This session explores the forefront of hybrid modeling that integrates process-based hydrologic and water quality models with AI and machine learning (ML) techniques to improve predictions and management of water resources under climate change stresses. Hybrid modeling leverages the physical realism of process-based models alongside the adaptive learning and data-driven capabilities of ML – including frontier AI such as foundation models and Large Language Models (LLMs) to overcome limitations such as data scarcity, structural deficiencies in process-based models, and the challenges of simulating non-linear and complex hydrological processes.
Contributions are sought to advance the conceptual and practical understanding of hybrid models applied to hydrologic and water quality simulation, especially those focusing on:
• Improving streamflow and pollutant transport predictions in diverse hydro-climatic and data-scarce regions
• Hybrid approaches for simulating nonpoint source pollution and watershed-scale water quality dynamics
• Regional and catchment-scale applications demonstrating scalability, transferability, and robustness of hybrid frameworks
• Real-time forecasting and operational water management enabled by hybrid modeling
• Policy-relevant applications linking model outputs to climate adaptation, water allocation, and resilience strategies
• Methodological challenges and solutions regarding model interpretability, uncertainty quantification, computational efficiency, and equitable technology access
This session will foster interdisciplinary dialogue on designing, implementing, and applying hybrid modeling approaches that enhance hydrologic prediction and water quality assessment to support sustainable water resource management and climate resilience.