July 22, 2019

Core themes

I. Assessing and Mitigating Natural Hazards and Climate Change

Although natural hazards, including those associated with climate change, can be enormously destructive, our ability to assess these hazards and develop strategies to adapt to and mitigate their consequences is far from adequate. One key need is to improve the analytical and computational methods used to model the chaotic and complex nonlinear systems involved. A second is to improve our ability to assess the hazards involved, including quantifying the uncertainties in estimating the occurrence of such extreme events and the resulting losses, both from a social and economic perspective. A third is to evaluate the costs and benefits of alternative adaptation and mitigation approaches and thus help society formulate strategies to address these problems.

II. Developing and Managing Natural Resources for Sustainability

Societal sustainability depends on responsible long-term stewardship of natural resources including fossil fuels and water. Essential components are exploration for resources and their careful development. These will require further advances in seismic imaging techniques used in oil and gas exploration. Collaborations between geoscientists, mathematicians, and computational scientists focused on the challenging problems of handling and analyzing enormous volumes of data are essential. Developing strategies to manage the development of energy, water, and other resources to maximize societal good while minimizing adverse effects, is a major challenge in complex systems analysis. This challenge is the reverse of protecting society from natural hazards, in that it involves developing cost-effective strategies for protecting nature and our environment in the possible presence of tipping points and regime transitions, yet it is amenable to similar approaches.

III. Exploring and Observing Earth Structure and Processes

Understanding earth structure and processes is a fundamental scientific goal with important consequences for studying many of the natural resources on which societal sustainability relies. The Earth system involves the interaction of many physical processes operating through a myriad of scales in a heterogeneous, complex environment Describing these processes and their interactions require descriptions of great complexity and detail. At the same time, certain aspects of the system are observable only indirectly or not at all, even with powerful new observing tools coming on line. Hence progress requires major innovative advances in data assimilation and experimental design, harnessing heterogeneous data types to improve and validate predictions, and computationally scalable solutions for massive inverse problems. These topics are at the intersection of data acquisition, inverse theory, uncertainty analysis, and optimization and thus offer the opportunity for significant advances in mathematics, statistics, and computational science.

IV. Modeling and Simulating Earth Structure and Processes

The crucial task of better predicting the behavior of the earth system will require advances in both the models for earth processes and structure, and the numerical techniques for simulating these models. The models will involve novel combinations of both deterministic and stochastic mechanisms, and integration of multiple spatial and temporal scales. Reliable simulations of these models will demand novel developments in both hardware technology and in numerical methods that can leverage such architectures. These methods must be highly accurate, cost-efficient, feature algorithmic simplicity while allowing for non-uniform spatial and temporal resolution. Concurrently, software libraries must be developed for geoscientists to easily implement such advances in computational mathematics on modern hardware platforms.

V. Developing & Strengthening Cross-Disciplinary Expertise

These tasks involve integrating the expertise of geoscientists together with that of mathematicians, statisticians and computer scientists. This requires a process in which different groups learn the others’ methods, including both strengths and limitations, to develop important insights and solve key challenges. Toward this end, we seek to promote both avenues for collaborative research and interdisciplinary education. The first involves approaches including educating and enabling software users, and developing libraries and community codes for reproducible research. The second involves both interdisciplinary graduate and postdoctoral training, and developing undergraduate curricula, with the goal of training a new generation of researchers with both the breadth and depth of expertise to make major advances.