除非我们迅速实现净零排放,否则即便在温室气体排放量减少之后,气候也将逐渐接近不可逆转的境地。
Modeling
Weathering of Rocks Can Release Carbon Dioxide
New research upends the notion that the weathering of rocks mainly removes CO2 from the atmosphere. Rocks can also be carbon sources, releasing as much CO2 as Earth’s volcanoes.
A Crystal Ball for the Carbon Cycle, But a Cloudy One
Carbon cycle models quantify relationships between emission scenarios and resulting atmospheric concentrations, but are the projections credible? New analyses find grounds for both hope and concern.
Climate Tipping Points Could Be Triggered by “Committed Warming”
Unless we rapidly reach net zero emissions, the climate will inch closer to a point of no return—even after greenhouse gas emissions are reduced.
Machine Learning Identifies Source Volcanoes of Ash Deposits
Tracing ash layers from explosive eruptions back to their source volcanoes is needed to evaluate hazards to population and aviation, a problem addressed by a new machine learning classification method.
Reporting Model Results Even When They Cannot (Yet) be Tested
Models simulating the nitrogen cycle track its multiple chemical forms but tend to report a subset that can be compared with available field measurements.
Carbon Dioxide’s Effect on Mountain Climate Systems
Greenhouse gases are rising in the atmosphere. But how will precipitation patterns change as climate systems rise over mountain chains?
Cumulus Cloud Botany in Large Domains
A new study provides a sample of shallow cumulus clouds simulated in domains 150-kilometers wide, enabling investigations of their structure and organization.
Humans Have Boosted Atmospheric Mercury Concentrations Sevenfold
A new baseline of volcanic contributions to the global mercury cycle reveals how drastically human activities have increased the element’s concentration in the atmosphere.
Advancing AI and Machine Learning Beyond Predictive Capabilities
A new cross-journal special collection invites contributions that unlock the next frontier in hydrology and Earth sciences through artificial intelligence and machine learning.