Developing an explainable deep learning module based on the LSTM framework for flood prediction

Long short-term memory (LSTM) networks have become indispensable tools in hydrological modeling due to their ability to capture long-term dependencies, handle non-linear relationships, and integrate multiple data sources but suffer from limited interpretability due to their black box nature. To addr...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhi Zhang, Dagang Wang, Yiwen Mei, Jinxin Zhu, Xusha Xiao
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Water
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/frwa.2025.1562842/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850274086993788928
author Zhi Zhang
Dagang Wang
Yiwen Mei
Jinxin Zhu
Xusha Xiao
author_facet Zhi Zhang
Dagang Wang
Yiwen Mei
Jinxin Zhu
Xusha Xiao
author_sort Zhi Zhang
collection DOAJ
description Long short-term memory (LSTM) networks have become indispensable tools in hydrological modeling due to their ability to capture long-term dependencies, handle non-linear relationships, and integrate multiple data sources but suffer from limited interpretability due to their black box nature. To address this limitation, we propose an explainable module within the LSTM framework, specifically designed for flood prediction across 531 catchments in the contiguous United States. Our approach incorporates a simplified gated module, which is interposed between the input data and the LSTM network, providing a transparent view of the module’s pattern recognition process. This gated module allows for easy identification of key variables and optimal lookback windows, and clusters the gated information into four categories: short-term and long-term impacts of precipitation and temperature. This categorization enhances our understanding of how the module utilizes input data and reveals underlying mechanisms in flood prediction. The modular design of our approach demonstrates high correlation with Saliency method, validating the credibility of its explanatory mechanisms, providing comparable interpretability features to LSTMs while illuminating key variables and optimal lookback windows considered most informative by hydrological models, and opening up new avenues for AI-assisted scientific discovery in the field.
format Article
id doaj-art-0c9e1e2b19b446f7bd769b8ad08fa851
institution OA Journals
issn 2624-9375
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Water
spelling doaj-art-0c9e1e2b19b446f7bd769b8ad08fa8512025-08-20T01:51:14ZengFrontiers Media S.A.Frontiers in Water2624-93752025-05-01710.3389/frwa.2025.15628421562842Developing an explainable deep learning module based on the LSTM framework for flood predictionZhi Zhang0Dagang Wang1Yiwen Mei2Jinxin Zhu3Xusha Xiao4Tourism and Historical Culture College, Zhaoqing University, Zhaoqing, ChinaSchool of Geography and Planning, Sun Yat-sen University, Guangzhou, ChinaSchool of Geography and Planning, Sun Yat-sen University, Guangzhou, ChinaSchool of Geography and Planning, Sun Yat-sen University, Guangzhou, ChinaSchool of Computer Science and Software, Zhaoqing University, Zhaoqing, ChinaLong short-term memory (LSTM) networks have become indispensable tools in hydrological modeling due to their ability to capture long-term dependencies, handle non-linear relationships, and integrate multiple data sources but suffer from limited interpretability due to their black box nature. To address this limitation, we propose an explainable module within the LSTM framework, specifically designed for flood prediction across 531 catchments in the contiguous United States. Our approach incorporates a simplified gated module, which is interposed between the input data and the LSTM network, providing a transparent view of the module’s pattern recognition process. This gated module allows for easy identification of key variables and optimal lookback windows, and clusters the gated information into four categories: short-term and long-term impacts of precipitation and temperature. This categorization enhances our understanding of how the module utilizes input data and reveals underlying mechanisms in flood prediction. The modular design of our approach demonstrates high correlation with Saliency method, validating the credibility of its explanatory mechanisms, providing comparable interpretability features to LSTMs while illuminating key variables and optimal lookback windows considered most informative by hydrological models, and opening up new avenues for AI-assisted scientific discovery in the field.https://www.frontiersin.org/articles/10.3389/frwa.2025.1562842/fullLSTMexplainable AIinterpretabilitygated modulecatchment analysisflood mechanisms
spellingShingle Zhi Zhang
Dagang Wang
Yiwen Mei
Jinxin Zhu
Xusha Xiao
Developing an explainable deep learning module based on the LSTM framework for flood prediction
Frontiers in Water
LSTM
explainable AI
interpretability
gated module
catchment analysis
flood mechanisms
title Developing an explainable deep learning module based on the LSTM framework for flood prediction
title_full Developing an explainable deep learning module based on the LSTM framework for flood prediction
title_fullStr Developing an explainable deep learning module based on the LSTM framework for flood prediction
title_full_unstemmed Developing an explainable deep learning module based on the LSTM framework for flood prediction
title_short Developing an explainable deep learning module based on the LSTM framework for flood prediction
title_sort developing an explainable deep learning module based on the lstm framework for flood prediction
topic LSTM
explainable AI
interpretability
gated module
catchment analysis
flood mechanisms
url https://www.frontiersin.org/articles/10.3389/frwa.2025.1562842/full
work_keys_str_mv AT zhizhang developinganexplainabledeeplearningmodulebasedonthelstmframeworkforfloodprediction
AT dagangwang developinganexplainabledeeplearningmodulebasedonthelstmframeworkforfloodprediction
AT yiwenmei developinganexplainabledeeplearningmodulebasedonthelstmframeworkforfloodprediction
AT jinxinzhu developinganexplainabledeeplearningmodulebasedonthelstmframeworkforfloodprediction
AT xushaxiao developinganexplainabledeeplearningmodulebasedonthelstmframeworkforfloodprediction