Spatial Correlation Research Articles

Enhanced Air Quality Prediction through Spatio-temporal Feature Sxtraction and Fusion: A Self-tuning Hybrid Approach with GCN and GRU

Accurate prediction of air quality change is essential for air pollution control and human daily mobility. Due to the strong spatial and temporal correlation of air quality changes, existing air quality prediction methods often face the problem of low prediction accuracy due to insufficient extraction of spatio-temporal features. In this paper, we proposed a self-tuning spatio-temporal neural network (ST2NN) to enhance air quality prediction. ST2NN model consisted of four modules. First, ST2NN model constructed a temporal feature extraction module and a spatial feature extraction module based on gated recurrent unit (GRU) and graph convolutional neural network (GCN), respectively, and the two feature extraction modules adopted a parallel structure, which could effectively extract the spatio-temporal features in data. Additionally, ST2NN model constructed a feature fusion module based on gating mechanism to delineate the contribution of spatio-temporal features to the predicted values. Further, ST2NN model constructed a Hyperband hyperparameter optimization module based on Hyperband optimization algorithm to automatically adjust the network hyperparameters. The structure of ST2NN model endowed it with excellent spatio-temporal feature extraction and parameter adaptability. ST2NN model was evaluated and compared with existing models, including convolutional long short-term memory neural network (ConvLSTM), GRU, combined convolutional neural network and long short-term memory neural network (CNN-LSTM), and GCN-LSTM for air quality index (AQI) prediction using data from twelve monitoring stations in Beijing, China. Across all four evaluation indexes, ST2NN model outperformed the comparative models, improving prediction accuracy by 0.51%-10.18% (measured using R2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${R}^{2}$$\\end{document}). From the experimental results, it can be seen that ST2NN model constructed from the perspective of spatio-temporal feature extraction has better prediction performance compared with the existing air quality prediction model, which provides a new method for air quality prediction and has certain application value.

From efficiency to resilience: unraveling the dynamic coupling of land use economic efficiency and urban ecological resilience in Yellow River Basin

This study investigates the Dynamic Coupling between Land Use Economic Efficiency (LUEE) and Urban Ecological Resilience (UER) in the Yellow River Basin (YRB). This exploration is pivotal for elucidating the interaction mechanisms between economic growth and ecological governance. Furthermore, understanding this relationship is essential for fostering high-quality, sustainable urban development in the YRB. Utilizing panel data from 56 cities spanning 2003 to 2020, this study employed the coupling coordination degree (CCD) model, spatial correlation analysis, Kernel density estimation, convergence model, and Geodetector to systematically analyze the spatio-temporal distribution, dynamic trend, and determinants of the CCD between LUEE and UER in the YRB. The findings indicate that: (1) A general upward trend in both LUEE and UER, accompanied by a steady improvement in their CCD. (2) Significant spatial disparities in their CCD, with higher levels in the lower reaches. (3) Marked positive spatial autocorrelation, predominantly characterized by clusters where high (low) values are surrounded by high (low) values. (4) Regarding the impact of individual factors, government fiscal budget expenditure demonstrates the most robust explanatory power for the CCD within the YRB. Concerning the effects of two-factor interactions, the interplay between industrial structure upgrading and government fiscal budget expenditure emerges as the most significant determinant in influencing the CCD between LUEE and UER. This study enhances our comprehensive understanding of the interplay between economic and ecological systems. It offers scientific insights and strategic direction for harmonizing ecological governance with urban economic growth at both the regional and global scales.

Spatial correlation between in vivo imaging and immunohistochemical biomarkers: A methodological study

In this study, we present a method that enables voxel-by-voxel comparison of in vivo imaging to immunohistochemistry (IHC) biomarkers. As a proof of concept, we investigated the spatial correlation between dynamic contrast enhanced (DCE-)CT parameters and IHC biomarkers Ki-67 (proliferation), HIF-1α (hypoxia), and CD45 (immune cells). 54 whole-mount tumor slices of 15 laryngeal and hypopharyngeal carcinomas were immunohistochemically stained and digitized. Heatmaps of biomarker positivity were created and registered to DCE-CT parameter maps. The adiabatic approximation to the tissue homogeneity model was used to fit the following DCE parameters: Ktrans (transfer constant), Ve (extravascular and extracellular space), and Vi (intravascular space). Both IHC and DCE maps were downsampled to 4 × 4 × 3 mm[3] voxels. The mean values per tumor were used to calculate the between-subject correlations between parameters. For the within-subject (spatial) correlation, values of all voxels within a tumor were compared using the repeated measures correlation (rrm). No between-subject correlations were found between IHC biomarkers and DCE parameters, whereas we found multiple significant within-subject correlations: Ve and Ki-67 (rrm = -0.17, P < .001), Ve and HIF-1α (rrm = -0.12, P < .001), Ktrans and CD45 (rrm = 0.13, P < .001), Vi and CD45 (rrm = 0.16, P < .001), and Vi and Ki-67 (rrm = 0.08, P = .003). The strongest correlation was found between IHC biomarkers Ki-67 and HIF-1α (rrm = 0.35, P < .001). This study shows the technical feasibility of determining the 3 dimensional spatial correlation between histopathological biomarker heatmaps and in vivo imaging. It also shows that between-subject correlations do not reflect within-subject correlations of parameters.

Spatial network characteristics of carbon balance in urban agglomerations– a case study in Beijing-Tianjin-Hebei city agglomeration

The cities within city clusters have strong trade connections, resulting in spatial heterogeneity and correlation of carbon balances. Previous studies have not deeply explored the characteristics of the spatial association network (SAN) of carbon balance in urban agglomerations, particularly regarding network structure, node effects, and spatial-temporal inhomogeneities. This study investigated the spatial network characteristics of carbon balance in the Beijing-Tianjin-Hebei (BTH) region from 2000 to 2019, employing a modified gravity model and social network analysis (SNA). The findings revealed the following: 1) Carbon emissions increased by 106.42%, and carbon sinks increased by 31.06%, displaying spatial-temporal heterogeneity and forming a multi-level, multi-nodal SAN of carbon balance centered around Beijing. 2) The carbon balance was mainly influenced by spatial spillovers occurring at different nodes, typically moving from lower-tier to higher-tier nodes. 3) Cities assumed four roles: bidirectional spillover, net spillover, primary beneficiary, and agent, with these roles being dynamic. 4) The spatial correlation of carbon balance was primarily influenced by economic development (max 0.602), spatial distance (max 0.331), and per capita ecological land (max 0.445). This approach would refine carbon management policies and deepen the understanding of the SAN and its influencing factors, providing insights for optimizing carbon emission reduction management.

Development Strategy Based on Combination Typologies of Building Carbon Emissions and Urban Vibrancy—A Multi-Sourced Data-Driven Approach in Beijing, China

When confronting the dual challenges of rapid urbanization and climate change, although extensive research has investigated the factors influencing urban carbon emissions and the practical strategies regarding urban vibrancy, the unclear mutual nexus between them and the development strategy for collaborative optimization requires further in-depth analysis. This study explores the delicate balance between urban vibrancy and low-carbon sustainability within the confines of Beijing’s Fifth Ring Road. By integrating OpenStreetMap, land use, population, and buildings’ carbon emission data, we have developed a reproducible method to estimate total carbon emissions and emission intensity. Furthermore, we have introduced vibrancy index data to distinguish the vibrancy evaluation of residential and non-residential land and applied cross-combinational classification technology to dissect the spatial correlation between urban carbon emissions and urban vibrancy. The results reveal that the four combination typologies show more significant differences and regularity in residential land. Based on the discovery of spatial correlation, this study puts forward corresponding development strategy suggestions for each of these four typologies based on the geographical location and requirements of urban development policies. In conclusion, our study highlights the importance of integrating carbon emissions and urban vibrancy comprehensively in sustainable urban planning and proposes that various land use combinations need targeted development strategies to achieve this goal, which need to consider population, energy, service facilities, and other diverse aspects.

Structure-function interrelationships and associated neurotransmitter profiles in drug-naïve benign childhood epilepsy with central-temporal spikes patients.

To explore the interrelationships between structural and functional changes as well as the potential neurotransmitter profile alterations in drug-naïve benign childhood epilepsy with central-temporal spikes (BECTS) patients. Structural magnetic resonance imaging (sMRI) and resting-state functional MRI data from 20 drug-naïve BECTS patients and 33 healthy controls (HCs) were acquired. Parallel independent component analysis (P-ICA) was used to identify covarying components among gray matter volume (GMV) maps and fractional amplitude of low-frequency fluctuations (fALFF) maps. Furthermore, we explored the spatial correlations between GMV/fALFF changes derived from P-ICA and neurotransmitter maps in JuSpace toolbox. A significantly positive correlation (p < 0.001) was identified between one structural component (GMV_IC6) and one functional component (fALFF_IC4), which showed significant group differences between drug-naïve BECTS patients and HCs (GMV_IC6: p < 0.01; fALFF_IC4: p < 0.001). GMV_IC6 showed increased GMV in the frontal lobe, temporal lobe, thalamus, and precentral gyrus as well as fALFF_IC4 had enhanced fALFF in the cerebellum in drug-naïve BECTS patients compared to HCs. Moreover, significant correlations between GMV alterations in GMV_IC6 and the serotonin (5HT1a: p < 0.001; 5HT2a: p < 0.001), norepinephrine (NAT: p < 0.001) and glutamate systems (mGluR5: p < 0.001) as well as between fALFF alterations in fALFF_IC4 and the norepinephrine system (NAT: p < 0.001) were detected. The current findings suggest co-altered structural/functional components that reflect the correlation of language and motor networks as well as associated with the serotonergic, noradrenergic, and glutamatergic neurotransmitter systems. The relationship between anatomical brain structure and intrinsic neural activity was evaluated using a multimodal fusion analysis and neurotransmitters which might provide an important window into the multimodal neural and underlying molecular mechanisms of benign childhood epilepsy with central-temporal spikes. Structure-function relationships in drug-naïve benign childhood epilepsy with central-temporal spikes (BECTS) patients were explored. The interrelated structure-function components were found and correlated with the serotonin, norepinephrine, and glutamate systems. Co-altered structural/functional components reflect the correlation of language and motor networks and correlate with the specific neurotransmitter systems.

The Impact of Green Innovation on CO2 Emissions in China: Evidence from Spatial Regression Model

Using data from USPTO, CEADs, the Chinese Statistical Yearbook, and spatial econometric models for the panel of the 30 Chinese provinces for the period 2000∼2019, this study explores and compares the impact of green and non-green innovation on CO2 emissions. The findings reveal a significant spatial correlation in CO2 emissions across provinces, validating the appropriateness of spatial econometric models for this analysis. The results indicate that both green and non-green innovations significantly reduce CO2 emissions, with green innovation having a more pronounced effect despite its lower prevalence. Additionally, patents and their citations notably enhance environmental quality in neighboring provinces through indirect effects. The empirical results highlight that China should focus on: (1) spatial spillover effects of technological, especially green, innovation; (2) removing barriers to green patent invention and citation; (3) transitioning from an energy-driven to a green innovation-driven economy; and (4) enhancing environmental quality by mitigating CO2 emissions.

Investigating the impact of urban growth on land use using spatial autocorrelation methods in Konya/Türkiye.

Land use land cover (LULC) change, global environmental change, and sustainable change are frequently discussed topics in research at the moment. It is important to determine the historical LULC change process for effective environmental planning and the most appropriate use of land resources. This study analysed the spatial autocorrelation of the land use structure in Konya between 1990 and 2018. For this, Global and Local Moran's I indices based on land use data from 122 neighbourhoods and hot spot analysis (Getis-Ord Gi*) methods were applied to measure the spatial correlation of changes and to determine statistically significant hot and cold spatial clusters. According to the research results, the growth of urban areas has largely destroyed the most productive agricultural lands in the region. This change showed high spatial clustering both on an area and a proportional basis in the northern and southern parts of the city. On the other hand, the growth in the industrial area suppressed the pasture areas the most in the north-eastern region of the city, and this region showed high spatial clustering on both spatial and proportional scales.

A multiobjective optimization framework for site investigation program based on Bayesian approach and NSGA‐II

AbstractSite investigation provides essential geotechnical parameter information for analysis and design. However, three conflicting objectives, namely exploration effort, robustness and parameter uncertainty, pose a challenge to the development of an optimal site investigation program. In this study, a three objective optimization framework for the site investigation program is proposed based on the Bayesian approach and the non‐dominated sorting genetic algorithm (NSGA‐II). The only inputs required by the proposed framework are prior distribution of geotechnical parameters and error information. The prior distribution of geotechnical parameters is derived from integrating engineering experience and measurements from basic exploration boreholes. The error information is obtained based on literature and expert judgment related to the specific project. Firstly, a design pool of candidate investigation programs is generated using Bayesian approach to determine the locations and number of exploration boreholes. The NSGA‐II is then applied to identify the optimal program that balances lower cost, higher robustness, and lower uncertainty. The proposed multiobjective optimization framework is illustrated and validated through a real site investigation case in Chongqing, China, aimed at determining the ultimate bearing capacity of the rock foundation. The spatial correlation of parameters within the study area is also considered. The optimal program is represented by the location and number of exploration boreholes. By comparing measurements with predictions from different site investigation programs, the efficiency of the proposed multiobjective framework is demonstrated. Additionally, the influence of engineering experience and random field modeling on the investigation program is discussed.

Exploring the dynamics and trends of carbon emission spatiotemporal patterns in the Chengdu–Chongqing Economic Zone, China, from 2000 to 2020

Analysis of the spatial–temporal pattern and trend of carbon emissions provides an important scientific basis for the development of a low-carbon economy. Based on the corrected NPP-VIIRS and DMSP/OLS nighttime light data, a carbon emission model for the Chengdu–Chongqing Economic Zone (CCEZ) in China is constructed. Furthermore, the article establishes an integrated qualitative and quantitative research system. The qualitative results show that at the city and county scales, the high carbon emission areas and counties are mainly distributed in Chengdu and Chongqing, while the low carbon emission areas are concentrated in the marginal cities of the CCEZ and the counties with low levels of industrialization around the Sichuan Basin. The high-carbon emission zone tended to expand to the north, and the low-carbon emission zone tended to expand to the south. At the grid scale, the carbon emissions of the CCEZ fluctuated and increased from 2000 to 2020, forming a trend connected with those of the central city, with high carbon emissions at the core and radiating outward expansion. Quantitative analysis revealed that carbon emissions at the county and grid scales exhibited a significant positive global spatial correlation, and the overall correlation degree exhibited an increasing trend.

Analysis of the Environmental Benefits and Driving Forces of the Development of the “Production–Living–Ecological Space” Pattern Based on the ERI-ESV Geodetector

Based on five periods of Landsat remote sensing data from 1980 to 2020, this study constructs a landscape ecological risk-ecosystem service value evaluation model and integrates it with a geodetector model to analyse the environmental benefits of the development of the “production–living–ecological space” pattern and its driving factors in the Xuzhou planning area. The results of the study are as follows: (1) Over the past 40 years, the expansion of living spaces has significantly encroached upon adjacent agricultural production areas and ecological spaces, such as forests and grasslands. Specifically, the areas of agricultural land, forests, and grassland have been diminished by 277.39 km2, 23.8 km2 and 12.93 km2, respectively; in contrast, urban and rural living spaces have increased by 238.62 km2 and 58.92 km2, alongside a rise in industrial production areas, water bodies, and other ecological spaces. (2) Throughout the 40-year period, both the landscape ecological risk (ERI) and ecosystem service value (ESV) in the study area have shown a decreasing trend. The proportion of high- and medium-high-risk areas of the ERI have decreased by 5.19% and 7.50%, respectively, while low, lower, and medium ecological risk areas have increased by 6.40%, 3.22% and 3.07%, respectively. In addition, low-ESV areas have increased by 14.22%, while the proportion of high- and medium-high-ESV areas have decreased by 1.16%. (3) There is a significant positive spatial correlation between the ERI and ESV. Regions with dense ecological spaces comprising forests, water bodies, and grasslands, particularly in the northeastern part of the Jiawang District and the southeastern part of the Tongshan District, demonstrate superior regional ecosystem service quality. The ERI and ESV are dominated by “high–high” and “low–high” aggregation. Conversely, in the southwestern part of the study area, the expansion of living space has led to the transformation of some agricultural land, forest land, and grassland into less risky construction land, resulting in a decline in the quality of regional ecosystem services. The local spatial correlation between the ERI and ESV changed from “high–high”, “low–low”, “low–high” agglomeration to “low–low” agglomeration. (4) Key factors influencing the spatial differentiation of the “production–living–ecological space” include the GDP, population density, soil type, and the distance to towns and roads. Among these, the interaction between population density and soil type has the most significant effect on the changes in the pattern of the “production–living–ecological space”.

Can urbanization improve carbon performance?

Cities are the core carriers and key positions to achieve the dual carbon goals. It is of great significance to explore whether promoting urbanization can improve carbon emission performance, which is of great significance to comprehensively promote the goal of carbon neutrality. Based on the panel data from 2006 to 2021, this paper analyzes the spatial autocorrelation of carbon emission performance per unit space and the impact mechanism of urbanization process on it. The fixed-effect model was further used to identify the influencing factors of spatial carbon emission performance. The results show that: 1) China’s carbon emission performance per unit space is declining year by year. 2) There is a strong positive spatial correlation and stable path dependence on the performance of carbon emissions per unit space in each region. 3) To a certain extent, increasing the level of urbanization will reduce the carbon emission performance per unit space. 4) The urbanization process has a spatial spillover effect on the carbon emission performance per unit space of surrounding provinces, and the spatial spillover effect of industrial structure and energy consumption structure is more obvious than that of economic level, population density and urbanization rate. Based on the conclusions, this paper puts forward specific policy suggestions to reduce the carbon emission performance per unit space to help the low-carbon development of cities.

Deep causal inference for understanding the impact of meteorological variations on traffic

Understanding the causal impact of meteorological variations on traffic conditions (e.g., traffic flow and speed) is crucial for effective traffic prediction and management, as well as the mitigation of adverse weather effects on traffic. However, many existing studies focused on establishing associations between meteorological situations and traffic, rather than delving into causal relationships, especially with deep learning techniques. Consequently, the ability to identify specific meteorological conditions that significantly contribute to traffic congestion or delays is still limited. To address this issue, this study proposes the Meteorological-Traffic Causal Inference Variational Auto-Encoder Model (MT-CIVAE) to estimate the causal impact of fine-grained meteorological variations (e.g., rain and temperature) on traffic. Specifically, MT-CIVAE is based on the Variational Auto-Encoder and consists of an encoder to recover the distribution of latent confounders and a decoder to estimate the conditional probabilities of treatments. Transformer encoder layers are incorporated to analyze the spatial and temporal correlations of historical traffic data to further enhance the inference capability. To evaluate the effectiveness of the proposed approach for causal inference, real-world traffic flow and speed datasets collected from California, along with corresponding fine-grained meteorological datasets, are employed. The counterfactual analysis is conducted using artificially generated meteorological conditions as treatments, which allows for the simulation of hypothetical meteorological scenarios and the evaluation of their potential impact on traffic conditions. This study develops deep learning methods for assessing the causal impact of meteorological variations on traffic dynamics, offering explanations and insights that can assist transportation institutions in guiding post-meteorology traffic management strategies.

Spatiotemporal evolution of carbon emissions and influencing factors in county-level based on nighttime lighting data: a case study in Huaihai economic zone core city

ABSTRACT Driven by China’s carbon peaking and carbon neutrality goals, this study constructs a carbon emission inversion model for 57 county-level in the core city cluster of the Huaihai Economic Zone. It analyzes the spatial-temporal characteristics of carbon emissions in these counties between 2001 and 2021. It also explores the impacts of different socio-economic factors on carbon emissions by adopting geographically and temporally weighted regression methods. The results showed that: (1) From 2001 to 2021, the total carbon emissions of counties increased from 63.42 Mt to 279.89 Mt, and the distribution pattern of the region is “high in the middle and low in the east and west.” (2) Carbon emissions have significant positive spatial correlations, with the global Moran’s I index showing a wave-like downward trend from 0.233 to 0.0954, indicating that the spatial concentration of carbon emissions in different counties is gradually decreasing, and the differences show a trend of narrowing. (3) Significant spatial and temporal heterogeneity is among the influencing factors. It is mainly manifested in population size and industrial structure, while economic scale and technological progress are the dominant factors in carbon emissions. The research results will inform county-level low-carbon development.

A machine learning-based framework for spatio-temporal extension and filling of SMOS surface soil moisture observations over Canada

ABSTRACT Advancements in global satellite missions have revolutionized the assessment of Surface Soil Moisture (SSM) at global to local scales. However, spatio-temporal data discontinuities in specific regions remain a challenge. This study proposes a Machine Learning (ML)-based framework to extend the Soil Moisture Ocean Salinity (SMOS) SSM product both in the spatial and temporal domains, over Canada. In the first phase of the proposed framework, ML models based on Random Forest (RF) and Convolutional Neural Networks (CNN) are trained and validated with SMOS SSM as target and SSM-relevant climatic variables and geophysical variables, obtained from fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis data (ERA5), for the 2011–2020 period, as predictors. Developed models, when tested on unseen data for the years 2021–2022, suggest slightly better performance of the RF model compared to CNN, with root mean square error (RMSE) of 0.033 and 0.056 respectively; prediction biases mostly noted for regions with large inter-annual variability. The spatial filling of SSM for grid cells that were excluded during the training process, with similar land types as those in the SMOS training data, yields reasonable performance, with RF (RMSE = 0.013) performing better than CNN (RMSE = 0.064). In the second phase, the RF model is selected to extend the SMOS dataset for the 2008–2010 period. The temporal correlation between extended SMOS and ASCAT (Advanced Scatterometer) SSM demonstrates a reasonable association, with correlation coefficient exceeding 0.6. Additionally, spatial correlation analysis reveals similar patterns between the two datasets, with smaller values for the summer season owing to the importance of local processes on SSM during this period. However, extending SMOS SSM spatially for surface types that were not included in the training process such as peatlands, remains a challenge warranting additional studies. The developed framework is robust and can address spatio-temporal discontinuities in other SSM products.

Critical dynamics and interictal epileptiform discharge: a comparative analysis with respect to tracking seizure risk cycles

Epilepsy is characterized by recurrent, unprovoked seizures. Accurate prediction of seizure occurrence has long been a clinical goal since this would allow to optimize patient treatment, prevent injuries due to seizures, and alleviate the patient burden of unpredictability. Advances in implantable electroencephalographic (EEG) devices, allowing for long-term interictal EEG recordings, have facilitated major progress in this field. Recently, it has been discovered that interictal brain activity demonstrates circadian and multi-dien cycles that are strongly aligned, or phase locked, with seizure risk. Thus, cyclical brain activity patterns have been used to forecast seizures. However, in the effort to develop a clinically useful EEG based seizure forecasting system, challenges remain. Firstly, multiple EEG features demonstrate cyclical patterns, but it remains unclear which feature is best suited for predicting seizures. Secondly, the technology for long-term EEG recording is currently limited in both spatial and temporal sampling resolution. In this study, we compare five established EEG metrics:synchrony, spatial correlation, temporal correlation, signal variance which have been motivated from critical dynamics theory, and interictal epileptiform discharge (IED) which are a traditional marker of seizure propensity. We assess their effectiveness in detecting 24-h and seizure cycles as well as their robustness under spatial and temporal subsampling. Analyzing intracranial EEG data from 23 patients, we report that all examined features exhibit 24-h cycles. Spatial correlation, signal variance, and synchrony showed the highest phase locking with seizures, while IED rates were the lowest. Notably, spatial and temporal correlation were also found to be highly correlated to each other, as were signal variance and IED—suggesting some features may reflect similar aspects of cortical dynamics, whereas others provide complementary information. All features proved robust under subsampling, indicating that the dynamic properties of interictal activity evolve slowly and are not confined to specific brain regions. Our results may aid future translational research by assisting in design and testing of EEG based seizure forecasting systems.

SwinVNETR: Swin V-net Transformer with non-local block for volumetric MRI Brain Tumor Segmentation

Brain Tumor Segmentation (BTS) and classification are important and growing research fields. Magnetic resonance imaging (MRI) is commonly used in the diagnosis of brain tumours owing to its low radiation exposure and high image quality. One of the current subjects in the field of medical imaging is how to quickly and precisely segment MRI scans of brain tumours. Unfortunately, most existing brain tumour segmentation algorithms use inadequate 2D picture segmentation methods and fail to capture the spatial correlation between features. In this study, we propose a segmentation model (SwinVNETR) Swin V-NetTRansformer-based architecture with a non-local block. This model was trained using the Brain Tumor Segmentation Challenge BraTS 2021 dataset. The Dice similarity coefficients for the enhanced tumour (ET), whole tumour (WT), and tumour core (TC) are 0.84, 0.91, and 0.87, respectively. By leveraging this methodology, we can segment brain tumours more accurately than ever before. In conclusion, we present the findings of our model through the application of the Grad-CAM methodology, an eXplainable Artificial Intelligence (XAI) technique utilized to elucidate the insights derived from the model, which helped in better understanding; doctors can better diagnose and treat patients with brain tumours.

The promotion of sustainable land use planning for the enhancement of ecosystem service capacity: Based on the FLUS-INVEST-RUSLE-CASA model.

Land Use/Land Cover (LULC) is one of the most significant human variables influencing the efficiency of Ecosystem Services (ESs) in terrestrial ecosystems. Theoretical and technical assistance for regional sustainable land use planning and management, as well as ecosystem conservation and restoration, is provided by investigating the influence of changes in the LULC pattern on the efficiency of ESs. This research focuses on the interactions between socioeconomic activities and natural ecological processes in the Three Gorges Reservoir Area (TGRA). We use LULC data from the TGRA for the years 1990, 2000, 2010, and 2020. The study includes the analysis and calculation of the spatiotemporal evolution features of the current LULC pattern and the efficiency of ESs, including their spatiotemporal distribution. Considering the TGRA's national development orientation and guidance, three potential LULC patterns are constructed under various develop-ment scenarios. To calculate the efficiency of ESs, the GeoSOS-FLUS future LULC simulation model is linked, and several methodologies such as INVEST, RUSLE, and CASA are used. The goal is to investigate the influence of future changes in LULC patterns on ESs efficiency. The findings show the following: (1) From 1990 to 2020, the values of water conservation services in the TGRA decreased and subsequently increased. High-value areas are primarily located in the reservoir's centre and eastern sections, whereas low-value areas are mostly found in the western section. Soil conservation service values initially declined and later climbed. The TGRA's carbon storage services have in-creased yearly, from 552.64 g/m2 in 2000 to 615.92 g/m2 in 2020. (2) In the ecological protection scenario, carbon storage and soil erosion increased compared to the ecosystem services in 2020. The ecological system service benefits are greater when compared to the natural development scenario. (3) The four ESs show positive spatial correlations across all three scenarios, and local spatial au-tocorrelation analysis findings demonstrate that carbon storage, water yield, and habitat quality have comparable spatial distributions across all three scenarios. To some extent, high-value areas for water conservation, soil retention, carbon storage, and habitat quality overlap.

Spatio-temporal Differentiation of County Carbon Budget and Territorial Space Optimization Zoning Strategy in Gansu Province from the Perspective of Carbon Neutrality

Under the background of comprehensively practicing the overall system concept of the "living community" in the new era, incorporating the carbon neutral development goal into the territorial spatial planning and construction and establishing the territorial spatial pattern and optimization strategy in line with the actual development of Gansu Province are of great significance for promoting the comprehensive green low-carbon transformation and high-quality development of regional economy and society. Taking counties in Gansu Province as an example, based on the perspective of carbon neutrality research, the land use carbon budget of 87 counties in Gansu Province in 2010, 2015, and 2021 was calculated and analyzed. GIS spatial analysis and social network analysis were used to further explore their spatial differentiation characteristics and the overall characteristics of the carbon emission spatial correlation network. At last, combined with the main function zoning, the low-carbon oriented land space optimization zoning was carried out, and differentiated low-carbon development strategies were proposed. The results were as follows： ① Carbon emissions in Gansu Province showed an upward trend, but the increase rate decreased, showing a spatial distribution of "high in the central and eastern part of the country, low in the southwest." Construction land was the main carbon source. The carbon uptake showed a spatial distribution of "high in the south and low in the north, high in the west and low in the east." Woodlands were the main carbon sinks. The net carbon emissions showed an increasing trend, and approximately 58.62% of the counties in the province were in a carbon imbalance situation. ② In 2021, the spatial network of county carbon emissions was closely related, showing a "core-edge" pattern. The Chenguan District and Qilihe District were in the core position of the network and received more correlation relationships in the network. The network contacts in Longzhong area were frequent, followed by the contacts in Longdongnan area. ③ Based on carbon emissions, carbon sequestration, and ecological carrying capacity coefficients and using the results of spatial correlation of social networks as role positions, the province was divided into four carbon-neutral sub-districts. At the same time, superimposed analysis of the main function zoning, the county area of the province was reconstructed into seven territorial space zones, and the differentiated regional low-carbon optimization development strategy was proposed for each zone.

Water yield forecast in Shandong Province based on InVEST model

A better understanding of the effects of land use change and climate change on water yield is highly important for water resource planning and sustainable management. Land use changes in Shandong Province were analyzed over recent years. The PLUS Model was applied to forecast the land use layout for 2032, with adjustments made based on actual conditions and utilizing various CMIP6 scenarios for precipitation and actual evapotranspiration. The 2032 Water Yield was quantified using the InVEST Model at both grid and administrative scales, and its temporal-spatial characteristics and spatial correlation were analyzed. The findings revealed several key points: first, a decrease of 16.92% in arable land area, along with diminishing trends in grassland and unused land, and an increase of 33.92% in built-up land area, accompanied by growth in forest and water area. Second, the water yield decreases from southeast to northwest in Shandong Province, with significant inter-annual variations. Third, future predictions suggest that the annual water yield generally increases during flat water years, with offshore areas exhibiting greater water yield depth compared to inland areas. Rizhao is projected to have the highest water yield depth, while Binzhou and Dongying are expected to have the lowest depths, with a spatially positive correlation between them.