Spatially Varying Relationships between Land Subsidence and Urbanization: A Case Study in Wuhan, China

Land subsidence has become an increasing global concern over the past few decades due to natural and anthropogenic factors. However, although several studies have examined factors affecting land subsidence in recent years, few have focused on the spatial heterogeneity of relationships between land subsidence and urbanization. In this paper, we adopted the small baseline subset-synthetic aperture radar interferometry (SBAS-InSAR) method using Sentinel-1 radar satellite images to map land subsidence from 2015 to 2018 and characterized its spatial pattern in Wuhan. The bivariate Moran’s I index was used to test and visualize the spatial correlations between land subsidence and urbanization. A geographically weighted regression (GWR) model was employed to explore the strengths and directions of impacts of urbanization on land subsidence. Our findings showed that land subsidence was obvious and unevenly distributed in the study area, the annual deformation rate varied from −42.85 mm/year to +29.98 mm/year, and its average value was −1.0 mm/year. A clear spatial pattern for land subsidence in Wuhan was mapped, and several apparent subsidence funnels were primarily located in central urban areas. All urbanization indicators were found to be significantly spatially correlated with land subsidence at different scales. In addition, the GWR model results showed that all urbanization indicators were significantly associated with land subsidence across the whole study area in Wuhan. The results of bivariate Moran’s I and GWR results confirmed that the relationships between land subsidence and urbanization spatially varied in Wuhan at multiple spatial scales. Although scale dependence existed in both the bivariate Moran’s I and GWR models for land subsidence and urbanization indicators, a “best” spatial scale could not be confirmed because the disturbance of factors varied over different sampling scales. The results can advance the understanding of the relationships between land subsidence and urbanization, and they will provide guidance for subsidence control and sustainable urban planning.

Does the Accessibility of Regional Internal and External Traffic Play the Same Role in Achieving Anti-Poverty Goals?

Traffic development can promote the flow of goods and people, which has long been widely considered to have a poverty reduction effect but, in fact, is not unbreakable. The development of traffic is similar to economic and social development, with internal and external characteristics, but few studies have explored the differences between the effects of their poverty reduction. Taking the land traffic of the Chengdu-Chongqing Economic Zone (CCEZ) as an example, this paper represents traffic accessibility at a county level by relying on the average internal and external travel times. Rural poverty was identified by the pentagon of livelihoods to measure the Multidimensional Development Index (MDI). Furthermore, a Geographically Weighted Regression (GWR) model was used to explore the relationship and spatial differentiation characteristics between county traffic accessibility and poverty. The results show that the traffic accessibility of the counties in the CCEZ was quite different. The average internal travel time was between 0.16 and 7 h, and the average external travel time was between 4.2 and 10.6 h. The radiation gradient structure centered on Chengdu municipal districts and the Chongqing main urban area, and the accessibility level needed to be improved. Furthermore, the MDI values of each county in the CCEZ showed the structural characteristics of “large bottom and small top”; additionally, the higher the high-value group of MDI, the stronger the spatial aggregation and the more obvious the characteristics of regional differentiation. Finally, the relationship between traffic accessibility and poverty in counties cannot be generalized. The improvement of external traffic accessibility obviously helped to improve the poverty situation in the CCEZ; the improvement of internal traffic accessibility had a multidimensional impact, but it was mainly due to the occupation or spillover of livelihood capital in rural areas; counties accounting for 82.74% would even reduce the MDI and, thus, aggravate poverty.

Prediction of Soil Particle Size Fraction using Geographically Weighted Regression and Random Forest

The development of spatial modeling for soil properties has progressed in recent decades. This responds to the growing demand for land spatial data and exact soil property prediction for agronomical reasons, particularly in precision farming, in order to speed up precision agricultural activities. In this regards a comparison of the GWR and RF models was carried out in order to determine which model is the best at forecasting surface soil texture and how dependable each model is at doing so. The purpose of this research is to get the best model in predicting particle soil fraction (PSF). 50 topsoil samples were collected from several locations in the Kalikonto Watershed, Indonesia, and the soil PSF (sand, silt, and clay) in the upper 10 cm varied. The LMV, slope, and elevation were calculated using DEM data and utilized as predictor variables. As a result, the weighting of the GWR model has a considerable impact on the final model, and all other factors have a major effect on the PSF determination. The RF, on the other hand, looks to be superior than the GWR variants. The RF model outperformed the other models in every PSF variable. This study reveals that topsoil quality and terrain attributes are linked, which may be assessed using field measurements and model projections. More research is needed to generate more efficient input parameters that will help with soil variability precision and accuracy of soil map products.

Exploring the Spatial Heterogeneity of Residents' Marginal Willingness to Pay for Clean Air in Shanghai

Previous studies have paid little attention to the spatial heterogeneity of residents' marginal willingness to pay (MWTP) for clean air at a city level. To fill this gap, this study adopts a geographically weighted regression (GWR) model to quantify the spatial heterogeneity of residents' MWTP for clean air in Shanghai. First, Shanghai was divided into 218 census tracts and each tract was the smallest research unit. Then, the impacts of air pollutants and other built environment variables on housing prices were chosen to reflect residents' MWTP and a GWR model was used to analyze the spatial heterogeneity of the MWTP. Finally, the total losses caused by air pollutants in Shanghai were estimated from the perspective of housing market value. Empirical results show that air pollutants have a negative impact on housing prices. Using the marginal rate of transformation between housing prices and air pollutants, the results show Shanghai residents, on average, are willing to pay 50 and 99 Yuan/m2 to reduce the mean concentration of PM2.5 and NO2 by 1 μg/m3, respectively. Moreover, residents' MWTP for clean air is higher in the suburbs and lower in the city center. This study can help city policymakers formulate regional air management policies and provide support for the green and sustainable development of the real estate market in China.

Exploring a Pricing Model for Urban Rental Houses from a Geographical Perspective

Models for estimating urban rental house prices in the real estate market continue to pose a challenging problem due to the insufficiency of algorithms and comprehensive perspectives. Existing rental house price models based on either the geographically weighted regression (GWR) or deep-learning methods can hardly predict very satisfactory prices, since the rental house prices involve both complicated nonlinear characteristics and spatial heterogeneity. The linear-based GWR model cannot characterize the nonlinear complexity of rental house prices, while existing deep-learning methods cannot explicitly model the spatial heterogeneity. This paper proposes a fully connected neural network–geographically weighted regression (FCNN–GWR) model that combines deep learning with GWR and can handle both of the problems above. In addition, when calculating the geographical location of a house, we propose a set of locational and neighborhood variables based on the quantities of nearby points of interests (POIs). Compared with traditional locational and neighborhood variables, the proposed “quantity-based” locational and neighborhood variables can cover more geographic objects and reflect the locational characteristics of a house from a comprehensive geographical perspective. Taking four major Chinese cities (Wuhan, Nanjing, Beijing, and Xi’an) as study areas, we compare the proposed method with other commonly used methods, and this paper presents a more precise estimation model for rental house prices. The method proposed in this paper may serve as a useful reference for individuals and enterprises in their transactions relevant to rental houses, and for the government in terms of the policies and positions of public rental housing.

Investigation on spatial variability and influencing factors of drinking water iodine in Xinjiang, China

Background and objectives Xinjiang is one of the areas in China with extremely severe iodine deficiency. The health of Xinjiang residents has been endangered for a long time. In order to provide reasonable suggestions for scientific iodine supplementation and improve the health and living standards of the people in Xinjiang, it is necessary to understand the spatial distribution of iodine content in drinking water and explore the influencing factors of spatial heterogeneity of water iodine content distribution. Methods The data of iodine in drinking water arrived from the annual water iodine survey in Xinjiang in 2017. The distribution of iodine content in drinking water in Xinjiang is described from three perspectives: sampling points, districts/counties, and townships/streets. ArcGIS was used for spatial auto-correlation analysis, mapping the distribution of iodine content in drinking water and visualizing the distribution of Geographically Weighted Regression (GWR) model parameter. Kriging method is used to predict the iodine content in water at non-sampling points. GWR software was used to build GWR model in order to find the factors affecting the distribution of iodine content in drinking water. Results There are 3293 sampling points in Xinjiang. The iodine content of drinking water ranges from 0 to 128 μg/L, the median is 4.15 μg/L. The iodine content in 78.6% of total sampling points are less than 10 μg/L, and only that in the 3.4% are more than 40 μg/L. Among 1054 towns’ water samples in Xinjiang, 88.9% of the samples’ water iodine content is less than 10 μg/L. Among the 94 studied areas, the median iodine content in drinking water in 87 areas was less than 10 μg/L, those values in 7 areas were between 10–40 μg/L, and the distribution of water iodine content in Xinjiang shows clustered. The GWR model established had found that the effects of soil type and precipitation on the distribution of iodine content in drinking water were statistically significant. Conclusions The iodine content of drinking water in Xinjiang is generally low, but there are also some areas which their drinking water has high iodine content. Soil type and precipitation are the factors affecting the distribution of drinking water iodine content, and are statistically significant (P<0.05).

Spatial variability of water vapour in south and southwest of Iran

In this study two regression models, ordinary least square and geographically weighted regression as widely applied techniques, were used in modeling the regression relationships between water vapour and related geographic features, i.e., longitude, latitude, elevation, slope and aspect. Accordingly, the water vapour data in south and southwest of Iran were collected in pixels in the time interval 1981-2010. According to the general OLS regression, the relationship between WV and latitude, elevation and aspect were reverse and with longitude and slope were positive. Analyzing the relationship between geographic features and WV by GWR model determined that greatest coefficients of explanatory variables were in longitude, latitude, slope, aspect and elevation, respectively. Regarding to the model performance, GWR showed an improvement over OLS in estimating the WV and provided more realistic and useful results. So that the R2, Adjusted R2 and AICc for GWR were 0.967, 0.968 and 9329.38, respectively while these factors for OLS were 0.8478, 0.8475 and 14559.04.

Analysis of Factors Influencing Forest Loss in South Korea: Statistical Models and Machine-Learning Model

Analyzing the current status of forest loss and its causes is crucial for understanding and preparing for future forest changes and the spatial pattern of forest loss. We investigated spatial patterns of forest loss in South Korea and assessed the effects of various factors on forest loss based on spatial heterogeneity. We used the local Moran’s I to classify forest loss spatial patterns as high–high clusters, low–low clusters, high–low outliers, and high–low outliers. Additionally, to assess the effect of factors on forest loss, two statistical models (i.e., ordinary least squares regression (OLS) and geographically weighted regression (GWR) models) and one machine-learning model (i.e., random forest (RF) model) were used. The accuracy of each model was determined using the R2, RMSE, MAE, and AICc. Across South Korea, the forest loss rate was highest in the Seoul–Incheon–Gyeonggi region. Moreover, high–high spatial clusters were found in the Seoul–Incheon–Gyeonggi and Daejeon–Chungnam regions. Among the models, the GWR model was the most accurate. Notably, according to the GWR model, the main factors driving forest loss were road density, cropland area, number of households, and number of tertiary industry establishments. However, the factors driving forest loss had varying degrees of influence depending on the location. Therefore, our findings suggest that spatial heterogeneity should be considered when developing policies to reduce forest loss.

Assessing the impact of land surface temperature on urban net primary productivity increment based on geographically weighted regression model

Urbanization had a huge impact on the regional ecosystem net primary productivity (NPP). Although the urban heat island (UHI) caused by urbanization has been found to have a certain promoting effect on urban vegetation NPP, the factors on the impact still are not identified. In this study, the impact of urbanization on NPP was divided into direct impact (NPPdir) and indirect impact (NPPind), taking Kunming city as a case study area. Then, the spatial heterogeneity impact of land surface temperature (LST) on NPPind was analyzed based on the geographically weighted regression (GWR) model. The results indicated that NPP, LST, NPPdir and NPPind in 2001, 2009 and 2018 had significant spatial autocorrelation in Kunming based on spatial analytical model. LST had a positive impact on NPPind in the central area of Kunming. The positively correlation areas of LST on NPPind increased by 4.56%, and the NPPind caused by the UHI effect increased by an average of 4.423 gC m−2 from 2009 to 2018. GWR model can reveal significant spatial heterogeneity in the impacts of LST on NPPind. Overall, our findings indicated that LST has a certain role in promoting urban NPP.

Spatial–Temporal Heterogeneity and Driving Factors of Rural Residents’ Food Consumption Carbon Emissions in China—Based on an ESDA-GWR Model

The increase in income among Chinese residents has been accompanied by dramatic changes in dietary structure, promoting a growth in carbon emissions. Therefore, in the context of building a beautiful countryside, it is of great significance to study the carbon emissions of rural residents’ food consumption to realize the goal of low-carbon food consumption. In this paper, the calculation of food consumption carbon emissions of Chinese rural residents is based on the carbon conversion coefficient method, and the spatial heterogeneity of influencing factors is analyzed with the aid of the ESDA-GWR model. The results indicate that the per capita food consumption carbon emissions of rural residents have increased by 1.68% annually, reaching 336.73 kg CO2-eq in 2020, which is 1.32 times that of 2002. Carbon emissions generated from rural residents’ food consumption have significant spatial agglomeration characteristics, showing the spatial distribution characteristics of a north–south confrontation, with a central area collapse. The influencing factors of food consumption carbon emissions have significant spatial heterogeneity, among which, as the main force to restrain the growth of food consumption carbon emissions, the price factor has a regression coefficient between −0.1 and −0.3, and its influence has weakened from northwest to southeast in 2020. The education–social factor is the main driving force for the growth of food consumption carbon emissions, with a regression coefficient between 0.58 and 0.99, and its influence has increased from east to west. In the future, formulating food consumption optimization policies should be based on the actual situation of food consumption carbon emissions in various regions to promote the realization of low-carbon food consumption.