Using artificial intelligence (AI) to help collect, understand, and analyze large sets of information has the potential to revolutionize our ability to monitor, understand, and predict processes in Earth’s systems.
Researchers and scientists are working together to apply artificial intelligence and modeling techniques such as machine learning (ML) to advance the Earth and Environmental sciences. Specifically, a group of scientists and experts aims to merge Modern technology in making Earth system models, observations, and theories – in addition to providing computational capabilities that can provide speed, accuracy, and more informed and agile decision-making.
in Collaborative effort between the US Department of Energy’s (DOE) Bureau of Biological and Environmental Research (BER) and the US Department of Energy’s (DOE) Advanced Scientific Computing Research Program, as well as with community experts, Artificial intelligence The Workshop on Earth System Predictability (AI4ESP) took place from October through December 2021. The five-week virtual workshop explored the challenges and infrastructure development that would better integrate a combination of technological capabilities and human activities in the field and laboratories with computational resources. . BER developed the process as the “Model Experiment” model, or ModEx.
“Effective improvements in Earth system predictability require dramatic developments across the ModEx environment. This workshop provided an interdisciplinary and cross-functional opportunity for the scientific and application communities to collaborate in order to understand the required developments,” said Niki Hickmon, AI4ESP Program Leader. Co-Director of Operations for the DOE’s Office of Atmospheric Radiometry at the Science User Facility at DOE’s Argonne National Laboratory.
According to a newly released report summarizing the AI4ESP workshop, the event brought together more than 700 participants from the private and public sectors, with representatives from Earth and Environmental Sciences, Computing and Artificial Intelligence. Together, some 100 experts designed the workshop based on 156 white papers submitted by 640 authors from 112 institutions around the world.
The information has been narrowed down to 17 topics related to the integral water cycle and extreme weather events in that cycle. Experts discussed nine focal points related to Earth System Forecasts, including sessions involving hydrology, watershed science, and coastal dynamics; atmosphere, land, ocean, and ice; Climate variability and extreme events. During the sessions, participants explored the potential of AI to unlock scientific discovery using tools such as neural networks, knowledge-informed machine learning, AI structures and co-design.
In each session, researchers identified challenges that underpin the need for a revolution in AI technology and the infrastructure that can be applied to manage complex business in the field of environmental science.
“We need new AI methodologies that incorporate an understanding of processes and respect physical laws to make predictions of Earth system behavior scalable, reliable, and workable under different systems,” said Charu Varadharajan, a research scientist at DOE’s Lawrence Berkeley National Laboratory who leads the lab’s Earth lab. future climate. Artificial intelligence and data program field. “This workshop is unique in discussing how AI can improve models, observations, and theory incorporating the DOE’s ModEx approach.”
“The workshop and report allowed us to develop two-year, five-year and 10-year goals for developing the integrative framework for each focal point. We also identified priorities for Earth science, computational science, programmatic and cultural changes that will include the AI4ESP mission.”
Experts have put together a comprehensive list of opportunities where AI research and development can help address some of the biggest challenges facing Earth sciences. These challenges include managing and analyzing large data sets to enhance the ability to monitor and predict extreme events and to promote the integration of human activities into theory and models.
“One of the most exciting modeling opportunities is the development of new hybrid models that incorporate both process-based and ML-based modules,” said Forrest Hoffman, group leader for computational geosciences at DOE’s Oak Ridge National Laboratory. “These modeling frameworks allow the incorporation of data about poorly understood processes that can improve accuracy and often improve computational performance of Earth system models, enabling further simulations and analysis within given resource limits.”
Workshop participants also identified several priorities for tackling computational challenges – including advances in both artificial intelligence and machine learning, algorithms, data management and more. The outcome of these priorities can help develop a technology infrastructure that is efficient, accurate, strategic, and relevant, and reaches beyond resources.
There is also a need for programmatic and cultural changes to support a more coherent mission across various scientific and government agencies, as well as a trained workforce that can successfully integrate technology into their research and humanitarian activities. Experts have identified solutions that would include AI research centers for the environmental sciences, frameworks that enable shared services across different communities, and ongoing training and support missions.
2021 AI4ESP Workshop Participants continue to discuss community computational activities, including those of the American Geophysical Union and the American Meteorological Society. Stay tuned for additional workshops and meetings in the near future – further collaboration, sharing and framework development will continue to advance AI4ESP’s mission.
Nikki Hickmon et al., Report of the Artificial Intelligence for Earth System Prediction (AI4ESP) Workshop, (2022). doi: 10.2172/1888810
Argonne National Laboratory
the quote: New Report Details AI Infrastructure for Earth System Predictability (2023, January 24) Retrieved January 24, 2023 from https://phys.org/news/2023-01-ai-infrastructure-earth.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.