Marcia McNutt (B.A. in physics, Colorado College; Ph.D. in Earth sciences, Scripps Institution of Oceanography) is a geophysicist and the 22nd president of the National Academy of Sciences.

From 2013 to 2016, she was editor-in-chief of Science journals. McNutt was director of the U.S. Geological Survey from 2009 to 2013, during which time USGS responded to a number of major disasters, including the Deepwater Horizon oil spill. For her work to help contain that spill, McNutt was awarded the U.S. Coast Guard’s Meritorious Service Medal.

She is a fellow of the American Geophysical Union (AGU), Geological Society of America, the American Association for the Advancement of Science, and the International Association of Geodesy. McNutt is a member of the American Philosophical Society and the American Academy of Arts and Sciences, and a Foreign Member of the Royal Society, UK, and the Russian Academy of Sciences.

In 1998, McNutt was awarded the AGU’s Macelwane Medal for research accomplishments by a young scientist, and she received the Maurice Ewing Medal in 2007 for her contributions to deep-sea exploration.

    However, massive increases in computing power will not be sufficient to overcome significant obstacles in running such models on exascale systems.  Issues include under- or unexposed parallelism, load-imbalance, and critical dependence on scarce memory bandwidth (low computational intensity).  Most weather prediction models use less than 5 percent of the peak computational capabilities of today’s CPU chips, and less than 2 percent of the latest generation GPU and MIC chips. Future chips are expected to continue this downward trend of diminishing computational benefit.  Growing awareness of these problems led to formation of a multi-agency (NOAA, NASA, NCAR, DoE, DoD) working group in the Earth System Prediction Capability (ESPC), and spawned meetings on the topic with industry.

    While hardware improvements will help, effective utilization of next generation systems with millions of compute cores will require adapting and rewriting applications to better prepare them for exascale. Weather and climate models are complex, multi-component models often designed separately, and then coupled or linked together. Lack of overall system design can limit parallelism, and make it difficult to understand, modify, test, and maintain them.  In addition, the scientific algorithms, grids, and formulations used are motivated by ease of development and scientific accuracy of the solution at the expense of computational efficiency, portability and maintainability. In the exascale era, a balance between accuracy and computational efficiency may be needed to further advance capabilities of our earth system models.

    To address these challenges requires cooperation and collaboration between modeling teams, computational experts and software designers. We invite your submissions in a wide range of topics including model and assimilation development, algorithms, software design, languages, high performance I/O, performance portability, and computing hardware.

Analogously, high-end scientific computing has entered a new disruptive period. As microprocessors have become ever smaller, they are approaching limitations dictated by power consumption and heat generation, requiring new and experimental processor designs that are the building blocks for the next generations of high-end computers. Achieving grand challenge science goals necessitates that the Earth system modeling community adapt to unprecedented changes in the computing landscape. These challenges include a plateau in system clock speeds and increased software complexity driven by computing system heterogeneity.

Developing and running ever more complex Earth system models requires modelers to support current machines while also preparing for Exascale architectures that will appear around 2020.  Success requires a mix of short-term optimization and long-term research in algorithms and computational approaches. We will solicit papers on all aspects of global climate and Earth system models on future Exascale computing systems. This includes fully-coupled models as well as their individual components, including the atmosphere, the ocean, sea ice, land ice, and the terrestrial biosphere. We will cover topics such as software complexity, programming models, algorithms for heterogeneous computers (e.g. CPU-GPU systems), data management, and analysis methods (both post simulation and in-situ).

Embracing the challenges now, rather than reacting later, is essential to sustain scientific progress. We hope this session will further enhance the exchange of ideas and identify new approaches to these complex problems at the intersection of the Earth system and computational sciences.