Spatiotemporal evolution analysis of social functions in multi-scale urban green spaces based on multimodal data fusion: A case study of shanghai from 2014 to 2023
Year: 2025
Type: article
Abstract: • Constructed a multi-scale dynamic research framework for urban green space functions. • A multi-modal data fusion urban green space classification model was constructed, named the MFUC model. • Combine urban green space functions and time series methods for dynamic analysis. Urban green spaces (UGS) are vital for sustainable urban development. However, research often views UGS as singular elements, ignoring their diverse functions and spatial dynamics, limiting a nuanced understanding of urban sustainability. To address this gap, this study proposes a multiple dynamic research framework for examining UGS functions. Firstly, establish a multi-scale analysis framework, which divides urban space into three different levels of spatial scale to meet the needs of different decision-making and analysis perspectives. The time period is ten years from 2014 to 2023. Secondly, establish a multi-modal data fusion model to map the social functions of urban green spaces. The semantic segmentation network U-NET is used to extract fine-grained, multi-period UGS data from remote sensing images. Furthermore, the Enhanced Representation through Knowledge Integration (ERNIE) model is employed within the natural language processing network to reclassify Points of Interest (POI) data text into distinct UGS function types. Moreover, OpenStreetMap (OSM) data are used to segment the internal space, and the reclassified POI data and UGS data are fused to extract multi-period UGS with functional types. The results show that the overall accuracy is 94.12%. Finally, the evolution of UGS was analyzed by combining spatial scale and time series, including the area and proportion of overall and functional UGS, as well as trend and cause analysis. The results show that over the past decade, Shanghai’s UGS system has shown a development trend of shifting from “incremental expansion” to “quality improvement”, with its functional uses becoming increasingly diverse and complex, reflecting its key role in supporting the city’s sustainable development. This study provides a basis for scientific analysis and decision-making on urban sustainable development.
Source: Ecological Indicators
Authors: Su Cao, Xuesheng Zhao, Shouhang Du, Siqi Yin
Institution: China University of Mining and Technology - Beijing
Language: English
Topic: Urban Green Space and Health
Subfield: Health, Toxicology and Mutagenesis
Field: Environmental Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,650
OpenAlex ID: W4414274653