The partnership to apply artificial intelligence to improve complex systems is part of a U.S. Department of Energy Office of Science $4.2 million, three-year grant.

PNNL scientists joined international leaders in artificial intelligence research to discuss the latest advances, opportunities, and challenges for neural information processing—the foundation for AI.

Red teaming for CPS, the process of challenging systems, involves a group of cybersecurity experts to emulate end-to-end cyberattacks following a set of realistic tactics, techniques, and procedures.

Comparing large datasets from coordinated aircraft flights provides critical information about data researcher use in a range of analyses.

Data Scientist Sam Chatterjee to chair security and defense specialty group of the Society for Risk Analysis.

Urbanization-induced land cover and anthropogenic aerosol changes result in stronger thunderstorms with enhanced precipitation in the Houston area.

The workshop brought together 184 scientists to explore research opportunities for nuclear physics in artificial intelligence.

Long term power grid planning needs to consider regional dependencies in climate adaption plans.

PNNL computational biologists, structural biologists, and analytical chemists are using their expertise to safely accelerate the design step of the COVID-19 drug discovery process.

Integrating information on irrigation processes into Earth system models allows for more accurate representation of the global water cycle.

The effects of too much CO2 can cause clouds covering the oceans to dissipate and be unable to reform.

Hector, a reduced complexity climate model, can be calibrated to emulate 30 different complex, computationally expensive Earth System Models.

Using public data from the entire 1,500-square-mile Los Angeles metropolitan area, PNNL researchers reduced the time needed to create a traffic congestion model by an order of magnitude, from hours to minutes.

New experiments show organic aerosol can be lost from the atmosphere faster than most models predict.

PNNL researchers have shown an improved binarized neural network can deliver a low-cost and low-energy computation to help the performance of smart devices and the power grid.

Historical weather data informs projections of global yields of economically important crops.

To accurately model future U.S. agricultural production, changes in international crop yields cannot be ignored.

The project received an Innovative and Novel Computational Impact on Theory and Experiment (INCITE) award, a highly competitive U.S. Department of Energy Office of Science program.

Researchers introduced a simulated carbon cycle to the Energy Exascale Earth System Model, broadening its utility and enabling new research directions.

Ecosystem feedback fire model opens new investigative pathways into impacts of fire on future climate.