Geographically weighted regression (GWR) introduces the distance weighted kernel function to examine the non-stationarity of geographical phenomena and improve the performance of global regression. However, GWR calibration becomes critical when using a serial computing mode to process large volumes of data. To address this problem, an improved approach based on the compute unified device architecture (CUDA) parallel architecture fast-parallel-GWR (FPGWR) is proposed in this paper to efficiently handle the computational demands of performing GWR over millions of data points. FPGWR is capable of decomposing the serial process into parallel atomic modules and optimizing the memory usage. To verify the computing capability of FPGWR, we designed simulation datasets and performed corresponding testing experiments. We also compared the performance of FPGWR and other GWR software packages using open datasets. The results show that the runtime of FPGWR is negatively correlated with the CUDA core number, and the calculation efficiency of FPGWR achieves a rate of thousands or even tens of thousands times faster than the traditional GWR algorithms. FPGWR provides an effective tool for exploring spatial heterogeneity for large-scale geographic data (geodata).

Forests play a fundamental role in stabilizing the global climate. Aboveground biomass (AGB) is important to the global carbon balance and environmental protection. Remotely sensed features such as single-band information, vegetation indices, texture features and terrain factors have been assessed to accurately estimate the forest biomass. A feature selection SVM-RFE (support vector machine recursive feature elimination) method is proposed to explore the relationship between the biomass and parameters derived from the Mount Tai area Landsat-8 imagery, thereby improving the AGB estimation accuracy. The least significant parameter is recursively removed according to a scoring function to determine the optimal subset of parameters. The performance of the SVM-RFE algorithm biomass feature selection method is then compared with the widely used stepwise regression method. The results show that the SVM-RFE method is superior to the stepwise linear regression method. And it could accurately estimate the AGB compared with field measurements in the context of the recent reducing emissions from forest deforestation and degradation (REDD) mechanism adopted by the United Nations.

Air quality degradation has triggered a large-scale public health crisis globally. Existing machine learning techniques have been used to attempt the remote sensing estimates of PM2.5. However, many machine learning models ignore the spatial non-stationarity of predictive variables. To address this issue, this study introduces a Flexible Geographically Weighted Neural Network (FGWNN) to estimate PM2.5 based on multi-source remote sensing data. FGWNN incorporates the Flexible Geographical Neuron (FGN) and Geographical Activation Function (GWAF) within the framework of Artificial Neural Network (ANN) to capture the intricate spatial non-stationary relationships among predictive variables. A robust air quality remote sensing estimation model was constructed using remote sensing data of Aerosol Optical Depth (AOD), Normalized Difference Vegetation Index (NDVI), Temperature (TMP), Specific Humidity (SPFH), Wind Speed (WIND), and Terrain Elevation (HGT) as inputs, and Ground-Based PM2.5 as the observation. The results indicated that FGWNN successfully generates PM2.5 remote sensing data with a 2.5 km spatial resolution for the contiguous United States (CONUS) in 2022. It exhibits higher regression accuracy compared to traditional ANN and Geographically Weighted Regression (GWR) models. FGWNN holds the potential for applications in high-precision and high-resolution remote sensing scenarios.

In the rapid development of air pollution over the past two decades in Shandong Province, it has played a detrimental role, causing severe damage to regional ecological security and public health. There has been little research at the county scale to explore the spatiotemporal causes and heterogeneity of PM2.5 pollution. This study utilizes a Geographically and Temporally Weighted Regression Model (GTWR) to environmentally model meteorological elements and socioeconomic conditions in Shandong Province from 2000 to 2020, aiming to identify the key driving factors of PM2.5 concentration changes across 136 counties. The results show that PM2.5 pollution in Shandong Province peaked in 2013, followed by a rapid decline in pollution levels. Geographically, counties in the western plains of Shandong generally exhibit higher pollution levels, while most counties in the central hills of Shandong and the Jiaodong Peninsula are in low pollution areas. Strong winds positively influence air quality in the southeast of Shandong; high temperatures can ameliorate air pollution in areas outside the southeast, whereas air pressure exhibits the opposite effect. Precipitation shows a significant negative correlation in the Laizhou Bay and central Shandong regions, while relative humidity primarily exerts a negative effect in coastal areas. The impact of fractional vegetation cover is relatively mild, with positive effects observed in southern Shandong and negative effects in other regions. Population density shows a significant positive correlation in the western plains of Shandong. Economic factors exhibit predominantly positive relationships, particularly in the northwest and the Jiaodong Peninsula. Electricity consumption in southern Shandong correlates positively, while industrial factors show positive effects province-wide. PM2.5 pollution in Shandong Province demonstrates significant spatiotemporal heterogeneity, aligning with governmental expectations for the effectiveness of air pollution control measures. The conclusions of this study can be utilized to assess the efficiency of air pollution abatement at the county level and provide quantitative data support for the revision of regional emission reduction policies.

The prolonged dependence on industrial development has accentuated the cumulative effects of pollutants. Simultaneously, influenced by land construction activities and green space depletion, the Urban Heat Island (UHI) effect in cities has intensified year by year, jeopardizing the foundation of sustainable urban development. Prudent urban spatial planning holds the potential to robustly ameliorate the persistent deterioration of the UHI phenomenon. This study selects Jinan City as a case study and employs spatial autocorrelation and spatial regression algorithms to explore the spatiotemporal evolution of urban-rural patterns at the township scale. The aim is to identify key factors driving the spatiotemporal differentiation of Land Surface Temperature (LST) from 2013 to 2022. The research reveals a trend of initially rising and subsequently falling LST in various townships, with low-temperature concentration areas in the southern mountainous region and the northern plain area. The "West-Central-East" main urban axis and the southeast Laiwu District exhibit high-temperature zones. Significant influences on LST are attributed to pollution levels, topographical factors, urbanization levels, and urban greenness. The global Moran’s Index for LST exceeds 0.7, indicating a strong positive spatial correlation. Cluster analysis results indicate High-High (HH) clustering in the central Shizhong District and Low-Low (LL) clustering in the northern Shanghe County. Multiscale Geographically Weighted Regression (MGWR) outperforms Geographically Weighted Regression (GWR) and Ordinary Linear Regression (OLR), providing a more accurate reflection of the regression relationships between variables. By investigating the spatiotemporal evolution of LST and its driving factors at the township scale, this study contributes insights for future urban planning and sustainable development.