Related Experiment Videos
Investigating LST evolution and heatwave patterns using machine learning in the Beijing-Tianjin-Hebei Urban
Chen Liu1,2, Maomao Zhang3,4
1College of Art, Hebei GEO University, Shijiazhuang 050000, China.
Iscience
|July 23, 2026
Summary
Urban heat risks are rising due to urbanization and climate change, worsening day-night temperature extremes. Nighttime temperatures increased significantly, expanding areas affected by compound heatwaves, highlighting the need for targeted urban heat management strategies.
Area of Science:
- Environmental Science
- Urban Climatology
- Remote Sensing
Background:
- Urban agglomerations face intensifying summer thermal risks from rapid urbanization and climate warming.
- A significant coupling exists between daytime heat exposure and insufficient nighttime cooling in cities.
Purpose of the Study:
- To analyze daytime and nighttime thermal evolution in the Beijing-Tianjin-Hebei urban agglomeration (BTHUA) from 2000 to 2024.
- To identify factors influencing urban heat distribution and trends.
Main Methods:
- Utilized MODIS land surface temperature (LST) data, heatwave metrics, and spatial statistics (Moran's I, Getis-Ord Gi*).
- Employed XGBoost-SHAP for interpreting the drivers of LST variations.
- Analyzed thermal evolution across the BTHUA over a 25-year period.
Main Results:
- Nighttime LST steadily increased by 2.3°C, while daytime LST showed weaker warming with fluctuations.
- Areas affected by compound day-night heatwaves expanded from 6.2% to 18.6%.
- Elevation was the primary cooling factor; vegetation and albedo influenced daytime LST, while nighttime lights and built-up land affected nighttime LST.
Conclusions:
- Compound heatwave exposure has significantly increased in the BTHUA.
- Distinct spatial patterns of heat hotspots and coldspots persist.
- Findings support the development of time-specific and regionally differentiated heat-risk governance strategies.
