Many environmental models struggle with capturing key hydrological processes during floods and droughts. Disconnection between process understanding and data driven extrapolation and challenges in monitoring and modeling of extreme events have highlighted the need to integrate diverse approaches to understand the role of surface water and groundwater in floods, droughts, and compound events. In this session, we welcome studies that leverage process based models, numerical frameworks, machine learning (ML), hybrid physics-ML approaches, and multi-model comparisons to address these gaps. Key themes include:  1) Temporal and spatial scaling challenges in heterogeneous hydrological systems, 2) Uncertainty quantification under changing climate, land use, and land cover scenarios, 3) Remote sensing and data assimilation techniques for extreme events, and 4) Role of snowmelt, frozen ground thaw, bank filtration, floodplain connectivity, and seasonal dynamics.  This session invites researchers applying integrated modeling approaches to water security, climate adaptation, and extreme event forecasting to submit their work.