The first machine learning-derived global-scale survey of Mercury’s hollows suggests they are young features that may be active and will continue to evolve.
machine learning & AI
Fast Adept Sea Ice Forecasts
Artificial intelligence facilitates an efficient, skillful surrogate of a coupled Arctic sea ice prediction model using generative diffusion.
Listening to Earth’s Subsurface with Distributed Acoustic Sensing
A new book examines how fiber-optic cables installed in boreholes can monitor seismic activity, fluid flow, subsurface temperatures, and more.
Machine Learning Enhances Image Analysis in Biogeosciences
Machine learning can enhance our ability to identify communities of microorganisms and how they change in response to climate change over time.
Unlocking the Power of Synthetic Aperture Radar for Geosciences
Due to its unique ability to monitor Earth’s surface, Synthetic Aperture Radar plays a pivotal role in revolutionizing the geosciences.
What’s On the Horizon for Open Access Geoscience Books?
On the first anniversary of their partnership, AGU and the Geological Society of London reflect on the GeoHorizons series and why open access books are valuable for the geoscience community.
Machine Learning Could Improve Extreme Weather Warnings
A deep learning technique could reduce the error in 10-day weather forecasts by more than 90%, allowing communities to better prepare for extreme events such as heat waves.
Forecasting Caldera Collapse Using Deep Learning
A deep learning model trained with geophysical data recorded during the well-documented 2018 Kilauea volcano eruption, Hawaii, predicts recurrent caldera collapse events.
Cultivating Trust in AI for Disaster Management
Artificial intelligence applied in disaster management must be reliable, accurate, and, above all, transparent. But what does transparency in AI mean, why do we need it, and how is it achieved?
Physics Meets Machine Learning for Better Cyclone Predictions
A new hybrid modeling approach combines physics-based and machine learning models to extend—and improve—path and intensity predictions of tropical cyclones.