Search Results (25)

Under the combined influence of climate change and human activities, the non-stationarity of reservoir runoff has significantly intensified, posing challenges for traditional statistical models to accurately capture its multi-scale abrupt changes. This study focuses on Qianping (QP) Reservoir and systematically integrates climate-driven mechanisms with machine learning approaches to uncover the patterns of runoff evolution and develop high-precision prediction models. The findings offer a novel paradigm for adaptive reservoir operation under non-stationary conditions. In this paper, we employ methods including extreme-point symmetric mode decomposition (ESMD), Bayesian ensemble time series decomposition (BETS), and cross-wavelet transform (XWT) to investigate the variation trends and mutation features of the annual runoff in QP Reservoir. Additionally, four models—ARIMA, LSTM, LSTM-RF, and LSTM-CNN—are utilized for runoff prediction and analysis. The results indicate that: (1) the annual runoff of QP Reservoir exhibits a quasi-8.25-year mid-short-term cycle and a quasi-13.20-year long-term cycle on an annual scale; (2) by using Bayesian estimators based on abrupt change year detection and trend variation algorithms, an abrupt change point with a probability of 79.1% was identified in 1985, with a confidence interval spanning 1984 to 1986; (3) cross-wavelet analysis indicates that the periodic associations between the annual runoff of QP Reservoir and climate-driving factors exhibit spatiotemporal heterogeneity: the AMO, AO, and PNA show multi-scale synergistic interactions; the DMI and ENSO display only phase-specific weak coupling; while solar sunspot activity modulates runoff over long-term cycles; and (4) The NSE of the ARIMA, LSTM, LSTM-RF, and LSTM-CNN models all exceed 0.945, the RMSE is below 0.477 × 10⁹ m³, and the MAE is below 0.297 × 10⁹ m³, Among them, the LSTM-RF model demonstrated the highest accuracy and the most stable predicted fluctuations, indicating that future annual runoff will continue to fluctuate but with a decreasing amplitude. Full article

The southern slope of the Tianshan Mountains features complex terrain and an arid climate, yet paradoxically experiences frequent extreme precipitation events (EPEs), which pose significant challenges for weather forecasting. This study investigates an EPE that occurred from 20 to 21 August 2019 using multi-source data to examine circulation patterns, mesoscale characteristics, moisture dynamics, and energy-instability mechanisms. The results reveal distinct spatiotemporal variability in precipitation, prompting a two-stage analytical framework: stage 1 (western plains), dominated by localized convective cells, and stage 2 (northeastern mountains), characterized by orographically enhanced precipitation clusters. The event was associated with a “two ridges and one trough” circulation pattern at 500 hPa and a dual-core structure of the South Asian high at 200 hPa. Dynamic forcing stemmed from cyclonic convergence, vertical wind shear, low-level convergence lines, water vapor (WV) transport, and jet-induced upper-level divergence. A stronger vorticity, divergence, and vertical velocity in stage 1 resulted in more intense precipitation. The thermodynamic analysis showed enhanced low-level cold advection in the plains before the event. Sounding data revealed increases in precipitable water and convective available potential energy (CAPE) in both stages. WV tracing showed vertical differences in moisture sources: at 3000 m, ~70% originated from Central Asia via the Caspian and Black Seas; at 5000 m, source and path differences emerged between stages. In stage 1, specific humidity along each vapor track was higher than in stage 2 during the EPE, with a 12 h pre-event enhancement. Both stages featured rapid convective cloud growth, with decreases in total black body temperature (TBB) associated with precipitation intensification. During stage 1, the EPE center aligned with a large TBB gradient at the edge of a cold cloud zone, where vigorous convection occurred. In contrast to typical northern events, which are linked to colder cloud tops and vigorous convection, the afternoon EPE in stage 2 formed near cloud edges with lesser negative TBB values. These findings advance the understanding of multi-scale extreme precipitation mechanisms in arid mountains, aiding improved forecasting in complex terrains. Full article

Investigating spatio-temporal differentiation patterns of land-use conflicts in mountainous and flatland regions provides critical insights for optimizing spatial regulation strategies and advancing sustainable regional development. Using the Urban Agglomeration in Central Yunnan (UACY) as a case study, the production–living–ecological space (PLES) was classified through land-use functional dominance analysis based on 2010–2020 geospatial datasets. Spatio-temporal evolution patterns and mountain–dam differentiation were analyzed using spatial superposition, dynamic degree analysis, transfer matrices, and geospatial TuPu methods. A multi-scale conflict index incorporating landscape metrics was developed to assess PLES conflict intensities across spatial scales, with contribution indices identifying key conflict-prone spatial types. Analysis revealed distinct regional differentiation in PLES distribution and evolutionary trajectories during 2010–2020. Forest Ecological Space (FES) and Agricultural Production Space (APS) dominated both the entire study area and mountainous zones, with APS exhibiting particular dominance in dam regions. Grassland Ecological Space (GES) and Other Ecological Space (OES) experienced rapid conversion rates, contrasting with stable or gradual expansion trends in other space types. Change intensity was significantly greater in mountainous zones compared to flatland area (FA). PLES conflict exhibited marked spatial heterogeneity. FA demonstrated substantially higher conflict levels than mountainous zones, with evident scale-dependent variations. Maximum conflict intensity occurred at the 4000 m scale, with all spatial scales demonstrating consistent escalation trends during the study period. ULS, FES, and WES predominantly occurred in low-conflict zones characterized by stability, whereas APS, Industrial and Mining Production Space (IMPS), RLS, GES, and OES were primarily associated with high-conflict areas, constituting principal conflict sources. Full article

Under the rapid urbanization process, PM_2.5 pollution has become an increasingly critical issue. Changes in land-use types will inevitably affect PM_2.5 concentration. Meanwhile, the problem of imbalance and inadequacy of regional development remains prominent. This study took the Huaihe River Ecological Economic Belt as the research object, integrating the spatial econometric model with the Geographically and Temporally Weighted Regression (GTWR) and Multiscale Geographically Weighted Regression (MGWR) models, to investigate the spatiotemporal heterogeneity and spillover effect of the association between PM_2.5 concentration and land use from 1998 to 2021. The main findings are as follows: (1) PM_2.5 concentration in the study area from 1998 to 2021 showed an upward and then a downward trend, taking 2013 as a turning point, with respective magnitudes of 50.4% and 42.1%; (2) land use exerts a significant spillover effect on PM_2.5 pollution. Except for grassland and cropland, the direct effect of each land type on PM_2.5 pollution exceeds its indirect effect; (3) the influence of land use on PM_2.5 pollution exhibits significant spatiotemporal variations. The impact coefficient of forests remains relatively consistent across the entire region, whereas that of cropland, water bodies, and impervious surfaces varies markedly across different regions, particularly in the northeastern and southern cities of the study area. The results of this study may give new ideas for collective governance and joint environmental remediation in different cities and probably provide some basis for the formulation of air pollution control policies and urban land planning. Full article

There is a close correlation between brain aging and age. However, traditional neural networks cannot fully capture the potential correlation between age and brain aging due to the limited receptive field. Furthermore, they are more concerned with deep spatial semantics, ignoring the fact that effective temporal information can enrich the representation of low-level semantics. To address these limitations, a local attention spatio-temporal graph inference network (LSTGINet) was developed to explore the details of the association between age and brain aging, taking into account both spatio-temporal and temporal perspectives. First, multi-scale temporal and spatial branches are used to increase the receptive field and model the age information simultaneously, achieving the perception of static correlation. Second, these spatio-temporal feature graphs are reconstructed, and large topographies are constructed. The graph inference node aggregation and transfer functions fully capture the hidden dynamic correlation between brain aging and age. A new local attention module is embedded in the graph inference component to enrich the global context semantics, establish dependencies and interactivity between different spatio-temporal features, and balance the differences in the spatio-temporal distribution of different semantics. We use a newly designed weighted loss function to supervise the learning of the entire prediction framework to strengthen the inference process of spatio-temporal correlation. The final experimental results show that the MAE on baseline datasets such as CamCAN and NKI are 6.33 and 6.28, respectively, better than the current state-of-the-art age prediction methods, and provides a basis for assessing the state of brain aging in adults. Full article

The expansion of road networks profoundly affects ecological systems by intensifying habitat fragmentation, altering hydrological processes, and exacerbating pollution. However, our understanding of the multi-scale spatiotemporal coupling between road networks and ecological quality remains limited. Thus, taking Fuzhou City in Southeastern China as a case study (~12,000 km²), we apply bivariate spatial autocorrelation, geographical detectors (GDs), and multi-scale geographically weighted regression (MGWR) to explore the multi-scale interactions between road networks and ecological quality. Results reveal the following: (1) From 2016 to 2021, kernel density estimation (KDE) analysis of the road network indicates coordinated growth in both urban and rural areas, with an increase of 0.759 km/km². Analysis based on the remote sensing-based ecological index (RSEI) shows a decrease from 2000 to 2016, and then an increase from 2016 to 2021, with a trend of increasing gradually from urban center to rural area. (2) Predominant tradeoff relationships exist between KDE and RSEI in 2016 and 2021, while notable synergistic relationships emerge between ΔKDE and ΔRSEI. (3) Multi-scale GD analysis identifies ΔKDE as a principal factor influencing ΔRSEI, and the MGWR reveals their significant synergistic associations at an optimal scale of 3000 m. These findings highlight the unequal impact of road network expansion on ecological quality, underscoring the pivotal role of road density changes in its spatiotemporal dynamics. They offer essential insights for sustainable transport and ecological planning. Full article

The Geiger-Mode Avalanche Photodiode (Gm-APD) LiDAR system demonstrates high-precision detection capabilities over long distances. However, the detection of occluded small objects at long distances poses significant challenges, limiting its practical application. To address this issue, we propose a multi-scale spatio-temporal object detection network (MSTOD-Net), designed to associate object information across different spatio-temporal scales for the effective detection of occluded small objects. Specifically, in the encoding stage, a dual-channel feature fusion framework is employed to process range and intensity images from consecutive time frames, facilitating the detection of occluded objects. Considering the significant differences between range and intensity images, a multi-scale context-aware (MSCA) module and a feature fusion (FF) module are incorporated to enable efficient cross-scale feature interaction and enhance small object detection. Additionally, an edge perception (EDGP) module is integrated into the network’s shallow layers to refine the edge details and enhance the information in unoccluded regions. In the decoding stage, feature maps from the encoder are upsampled and combined with multi-level fused features, and four prediction heads are employed to decode the object categories, confidence, widths and heights, and displacement offsets. The experimental results demonstrate that the MSTOD-Net achieves mAP₅₀ and mAR₅₀ scores of 96.4% and 96.9%, respectively, outperforming the state-of-the-art methods. Full article

Environmental particulate pollution is a major global environmental health risk factor, which is associated with numerous adverse health outcomes, negatively impacting public health in many countries, including China. Despite the implementation of strict air quality management policies in China and a significant reduction in PM_2.5 concentrations in recent years, the health burden caused by PM_2.5 pollution has not decreased as expected. Therefore, a comprehensive analysis of the health burden caused by PM_2.5 is necessary for more effective air quality management. This study makes an innovative contribution by integrating the Enhanced Vegetation Index (EVI), Normalized Difference Vegetation Index (NDVI), and Soil-Adjusted Vegetation Index (SAVI), providing a comprehensive framework to assess the health impacts of green space coverage, promoting healthy urban environments and sustainable development. Using Nanjing, China, as a case study, we constructed a health impact assessment system based on PM_2.5 concentrations and quantitatively analyzed the spatiotemporal evolution of premature deaths caused by PM_2.5 from 2000 to 2020. Using Multiscale Geographically Weighted Regression (MGWR), we explored the impact of greening improvement on premature deaths attributed to PM_2.5 and proposed relevant sustainable governance strategies. The results showed that (1) premature deaths caused by PM_2.5 in Nanjing could be divided into two stages: 2000–2015 and 2015–2020. During the second stage, deaths due to respiratory and cardiovascular diseases decreased by 3105 and 1714, respectively. (2) The spatial variation process was slow, with the overall evolution direction predominantly from the southeast to northwest, and the spatial distribution center gradually shifted southward. On a global scale, the Moran’s I index increased from 0.247251 and 0.240792 in 2000 to 0.472201 and 0.468193 in 2020. The hotspot analysis revealed that high–high correlations slowly gathered toward central Nanjing, while the proportion of cold spots increased. (3) The MGWR results indicated a significant negative correlation between changes in green spaces and PM_2.5-related premature deaths, especially in densely vegetated areas. This study comprehensively considered the spatiotemporal changes in PM_2.5-related premature deaths and examined the health benefits of green space improvement, providing valuable references for promoting healthy and sustainable urban environmental governance and air quality management. Full article

Regarding Alzheimer’s disease (AD), specific neuronal populations and brain regions exhibit selective vulnerability. Understanding the basis of this selective neuronal and regional vulnerability is essential to elucidate the molecular mechanisms underlying AD pathology. However, progress in this area is currently hindered by the incomplete understanding of the intricate functional and spatial diversity of neuronal subtypes in the human brain. Previous studies have demonstrated that neuronal subpopulations with high neuropeptide (NP) co-expression are disproportionately absent in the entorhinal cortex of AD brains at the single-cell level, and there is a significant decline in hippocampal NP expression in naturally aging human brains. Given the role of NPs in neuroprotection and the maintenance of microenvironments, we hypothesize that neurons expressing higher levels of NPs (HNP neurons) possess unique functional characteristics that predispose them to cellular abnormalities, which can manifest as degeneration in AD with aging. To test this hypothesis, multiscale and spatiotemporal transcriptome data from ~1900 human brain samples were analyzed using publicly available datasets. The results indicate that HNP neurons experienced greater metabolic burden and were more prone to protein misfolding. The observed decrease in neuronal abundance during stages associated with a higher risk of AD, coupled with the age-related decline in the expression of AD-associated neuropeptides (ADNPs), provides temporal evidence supporting the role of NPs in the progression of AD. Additionally, the localization of ADNP-producing HNP neurons in AD-associated brain regions provides neuroanatomical support for the concept that cellular/neuronal composition is a key factor in regional AD vulnerability. This study offers novel insights into the molecular and cellular basis of selective neuronal and regional vulnerability to AD in human brains. Full article

Moving traces are essential data for target detection and associated behavior recognition. Previous studies have used time–location sequences, route maps, or tracking videos to establish mathematical recognition models for behavior recognition. The multimodal approach has seldom been considered because of the limited modality of sensing data. With the rapid development of natural language processing and computer vision, the multimodal model has become a possible choice to process multisource data. In this study, we have proposed a mathematical model for aircraft behavior recognition with joint data manners. The feature abstraction, cross-modal fusion, and classification layers are included in the proposed model for obtaining multiscale features and analyzing multimanner information. Attention has been placed on providing self- and cross-relation assessments on the spatiotemporal and geographic data related to a moving object. We have adopted both a feedforward network and a softmax function to form the classifier. Moreover, we have enabled a modality-increasing phase, combining longitude and latitude sequences with related geographic maps to avoid monotonous data. We have collected an aircraft trajectory dataset of longitude and latitude sequences for experimental validation. We have demonstrated the excellent behavior recognition performance of the proposed model joint with the modality-increasing phase. As a result, our proposed methodology reached the highest accuracy of 95.8% among all the adopted methods, demonstrating the effectiveness and feasibility of trajectory-based behavior recognition. Full article

In the era of big data, a significant volume of spatiotemporal data exists in a multiscale format, describing diverse phenomena in the objective world across different spatial and temporal scales. While existing methods focus on analyzing the features and connections of spatiotemporal data at various scales, they often overlook the consideration of uncertainty in spatiotemporal information within the context of multiscale meaning. To effectively harness the potential of spatiotemporal data, it becomes crucial to capture the fuzzy spatiotemporal information inherent in multiscale datasets. This paper proposes a novel multiscale spatiotemporal correlation method that accounts for and quantifies the uncertainty of spatiotemporal information. Spatiotemporal information is categorized into two types, explicit information and implicit information, based on respective levels of uncertainty. The method employs spatiotemporal cubes to interpret the spatiotemporal items within the data, followed by the introduction of a benchmark scale to determine the certainty of each spatiotemporal item based on its range and topological relationships. Subsequently, spatiotemporal confidence and correlation index are proposed to gauge the significance of geographical elements and their interrelationships. To validate the proposed method, a multiscale spatiotemporal transaction dataset is generated and utilized in the experiment. The experimental results demonstrate that the proposed method effectively captures spatiotemporal implicit information and enables better utilization of multiscale spatiotemporal data. Notably, the importance of each object of study varies when analyzed using different benchmark scales, providing valuable insights for professionals to identify novel objects and associations worthy of consideration. The obtained results can be used to construct spatiotemporal knowledge graphs. Full article

The association between rural public services (RPSs) and population migration (PM) has become a key aspect of rapid urbanization in developing countries and an important breakthrough for improving rural–urban relations. An in-depth analysis of the heterogeneity of the weakening effect of RPSs on PM at different transformation phases and the internal mechanism of the evolution of association patterns driven by RPSs and PM helps to ensure better co-ordinated urban and rural development. This paper establishes an interactive analysis framework for measuring the spatiotemporal association and regional differences between RPSs and PM in the Sichuan–Chongqing region (SCR), and reveals the influence mechanism by employing multiscale geographically weighted regression (MGWR). The results indicate that the association rapidly increased with clear spatial heterogeneity across topographic units and the weakening effect of RPSs on PM begin to diverge during the urban–rural transition. The natural, economic, social, and urban–rural disparity factors in terms of the association exhibit significant spatial variability. In mountainous areas, where topography dominates, RPSs fail to effectively weaken rural migration. However, in the plain areas, urbanization is the main driver of urban–rural transition, and the adaptive upgrading and transformation of RPSs has made their weakening effect stronger, thus alleviating rural exodus and increasing population concentration. All these findings show that differentiated optimization strategies adhering to the association trends should be proposed for a deeper integration of rural revitalization and new urbanization in the SCR. Full article

Spatio-temporal association analysis has attracted attention in various fields, such as urban computing and crime analysis. The proliferation of positioning technology and location-based services has facilitated the expansion of association analysis across spatio-temporal scales. However, existing methods inadequately consider the scale differences among spatio-temporal objects during analysis, leading to suboptimal precision in association analysis results. To remedy this issue, we propose a multiscale spatio-temporal object representation method, STO2Vec, for association analysis. This method comprises of two parts: graph construction and embedding. For graph construction, we introduce an adaptive hierarchical discretization method to distinguish the varying scales of local features. Then, we merge the embedding method for spatio-temporal objects with that for discrete units, establishing a heterogeneous graph. For embedding, to enhance embedding quality for homogeneous and heterogeneous data, we use biased sampling and unsupervised models to capture the association strengths between spatio-temporal objects. Empirical results using real-world open-source datasets show that STO2Vec outperforms other models, improving accuracy by 16.25% on average across diverse applications. Further case studies indicate STO2Vec effectively detects association relationships between spatio-temporal objects in a range of scenarios and is applicable to tasks such as moving object behavior pattern mining and trajectory semantic annotation. Full article

Understanding the characteristics of PM2.5 and its socioeconomic factors is crucial for managing air pollution. Research on the socioeconomic influences of PM2.5 has yielded several results. However, the spatial heterogeneity of the effect of various socioeconomic factors on PM2.5 at different scales has yet to be studied. This paper collated PM2.5 data for 359 cities in China from 2005 to 2020, as well as socioeconomic data: GDP per capita (GDPP), secondary industry proportion (SIP), number of industrial enterprise units above the scale (NOIE), general public budget revenue as a proportion of GDP (PBR), and population density (PD). The spatial autocorrelation and multiscale geographically weighted regression (MGWR) model was used to analyze the spatiotemporal heterogeneity of PM2.5 and explore the impact of different scales of economic factors. Results show that the overall economic level was developing well, with a spatial distribution trend of high in the east and low in the west. With a large positive spatial correlation and a highly concentrated clustering pattern, the PM2.5 concentration declined in 2020. Secondly, the OLS model’s statistical results were skewed and unable to shed light on the association between economic factors and PM2.5. Predictions from the GWR and MGWR models may be more precise than those from the OLS model. The scales of the effect were produced by the MGWR model’s variable bandwidth and regression coefficient. In particular, the MGWR model’s regression coefficient and variable bandwidth allowed it to account for the scale influence of economic factors; it had the highest adjusted R² values, smallest AICc values, and residual sums of squares. Lastly, the PBR had a clear negative impact on PM2.5, whereas the negative impact of GDPP was weak and positively correlated in some western regions, such as Gansu and Qinghai provinces. The SIP, NOIE, and PD were positively correlated with PM2.5 in most regions. Our findings can serve as a theoretical foundation for researching the associations between PM2.5 and socioeconomic variables, and for encouraging the coequal growth of the economy and the environment. Full article

The COVID-19 pandemic has posed numerous challenges to human society. Previous studies explored multiple factors in virus transmission. Yet, their impacts on COVID-19 are not universal and vary across geographical regions. In this study, we thoroughly quantified the spatiotemporal associations of 49 health, socioeconomic, demographic, and environmental factors with COVID-19 at the county level in Arkansas, US. To identify the associations, we applied the ordinary least squares (OLS) linear regression, spatial lag model (SLM), spatial error model (SEM), and multiscale geographically weighted regression (MGWR) model. To reveal how such associations change across different COVID-19 times, we conducted the analyses for each season (i.e., spring, summer, fall, and winter) from 2020 to 2021. We demonstrate that there are different driving factors along with different COVID-19 variants, and their magnitudes change spatiotemporally. However, our results identify that adult obesity has a positive association with the COVID-19 incidence rate over entire Arkansas, thus confirming that people with obesity are vulnerable to COVID-19. Humidity consistently negatively affects COVID-19 across all seasons, denoting that increasing humidity could reduce the risk of COVID-19 infection. In addition, diabetes shows roles in the spread of both early COVID-19 variants and Delta, while humidity plays roles in the spread of Delta and Omicron. Our study highlights the complexity of how multifactor affect COVID-19 in different seasons and counties in Arkansas. These findings are useful for informing local health planning (e.g., vaccine rollout, mask regulation, and testing/tracing) for the residents in Arkansas. Full article