This study combines official records, news reports, social media posts, and search data to track urban flood events in twenty major Chinese cities from 2000 to 2024. It builds a comparable database and uses an interpretable machine learning framework to measure flood severity across cities and years. Results show a clear north south contrast: southern river and coastal cities face frequent severe floods, while northern cities usually face lower risk but suffer sudden extremes during rare storms.

Our study developed a preliminary dataset of catchment attributes and meteorological variables covering 544 basins in China, and evaluated the applicability of process-based models, the long short-term memory model, and hybrid modeling methods. Results highlight the critical role of meteorological data quality and the potential of hybrid approaches. Our findings support model selection and offer reference for regions with limited observational data.

Roadside noise barriers (RNBs) are important urban infrastructures to ensure a city is liveable. This study provides the first reliable and nationwide vectorized RNB dataset with street view imagery in China. The generated RNB dataset is evaluated in terms of two aspects, i.e., the detection accuracy and the completeness and positional accuracy. The method is based on a developed geospatial artificial intelligence framework.