the watershed scale. GWR model performed on par with the RK and better than the SMLR or RF model. Silt content, NDVI, and plan curvature showed both significant negative and positive effects on BD. Spatial non‐stationarity of BD should be considered for improving the performance of hydrological modeling.

Influence of Built Environment on Subway Trip Origin and Destination: Insights Based on Mobile Positioning Data

An accurate understanding of the relationship between subway trips and the built environment is crucial to meet people’s travel demands and promote the coordinated development of urban land. However, existing literature mainly examines this relationship at the station level, ignoring the variations between the surrounding areas of the station. Therefore, the present study aims to explore the influencing factors and spatial variations of subway trips at the grid level. First, a method was proposed to extract the subway trips using mobile positioning data in Chengdu, China. Then, two geographically weighted regression (GWR) models were adopted to examine the spatially varying relationships between the subway trip origin and destination and selected explanatory variables. The results show that the hotel, company, residential, tourist, bus, subway, road density, and transfer stations variables positively affect trip origin and destination. However, distance to the nearest subway station has a negative impact. Besides, the goodness of fit of the GWR model is better than that of the global regression model, indicating that the influence of the built environment on trip origin and destination varies across space. This study can guide planning departments and transportation agencies to implement target policies and create a convenient travel environment at the micro-level.

Spatiotemporal changes and driving forces of ecosystem vulnerability in the Yangtze River Basin, China: Quantification using habitat-structure-function framework

Ecosystem vulnerability is the degree to which an ecosystem is susceptible to adverse effects of external disturbances. Exploring the pattern of ecosystem vulnerability and its driving mechanism is important for regional ecological protection and management. A little study has conducted the ecosystem vulnerability assessment from the perspective of multiple ecosystems characteristics, and the spatial heterogeneity impacts of climate change and human activities on ecosystem vulnerability variation need to be further explored. In this study, a habitat-structure-function framework was proposed to evaluate ecosystem vulnerability pattern of the Yangtze River Basin (YRB) in China from 1990 to 2018. Then, the spatial heterogeneity impacts of various factors on ecosystem vulnerability changes were examined utilizing the Geographically Weighted Regression model. Results show that the ecosystem vulnerability index (EVI) pattern in the YRB decreased from upstream to downstream. There was 63.85% of the basin area experiencing a decline in EVI from 1990 to 2018, which was primarily found in the source, southwest and north regions, while the southeast and east regions have suffered an increase in EVI. The impact of climate change on EVI changes increased as time scales increase, while, human activities were still the dominant factor leading EVI changes. Overall, areas with great impact of climate change on EVI variation were concentrated in the source region and upper reaches, while the remarkable impact of human activities occurred in the whole basin. The enhancement of climate warming and humid trend and the strengthen of ecological protection were benefit to the decline of EVI. The proposed framework can be extended to assess vulnerability in other areas or specific ecosystem types, and the findings are expected to provide policy recommendations for ecosystem conservation and management in the YRB.

Exploring spatial patterns of environmental noise and perceived sound source dominance in urban areas. Case study: the city of Athens, Greece

The aim of the present study is to map spatial patterns related to noise pollution and the acoustic environment -in a broader context- in the urban area of Athens, Greece. The primary goal of this thesis is to present a comprehensive approach that combines elements of two basic methodologies related to acoustic environment studies: a) noise mapping and b) the soundscape approach. The main inputs are environmental noise measurements and perceptual sound source-related observations. The results feature three noise pollution maps (LAeq,30 sec, L10, and L90 indices) and three sound source maps which reflect the way in which the human ear perceives the presence of sounds. Additionally, the question of whether the spatial distribution of sound source dominance can be explained by the dispersion of environmental noise levels was examined using geographically weighted regressions (GWR). The GWR models showed that sound source-related observations are explained to a significant extent by all three indicators. Four important findings emerge from the analysis. Firstly, areas with high levels of noise pollution are characterized by high to moderate presence of technological and absence of anthropic and natural sounds. Secondly, regions, where there is a simultaneous presence of all sound sources, are characterized by moderate to low noise levels. Thirdly, the absence of technological sounds is observed in quiet areas. Finally, areas featuring a moderate presence of technological and natural sounds are mostly urban green spaces built-in proximity to the main road network.

지역별 주택 공급과 시장가격 변화는 신혼부부 증감에 영향을 미치는가?

This study aimed to identify the effects of housing supply and price changes by regions for the increase and decrease in newlyweds and spatial causation with demographic-sociological factors for the quality of life and to present implications. To do so, this study used the ordinary least squares regression (OLS) and geographic weighted regression (GWR) models. The main results are as follows. First, the variables in housing markets significantly affected the increase and decrease in newlyweds

코로나바이러스감염증-19 발생이 서울시 외식업 매출에 미치는 영향*

This study aims to identify the effects of the COVID-19 pandemic on the food business and come up with more efficient bases to set measures for promoting local economy. To the end, this study used the sales of food business as a dependent variable and locational factors and data related to the COVID-19 as an independent variable for 424 administrative dongs in Seoul in order to analyze the ordinary least squares (OLS) and the geographically weighted regression (GWR) models in 2019 and 2020. The results of the analysis are as follows. Firstly, factors turned out to change for influencing the sales of food business caused by the pandemic. The influence of the factors has decreased since the outbreak of the COVID-19, though, they had positive relations with the sales around the outbreak. However, the variables related to social distancing had negative relations with the sales around the outbreak of the COVID-19 and the variable of COVID-19 patients had stronger relations on the 2020 model. Secondly, the GWR model demonstrated that the influences caused by the COVID-19 are not equal to every dong in Seoul because there are both positive and negative relations. However, this study has a limitation as it did not consider when the variables could influence on the sales of the food business. Accordingly, this study suggests a follow-up researches with more concrete explanations about time and space for other local areas.

Effects of block morphology on the surface thermal environment and the corresponding planning strategy using the geographically weighted regression model

Although an increasing number of studies have explored how spatial morphology influences the urban heat island (UHI) effect in different regions with the spatial regression methods, few studies further explored the potential significance of the identified spatial differences at the block level. In this study, 410 regulatory management units with diverse block features in Xi'an, China were screened as the research samples, and 14 block morphological indicators from three categories of land cover, building group, and distance to cold and heat sources were calculated with multi-source data. The Geographically Weighted Regression (GWR) model combined with spatial autocorrelation analysis was applied to quantitatively explore the spatial heterogeneity of land surface temperature (LST) and its relationship to various morphological factors. Results confirm that the GWR model with the better modeling fit (R2 = 0.836, Adj.R2 = 0.785, AICc = 977.510, RMSE = 0.696, and MAPE = 0.013) is an effective method for addressing the spatial relationship between block morphology and LST. Furthermore, the target control strategy was finally proposed in five zoning blocks according to the contribution of morphological variables. ISR is the most influential factor that intensifies the UHI effect, while GR is a critical factor that form the cool urban island in different blocks. Immediate improvement measures should be taken for Zone 1 with the most severe UHI effect by controlling the key indicators of ISR, BD, and GR. This study sheds new light on the spatial variations of the morphology-LST linkage at the block level and provides heat-mitigation regulation recommendations for urban planners and policy makers.

Spatio-temporal heterogeneity analysis of energy use in residential buildings

Conventional building energy dynamic analysis relies on climatic and geographical statistics of buildings. A large number of studies focus on investigating the major influencing factors and the energy use with engineering methods and statistical methods to predict buildings' energy use. However, due to the lack of temporal statistics when analyzing buildings' energy use intensity (EUI), very few studies concerned the energy consumption from spatial and temporal heterogeneity simultaneously. This study extends the traditional regression model by introducing the temporal variations into the geographically weighted regression (GWR) model and developed a geographically and temporally weighted regression (GTWR) model. A case study, which is based on four years’ urban building energy consumption (from 2015 to 2018) in the city of Seattle, is used to validate the performance of the proposed regression model. Meanwhile, the ordinary least square (OLS), GWR, and temporally weighted regression (TWR) models are developed for making cross-comparison of the goodness-of-fit. The results show the GTWR and TWR models reduce the mean absolute error by 13% and 7%, respectively, relative to the GWR model. Moreover, the GTWR model has a substantial improvement of goodness-of-fit, which can be referred from the value of R2 increasing by 49% in OLS, 16% in GWR, and 6% in TWR. Two driving forces, the number of floors (NF) and energy stars (ES), are identified to have the most significant coefficients in evaluating the site EUI. The novelty of this research lies in achieving better reliability and accuracy in building energy modeling by GTWR, which provides an assisting tool for government and local authorities to make the master plan and sustainability strategies for the new buildings in the city.

Spatial Downscaling Model Combined with the Geographically Weighted Regression and Multifractal Models for Monthly GPM/IMERG Precipitation in Hubei Province, China

High spatial resolution (1 km or finer) precipitation data fields are crucial for understanding the Earth’s water and energy cycles at the regional scale for applications. The spatial resolution of the Global Precipitation Measurement (GPM) mission (IMERG) satellite precipitation products is 0.1° (latitude) × 0.1° (longitude), which is too coarse for regional-scale analysis. This study combined the Geographically Weighted Regression (GWR) and the Multifractal Random Cascade (MFRC) model to downscale monthly GPM/IMERG precipitation products from 0.1° × 0.1° (approximately 11 km × 11 km) to 1 km in Hubei Province, China. This work’s results indicate the following: (1) The original GPM product can accurately express the precipitation in the study area, which highly correlates with the site data from 2015 to 2017 (R2 = 0.79) and overall presents the phenomenon of overestimation. (2) The GWR model maintains the precipitation field’s overall accuracy and smoothness, with even improvements in accuracy for specific months. In contrast, the MFRC model causes a slight decrease in the overall accuracy of the precipitation field but performs better in reducing the bias. (3) The GWR-MF combined with the GWR and MFRC model improves the observation accuracy of the downscaling results and reduces the bias value by introducing the MFRC to correct the deviation of GWR. The conclusion and analysis of this paper can provide a meaningful experience for 1 km high-resolution data to support related applications.

Spatial Non-Stationarity-Based Landslide Susceptibility Assessment Using PCAMGWR Model

Landslide Susceptibility Assessment (LSA) is a fundamental component of landslide risk management and a substantial area of geospatial research. Previous researchers have considered the spatial non-stationarity relationship between landslide occurrences and Landslide Conditioning Factors (LCFs) as fixed effects. The fixed effects consider the spatial non-stationarity scale between different LCFs as an average value, which is represented by a single bandwidth in the Geographically Weighted Regression (GWR) model. The present study analyzes the non-stationarity scale effect of the spatial relationship between LCFs and landslides and explains the influence of factor correlation on the LSA. A Principal-Component-Analysis-based Multiscale GWR (PCAMGWR) model is proposed for landslide susceptibility mapping, in which hexagonal neighborhoods express spatial proximity and extract LCFs as the model input. The area under the receiver operating characteristic curve and other statistical indicators are used to compare the PCAMGWR model with other GWR-based models and global regression models, and the PCAMGWR model has the best prediction effect. Different spatial non-stationarity scales are obtained and improve the prediction accuracy of landslide susceptibility compared to a single spatial non-stationarity scale.

An Application of Geographically Weighted Regression for Assessing Water Polution in Pontianak, Indonesia

Geographically weighted regression (GWR) is an exploratory analytical tool that creates a set of location-specific parameter estimates. The estimates can be analysed and represented on a map to provide information on spatial relationships between the dependent and the independent variables. A problem that is faced by the GWR users is how best to map these parameter estimates. This paper introduces a simple mapping technique that plots local t-values of the parameters on one map. This study employed GWR to evaluate chemical parameters of water in Pontianak City. The chemical oxygen demand (COD) was used as the dependent variable as an indicator of water polution. Factors used for assessing water pollution were pH (X1), iron (X2), fluoride (X3), water hardness (X4), nitrate (X5), nitrite (X6), detergents (X7) and dissolved oxygen, DO, (X8). Samples were taken from 42 locations. Chemical properties were measured in the laboratory. The parameters of the GWR model from each site were estimated and transformed using Geographic Information Systems (GIS). The results of the analysis show that X1, X2, X3, X5, and X7 influence the amount of COD in water. The resulting map can assist the exploration and interpretation of data.

Geospatial Assessment of Pesticide Concentration in Ambient Air and Colorectal Cancer Incidence in Arkansas, 2013-2017.

Exposure to various agricultural pesticides has been linked to colorectal cancer (CRC), mostly among farmworkers and applicators. Given the potential pesticide drift in ambient air, residents near farmland may be exposed to carcinogenic pesticides even if they are not actively engaged in pesticide application. Pesticide air pollution at the county level was estimated using the 2014 National Air Toxics Assessment. CRC incidence data were acquired from the Arkansas Central Cancer Registry for 2013–2017. We ran ordinary least squares (OLS) regression models, finding significant spatial autocorrelation of residuals for most models. Using geographically weighted regression (GWR) we found age-adjusted CRC incidence rates vary in an increasing west-to-east gradient, with the highest rates in the Arkansas Delta region. A similar gradient was observed in the distribution of the population living below the poverty line and the population percentage of Black people. Significant associations between Trifluralin (crude model only), Carbon Tetrachloride, and Ethylene Dibromide with CRC incidence rates in OLS models only explained 5–7% of the variation and exhibited spatial autocorrelation of residuals. GWR models explained 24–32% (adjusted r2 9–16%) of CRC incidence rate variation, suggesting additional factors may contribute to the association between pesticides and CRC.

GaWRDenMap: a quantitative framework to study the local variation in cell\u2013cell interactions in pancreatic disease subtypes

Spatial pattern modelling concepts are being increasingly used in capturing disease heterogeneity. Quantification of heterogeneity in the tumor microenvironment is extremely important in pancreatic ductal adenocarcinoma (PDAC), which has been shown to co-occur with other pancreatic diseases and neoplasms with certain attributes that make visual discrimination difficult. In this paper, we propose the GaWRDenMap framework, that utilizes the concepts of geographically weighted regression (GWR) and a density function-based classification model, and apply it to a cohort of multiplex immunofluorescence images from patients belonging to six different pancreatic diseases. We used an internal cohort of 228 patients comprised of 34 Chronic Pancreatitis (CP), 71 PDAC, 70 intraductal papillary mucinous neoplasm (IPMN), 16 mucinous cystic neoplasm (MCN), 29 pancreatic intraductal neoplasia (PanIN) and 8 IPMN-associated PDAC patients. We utilized GWR to model the relationship between epithelial cells and immune cells on a spatial grid. The GWR model estimates were used to generate density signatures which were used in subsequent pairwise classification models to distinguish between any two pairs of disease groups. Image-level, as well as subject-level analysis, were performed. When applied to this dataset, our classification model showed significant discrimination ability in multiple pairwise comparisons, in comparison to commonly used abundance-based metrics, like the Morisita-Horn index. The model was able to best discriminate between CP and PDAC at both the subject- and image-levels. It was also able to reasonably discriminate between PDAC and IPMN. These results point to a potential difference in the spatial arrangement of epithelial and immune cells between CP, PDAC and IPMN, that could be of high diagnostic significance. Further validation on a more comprehensive dataset would be warranted.

Abstract EP17: Associations Between Neighborhood Child Opportunity And Cardiovascular Fitness For New York City Public School Youth

Introduction: Fewer than 1/4th of US youth meet physical activity guidelines, leading to physical fitness-related cardiovascular disease disparities tracking into adulthood. Neighborhood opportunity may serve as a critical modifiable factor to reduce youth cardiovascular health-related disparities. The present analysis uses the Child Opportunity Index (COI), a comprehensive measure of children’s neighborhood opportunity, to examine associations between neighborhood context and multiple youth cardiovascular fitness outcomes. Methods: Data were drawn from the NYC FITNESSGRAM (n = 300,000), comprising of fitness data from New York City (NYC) public school youth (grades K-12). Generalized linear mixed models were run to estimate overall and sex-stratified associations between overall COI and individual COI indicators (greenspace, healthy food, walkability, commute time) and youth cardiovascular fitness (body mass index (BMI) percentiles, curl-ups, push-ups, PACER, and sit and reach). Geographically weighted regression (GWR) models explored spatial variation of COI-fitness associations across NYC. Results: Overall COI was associated with improved youth cardiovascular fitness outcomes, with the strongest magnitude of effects across aerobic capacity and muscular strength and endurance measures (PACER: β: 0.15, 95% CI: 0.14, 0.15, curl-ups: β: 0.13, 95% CI: 0.12, 1490.13, and push-ups: β: 0.13, 95% CI: 0.12, 0.13). Greenspace was positively associated with muscular endurance (curl-ups: β: 1.13, 95% CI:0.76, 1.50; push-ups: β: 1.53, 95% CI: 1.21, 1.84), and walkability and commuting time were positively associated with aerobic capacity (PACER: β: 2.11, 95% CI: 1.72, 2.50; β: 2.03, 95% CI: 1.87, 2.19, respectively). Stratified models showed strengthened COI-fitness associations among girls versus boys. GWR models were largely consistent with findings from multilevel models, except for overall COI-BMI associations, which varied in magnitude and direction across regions in NYC. Conclusion: Neighborhood opportunity was associated with multiple measures of youth cardiovascular fitness, though geographic models demonstrate that certain associations vary in magnitude and direction across NYC. Continued research on neighborhood factors and cardiovascular fitness may better inform public health efforts to reduce health disparities modifiable through comprehensive place-based interventions.

The Impact of Tourism on Ecosystem Services Value: A Spatio-Temporal Analysis Based on BRT and GWR Modeling

The healthy development of the ecosystem and tourism in destinations plays an essential role in sustainable development. Taking Shennongjia as an example, we analyzed the spatial–temporal variation in the ecosystem services value (ESV) and investigated the impacts of tourism on ESV and their spatial heterogeneity using the geographically weighted regression (GWR) and boosting regression tree (BRT) models. The results showed that (1) the types of ecosystem services (ESs) were dominated by climate regulation and biodiversity. The ESV increased from 3.358 billion yuan to 8.910 billion yuan from 2005 to 2018 and showed significant spatial divergence, maintaining a long-term distribution pattern of high in the center and low at the border. (2) The GWR and BRT results showed that the Distance to Scenic Spots (DSS) and the Distance to Residential Areas (DRA) are important factors influencing ESV, with the Distance to Hotels (DH) and the Distance to Roads (DR) having a relatively weak influence on ESV. (3) The influencing factors presented positive and negative effects, and the degree of influence has spatial heterogeneity. The DRA and DH inhibited the increase in ESV in nearby areas, while DR was the driving factor for increasing ESV. The assessment results of DSS vary according to the models.