Data acquisition and assimilation enabled by machine learning (ML), artificial intelligence (AI), and advanced methods including experimental/network design/optimization, unsupervised learning (deep learning), leveraging advanced hardware (e.g., edge computing). This development is to enable solving the science questions regarding the human and climatic factors that interact with and drive global water scarcity.
Science Challenge:Data-model integration in the context of complex high dimensional hierarchical nonlinear structure. A self-evolution, comprehensive data fusion platform; building data fusion approaches from the ground up into a web-based, globally accessible platform.
Rationale:AI's superpower is fueled by data. The successes of applying AI for building integrative predictive models rely on data accessibility, adequacy, and quality, which are generally promoted for data managed in adherence to FAIR (findable, accessible, interoperable, and reusable) guiding principles. Even considering the ability to employ FAIR data, it is still non-trivial to integrate data for scientific discovery (data fusion) given the usually high-dimensional and hierarchical nonlinear cross-dependence structure nature of some domains (e.g., complex large-scale water modeling systems). Recent advances in computational resources, observation and monitoring methods, and sensing tools, bring unprecedented opportunities in data-model integration or data fusion but introduce new challenges that require a systematic approach to provide heuristic value.