A0196
Title: Managing water under a changing climate, helped by models
Authors: Marian Scott - University of Glasgow (United Kingdom) [presenting]
Abstract: Water is a critical national (and planetary) infrastructure, yet water quantity and quality continue to be modeled separately. Water is used for many purposes- from generating electricity to irrigating crops to providing recreational benefits, to goods production, and as part of cultural heritage, so it interacts with many parts of engineered and natural systems. Data on water quantity and quality are becoming increasingly rich with new real-time sensors and satellite missions (a data deluge), but water data are disparate, held in many places, and sometimes behind barriers, so the first statistical challenge is data linkage. The second gap is the underpinning science; there is no holistic view of hydrological cycles and their interactions, whether this be rainfall or groundwater, and the different scales at which they operate. Integrated water models (global and regional) are essential tools to understand the impacts of climate change, and one such type of model is a digital twin. A digital twin is a dynamic virtual representation of a system, which is fed by data and, at the heart, contains potentially multiple models, both physics-based and data-driven, as well as hybrid models. There are feedback loops, data assimilation, and updating to give temporal and spatial predictions of future states under different scenarios. Uncertainty remains the third statistical challenge, given the different unknowns.
