AI for Water

Abstract:
Environmental sustainability is among the greatest challenges in the 21^st century. Recent IT and data revolution offers unprecedented opportunities for potential breakthroughs in our ability to understand and manage complex systems, interactions, and sustainability. These possibilities, however, are largely unrealized, primarily because I argue the technologies to produce big data are not matched by technologies to analyze them. We have reached a point in the evolution of science where many of the complex problems that exist can only be solved with a change in the fundamental scientific and technological infrastructure.

In this talk, I present an integrated, cross-cutting research initiative on “big data” and data-enabled water science and engineering that has led to the development of the first ever global water modeling technology – MAGNET 4 WATER (Multiscale, Adaptive, Global Network 4 Water). The technology has the potential to shift the water research, education, and management paradigm, drastically reduce the costs of water system investigations, and transform how water is managed, how knowledge is created, and even how people learn and collaborate. 

MAGNET 4 WATER is built on the achievements at the interface environmental science and engineering, data science, IT and computer science and engineering, human cognitive science, and emerging concepts in economics. The platform integrates fundamental and technological advances in environmental and hydrologic modeling, remote sensing and geographical information systems, networking and communication, data intensive and distributed computing, and an international network of data services, including high resolution data services provided by the US Geological Survey, NASA, National Oceanic and Atmospheric Administration, US Department of Agriculture, US Environmental Protection Agency, Federal Emergency Management Agency, US Army Corps of Engineers, US Fish & Wildlife Service, Environment Canada, British Geological Survey, German Federal Institute for Geosciences and Natural Resources, French Geological Survey, Geoscience Australia, and United Nations Educational, Scientific, and Cultural Organization.

The platform allows, for the first time, realtime and participatory modeling, visualization, analysis, reporting, and publication, providing a new, unique problem-solving environment and collaborative workspace for global environmental and water community – for students, researchers, professors, consultants, citizens, planners, managers, regulators, and decision makers. Users can zoom in anywhere in the world to use big data to perform a “virtual site visit” or build almost instantly a model, analysis, and visualization that can be further refined with the user’s own data. Modeling regimes include: groundwater, watersheds, river and channels, lakes and reservoirs, storm water, sanitary and urban catchments, and water distribution systems.

The platform further allows users to publish instantly data and model results to the world to showcase their capabilities and achievements – in high impact 3D visualization, animation, and intelligent report. The instant publication provides an effective mechanism of spontaneous, mass collaboration in developing cyber-enabled environmental observatory network from the bottom-up by the global community.

The ability to visualize surface and subsurface, groundwater flow and surface water flow and quality, and contaminant transport and geology in a real life setting sparks pivotal insights into the complex interrelationships among components of the environment and human activities and an intuitive grasp of implications of management actions and policy decisions that can’t be readily obtained otherwise.

Dr Li will demonstrate the platform at the conference with applications related to sustainable water resources management, pollution control, groundwater contamination, flood inundation, and green infrastructure.