Yuxue Guo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yuxue Guo has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Yue-Ping Xu
Institute of Water Science and Engineering, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China
10 shared publications Xinting Yu
Institute of Water Science and Engineering, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China
7 shared publicationsHaiting Gu
Institute of Water Science and Engineering, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China
6 shared publications- Siwei Chen
Institute of Water Science and Engineering, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China
6 shared publications Akash Koppa
Department of Environmental Science & Technology, University of Maryland, College Park, USA
4 shared publicationsChangmin Du
Institute of Water Science and Engineering, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China
4 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Copernicus GmbH
- Vol. 2026, 1-24
Comment on egusphere-2026-5551
Ecosystem water use efficiency (WUE), the ratio of gross primary productivity (GPP) to evapotranspiration (ET), couples the land carbon and water cycles and is a key indicator of water–carbon coupling and ecosystem resilience. Yet how WUE responds to intensifying drought has remained unresolved, with studies reporting both increases and decreases. Here, combining multi-source global satellite, reanalysis, and model-fusion datasets, we show that across 57.5 % of vegetated land WUE first rises and then falls as drought deepens, reversing only once drought reaches extreme severity. This reconciles the apparent contradiction in earlier work, whose reported increases and decreases are the rising and falling limbs of one threshold response rather than competing outcomes. The reversal arises from an ecohydrological asymmetry, in which the early rise is driven by ET declining faster than GPP while continued GPP loss under severe drought overturns it. The depth of the threshold is not uniform but set by aridity, radiation, leaf area, and atmospheric dryness overlaid on vegetation type, so that denser, more productive ecosystems tolerate more severe drought before reversing. Because these thresholds cluster in the semi-arid and transitional biomes that dominate the variability of the land carbon sink, a rising WUE signals only transient coping rather than durable resilience, and the threshold marks an early indicator of where intensifying drought begins to erode ecosystem carbon uptake.
- 9 November 2026
- Published by Copernicus GmbH
- Vol. 2026, 1-24
Open AccessComment on egusphere-2026-5551
Ecosystem water use efficiency (WUE), the ratio of gross primary productivity (GPP) to evapotranspiration (ET), couples the land carbon and water cycles and is a key indicator of water–carbon coupling and ecosystem resilience. Yet how WUE responds to intensifying drought has remained unresolved, with studies reporting both increases and decreases. Here, combining multi-source global satellite, reanalysis, and model-fusion datasets, we show that across 57.5 % of vegetated land WUE first rises and then falls as drought deepens, reversing only once drought reaches extreme severity. This reconciles the apparent contradiction in earlier work, whose reported increases and decreases are the rising and falling limbs of one threshold response rather than competing outcomes. The reversal arises from an ecohydrological asymmetry, in which the early rise is driven by ET declining faster than GPP while continued GPP loss under severe drought overturns it. The depth of the threshold is not uniform but set by aridity, radiation, leaf area, and atmospheric dryness overlaid on vegetation type, so that denser, more productive ecosystems tolerate more severe drought before reversing. Because these thresholds cluster in the semi-arid and transitional biomes that dominate the variability of the land carbon sink, a rising WUE signals only transient coping rather than durable resilience, and the threshold marks an early indicator of where intensifying drought begins to erode ecosystem carbon uptake.
- Open Access
- 27 October 2026
- Published by Copernicus GmbH
- Vol. 2026, 1-24
Comment on egusphere-2026-5551
Ecosystem water use efficiency (WUE), the ratio of gross primary productivity (GPP) to evapotranspiration (ET), couples the land carbon and water cycles and is a key indicator of water–carbon coupling and ecosystem resilience. Yet how WUE responds to intensifying drought has remained unresolved, with studies reporting both increases and decreases. Here, combining multi-source global satellite, reanalysis, and model-fusion datasets, we show that across 57.5 % of vegetated land WUE first rises and then falls as drought deepens, reversing only once drought reaches extreme severity. This reconciles the apparent contradiction in earlier work, whose reported increases and decreases are the rising and falling limbs of one threshold response rather than competing outcomes. The reversal arises from an ecohydrological asymmetry, in which the early rise is driven by ET declining faster than GPP while continued GPP loss under severe drought overturns it. The depth of the threshold is not uniform but set by aridity, radiation, leaf area, and atmospheric dryness overlaid on vegetation type, so that denser, more productive ecosystems tolerate more severe drought before reversing. Because these thresholds cluster in the semi-arid and transitional biomes that dominate the variability of the land carbon sink, a rising WUE signals only transient coping rather than durable resilience, and the threshold marks an early indicator of where intensifying drought begins to erode ecosystem carbon uptake.
- 27 October 2026
- Published by Copernicus GmbH
- Vol. 2026, 1-24
Open AccessComment on egusphere-2026-5551
Ecosystem water use efficiency (WUE), the ratio of gross primary productivity (GPP) to evapotranspiration (ET), couples the land carbon and water cycles and is a key indicator of water–carbon coupling and ecosystem resilience. Yet how WUE responds to intensifying drought has remained unresolved, with studies reporting both increases and decreases. Here, combining multi-source global satellite, reanalysis, and model-fusion datasets, we show that across 57.5 % of vegetated land WUE first rises and then falls as drought deepens, reversing only once drought reaches extreme severity. This reconciles the apparent contradiction in earlier work, whose reported increases and decreases are the rising and falling limbs of one threshold response rather than competing outcomes. The reversal arises from an ecohydrological asymmetry, in which the early rise is driven by ET declining faster than GPP while continued GPP loss under severe drought overturns it. The depth of the threshold is not uniform but set by aridity, radiation, leaf area, and atmospheric dryness overlaid on vegetation type, so that denser, more productive ecosystems tolerate more severe drought before reversing. Because these thresholds cluster in the semi-arid and transitional biomes that dominate the variability of the land carbon sink, a rising WUE signals only transient coping rather than durable resilience, and the threshold marks an early indicator of where intensifying drought begins to erode ecosystem carbon uptake.
- Open Access
- 14 October 2026
- Published by Copernicus GmbH
Supplementary material to "Drought thresholds shape divergent responses in ecosystem water use efficiency"
- 14 October 2026
- Published by Copernicus GmbH
Open AccessSupplementary material to "Drought thresholds shape divergent responses in ecosystem water use efficiency"
- Open Access
- 14 October 2026
- Published by Copernicus GmbH in EGUsphere
- Vol. 2026, 1-24
Drought thresholds shape divergent responses in ecosystem water use efficiency
Ecosystem water use efficiency (WUE), the ratio of gross primary productivity (GPP) to evapotranspiration (ET), couples the land carbon and water cycles and is a key indicator of water–carbon coupling and ecosystem resilience. Yet how WUE responds to intensifying drought has remained unresolved, with studies reporting both increases and decreases. Here, combining multi-source global satellite, reanalysis, and model-fusion datasets, we show that across 57.5 % of vegetated land WUE first rises and then falls as drought deepens, reversing only once drought reaches extreme severity. This reconciles the apparent contradiction in earlier work, whose reported increases and decreases are the rising and falling limbs of one threshold response rather than competing outcomes. The reversal arises from an ecohydrological asymmetry, in which the early rise is driven by ET declining faster than GPP while continued GPP loss under severe drought overturns it. The depth of the threshold is not uniform but set by aridity, radiation, leaf area, and atmospheric dryness overlaid on vegetation type, so that denser, more productive ecosystems tolerate more severe drought before reversing. Because these thresholds cluster in the semi-arid and transitional biomes that dominate the variability of the land carbon sink, a rising WUE signals only transient coping rather than durable resilience, and the threshold marks an early indicator of where intensifying drought begins to erode ecosystem carbon uptake.
- 14 October 2026
- Published by Copernicus GmbH in EGUsphere
- Vol. 2026, 1-24
Open AccessDrought thresholds shape divergent responses in ecosystem water use efficiency
Ecosystem water use efficiency (WUE), the ratio of gross primary productivity (GPP) to evapotranspiration (ET), couples the land carbon and water cycles and is a key indicator of water–carbon coupling and ecosystem resilience. Yet how WUE responds to intensifying drought has remained unresolved, with studies reporting both increases and decreases. Here, combining multi-source global satellite, reanalysis, and model-fusion datasets, we show that across 57.5 % of vegetated land WUE first rises and then falls as drought deepens, reversing only once drought reaches extreme severity. This reconciles the apparent contradiction in earlier work, whose reported increases and decreases are the rising and falling limbs of one threshold response rather than competing outcomes. The reversal arises from an ecohydrological asymmetry, in which the early rise is driven by ET declining faster than GPP while continued GPP loss under severe drought overturns it. The depth of the threshold is not uniform but set by aridity, radiation, leaf area, and atmospheric dryness overlaid on vegetation type, so that denser, more productive ecosystems tolerate more severe drought before reversing. Because these thresholds cluster in the semi-arid and transitional biomes that dominate the variability of the land carbon sink, a rising WUE signals only transient coping rather than durable resilience, and the threshold marks an early indicator of where intensifying drought begins to erode ecosystem carbon uptake.
- Open Access
- 17 August 2026
- Published by IWA Publishing in Journal of Water and Climate Change
- Vol. 17 (10) , 2767-2793
A spatial-temporal AI model with a dynamic edge-weighted GCN for both deterministic and probabilistic flood forecasting in the Shifeng River, Southeastern China
Three types of watershed spatial attributes are separately used to guide dynamic edge-weighted GCN (DynE-GCN) in learning spatial dependencies between each station's observed rainfall. Temporal dynamics are modeled by LSTM, followed by a VAE that compresses high-dimensional spatiotemporal features and reconstructs new data. Finally, by modulating the VAE output and sampling frequency, both deterministic and probabilistic flood forecasting can be performed. AI-based forecasting models have emerged as a hot topic in flood forecasting. To address the limitations of single models in uncertainty quantification and flood peak prediction, this study proposes a novel spatial-temporal AI model, GCN-VAE-ResLSTM. The model improves LSTM by utilizing a Graph Convolutional Network (GCN) with dynamic edge weights, termed DynE-GCN, to learn spatial dependencies of rainfall. Three residual connections are employed to strengthen information flow. A Variational Autoencoder (VAE) is incorporated for both extreme trend learning and uncertainty modeling. Seven series models are constructed to investigate the contributions of all modules to LSTM. Taking the Shifeng River basin in Zhejiang Province, China, as the study area, this study incorporates three watershed spatial attributes, namely hydrological network distribution, topography, and distance. Results indicate: (1) The DynE-GCN exhibits superior adaptability to spatial feature inputs. (2) The GCN-VAE-ResLSTM delivers optimal performance for deterministic predictions, reducing RMSE and PRE by 56.4 and 60.7%, respectively, compared to LSTM at a 1-hour lead time. And the GCN-VAE-LSTM delivers robust and accurate flood peak predictions at all lead times. (3) In probabilistic forecasting, the GCN-VAE-ResLSTM demonstrates the highest accuracy, while the VAE-ResLSTM achieves better sharpness, reliability, and timeliness.
- 17 August 2026
- Published by IWA Publishing in Journal of Water and Climate Change
- Vol. 17 (10) , 2767-2793
Open AccessA spatial-temporal AI model with a dynamic edge-weighted GCN for both deterministic and probabilistic flood forecasting in the Shifeng River, Southeastern China
Three types of watershed spatial attributes are separately used to guide dynamic edge-weighted GCN (DynE-GCN) in learning spatial dependencies between each station's observed rainfall. Temporal dynamics are modeled by LSTM, followed by a VAE that compresses high-dimensional spatiotemporal features and reconstructs new data. Finally, by modulating the VAE output and sampling frequency, both deterministic and probabilistic flood forecasting can be performed. AI-based forecasting models have emerged as a hot topic in flood forecasting. To address the limitations of single models in uncertainty quantification and flood peak prediction, this study proposes a novel spatial-temporal AI model, GCN-VAE-ResLSTM. The model improves LSTM by utilizing a Graph Convolutional Network (GCN) with dynamic edge weights, termed DynE-GCN, to learn spatial dependencies of rainfall. Three residual connections are employed to strengthen information flow. A Variational Autoencoder (VAE) is incorporated for both extreme trend learning and uncertainty modeling. Seven series models are constructed to investigate the contributions of all modules to LSTM. Taking the Shifeng River basin in Zhejiang Province, China, as the study area, this study incorporates three watershed spatial attributes, namely hydrological network distribution, topography, and distance. Results indicate: (1) The DynE-GCN exhibits superior adaptability to spatial feature inputs. (2) The GCN-VAE-ResLSTM delivers optimal performance for deterministic predictions, reducing RMSE and PRE by 56.4 and 60.7%, respectively, compared to LSTM at a 1-hour lead time. And the GCN-VAE-LSTM delivers robust and accurate flood peak predictions at all lead times. (3) In probabilistic forecasting, the GCN-VAE-ResLSTM demonstrates the highest accuracy, while the VAE-ResLSTM achieves better sharpness, reliability, and timeliness.
- Open Access
- 13 March 2026
- Published by Copernicus GmbH
Multi-Objective Joint Robust Optimization for Flood Control Operation of Reservoir Groups under Uncertaint
To address multi-objective conflicts and hydrological uncertainty in joint flood control operation of reservoir groups, this study develops an uncertainty-aware optimization framework. An improved Vine Copula method with variable selection and structural simplification (RDV-Copula) is first introduced to describe the spatiotemporal dependence of multi-site flood processes. By simplifying the dependence structure, the method alleviates the complexity of high-dimensional modeling and generates stochastic inflow scenarios for reservoir operation. On this basis, a two-layer hedging–robust optimization model (TL-HRO) is formulated, in which hedging strategies are combined with robust optimization to coordinate flood control and hydropower generation objectives across current and future operation stages. The framework is applied to the Shifengxi Basin in Zhejiang Province, China. The analysis shows pronounced spatial dependence among flood processes, with a four-site flood synchronization probability of 41.92% and an average pairwise synchronization frequency of 65.87%. Compared with conventional approaches, the RDV-Copula achieves improvements in simulation accuracy of approximately 15.0%–61.2% while reducing model complexity, providing reliable stochastic inflow inputs for reservoir operation. Using these stochastic scenarios, the TL-HRO model is evaluated against a conventional multi-objective robust optimization (MORO) model. The results indicate that TL-HRO performs better in terms of both flood risk control and hydropower generation, with an average reduction of 58.26% in the optimization reference value relative to MORO. These findings suggest that the proposed approach can improve the overall performance of reservoir group operation under flood-related uncertainty and support flood control decision-making.
- 13 March 2026
- Published by Copernicus GmbH
Open AccessMulti-Objective Joint Robust Optimization for Flood Control Operation of Reservoir Groups under Uncertaint
To address multi-objective conflicts and hydrological uncertainty in joint flood control operation of reservoir groups, this study develops an uncertainty-aware optimization framework. An improved Vine Copula method with variable selection and structural simplification (RDV-Copula) is first introduced to describe the spatiotemporal dependence of multi-site flood processes. By simplifying the dependence structure, the method alleviates the complexity of high-dimensional modeling and generates stochastic inflow scenarios for reservoir operation. On this basis, a two-layer hedging–robust optimization model (TL-HRO) is formulated, in which hedging strategies are combined with robust optimization to coordinate flood control and hydropower generation objectives across current and future operation stages. The framework is applied to the Shifengxi Basin in Zhejiang Province, China. The analysis shows pronounced spatial dependence among flood processes, with a four-site flood synchronization probability of 41.92% and an average pairwise synchronization frequency of 65.87%. Compared with conventional approaches, the RDV-Copula achieves improvements in simulation accuracy of approximately 15.0%–61.2% while reducing model complexity, providing reliable stochastic inflow inputs for reservoir operation. Using these stochastic scenarios, the TL-HRO model is evaluated against a conventional multi-objective robust optimization (MORO) model. The results indicate that TL-HRO performs better in terms of both flood risk control and hydropower generation, with an average reduction of 58.26% in the optimization reference value relative to MORO. These findings suggest that the proposed approach can improve the overall performance of reservoir group operation under flood-related uncertainty and support flood control decision-making.
- Open Access
- 15 January 2025
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 29 (1) , 179-214
Synchronization frequency analysis and stochastic simulation of multi-site flood flows based on the complicated vine copula structure
Accurately modeling and predicting flood flows across multiple sites within a watershed presents significant challenges due to potential issues of insufficient accuracy and excessive computational demands in existing methodologies. In response to these challenges, this study introduces a novel approach centered around the use of vine copula models, termed RDV-Copula (reduced-dimension vine copula construction approach). The core of this methodology lies in its ability to integrate and extract complex data before constructing the copula function, thus preserving the intricate spatial–temporal connections among multiple sites while substantially reducing the vine copula's complexity. This study performs a synchronization frequency analysis using the devised copula models, offering valuable insights into flood encounter probabilities. Additionally, the innovative approach undergoes validation by comparison with three benchmark models which vary in dimensions and nature of variable interactions. Furthermore, the study conducts stochastic simulations, exploring both unconditional and conditional scenarios across different vine copula models. Applied in the Shifeng Creek watershed, China, the findings reveal that vine copula models are superior in capturing complex variable relationships, demonstrating significant spatial interconnectivity crucial for flood risk prediction in heavy-rainfall events. Interestingly, the study observes that expanding the model's dimensions does not inherently enhance simulation precision. The RDV-Copula method not only captures comprehensive information effectively but also simplifies the vine copula model by reducing its dimensionality and complexity. This study contributes to the field of hydrology by offering a refined method for analyzing and simulating multi-site flood flows.
- 15 January 2025
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 29 (1) , 179-214
Open AccessSynchronization frequency analysis and stochastic simulation of multi-site flood flows based on the complicated vine copula structure
Accurately modeling and predicting flood flows across multiple sites within a watershed presents significant challenges due to potential issues of insufficient accuracy and excessive computational demands in existing methodologies. In response to these challenges, this study introduces a novel approach centered around the use of vine copula models, termed RDV-Copula (reduced-dimension vine copula construction approach). The core of this methodology lies in its ability to integrate and extract complex data before constructing the copula function, thus preserving the intricate spatial–temporal connections among multiple sites while substantially reducing the vine copula's complexity. This study performs a synchronization frequency analysis using the devised copula models, offering valuable insights into flood encounter probabilities. Additionally, the innovative approach undergoes validation by comparison with three benchmark models which vary in dimensions and nature of variable interactions. Furthermore, the study conducts stochastic simulations, exploring both unconditional and conditional scenarios across different vine copula models. Applied in the Shifeng Creek watershed, China, the findings reveal that vine copula models are superior in capturing complex variable relationships, demonstrating significant spatial interconnectivity crucial for flood risk prediction in heavy-rainfall events. Interestingly, the study observes that expanding the model's dimensions does not inherently enhance simulation precision. The RDV-Copula method not only captures comprehensive information effectively but also simplifies the vine copula model by reducing its dimensionality and complexity. This study contributes to the field of hydrology by offering a refined method for analyzing and simulating multi-site flood flows.
- Open Access
- 25 November 2024
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 28 (22) , 4989-5009
Future response of ecosystem water use efficiency to CO2 effects in the Yellow River Basin, China
Ecosystem water use efficiency (WUE) is pivotal for understanding carbon–water cycle interplay. Current research seldom addresses how WUE might change under future elevated CO2 concentrations, limiting our understanding of regional ecohydrological effects. We present a land–atmosphere attribution framework for WUE in the Yellow River basin (YRB), integrating the Budyko model with global climate models (GCMs) to quantify the impacts of climate and underlying surface changes induced by CO2. Additionally, we further quantitatively decoupled the direct and secondary impacts of CO2 radiative and biogeochemical effects. Attribution results indicate that WUE in the YRB is projected to increase by 0.36–0.84 gC kg−1H2O in the future, with climate change being the predominant factor (relative contribution rate of 77.9 %–101.4 %). However, as carbon emissions intensify, the relative importance of land surface changes becomes increasingly important (respective contribution rates of −1.4 %, 14.9 %, 16.9 %, and 22.1 % in SSP126, SSP245, SSP370, and SSP585). Typically, WUE is considered a reflection of an ecosystem's adaptability to water stress. Thus, we analyzed the response of WUE under different scenarios and periods and various drought conditions. The results show a distinct “two-stage” response pattern of WUE to drought in the YRB, where WUE increases under moderate–severe drought conditions but decreases as drought intensifies across most areas. Furthermore, GCM projections suggest that plant adaptability to water stress may improve under higher-carbon-emission scenarios. Our findings enhance the understanding of regional ecohydrological processes and provide insights for future predictions of drought impacts on terrestrial ecosystems.
- 25 November 2024
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 28 (22) , 4989-5009
Open AccessFuture response of ecosystem water use efficiency to CO2 effects in the Yellow River Basin, China
Ecosystem water use efficiency (WUE) is pivotal for understanding carbon–water cycle interplay. Current research seldom addresses how WUE might change under future elevated CO2 concentrations, limiting our understanding of regional ecohydrological effects. We present a land–atmosphere attribution framework for WUE in the Yellow River basin (YRB), integrating the Budyko model with global climate models (GCMs) to quantify the impacts of climate and underlying surface changes induced by CO2. Additionally, we further quantitatively decoupled the direct and secondary impacts of CO2 radiative and biogeochemical effects. Attribution results indicate that WUE in the YRB is projected to increase by 0.36–0.84 gC kg−1H2O in the future, with climate change being the predominant factor (relative contribution rate of 77.9 %–101.4 %). However, as carbon emissions intensify, the relative importance of land surface changes becomes increasingly important (respective contribution rates of −1.4 %, 14.9 %, 16.9 %, and 22.1 % in SSP126, SSP245, SSP370, and SSP585). Typically, WUE is considered a reflection of an ecosystem's adaptability to water stress. Thus, we analyzed the response of WUE under different scenarios and periods and various drought conditions. The results show a distinct “two-stage” response pattern of WUE to drought in the YRB, where WUE increases under moderate–severe drought conditions but decreases as drought intensifies across most areas. Furthermore, GCM projections suggest that plant adaptability to water stress may improve under higher-carbon-emission scenarios. Our findings enhance the understanding of regional ecohydrological processes and provide insights for future predictions of drought impacts on terrestrial ecosystems.
- Open Access
- 25 November 2022
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 26 (22) , 5933-5954
Monitoring the extreme flood events in the Yangtze River basin based on GRACE and GRACE-FO satellite data
Gravity Recovery and Climate Experiment (GRACE) and its successor GRACE Follow-on (GRACE-FO) satellite provide terrestrial water storage anomaly (TWSA) estimates globally that can be used to monitor flood in various regions at monthly intervals. However, the coarse temporal resolution of GRACE and GRACE-FO satellite data has been limiting their applications at finer temporal scales. In this study, TWSA estimates have been reconstructed and then temporally downscaled into daily values based on three different learning-based models, namely a multi-layer perceptron (MLP) model, a long-short term memory (LSTM) model and a multiple linear regression (MLR) model. Furthermore, a new index incorporating temporally downscaled TWSA estimates combined with daily average precipitation anomalies is proposed to monitor the severe flood events at sub-monthly timescales for the Yangtze River basin (YRB), China. The results indicated that (1) the MLP model shows the best performance in reconstructing the monthly TWSA with root mean square error (RMSE) = 10.9 mm per month and Nash–Sutcliffe efficiency (NSE) = 0.89 during the validation period; (2) the MLP model can be useful in temporally downscaling monthly TWSA estimates into daily values; (3) the proposed normalized daily flood potential index (NDFPI) facilitates robust and reliable characterization of severe flood events at sub-monthly timescales; (4) the flood events can be monitored by the proposed NDFPI earlier than traditional streamflow observations with respect to the YRB and its individual subbasins. All these findings can provide new opportunities for applying GRACE and GRACE-FO satellite data to investigations of sub-monthly signals and have important implications for flood hazard prevention and mitigation in the study region.
- 25 November 2022
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 26 (22) , 5933-5954
Open AccessMonitoring the extreme flood events in the Yangtze River basin based on GRACE and GRACE-FO satellite data
Gravity Recovery and Climate Experiment (GRACE) and its successor GRACE Follow-on (GRACE-FO) satellite provide terrestrial water storage anomaly (TWSA) estimates globally that can be used to monitor flood in various regions at monthly intervals. However, the coarse temporal resolution of GRACE and GRACE-FO satellite data has been limiting their applications at finer temporal scales. In this study, TWSA estimates have been reconstructed and then temporally downscaled into daily values based on three different learning-based models, namely a multi-layer perceptron (MLP) model, a long-short term memory (LSTM) model and a multiple linear regression (MLR) model. Furthermore, a new index incorporating temporally downscaled TWSA estimates combined with daily average precipitation anomalies is proposed to monitor the severe flood events at sub-monthly timescales for the Yangtze River basin (YRB), China. The results indicated that (1) the MLP model shows the best performance in reconstructing the monthly TWSA with root mean square error (RMSE) = 10.9 mm per month and Nash–Sutcliffe efficiency (NSE) = 0.89 during the validation period; (2) the MLP model can be useful in temporally downscaling monthly TWSA estimates into daily values; (3) the proposed normalized daily flood potential index (NDFPI) facilitates robust and reliable characterization of severe flood events at sub-monthly timescales; (4) the flood events can be monitored by the proposed NDFPI earlier than traditional streamflow observations with respect to the YRB and its individual subbasins. All these findings can provide new opportunities for applying GRACE and GRACE-FO satellite data to investigations of sub-monthly signals and have important implications for flood hazard prevention and mitigation in the study region.
- Open Access
- 18 November 2021
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 25 (11) , 5951-5979
AI-based techniques for multi-step streamflow forecasts: application for multi-objective reservoir operation optimization and performance assessment
Streamflow forecasts are traditionally effective in mitigating water scarcity and flood defense. This study developed an artificial intelligence (AI)-based management methodology that integrated multi-step streamflow forecasts and multi-objective reservoir operation optimization for water resource allocation. Following the methodology, we aimed to assess forecast quality and forecast-informed reservoir operation performance together due to the influence of inflow forecast uncertainty. Varying combinations of climate and hydrological variables were input into three AI-based models, namely a long short-term memory (LSTM), a gated recurrent unit (GRU), and a least-squares support vector machine (LSSVM), to forecast short-term streamflow. Based on three deterministic forecasts, the stochastic inflow scenarios were further developed using Bayesian model averaging (BMA) for quantifying uncertainty. The forecasting scheme was further coupled with a multi-reservoir optimization model, and the multi-objective programming was solved using the parameterized multi-objective robust decision-making (MORDM) approach. The AI-based management framework was applied and demonstrated over a multi-reservoir system (25 reservoirs) in the Zhoushan Islands, China. Three main conclusions were drawn from this study: (1) GRU and LSTM performed equally well on streamflow forecasts, and GRU might be the preferred method over LSTM, given that it had simpler structures and less modeling time; (2) higher forecast performance could lead to improved reservoir operation, while uncertain forecasts were more valuable than deterministic forecasts, regarding two performance metrics, i.e., water supply reliability and operating costs; (3) the relationship between the forecast horizon and reservoir operation was complex and depended on the operating configurations (forecast quality and uncertainty) and performance measures. This study reinforces the potential of an AI-based stochastic streamflow forecasting scheme to seek robust strategies under uncertainty.
- 18 November 2021
- Published by Copernicus GmbH in Hydrology and Earth System Sciences
- Vol. 25 (11) , 5951-5979
Open AccessAI-based techniques for multi-step streamflow forecasts: application for multi-objective reservoir operation optimization and performance assessment
Streamflow forecasts are traditionally effective in mitigating water scarcity and flood defense. This study developed an artificial intelligence (AI)-based management methodology that integrated multi-step streamflow forecasts and multi-objective reservoir operation optimization for water resource allocation. Following the methodology, we aimed to assess forecast quality and forecast-informed reservoir operation performance together due to the influence of inflow forecast uncertainty. Varying combinations of climate and hydrological variables were input into three AI-based models, namely a long short-term memory (LSTM), a gated recurrent unit (GRU), and a least-squares support vector machine (LSSVM), to forecast short-term streamflow. Based on three deterministic forecasts, the stochastic inflow scenarios were further developed using Bayesian model averaging (BMA) for quantifying uncertainty. The forecasting scheme was further coupled with a multi-reservoir optimization model, and the multi-objective programming was solved using the parameterized multi-objective robust decision-making (MORDM) approach. The AI-based management framework was applied and demonstrated over a multi-reservoir system (25 reservoirs) in the Zhoushan Islands, China. Three main conclusions were drawn from this study: (1) GRU and LSTM performed equally well on streamflow forecasts, and GRU might be the preferred method over LSTM, given that it had simpler structures and less modeling time; (2) higher forecast performance could lead to improved reservoir operation, while uncertain forecasts were more valuable than deterministic forecasts, regarding two performance metrics, i.e., water supply reliability and operating costs; (3) the relationship between the forecast horizon and reservoir operation was complex and depended on the operating configurations (forecast quality and uncertainty) and performance measures. This study reinforces the potential of an AI-based stochastic streamflow forecasting scheme to seek robust strategies under uncertainty.