Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau

Abstract Interpretability of deep learning (DL) poses a significant challenge in hydrology modeling, particularly under the complex and frigid conditions of the Tibetan Plateau (TP), which further restricts its application. In this study, we developed a cascade‐style hybrid modeling framework by int...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhanliang Zhu, Xiongpeng Tang, Jianyun Zhang, Yehai Tang, Lei Liu, Chao Gao, Silong Zhang, Yanli Liu, Junliang Jin, Cuishan Liu, Bikui Zhao, Guoqing Wang
Format: Article
Language:English
Published: Wiley 2025-08-01
Series:Water Resources Research
Subjects:
Online Access:https://doi.org/10.1029/2024WR038515
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846174012655271936
author Zhanliang Zhu
Xiongpeng Tang
Jianyun Zhang
Yehai Tang
Lei Liu
Chao Gao
Silong Zhang
Yanli Liu
Junliang Jin
Cuishan Liu
Bikui Zhao
Guoqing Wang
author_facet Zhanliang Zhu
Xiongpeng Tang
Jianyun Zhang
Yehai Tang
Lei Liu
Chao Gao
Silong Zhang
Yanli Liu
Junliang Jin
Cuishan Liu
Bikui Zhao
Guoqing Wang
author_sort Zhanliang Zhu
collection DOAJ
description Abstract Interpretability of deep learning (DL) poses a significant challenge in hydrology modeling, particularly under the complex and frigid conditions of the Tibetan Plateau (TP), which further restricts its application. In this study, we developed a cascade‐style hybrid modeling framework by integrating the Variable Infiltration Capacity (VIC) model with a two‐dimensional grid long short‐term memory (termed VIC‐LSTM), and incorporated a dual‐layer probe for training and investigating non‐linear hydrological processes. Our objective was to explore the framework's potential for enhancing hydrological simulation accuracy and expanding interpretability. This framework was adopted for the Yarlung Zangbo River Basin (above the Lazi guaged station) in the TP. The results of the VIC‐LSTM demonstrated its effectiveness, achieving a simulated daily streamflow NSE of 0.78 compared to 0.69 for the VIC model during the training period. Moreover, the probe experiments, aided by remote‐sensed images, successfully deciphered the timing signals of snowmelt and glacier melt, with ablation and duration time errors within 2 weeks. The average errors for snowmelt and glacier melt were approximately 7 and 10 days, respectively. The combined spatial and temporal feature quantification indicated that snowmelt and glacier melt contributed 23.7% and 7.7% to the total streamflow, respectively. Additionally, VIC‐LSTM identified that snowmelt signals generally preceded glacier melt signals. These findings underscore the potential of VIC‐LSTM to interpret hydrological processes, increasing confidence in using DL approaches. Therefore, it offers a new perspective for quantifying water resources and understanding hydrological processes in high‐elevation, glacier‐snow coexisting basins.
format Article
id doaj-art-6e4d6dabc26647699a19ec6409d0b0a3
institution Kabale University
issn 0043-1397
1944-7973
language English
publishDate 2025-08-01
publisher Wiley
record_format Article
series Water Resources Research
spelling doaj-art-6e4d6dabc26647699a19ec6409d0b0a32025-08-26T12:02:53ZengWileyWater Resources Research0043-13971944-79732025-08-01618n/an/a10.1029/2024WR038515Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan PlateauZhanliang Zhu0Xiongpeng Tang1Jianyun Zhang2Yehai Tang3Lei Liu4Chao Gao5Silong Zhang6Yanli Liu7Junliang Jin8Cuishan Liu9Bikui Zhao10Guoqing Wang11Guangdong‐Hong Kong Joint Laboratory for Water Security Beijing Normal University Zhuhai ChinaGuangdong‐Hong Kong Joint Laboratory for Water Security Beijing Normal University Zhuhai ChinaNanjing Hydraulic Research Institute Nanjing ChinaGuangdong‐Hong Kong Joint Laboratory for Water Security Beijing Normal University Zhuhai ChinaGuangdong‐Hong Kong Joint Laboratory for Water Security Beijing Normal University Zhuhai ChinaGuangdong‐Hong Kong Joint Laboratory for Water Security Beijing Normal University Zhuhai ChinaGuangdong‐Hong Kong Joint Laboratory for Water Security Beijing Normal University Zhuhai ChinaNanjing Hydraulic Research Institute Nanjing ChinaNanjing Hydraulic Research Institute Nanjing ChinaNanjing Hydraulic Research Institute Nanjing ChinaGuangdong Research Institute of Water Resources and Hydropower Guangzhou ChinaNanjing Hydraulic Research Institute Nanjing ChinaAbstract Interpretability of deep learning (DL) poses a significant challenge in hydrology modeling, particularly under the complex and frigid conditions of the Tibetan Plateau (TP), which further restricts its application. In this study, we developed a cascade‐style hybrid modeling framework by integrating the Variable Infiltration Capacity (VIC) model with a two‐dimensional grid long short‐term memory (termed VIC‐LSTM), and incorporated a dual‐layer probe for training and investigating non‐linear hydrological processes. Our objective was to explore the framework's potential for enhancing hydrological simulation accuracy and expanding interpretability. This framework was adopted for the Yarlung Zangbo River Basin (above the Lazi guaged station) in the TP. The results of the VIC‐LSTM demonstrated its effectiveness, achieving a simulated daily streamflow NSE of 0.78 compared to 0.69 for the VIC model during the training period. Moreover, the probe experiments, aided by remote‐sensed images, successfully deciphered the timing signals of snowmelt and glacier melt, with ablation and duration time errors within 2 weeks. The average errors for snowmelt and glacier melt were approximately 7 and 10 days, respectively. The combined spatial and temporal feature quantification indicated that snowmelt and glacier melt contributed 23.7% and 7.7% to the total streamflow, respectively. Additionally, VIC‐LSTM identified that snowmelt signals generally preceded glacier melt signals. These findings underscore the potential of VIC‐LSTM to interpret hydrological processes, increasing confidence in using DL approaches. Therefore, it offers a new perspective for quantifying water resources and understanding hydrological processes in high‐elevation, glacier‐snow coexisting basins.https://doi.org/10.1029/2024WR038515deep learninghybrid modelinghydrological processsnowmelt runoffTibetan Plateau
spellingShingle Zhanliang Zhu
Xiongpeng Tang
Jianyun Zhang
Yehai Tang
Lei Liu
Chao Gao
Silong Zhang
Yanli Liu
Junliang Jin
Cuishan Liu
Bikui Zhao
Guoqing Wang
Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau
Water Resources Research
deep learning
hybrid modeling
hydrological process
snowmelt runoff
Tibetan Plateau
title Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau
title_full Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau
title_fullStr Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau
title_full_unstemmed Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau
title_short Deciphering Nonlinear Hydrological Process by a Coupled Deep Learning and Physical Based Model in Southern Tibetan Plateau
title_sort deciphering nonlinear hydrological process by a coupled deep learning and physical based model in southern tibetan plateau
topic deep learning
hybrid modeling
hydrological process
snowmelt runoff
Tibetan Plateau
url https://doi.org/10.1029/2024WR038515
work_keys_str_mv AT zhanliangzhu decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT xiongpengtang decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT jianyunzhang decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT yehaitang decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT leiliu decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT chaogao decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT silongzhang decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT yanliliu decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT junliangjin decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT cuishanliu decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT bikuizhao decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau
AT guoqingwang decipheringnonlinearhydrologicalprocessbyacoupleddeeplearningandphysicalbasedmodelinsoutherntibetanplateau