How do you determine the measurable "things" that describe the nature of our universe? To answer that question, researchers used CosmoFlow, a deep learning technique, running on a National Energy Research Scientific Computing Center supercomputer. They analyzed large, complex data sets from 3-D simulations of the distribution of matter to answer that question. The team showed that CosmoFlow offers a new platform to gain a deeper understanding of the universe.
American ingenuity is providing radical productivity improvements from advanced materials and robotic systems developed at the Department of Energy's Manufacturing Demonstration Facility at Oak Ridge National Laboratory.
Engineers can model heat distribution in reactor designs with fewer or no approximations.
The behavior of active magnetic liquids suggests new pathways to transport particles across surfaces and build materials that self-heal.
Researchers demystify how the nitrogenase enzyme breaks bonds to learn a better way to make ammonia.
The symmetrical light-gathering, energy-producing complex offers insights into how modern photosystems evolved.
Three types of water molecules form around a platinum-based ion, offering insights for waste processing and metal refining.
New technique enables more efficient and precise estimates of trends in ozone and other atmospheric constituents within selected geographical regions and timeframes.
Global data set shows monthly water use by irrigation, manufacturing, and other uses, helping researchers to analyze water use by region and season.
Researchers link root water uptake to root traits and assess (poor) performance of common models.
Ions at the edge of water, exposed to air, don't separate like they do when surrounded by water, offering insights for desalination and corrosion.
Predictions of the direct impacts of greenhouse gases must account for local temperature and humidity conditions.
Nutrients increasingly moving to the deep ocean with strong climate warming could lead to drastic drops in surface ocean life and fishery yields.
The MicroBooNE experiment demonstrates the use of machine learning to interpret images made by a liquid-argon particle detector.
Scientists developed a method to better distinguish the tracks that particles leave behind in liquid argon.
Six cameras are revolutionizing observations of shallow cumulus clouds.
Widespread fracturing during lake drainage triggers vertical shafts to form that affect the Greenland Ice Sheet.
The data system will allow for more detailed, consistent, and up-to-date global emissions trends that will aid in understanding aerosol effects.
Researchers can precisely study how different genes affect key properties in a yeast used industrially to produce fuel and chemicals.
Artificial intelligence on Summit to discover atomic-scale structures.
New model provides more accurate estimates of how fast microbes produce a mercury-based neurotoxin.
Scientists determine the accuracy of computational methods used to study the sulfate radical approach to purifying water.
Ultrafine aerosol particles produce bigger storm clouds and more precipitation than larger aerosols in pristine conditions.
Distortion of water droplet surface may increase the likelihood of the droplet freezing.
Unexpected molecular interactions involving water clusters have a subtle, yet profound, effect on extractants picking their targets.