Urban ecological environment quality evaluation using deep autoencoder integration of multisource remote sensing data
Urban ecological environmental quality (UEEQ) critically impacts human well-being and ecosystem sustainability. Current quantitative assessments of UEEQ face challenges due to complex interactions among air pollution, urban heat islands, forest degradation, and water scarcity. To address challenges,...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-12-01
|
| Series: | Ecological Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1574954125003565 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Urban ecological environmental quality (UEEQ) critically impacts human well-being and ecosystem sustainability. Current quantitative assessments of UEEQ face challenges due to complex interactions among air pollution, urban heat islands, forest degradation, and water scarcity. To address challenges, we develop a novel deep autoencoder-based framework that integrates multi-source remote sensing data to evaluate UEEQ dynamics systematically and identify its key driving factors. Compared to traditional methods, this framework provides a comprehensive and interpretable assessment process, encompassing feature selection (ecological, environmental, and social factors), methodological selection and construction, model training and evaluation, interpretability analysis, and trend analysis. Using this framework, the spatiotemporal variations in Beijing's UEEQ (2015–2020) were calculated, the primary driving factors were identified, and the spatiotemporal trends were analyzed. The results indicated that Beijing's UEEQ exhibited overall stability with localized variations, with 85.7 % of the area remaining unchanged. Ecological conditions in the urban core areas have improved significantly, while some degradation has occurred in the ecologically sensitive peripheral regions. SHAP analysis revealed that air quality factors, particularly AOD, PM2.5, and CO2, ranked among the top five out of the ten selected factors and were the primary drivers of UEEQ changes, showing a strong negative correlation with UEEQ. Improvements in urban air quality contributed to enhanced UEEQ, whereas pollution accumulation in surrounding areas led to delayed ecological degradation. These findings provide scientific support for zoned environmental governance, urban ecosystem management, and sustainable urban development strategies. |
|---|---|
| ISSN: | 1574-9541 |