Geospatial Data Research Articles

Analysis of secondary trauma transfers within a Canadian regional trauma network: room for improvement?

This study examines secondary trauma transfers of critically injured patients to an adult regional trauma centre in a mixed urban-suburban setting, to examine if these could be avoided through the provision of prehospital critical care at the scene of injury. This is a cohort study of trauma activations at an adult regional trauma centre in Toronto, Canada, over a 5-year period. We included all secondary trauma transfers of patients who were either admitted to the ICU, had surgery within 4h of arrival or died within 48h of admission. Baseline demographics, injury data, geospatial data and interventions provided were extracted from the hospital's trauma registry. 659 cases met the inclusion criteria during the five-year study period. 364 (55%) patients underwent secondary transfer from non-trauma centres located in relatively close proximity of 80km or less. Within this group, patients had a median injury severity score of 22 (IQR 16-29) and the mortality was 17%. 188 (52%) received at least one critical care intervention at the sending facility prior to secondary transfer to the trauma centre. The most frequently performed interventions were emergency anesthesia and intubation (37%), blood transfusion (27%), and finger and/or tube thoracostomy (13%). A significant proportion of critically injured patients in our mixed urban-suburban trauma network are transferred from non-trauma hospitals in relatively close proximity to the trauma centre. Non-trauma hospitals frequently provide time-critical and life-saving interventions prior to secondary transfer. A prehospital critical care scene response for major trauma should be explored as an option to deliver critical care interventions at the scene, followed by direct transport to a trauma centre.

Geo-hydrological analysis and watershed delineation of the Sarayan River, Ganga Plain, India: integrating geospatial data for water resource management

Geo-hydrological analysis and watershed delineation of the Sarayan River, Ganga Plain, India: integrating geospatial data for water resource management

GeoAggregator: An Efficient Transformer Model for Geo-Spatial Tabular Data

Modeling geospatial tabular data with deep learning has become a promising alternative to traditional statistical and machine learning approaches. However, existing deep learning models often face challenges related to scalability and flexibility as datasets grow. To this end, this paper introduces GeoAggregator, an efficient and lightweight algorithm based on transformer architecture designed specifically for geospatial tabular data modeling. GeoAggregators explicitly account for spatial autocorrelation and spatial heterogeneity through Gaussian-biased local attention and global positional awareness. Additionally, we introduce a new attention mechanism that uses the Cartesian product to manage the size of the model while maintaining strong expressive power. We benchmark GeoAggregator against spatial statistical models, XGBoost, and several state-of-the-art geospatial deep learning methods using both synthetic and empirical geospatial datasets. The results demonstrate that GeoAggregators achieve the best or second-best performance compared to their competitors on nearly all datasets. GeoAggregator's efficiency is underscored by its reduced model size, making it both scalable and lightweight. Moreover, ablation experiments offer insights into the effectiveness of the Gaussian bias and Cartesian attention mechanism, providing recommendations for further optimizing the GeoAggregator's performance.

Dynamic Assessment of Population Exposure to Urban Flooding Considering Building Characteristics

Under intensifying climate change impacts, accurate quantification of population exposure to urban flooding has become an imperative component of risk mitigation strategies, particularly when considering the dynamic nature of human mobility patterns. Previous assessments relying on neighborhood block-scale population estimates derived from conventional census data have been constrained by significant spatial aggregation errors. This study presents methodological advancements through the integration of social sensing data analytics, enabling unprecedented spatial resolution at the building scale while capturing real-time population dynamics. We developed an agent-based simulation framework that incorporates (1) building-based urban environment, (2) hydrodynamic flood modeling outputs, and (3) empirically grounded human mobility patterns derived from multi-source geospatial big data. The implemented model systematically evaluates transient population exposure through spatiotemporal superposition analysis of flood characteristics and human occupancy patterns across different urban functional zones in Lishui City, China. Firstly, multi-source points of interest (POIs) data are aggregated to acquire activated time of buildings, and an urban environment system at the building scale is constructed. Then, with population, buildings, and roads as the agents, and population behavior rules, activity time of buildings, and road accessibility as constraints, an agent-based model in an urban flood scenario is designed to dynamically simulate the distribution of population. Finally, the population dynamics of urban flood exposure under a flood scenario with a 50-year return is simulated. We found that the traditional exposure assessment method at the block scale significantly overestimated the exposure, which is four times of our results based on building scale. The proposed method enables a clearer portrayal of the disaster occurrence process at the urban local level. This work, for the first time, incorporates multi-source social sensing data and the triadic relationship between human activities, time, and space in the disaster process into flood exposure assessment. The outcomes of this study can contribute to estimate the susceptibility to urban flooding and formulate emergency response plans.

A high spatial resolution suitability layers to support feasible power plant site selection in China

China is undergoing significant energy system transitions to meet carbon neutrality targets, which requires the rapid deployment of new power plants, driven by the need for large-scale renewable energy expansion and increasing electricity demand. Moreover, the policy context emphasizes the critical need for effective local land planning and integrated ecological conservation. However, current capacity expansion planning models primarily focus on provincial or regional scales and overlook key location- and technology-specific factors for feasible power plant site selection. To effectively address these challenges, we use a transparent and comprehensive assessment framework that supports high-resolution spatial analysis of power generation technologies in mainland China, named Geospatial Raster Input Data for Capacity Expansion Regional Feasibility in China (GRIDCERF-China). It provides suitability layers for 7 technologies in GeoTIFF format that can be easily integrated with various modeling tools. GRIDCERF-China fills a significant open-source data gap, offering valuable insights for regional feasible planning and decision-making while addressing diverse scientific objectives across future scenarios.

Enhancing Railway Safety is Leveraging Local and Global Information for Obstacle Detection

Enhancing railway safety by leveraging local and global information for obstacle detection using OpenCV plays a vital role in preventing accidents and ensuring efficient railway operations. Our approach utilizes computer vision techniques and real-time image processing to detect obstacles on railway tracks with high accuracy. After performing the experiment, we achieved improved detection accuracy, faster processing time, and minimized false alarms, making the system more reliable for real-world applications. This innovative approach integrates OpenCV with AI-driven predictive analysis and cloud-based monitoring, offering a scalable and cost-effective solution compared to conventional obstacle detection methods. Railway safety is a critical concern for transportation networks worldwide. With increasing rail traffic and growing concerns about accidents caused by obstacles on tracks, efficient detection and mitigation strategies are essential. This research paper explores the integration of local and global information for obstacle detection, leveraging advanced technologies such as artificial intelligence (AI), and geospatial data analytics. By combining real-time local sensor inputs with global datasets, railway safety can be significantly enhanced, reducing the risk of accidents and improving operational efficiency. The paper further discusses methodologies, case studies, results, and future work to create a comprehensive safety framework.

SatCLIP: Global, General-Purpose Location Embeddings with Satellite Imagery

Geographic information is essential for modeling tasks in fields ranging from ecology to epidemiology. However, extracting relevant location characteristics for a given task can be challenging, often requiring expensive data fusion or distillation from massive global imagery datasets. To address this challenge, we introduce Satellite Contrastive Location-Image Pretraining (SatCLIP). This global, general-purpose geographic location encoder learns an implicit representation of locations by matching CNN and ViT inferred visual patterns of openly available satellite imagery with their geographic coordinates. The resulting SatCLIP location encoder efficiently summarizes the characteristics of any given location for convenient use in downstream tasks. In our experiments, we use SatCLIP embeddings to improve performance on nine diverse geospatial prediction tasks including temperature prediction, animal recognition, and population density estimation. Across tasks, SatCLIP consistently outperforms alternative location encoders and shows promise for improving geographic domain adaptation. These results demonstrate the potential of vision-location models to learn meaningful representations of our planet from the vast, varied, and largely untapped modalities of geospatial data.

Evaluation of surface irrigation potential areas using geospatial techniques: the case of the Nashe watershed, western Ethiopia

ABSTRACT One of the key strategies for addressing the first and second sustainable development Goals (zero hunger and poverty) defined by the UN as being accomplished by 2030 is surface irrigation. This study aimed to determine potentially suitable areas for surface irrigation in the Nashe watershed by integrating geospatial techniques with the analytical hierarchy process method. The study used eight factors, including land use/land cover, soil type, soil depth, soil texture, soil drainage, slope, and distance from rivers and roads in the study area. Unlike conventional studies, this research leverages high-resolution geospatial data and employs a multifactorial assessment to improve the accuracy of irrigation suitability classification. The findings indicated that 18.9% is highly suitable for irrigation, whereas 12.2% is unsuitable. The moderately suitable area for irrigation accounted for a substantial amount 68.8% of the study area. The novelty of this study lies in its integration of GIS-based modeling with validation techniques, as the model's accuracy was assessed by superimposing the study area's preexisting irrigation plan, which predominantly falls within the highly and moderately suitable categories. This approach provides a reliable decision-support tool for policymakers and farmers, enhancing sustainable water resource management and agricultural planning in Ethiopia.

Clustering-Based Urban Driving Cycle Generation: A Data-Driven Approach for Traffic Analysis and Sustainable Mobility Applications in Ecuador

A representative urban driving cycle was developed for Quito, Ecuador, using the K-Means clustering method. From 64 samples and 188,713 geospatial and speed data points, a 2870 s driving cycle was constructed to capture real-world traffic characteristics. Key parameters include an average speed of 22.68 km/h, acceleration and deceleration rates of 0.55 m/s2 and −0.57 m/s2, and a dwell time of 9.66%. Due to Quito’s linear urban development, where mobility is limited to north–south/south–north corridors, the driving cycle reflects frequent accelerations and decelerations along congested arterial roads. A comparative analysis with international driving cycles revealed that Quito’s traffic follows a unique pattern shaped by its geographic constraints. The HK cycle in China showed the greatest similarities, although differences in instantaneous speeds highlight the need for localized models. While this study primarily focuses on methodological robustness, the developed driving cycle provides a foundational dataset for future research on traffic flow optimization, emissions estimation, and sustainable urban mobility strategies. These insights contribute to data-driven decision-making for improving transportation efficiency and environmental impact assessment in cities with similar urban structures.

Comprehensive Scale Fusion Networks with High Spatiotemporal Feature Correlation for Air Quality Prediction

The escalation of industrialization has worsened air quality, underscoring the essential need for accurate forecasting to inform policies and protect public health. Current research has primarily emphasized individual spatiotemporal features for prediction, neglecting the interconnections between these features. To address this, we proposed the generative Comprehensive Scale Spatiotemporal Fusion Air Quality Predictor (CSST-AQP). The novel dual-branch architecture combines multi-scale spatial correlation analysis with adaptive temporal modeling to capture the complex interactions in pollutant dispersion and enhanced pollution forecasting. Initially, a fusion preprocessing module based on localized high-correlation spatiotemporal features encodes multidimensional air quality indicators and geospatial data into unified spatiotemporal features. Then, the core architecture employs a dual-branch collaborative framework: a multi-scale spatial processing branch extracts features at varying granularities, and an adaptive temporal enhancement branch concurrently models local periodicities and global evolutionary trends. The feature fusion engine hierarchically integrates spatiotemporally relevant features at individual and regional scales while aggregating local spatiotemporal features from related sites. In experimental results across 14 Chinese regions, CSST-AQP achieves state-of-the-art performance compared to LSTM-based networks with RMSE 6.11–9.13 μg/m3 and R2 0.91–0.93, demonstrating highly robust 60 h forecasting capabilities for diverse pollutants.

Integrating Machine Learning and Geospatial Data for Mapping Socioeconomic Vulnerability to Urban Natural Hazard

Rapid urbanization and climate change are increasing the risks associated with natural hazards, especially in cities where socio-economic disparities are significant. Current hazard risk assessment frameworks fail to consider socio-economic factors, which limits their ability to effectively address vulnerabilities at the community level. This study introduces a machine learning framework designed to assess flood susceptibility and socio-economic vulnerability, particularly in urban areas with limited data. Using Kigali, Rwanda, as a case study, we quantified socio-economic vulnerability through a composite index that includes indicators of sensitivity and adaptive capacity. We utilized a variety of data sources, such as demographic, environmental, and remotely sensing datasets, applying machine learning algorithms like Multilayer Perceptron (MLP), Random Forest, Support Vector Machine (SVM), and XGBoost. Among these, MLP achieved the best predictive performance, with an AUC score of 0.902 and an F1-score of 0.86. The findings indicate spatial differences in socio-economic vulnerability, with central and southern Kigali showing greater vulnerability due to a mix of socio-economic challenges and high flood risk. The vulnerability maps created were validated against historical flood records, socio-economic research, and expert insights, confirming their accuracy and relevance for urban risk assessment. Additionally, we tested the framework’s scalability and adaptability in Kampala, Uganda, and Dar es Salaam, Tanzania, showing that making context-specific adjustments to the model improves its transferability. This study offers a solid, data-driven approach for combining assessments of flood susceptibility and socio-economic vulnerability, filling important gaps in urban resilience planning. The results support the advancement of risk-informed decision-making, especially in areas with limited access to detailed socio-economic information.

A machine learning approach for type 2 diabetes diagnosis and prognosis using tailored heterogeneous feature subsets.

Type 2 diabetes (T2D) is becoming one of the leading health problems in Western societies, diminishing quality of life and consuming a significant share of healthcare resources. This study presents machine learning models for T2D diagnosis and prognosis, developed using heterogeneous data from a Spanish population dataset (Di@bet.es study). The models were trained exclusively on individuals classified as controls and undiagnosed diabetics, ensuring that the results are not influenced by treatment effects or behavioral changes due to disease awareness. Two data domains are considered: environmental (patient lifestyle questionnaires and measurements) and clinical (biochemical and anthropometric measurements). The preprocessing pipeline consists of four key steps: geospatial data extraction, feature engineering, missing data imputation, and quasi-constancy filtering. Two working scenarios (Environmental and Healthcare) are defined based on the features used, and applied to two targets (diagnosis and prognosis), resulting in four distinct models. The feature subsets that best predict the target have been identified based on permutation importance and sequential backward selection, reducing the number of features and, consequently, the cost of predictions. In the Environmental scenario, models achieved an AUROC of 0.86 for diagnosis and 0.82 for prognosis. The Healthcare scenario performed better, with an AUROC of 0.96 for diagnosis and 0.88 for prognosis. A partial dependence analysis of the most relevant features is also presented. An online demo page showcasing the Environmental and Healthcare T2D prognosis models is available upon request.

The first urban open space product of global 169 megacities using remote sensing and geospatial data

Urban open space (UOS) plays an important environmental role, especially in areas characterized by intense social and economic activity. However, the high interclass similarities, complex surroundings, and scale variations of UOS lead to unsatisfactory UOS mapping performance, and UOS mapping products for major cities around the world are lacking. To fill this gap, we used a deep learning-based method based on a tiny-manual annotation strategy and optical remote sensing imagery to produce a 1.19 m resolution UOS map of 169 megacities, namely the OpenspaceGlobal product. We generated the OpenspaceGlobal product with five urban open space categories. To obtain the final OpenspaceGlobal product, we processed over 8.5 TB of remote sensing images and nearly 90 million polygons in crowdsourced geospatial data. The validation results showed that the OpenspaceGlobal product had an overall accuracy of 79.13 % and a kappa coefficient of 73.47 %. The OpenspaceGlobal product can promote a better understanding of human-made space surfaces in major cities around the world.

Moran's I and Geary's C: investigation of the effects of spatial weight matrices for assessing the distribution of infectious diseases.

The COVID-19 outbreak has precipitated severe occurrences on a global scale. Hence, spatial analysis is crucial in determining the relationships and patterns of geospatial data. Moran's I and Geary's C are prominent methodologies used to measure the spatial autocorrelation of geographical data. Both measure the degree of similarity or dissimilarity between nearby locations based on attribute values in such a way that the selection of distance techniques and weight matrices significantly impact the spatial autocorrelation results. This paper aimed at carrying out the spatial epidemiological characteristics analysis of the pandemic comparing the results of Moran's I and Geary's C with different parameters to gain a comprehensive understanding of the spatial relationship of COVID-19 cases. We employed distance-based techniques, K-nearest neighbour, and Queen contiguity techniques to assess the sensitivity of the different parameter configurations for both Moran's I and Geary's C. The findings revealed that former provided more reliable and robust results compared to the latter, with consistent results of spatial autocorrelation (positive spatial autocorrelation). The distance weight of 0.05 using the Manhattan method of Moran's I is the recommended distance weight, as it outperformed other weight matrices (Moran's I = 0.0152, Z-value= 110.8844 and p-value=0.001).

An open-access database of nature-based carbon offset project boundaries

Nature-based climate solutions (NBS) have become an important component of strategies aiming to reduce atmospheric CO2 and mitigate climate change impacts. Carbon offsets have emerged as one of the most widely implemented NBS strategies, however, these projects have also been criticized for exaggerating offsets. Verifying the efficacy of NBS-derived carbon offset is complicated by a lack of readily available geospatial boundary data. Herein, we detail methods and present a database of nature-based offset project boundaries. This database provides the locations of 575 NBS projects distributed across 55 countries. Geospatial boundaries were aggregated using a combination of scraping data from carbon project registries (n = 433, 75.3%) as well as manual georeferencing and digitization (n = 127, 22.1%). Database entries include three varieties of carbon projects: avoided deforestation, afforestation, reforestation and re-vegetation, and improved forest management. An accuracy assessment of the georeferencing and digitizing process indicated a high degree of accuracy (intersection over union score of 0.98 ± 0.015).

Analysis and Evaluation of Land Suitability in Kupang City, Indonesia for Sustainable Settlement Development

Kupang City, as the capital of Nusa Tenggara Timur Province, faces significant challenges in land management and settlement development due to rapid population growth and urbanization. This study aims to analyze and evaluate land suitability for sustainable settlement development, considering relevant environmental factors. The method used is Spatial Multi-Criteria Analysis (SMCA), which integrates geospatial data from the Digital Elevation Model (DEM), including slope, road networks, rivers, economic activity centers, and coastlines. The analysis results indicate that out of the total area analyzed, 7,227.04 hectares have high suitability, 4,341.12 hectares have moderate suitability, and 3,967.85 hectares have low suitability. Areas with high suitability are generally located near roads, economic activity centers, have lower slope gradients, and are far from rivers and coastlines, making them ideal for infrastructure development. The discussion emphasizes the importance of spatial and participatory-based spatial planning, as well as the need for monitoring low-suitability areas to prevent disaster risks. In conclusion, this study provides policy recommendations for more sustainable settlement development in Kupang City, integrating land suitability analysis and involving the community in the planning process, thereby creating

Automated Forest Health Monitoring and Optimal Harvest Prediction System for Sustainable Resource Management

Predicting the best times to harvest trees is crucial for managing forests sustainably, preventing illegal logging, and increasing financial gains. Manual surveys and static satellite imaging are the mainstays of traditional monitoring techniques, which have drawbacks such as inefficiency, geographic restrictions, and a delayed ability to identify ecological changes. These difficulties frequently lead to financial losses, ecological imbalances, and early or unlawful harvesting. This study suggests an AI-driven architecture [Figure.2], to automate forest health monitoring and improve harvest forecasts by combining high-resolution satellite imagery, geospatial data, and sophisticated machine learning algorithms. This study employs the CRISP-ML(Q) [Figure.1], methodology to develop a scalable framework for automated forest health monitoring and harvest prediction. By utilizing Google Earth Engine (GEE) APIs, the system collects multi-temporal and multi-spectral satellite imagery to enhance monitoring precision. Tree canopy segmentation is performed using polygon-based annotation techniques, while geospatial referencing of latitude-longitude coordinates ensures accurate mapping. The framework integrates Mask R-CNN [Figure.3], for tree detection and segmentation, estimating canopy diameters through pixel-to-meter ratio analysis. Additionally, LSTM networks are deployed to forecast tree growth patterns and determine optimal harvest times based on historical and real-time observations. To facilitate decision-making, an interactive web-based UI is designed to dynamically map tree locations, display predictive insights, and send real-time alerts to stakeholders. The dataset comprises high-resolution geotagged images annotated with precise growth metrics and enriched with vegetation indices such as NDVI and EVI, improving model reliability across diverse environments. By combining deep learning, geospatial analytics, and predictive modelling, this research establishes a data-driven, AI-powered framework for sustainable forestry management and biodiversity conservation.

Assessing the Supply–Demand Matching and Spatial Flow of Urban Cultural Ecosystem Services: Based on Geospatial Data and User Interaction Data

Cultural ecosystem services (CESs) reflect the interaction between ecosystems and human well-being. Owing to constraints in data availability and existing methodological limitations, deriving information from non-material ecosystem attributes was inadequate. We took Yulin City, located in the northern Shaanxi Loess Plateau, as a case study. Based on open-source geospatial data and user interaction data from social media, a coupled multi-source model was applied to elucidate the spatial distribution of CESs’ supply–demand flow. The Maxent and LDA model were utilized to quantify CES supply–demand, whereas the breakpoint and gravity model were applied to explain the direction and intensity of CES flow. The results indicated the following: (1) aesthetic was the most perceivable CES in Yulin, with 27% high supply areas and four demand topics. And the perception of the educational CES was the least pronounced, with only 2% of high supply areas and two demand topics. (2) Yulin exhibited a notable mismatching in CES supply–demand, with the supply–demand matching area constituting only approximately 10%. In the center of the city, CESs displayed a spatial pattern of a supply–demand deficit, while areas farther from the city center presented a spatial pattern of a supply–demand surplus. (3) The flow of CESs followed a pattern of movement from peripheral counties to central counties and from less developed counties to more developed counties. We proposed the following targeted recommendations: introducing low-perception CESs to promote the enhancement of ecosystem services (ESs); and alleviating CES supply–demand mismatches by enhancing transportation accessibility and protecting the ecological environment. Simultaneously, attention should be directed towards the developmental disparities between counties, providing differentiated guidance for CES spatial flow. Our study provided a theoretical foundation for understanding CES supply–demand flow and offered scientific insights for the spatial development of urban CES.

Bathymetry and morphometric analysis of Greek natural lakes through a hybrid GIS-acoustic methodology

ABSTRACT Homogeneous and up-to-date bathymetric information on lakes is essential to improve limnological knowledge and produce evidence-based lake monitoring, assessment methods, and river basin management plans according to Water Framework Directive (WFD). The present study focused on the production of updated bathymetric information on the main natural Greek lakes and their morphometric analysis. The aim was to address the knowledge gap by providing updated homogeneous and accurate bathymetric information for the main freshwater natural lakes of Greece as part of the National Monitoring Network of Lakes in the context of WFD. We surveyed 14 freshwater lakes in mainland Greece using a dual frequency echosounder. Using the acoustic data collected during the surveys, bathymetric digital terrain models (DTMs) were produced for each lake and used to estimate morphometric characteristics (mean/maximum depth, volume, lake depth ratio, volume development, mean lake diameter, relative depth ratio, shoreline development factor, hypsographic curve). This analysis demonstrated the potential of the morphometric approach to generate valuable information using up-to-date bathymetric DTMs. The geospatial data derived by the present study are available through an online repository, the Greek National Monitoring Network of Lakes (http://ekbygis.biodiversity-info.gr/geoserver/web/). These data provide a valuable database for the advancement of limnological knowledge, the implementation of comprehensive monitoring programmes, and informed decision making in water resources management.

Integrating geospatial data and analytic hierarchy process for flood-prone zones mapping in the Upper Draa basin, Morocco

Integrating geospatial data and analytic hierarchy process for flood-prone zones mapping in the Upper Draa basin, Morocco