Assessing the Runoff Response to Vegetation Cover and Climate Change in a Typical Forested Headwater Watershed

Water Resources Management, Год журнала: 2025, Номер unknown

Опубликована: Март 1, 2025

Язык: Английский

---

Exploring Runoff Response to Simulated Rainfall: A Study of the Rising Limb of a Hydrograph on Sandy Slopes

Water Resources Management, Год журнала: 2025, Номер unknown

Опубликована: Янв. 27, 2025

Язык: Английский

---

Three-step Merging of Daily Multi-satellite Rainfall Estimates Based on Probability Density Function Matching and Dynamic Bayesian Model Averaging

Water Resources Management, Год журнала: 2025, Номер unknown

Опубликована: Март 19, 2025

Язык: Английский

---

Improving rainfall-runoff modelling using the fusion of satellite-based and gauge precipitation products in a data-sparse region

Hydrological Sciences Journal, Год журнала: 2025, Номер unknown

Опубликована: Апрель 17, 2025

Язык: Английский

---

Towards Interpreting Machine‐Learning Models for Multi‐Step Ahead Daily Streamflow Forecasting

Hydrological Processes, Год журнала: 2025, Номер 39(5)

Опубликована: Май 1, 2025

ABSTRACT Streamflow forecasting using interpretable machine learning methods (MLs) for exploring runoff processes has received a lot of attention. However, multi‐step ahead daily streamflow considering antecedent as an input various MLs is very limited. Thus, three in the Huaihe River basin China during 2002–2020, including eXtreme Gradient Boosting (XGBoost), long short‐term memory neural network (LSTM) and convolutional (CNN) with SHapley Additive exPlanations (SHAP) method, were implemented to study role potential controlling factors, streamflow, soil moisture vegetation growth, at lead times 0–6 days. The performances decreased times. Specifically, LSTM model performed best 0–3 days, followed by CNN XGBoost. was superior XGBoost models when time greater than 3 optimal 0.71–0.97, 311.45–674.27 m /s, 0.84–0.97 0.75–0.97 according Nash‐Sutclife efficiency, root‐mean‐square error, correlation coefficient Kling‐Gupta respectively. results varied across different consistently dominated processes, particularly models. significant depth 28–100 cm leaf area index low gradually emerged increased models, even outranking importance streamflow. Furthermore, interpretability demonstrated validated through infiltration uncertainty analysis. Overall, great enhance our understanding basin‐scale processes.

Язык: Английский

---

Prediction of non-stationary daily streamflow series based on ensemble learning: a case study of the Wei River Basin, China

Stochastic Environmental Research and Risk Assessment, Год журнала: 2024, Номер unknown

Опубликована: Дек. 18, 2024

Язык: Английский

---