A new study couples an emission and transport scheme of pollen from vegetation, and explores pollen’s evolution in different atmospheric conditions and its impacts on clouds and precipitation.
A new non-column based spectral element implementation of cloud microphysics enables full 3D flexibility in computing clouds and improves computational efficiency.
Automated Machine Learning liberates domain scientists from selecting learners and hyperparameters and discovers the importance of atmospheric trace gases for improving surface PM2.5 estimates.
A new numerical setup demonstrates that aerosols could affect clouds, and hence the radiation budget, thousands of kilometers from their location.
The capability to fully model the global methane cycle advances the international climate science community’s ability of providing essential evidence to underpin climate mitigation policy.
A new-generation weather radar and a massive supercomputing system enables forecasts of storms refreshed every 30 seconds, a significant development in severe weather prediction.
Online training produces more accurate and stable machine-learned models than classic offline learning from big data sets.
Using machine learning to represent sub-grid processes in weather and climate models holds promise, but also faces challenges. Incorporating physical knowledge can help.
Researchers present a new approach to modeling the stochastic mixing process of convection using a machine learning technique.
A probabilistic deep learning methodology that learns from climate simulation big data offers advantageous seasonal forecasting skill and crucial climate model diagnosis information at a global scale.