Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea

Jin, Youngkyu; Jeong, Taekmun; Gwon, Yonghyeon; Park, Jongpyo; Shin, Hyungjin; Lim, Heesung; Park, Sang I.

doi:10.3390/app16041792

Open AccessArticle

Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea

by

Youngkyu Jin

¹

,

Taekmun Jeong

²

,

Yonghyeon Gwon

²

,

Jongpyo Park

^2,*

,

Hyungjin Shin

¹

,

Heesung Lim

¹

and

Sang I. Park

³

¹

Rural Research Institute, Korea Rural Community Corporation, 870, Haean-ro, Sangnok-gu, Ansan 15634, Republic of Korea

²

Department of Water Resources and Environmental Engineering, HECOREA Inc., 233, Gasan Digital 1-ro, Geumcheon-gu, Seoul 08051, Republic of Korea

³

KALIS Institute of Technology, Korea Authority of Land & Infrastructure Safety (KALIS), Jinju 52856, Republic of Korea

^*

Author to whom correspondence should be addressed.

Appl. Sci. 2026, 16(4), 1792; https://doi.org/10.3390/app16041792

Submission received: 9 December 2025 / Revised: 18 January 2026 / Accepted: 3 February 2026 / Published: 11 February 2026

(This article belongs to the Special Issue Flood Risk and Geo-Hazards: The Strategy for Prevention and Mitigation)

Download

Browse Figures

Versions Notes

Abstract

Urban pluvial flooding in highly developed basins is challenging to forecast in real time because detailed 1D–2D hydraulic models are computationally expensive, while purely data-driven approaches often lack physical consistency. This study aims to enable operational urban flood nowcasting by proposing a model-informed AI framework for short-term stream-stage and urban inundation prediction in the Bisan-dong district of Anyang, South Korea, where the Anyang and Hagui Streams frequently overflow. A gated recurrent unit (GRU) network was trained on 10 min rainfall and stream-stage observations from 2011 to 2018 and independently validated on 2019–2022 data at four gauges to forecast stream stage at lead times of 10–60 min. In parallel, an ANN–CNN inundation surrogate was trained on 864 XP-SWMM 1D–2D simulation scenarios, forced by design storms and downstream water-level boundary conditions, to produce 256 × 256 maps of maximum inundation depth. The GRU model achieved

R^{2}

and Nash–Sutcliffe efficiency values generally above 0.95, with a mean absolute percentage error (MAPE) below approximately 5% for 10–30-min lead times; performance decreased but remained useful at 60 min. The inundation surrogate reproduced XP-SWMM results with an MAPE of 8.89% for inundation area and 19.49% for grid-based depth. Together, the ANN–CNN system enables rapid generation of high-resolution flood maps and provides a practical basis for AI-assisted urban flood nowcasting and risk management.

Keywords: urban flooding; stream-stage forecasting; gated recurrent unit; convolutional neural network; surrogate inundation model; dual-drainage modeling

Share and Cite

MDPI and ACS Style

Jin, Y.; Jeong, T.; Gwon, Y.; Park, J.; Shin, H.; Lim, H.; Park, S.I. Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea. Appl. Sci. 2026, 16, 1792. https://doi.org/10.3390/app16041792

AMA Style

Jin Y, Jeong T, Gwon Y, Park J, Shin H, Lim H, Park SI. Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea. Applied Sciences. 2026; 16(4):1792. https://doi.org/10.3390/app16041792

Chicago/Turabian Style

Jin, Youngkyu, Taekmun Jeong, Yonghyeon Gwon, Jongpyo Park, Hyungjin Shin, Heesung Lim, and Sang I. Park. 2026. "Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea" Applied Sciences 16, no. 4: 1792. https://doi.org/10.3390/app16041792

APA Style

Jin, Y., Jeong, T., Gwon, Y., Park, J., Shin, H., Lim, H., & Park, S. I. (2026). Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea. Applied Sciences, 16(4), 1792. https://doi.org/10.3390/app16041792

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Deep Learning-Based Short-Term Stream-Stage and Urban Inundation Prediction in a Highly Urbanized Basin: A Case Study of Bisan-dong, Anyang, South Korea

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI