EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

A Machine Learning Approach to Predict Urban Flood Risk under Future Climate Change Scenarios: A Case Study of Jakarta, Indonesia

Minju Jeong, Dakota McCarty, Hyun Woo Kim

Journal of the Korean Society of Hazard Mitigation · 30.04.2026 · 26 (2) · 53-67

DOI: 10.9798/kosham.2026.26.2.53 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the Korean Society of Hazard Mitigation, erschienen am 30.04.2026. Verfasst von Minju Jeong und 2 weiteren Autor:innen. Die Arbeit nennt 65 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: University of Seoul.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Jakarta, a rapidly urbanizing coastal city, faces increasing flood risks because of climate change. Many studies have analyzed the city’s flood risk, but they have mostly relied on surveys or aggregated data, limiting the assessment of persistent flood risks at finer spatial scales. This study develops a Flood Risk Index (FRI) to assess and predict flood risks in Jakarta, a highly flood-prone city, under projected climate change scenarios for 2050. By integrating physical and meteorological factors with historical flood occurrence data from inaRISK and PODES, the FRI provides a granular grid-based analysis of flood risk. The predictive model, trained on data from 2015-2020 and validated using 2021 data, achieved high accuracy (R2 = 88.85%) using the Extra Trees Regressor. The results identified significant flood-prone areas, including Danau Sunter Dll, Rawa Buaya, and Bungur. These areas are particularly vulnerable because of a combination of coastal proximity, land subsidence, and insufficient green infrastructure. Our findings emphasize the need for proactive flood-mitigation strategies, such as expanding green infrastructure, enhancing drainage systems, and engaging communities in resilience-building efforts. This study highlights the growing flood risks in Jakarta due to climate change and urbanization, offering a scalable and data-driven framework for identifying high-priority areas. It provides valuable insights for developing inclusive and adaptive urban policies to build long-term resilience in Jakarta and other climate-vulnerable cities worldwide.

Förderung

  • University of Seoul

Zitieren

APA

Jeong, M., McCarty, D., & Kim, H. W. (2026). A Machine Learning Approach to Predict Urban Flood Risk under Future Climate Change Scenarios: A Case Study of Jakarta, Indonesia. Journal of the Korean Society of Hazard Mitigation, 26(2), 53-67. https://doi.org/10.9798/kosham.2026.26.2.53

Vancouver

Jeong M, McCarty D, Kim HW. A Machine Learning Approach to Predict Urban Flood Risk under Future Climate Change Scenarios: A Case Study of Jakarta, Indonesia. Journal of the Korean Society of Hazard Mitigation. 2026;26(2):53-67. doi:10.9798/kosham.2026.26.2.53

BibTeX
@article{Jeong2026,
  author = {Jeong, Minju and McCarty, Dakota and Kim, Hyun Woo},
  title = {A Machine Learning Approach to Predict Urban Flood Risk under Future Climate Change Scenarios: A Case Study of Jakarta, Indonesia},
  journal = {Journal of the Korean Society of Hazard Mitigation},
  year = {2026},
  volume = {26},
  number = {2},
  pages = {53--67},
  doi = {10.9798/kosham.2026.26.2.53}
}
Kostenlos starten →