IUTF Dataset(Enhanced): Enabling Cross-Border Resource for Analysing the Impact of Rainfall on Urban Transportation Systems

IUTF Dataset: A Cross-Border Resource for Analysing the Impact of Rainfall on Urban Transportation SystemsDescriptionThe Integrated Urban Traffic-Flood (IUTF) dataset is a comprehensive, open-access resource that addresses a critical gap in urban transportation and climate resilience research. This dataset uniquely integrates high-resolution traffic measurements, detailed precipitation data, and road network topology across 40 major cities spanning Europe, North America, and Asia.Dataset OverviewUnderstanding the impact of extreme weather, particularly rainfall, on urban transportation systems is essential for enhancing city resilience and traffic management. However, research has been hampered by the lack of datasets that comprehensively integrate detailed traffic dynamics, high-resolution weather information, and road network topology across multiple diverse urban environments.The IUTF dataset provides:<b>Traffic Data</b>: High-resolution measurements from 23627 sensors across 40 cities, with raw data at 5-minute intervals harmonized to hourly resolution<b>Precipitation Data</b>: Detailed hourly precipitation information from ERA5 reanalysis, spatially aligned with transportation infrastructure<b>Road Network Data</b>: Comprehensive topology information for over 1 million road segments processed from OpenStreetMap<b>Temporal Coverage</b>: Three years of data (2015-2017) providing 411,631 temporal observations per sensor<b>Spatial Coverage</b>: 40 diverse global cities across multiple continents and climate zonesKey Features<b>Unprecedented Scale and Integration</b>: The dataset covers 40 major cities with 23627 traffic sensors, providing the largest integrated traffic-weather dataset publicly available for urban resilience research.<b>Rigorous Harmonization Methodology</b>: All data components undergo comprehensive spatio-temporal harmonization using a novel framework that ensures consistent geographical representation and temporal alignment across diverse urban contexts.<b>Multi-Resolution Data Access</b>: Traffic measurements are provided in both original 5-minute resolution and hourly aggregations, enabling analyses at different temporal scales while maintaining alignment with meteorological data.<b>Cross-Continental Diversity</b>: Cities span multiple continents, climate zones, and urban morphologies, enabling robust comparative studies and generalization of findings across different urban contexts.<b>Quality Assurance</b>: Comprehensive technical validation demonstrates the dataset's integrity, sensitivity to rainfall impacts, and capability to reveal complex traffic-weather interaction patterns.Data StructureThe dataset is organized into four primary components:<b>Road Network Data</b>: Topological representations including spatial geometry, functional classification, and connectivity information<b>Traffic Sensor Data</b>: Sensor metadata, locations, and measurements at both 5-minute and hourly resolutions<b>Precipitation Data</b>: Hourly meteorological information with spatial grid cell metadata<b>Derived Analytical Matrices</b>: Pre-computed structures for advanced spatial-temporal modelling and network analysesFile Formats<b>Tabular Data</b>: Apache Parquet format for optimal compression and fast query performance<b>Numerical Matrices</b>: NumPy NPZ format for efficient scientific computing<b>Total Size</b>: Approximately 2 GB uncompressedApplicationsThe IUTF dataset enables diverse analytical applications including:<b>Traffic Flow Prediction</b>: Developing weather-aware traffic forecasting models<b>Infrastructure Planning</b>: Identifying vulnerable network components and prioritizing investments<b>Resilience Assessment</b>: Quantifying system recovery curves, robustness metrics, and adaptive capacity<b>Climate Adaptation</b>: Supporting evidence-based transportation planning under changing precipitation patterns<b>Emergency Management</b>: Improving response strategies for weather-related traffic disruptionsMethodologyThe dataset creation involved three main stages:<b>Data Collection</b>: Sourcing traffic data from UTD19, road networks from OpenStreetMap, and precipitation data from ERA5 reanalysis<b>Spatio-Temporal Harmonization</b>: Comprehensive integration using novel algorithms for spatial alignment and temporal synchronization<b>Quality Assurance</b>: Rigorous validation and technical verification across all cities and data componentsCode AvailabilityProcessing code is available at: https://github.com/viviRG2024/IUTDF_processing