Sediment is a fundamental yet under-quantified component of Earth's hydrological and environmental systems, shaping ecosystems, water quality, agriculture, infrastructure, and hazard risk. While water flow is now modeled operationally at continental scales (e.g., NOAA's National Water Model), sediment dynamics remain poorly constrained across scales. This session invites contributions advancing sediment transport modeling and prediction from the particle and reach scales through watershed, regional, national, and global scales. We welcome physics-based mechanistic models, data-driven and machine-learning approaches (e.g., LSTM, graph neural networks, foundation models), and hybrid physics-informed frameworks, along with studies on scale bridging, uncertainty quantification, data assimilation, and integration with hydrological and land-surface models. Observational, remote-sensing, and cyberinfrastructure contributions that enable cross-scale prediction, including efforts toward a national-scale sediment model or similar continental-scale systems, are especially encouraged. The session aims to catalyze a community dialogue on convergent approaches to sediment prediction.

• B - Biogeosciences
• C - Cryosphere
• H - Hydrology
• NH - Natural Hazards