QDeepGR4J: Quantile-based ensemble of deep learning and GR4J hybrid rainfall-runoff models for extreme flow prediction with uncertainty quantification

2025-10-06

AI Summary

Hydrological models often exhibit low accuracy in extreme flow prediction and insufficient uncertainty quantification. To address this, we propose Quantile-Enhanced DeepGR4J—a hybrid modeling paradigm that embeds the conceptual GR4J model into a deep learning framework, replaces its routing module with a neural network surrogate, and integrates quantile regression with ensemble learning to enable multi-step streamflow forecasting and probabilistic uncertainty modeling. Evaluated on the CAMELS-Aus dataset, our method significantly improves extreme event detection capability and enhances probabilistic forecast calibration, as evidenced by improved prediction interval coverage probability (PICP) and mean prediction interval width (MPIW). Compared to benchmark models, it achieves superior uncertainty interval coverage, more reasonable interval width, and retains physical interpretability. Results demonstrate the framework’s effectiveness and practical utility for flood early warning systems.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Conceptual rainfall-runoff models aid hydrologists and climate scientists in modelling streamflow to inform water management practices. Recent advances in deep learning have unravelled the potential for combining hydrological models with deep learning models for better interpretability and improved predictive performance. In our previous work, we introduced DeepGR4J, which enhanced the GR4J conceptual rainfall-runoff model using a deep learning model to serve as a surrogate for the routing component. DeepGR4J had an improved rainfall-runoff prediction accuracy, particularly in arid catchments. Quantile regression models have been extensively used for quantifying uncertainty while aiding extreme value forecasting. In this paper, we extend DeepGR4J using a quantile regression-based ensemble learning framework to quantify uncertainty in streamflow prediction. We also leverage the uncertainty bounds to identify extreme flow events potentially leading to flooding. We further extend the model to multi-step streamflow predictions for uncertainty bounds. We design experiments for a detailed evaluation of the proposed framework using the CAMELS-Aus dataset. The results show that our proposed Quantile DeepGR4J framework improves the predictive accuracy and uncertainty interval quality (interval score) compared to baseline deep learning models. Furthermore, we carry out flood risk evaluation using Quantile DeepGR4J, and the results demonstrate its suitability as an early warning system.

Problem

• Predicting extreme streamflow events with uncertainty quantification
• Improving rainfall-runoff model accuracy using deep learning ensembles
• Developing flood early warning systems through uncertainty bounds

Innovation

• Hybrid deep learning with GR4J hydrological model
• Quantile regression ensemble for uncertainty quantification
• Multi-step prediction with uncertainty bounds for floods