An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.

Debris flows represent a persistent challenge for disaster prediction in mountainous regions due to their highly nonlinear and multivariate triggering mechanisms. This study proposes an explainable deep learning framework, the Improved Kepler Optimization Algorithm-Convolutional Neural Network-Bidir...

Full description

Saved in:
Bibliographic Details
Main Authors: Hao Yang, Tianlong Wang, Nikita Igorevich Fomin, Shuoting Xiao, Liang Liu
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0326587
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849419031527817216
author Hao Yang
Tianlong Wang
Nikita Igorevich Fomin
Shuoting Xiao
Liang Liu
author_facet Hao Yang
Tianlong Wang
Nikita Igorevich Fomin
Shuoting Xiao
Liang Liu
author_sort Hao Yang
collection DOAJ
description Debris flows represent a persistent challenge for disaster prediction in mountainous regions due to their highly nonlinear and multivariate triggering mechanisms. This study proposes an explainable deep learning framework, the Improved Kepler Optimization Algorithm-Convolutional Neural Network-Bidirectional Gated Recurrent Unit-Attention (IKOA-CNN-BiGRU-Attention) model, for precise debris flow hazard prediction in the Yunnan section of the Nujiang River Basin, China. The model is developed and validated using data from 159 debris flow-prone gullies, integrating deep convolutional, recurrent, and attention-based architectures, with hyperparameters autonomously optimized by IKOA. Model explainability is enhanced using SHapley Additive exPlanations (SHAP), which quantify the influence of key factors. The IKOA-CNN-BiGRU-Attention framework consistently outperforms 13 benchmark models, achieving a root mean square error of 2.33 × 10-6, mean absolute error of 1.51 × 10-6, and mean absolute percentage error of 0.006%. The model maintains high stability across 50 repeated experiments, strong resilience to 20% input noise, and robust generalizability under five-fold cross-validation. Interpretability analysis identifies potential source energy and maximum 24-hour rainfall as primary determinants and uncovers a dual-threshold physical mechanism underlying debris flow initiation. These findings provide a quantitative basis for adaptive early warning and targeted risk mitigation, and establish a transferable framework for explainable geohazard prediction.
format Article
id doaj-art-fb421cdf5dde46f99335b6bbb7122020
institution Kabale University
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-fb421cdf5dde46f99335b6bbb71220202025-08-20T03:32:15ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01206e032658710.1371/journal.pone.0326587An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.Hao YangTianlong WangNikita Igorevich FominShuoting XiaoLiang LiuDebris flows represent a persistent challenge for disaster prediction in mountainous regions due to their highly nonlinear and multivariate triggering mechanisms. This study proposes an explainable deep learning framework, the Improved Kepler Optimization Algorithm-Convolutional Neural Network-Bidirectional Gated Recurrent Unit-Attention (IKOA-CNN-BiGRU-Attention) model, for precise debris flow hazard prediction in the Yunnan section of the Nujiang River Basin, China. The model is developed and validated using data from 159 debris flow-prone gullies, integrating deep convolutional, recurrent, and attention-based architectures, with hyperparameters autonomously optimized by IKOA. Model explainability is enhanced using SHapley Additive exPlanations (SHAP), which quantify the influence of key factors. The IKOA-CNN-BiGRU-Attention framework consistently outperforms 13 benchmark models, achieving a root mean square error of 2.33 × 10-6, mean absolute error of 1.51 × 10-6, and mean absolute percentage error of 0.006%. The model maintains high stability across 50 repeated experiments, strong resilience to 20% input noise, and robust generalizability under five-fold cross-validation. Interpretability analysis identifies potential source energy and maximum 24-hour rainfall as primary determinants and uncovers a dual-threshold physical mechanism underlying debris flow initiation. These findings provide a quantitative basis for adaptive early warning and targeted risk mitigation, and establish a transferable framework for explainable geohazard prediction.https://doi.org/10.1371/journal.pone.0326587
spellingShingle Hao Yang
Tianlong Wang
Nikita Igorevich Fomin
Shuoting Xiao
Liang Liu
An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.
PLoS ONE
title An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.
title_full An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.
title_fullStr An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.
title_full_unstemmed An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.
title_short An integrated IKOA-CNN-BiGRU-Attention framework with SHAP explainability for high-precision debris flow hazard prediction in the Nujiang river basin, China.
title_sort integrated ikoa cnn bigru attention framework with shap explainability for high precision debris flow hazard prediction in the nujiang river basin china
url https://doi.org/10.1371/journal.pone.0326587
work_keys_str_mv AT haoyang anintegratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT tianlongwang anintegratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT nikitaigorevichfomin anintegratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT shuotingxiao anintegratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT liangliu anintegratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT haoyang integratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT tianlongwang integratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT nikitaigorevichfomin integratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT shuotingxiao integratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina
AT liangliu integratedikoacnnbigruattentionframeworkwithshapexplainabilityforhighprecisiondebrisflowhazardpredictioninthenujiangriverbasinchina