- 1Seoul National University, College of Agriculture and Life sciences, Landscape Architecture, Seoul, Korea, Republic of (sojin1427@snu.ac.kr)
- 2Seoul National University, Graduate School of Environmental Studies, Interdisciplinary Program in Landscape Architecture, Seoul, Korea, Republic of (kimsanghyuck@snu.ac.kr)
- 3Research Institute of Agriculture and Life Science, Seoul National University, 1 Gwanak-gu, Seoul, 08826, Republic of Korea (dklee7@snu.ac.kr)
- 4Professor, Department of Landscape Architecture and Rural Systems Engineering, college of Agriculture and Life Sciences, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul, 08826, Republic of Korea
Urban Heat Islands (UHIs) significantly impact urban climate resilience, with both beneficial and adverse effects depending on seasonal and spatial factors. This study evaluates the influence of cool roofs and green roofs, designed to mitigate summer UHI intensity, on winter UHI dynamics in Seoul, Korea. A deep learning framework, incorporating temporal and spatial models, was developed to forecast UHI intensity and propose balanced seasonal mitigation strategies.
The temporal model used meteorological data collected from 54 Automatic Weather Stations (AWSs) over a 10-year period (2014–2023) and accounted for variables such as temperature, humidity, wind speed, and solar radiation. The spatial model incorporated GIS-derived data, including building density, vegetation coverage, and road imperviousness, along with satellite-obtained albedo and radiance information. Both models were combined into a hybrid system to predict seasonal UHI patterns.
According to previous research, cool roofs alleviated the urban heat island intensity in summer by an average of 2.5°C, and green roofs showed a mitigation effect of 1.8°C. These two strategies had the greatest impact mainly during the noon hour (12:00–15:00). On the other hand, cool roofs in winter had the side effect of increasing heating energy demand by about 5%, but green roofs offset this effect, limiting temperature drops to an average of 1°C and suppressing additional heating demand to 2%. Spatial analysis indicated that high-density urban areas were the main targets of mitigation strategies, with marked differences in seasonal UHI characteristics.
This research provides actionable insights for urban climate resilience planning, demonstrating the potential of deep learning models to inform policy and design interventions. The findings underscore the importance of spatially and temporally adaptive strategies, such as targeted cool roof and green roof installations, to achieve sustainable urban heat management across seasons.
This work was supported by Korea Environment Industry &Technology Institute (KEITI) through "Climate Change R&D Project for New Climate Regime.", funded by Korea Ministry of Environment (MOE) (RS-2022-KE002102)
This work was supported by Korea Environment Industry & Technology Institute(KEITI) through Climate Change R&D Project for New Climate Regime Program, funded by Korea Ministry of Environment(MOE)(RS-2023-00221110)
How to cite: Jun, S., Kim, S. H., and Lee, D. K.: Evaluating the Seasonal Effects of Cool Roofs and Green Roofs on Urban Heat Island effect Using Deep Learning Models in Seoul, Korea, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8968, https://doi.org/10.5194/egusphere-egu25-8968, 2025.