深度学习地球系统模型与CMIP6模型相比更具竞争力,并且使用的计算资源更少。
machine learning & AI
Unveiling What’s Under the Hood in AI Weather Models
Artificial intelligence models have improved weather forecasting, but their inner workings are largely opaque. A new approach could make their predictions more interpretable by scientists.
Unexpected Carbonate Phase Revealed by Advanced Simulations
Advanced simulations reveal a new calcium carbonate phase whose unusual elastic behavior may explain puzzling seismic and electrical anomalies beneath ancient continents.
Deep Learning Goes Multi-Tasking
In hydrological modeling, predicting multiple tasks helps in identifying physical rules and generalizations.
As Simple as Possible: The Importance of Idealized Climate Models
As models that simulate Earth’s climate system become increasingly complex, the use of simpler and more flexible idealized models remains important for science and education.
Machine Learning Simulates 1,000 Years of Climate
The Deep Learning Earth System Model is competitive with CMIP6 models and uses less computational power.
Nearly 94 Million Boulders Mapped on the Moon Using Deep Learning
Scientists used a deep learning algorithm to map the size and location of nearly 94 million boulders on the lunar surface, highlighting differences in boulder densities and size distributions.
Machine Learning Model Flags Early, Invisible Signs of Marsh Decline
Decreases in underground plant biomass could signal future marsh loss and prompt conservation measures.
Rock Solid Augmentation: AI-Driven Digital Rock Analysis
Boosting digital rock images with AI-powered augmentation and quality analysis could improve subsurface engineering decisions.
Storm Prediction Gets 10 Times Faster Thanks to AI
Forecasters hope new algorithms will lead to earlier warnings of when dangerous weather is on the way.
