Screening Hydrological Boundaries and Identifying Spatiotemporal Response Disparities for Lake Water Level Forecasting in Regulated River–Lake Systems: Evidence from Poyang Lake before and after Three Gorges Regulation

Abstract Reliable water level forecasting for regulated river–lake systems requires systematic screening of external hydrological boundaries with effective, nonredundant predictive information. However, inconsistent criteria for exogenous boundary variable selection in existing studies hinder cross-region and cross-period comparability of water level modeling. Taking the Three Gorges Reservoir (TGR)-regulated Poyang Lake as empirical evidence, this study conducted systematic hydrological boundary screening and identified spatiotemporal disparities in lake water level responses under altered river–lake interactions. Monthly water level data from four representative stations (Xingzi, Duchang, Tangyin, and Kangshan) served as response variables. Candidate predictors include water level and discharge at Jiujiang, Datong, and Hukou, as well as total inflow from the five major tributaries. The study period was divided into pre- and post-TGR regulation phases with 2003 as the breakpoint. Using the well-established seasonal autoregressive integrated moving average with exogenous variables (SARIMAX) model as a unified benchmark, this study adopted a hydrologically informed forward selection framework to quantitatively evaluate the marginal predictive contribution of each boundary variable and diagnose information overlap and redundancy. Results revealed distinct spatial heterogeneity and temporal variations in optimal boundary configurations. Hukou-related variables were consistently included in optimal models for all stations across both periods, with Hukou water level becoming the dominant single predictor after TGR regulation. Jiujiang variables provided complementary predictive information for the northern lake region, and Datong discharge and five-tributary inflow were critical for water level simulation in the central and southern lake areas. The findings demonstrate that no universal boundary configuration can represent the spatially and temporally differentiated water level dynamics of regulated river–lake systems. Instead, a common outlet boundary signal coexists with location- and period-dependent predictive information from mainstream and catchment inflows. This new understanding provides guidance for identifying priority boundary stations, optimizing hydrological monitoring, and selecting parsimonious predictors for water level forecasting in river-connected lakes affected by upstream reservoir regulation.

Authors

Institutions

Publication Details

Journal
Journal of Hydrologic Engineering
Published
2026-11-12
DOI
https://doi.org/10.1061/jhyeff.heeng-7038
Primary Topic
Hydrology and Watershed Management Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Screening Hydrological Boundaries and Identifying Spatiotemporal Response Disparities for Lake Water Level Forecasting in Regulated River–Lake Systems: Evidence from Poyang Lake before and after Three Gorges Regulation

Ranhang Zhao, Yuxuan Zhang, Xinlin Xie, Shuanghu Zhang et al.
Journal of Hydrologic Engineering
Hydrology and Watershed Management Studies
article

Screening Hydrological Boundaries and Identifying Spatiotemporal Response Disparities for Lake Water Level Forecasting in Regulated River–Lake Systems: Evidence from Poyang Lake before and after Three Gorges Regulation

Ranhang Zhao, Yuxuan Zhang, Xinlin Xie, Shuanghu Zhang, Hao Wang, Xiangang Jiang
article en

Abstract

Abstract Reliable water level forecasting for regulated river–lake systems requires systematic screening of external hydrological boundaries with effective, nonredundant predictive information. However, inconsistent criteria for exogenous boundary variable selection in existing studies hinder cross-region and cross-period comparability of water level modeling. Taking the Three Gorges Reservoir (TGR)-regulated Poyang Lake as empirical evidence, this study conducted systematic hydrological boundary screening and identified spatiotemporal disparities in lake water level responses under altered river–lake interactions. Monthly water level data from four representative stations (Xingzi, Duchang, Tangyin, and Kangshan) served as response variables. Candidate predictors include water level and discharge at Jiujiang, Datong, and Hukou, as well as total inflow from the five major tributaries. The study period was divided into pre- and post-TGR regulation phases with 2003 as the breakpoint. Using the well-established seasonal autoregressive integrated moving average with exogenous variables (SARIMAX) model as a unified benchmark, this study adopted a hydrologically informed forward selection framework to quantitatively evaluate the marginal predictive contribution of each boundary variable and diagnose information overlap and redundancy. Results revealed distinct spatial heterogeneity and temporal variations in optimal boundary configurations. Hukou-related variables were consistently included in optimal models for all stations across both periods, with Hukou water level becoming the dominant single predictor after TGR regulation. Jiujiang variables provided complementary predictive information for the northern lake region, and Datong discharge and five-tributary inflow were critical for water level simulation in the central and southern lake areas. The findings demonstrate that no universal boundary configuration can represent the spatially and temporally differentiated water level dynamics of regulated river–lake systems. Instead, a common outlet boundary signal coexists with location- and period-dependent predictive information from mainstream and catchment inflows. This new understanding provides guidance for identifying priority boundary stations, optimizing hydrological monitoring, and selecting parsimonious predictors for water level forecasting in river-connected lakes affected by upstream reservoir regulation.

Journal of Hydrologic EngineeringVol. 32(1)
Shandong University (CN), Sichuan Agricultural University (CN), China Institute of Water Resources and Hydropower Research (CN)
Clean water and sanitation
Hydrology and Watershed Management Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon