Kim et al. (2026) Sequence Reconstruction for River Water Level Anomaly Correction Using a Simplified Bidirectional LSTM Autoencoder

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Water
• Year: 2026
• Date: 2026-09-15
• Authors: Chung-Soo Kim, Kahhoong Kok
• DOI: 10.3390/w18182301

Research Groups

• Department of Civil and Environmental Engineering, Seoul National University
• Korea Institute of Construction Technology (KICT)

Short Summary

This study proposes a Bidirectional Long Short-Term Memory (BiLSTM) Autoencoder framework to correct anomalous observations in river water level data. The proposed method demonstrates superior performance compared to conventional regression models.

Objective

• Investigate the effectiveness of a reconstruction-oriented BiLSTM Autoencoder for correcting anomalies in hourly observed water level data

Study Configuration

• Spatial Scale: Hourly observed water level data from the Han River, Republic of Korea
• Temporal Scale: Long-term (unseen) and short-term (calibration) periods

Methodology and Data

• Models used: Bidirectional Long Short-Term Memory (BiLSTM) Autoencoder, conventional first-, second-, and third-order polynomial and exponential regression models
• Data sources: Hourly observed water level data from the Han River, Republic of Korea

Main Results

• The proposed BiLSTM Autoencoder achieved reconstruction accuracy comparable to conventional regression models during calibration while exhibiting superior generalization to unseen validation datasets.
• Reconstruction accuracy deteriorated with increasing training data contamination.

Contributions

• This study proposes a novel sequence-reconstruction approach for offline river water level quality control, demonstrating potential as an effective method for correcting anomalies in water level data.

Funding

• This research was supported by the Korea Institute of Construction Technology (KICT) under the Ministry of Land, Infrastructure and Transport (Grant No. 21A06-032).

Citation

@article{Kim2026Sequence,
  author = {Kim, Chung-Soo and Kok, Kahhoong},
  title = {Sequence Reconstruction for River Water Level Anomaly Correction Using a Simplified Bidirectional LSTM Autoencoder},
  journal = {Water},
  year = {2026},
  doi = {10.3390/w18182301},
  url = {https://doi.org/10.3390/w18182301}
}