As the climate changes, many are studying sunlight reflection as a potential method of cooling. Such climate intervention could be deployed in many possible ways, including in scenarios where not every actor agrees on the strategy of cooling. These scenarios are so diverse that to explore all of them using earth system models proves to be too costly. In this paper, we develop a simplified climate model that allows users to easily explore climate intervention scenarios of their choice.

The paper addresses a two-order-of-magnitude discrepancy in deposition velocities of accumulation-mode aerosols measured with different methods. This uncertainty affects current atmospheric deposition models. By explicitly accounting for gas-particle transition and explicitly evaluating the observed quantities we could reproduce the observations . Resolving the discrepancy, reduces uncertainties in simulated concentrations and fallout.

Groundwater plays a key role in land–atmosphere water and energy exchange, yet it is often simplified in large-scale Earth system models. We review 20 years of efforts to couple the groundwater model ParFlow with land surface and atmospheric models, showing how groundwater dynamics shape terrestrial fluxes. We also present an updated coupling framework that enhances model performance and flexibility, and outline a modular strategy to guide future development.

Chemistry transport models (CTMs) are critical in environmental assessments, but their computational cost often limits direct use in decision-making. We evaluate data-driven model discovery and reduction methods as reduced-order models for CTM simulations, showing they can reconstruct and forecast changes in global ozone distribution from supersonic aircraft emissions for several years at a fraction of the CTM cost while also being more accessible.

Satellite-based greenhouse gas measurements can be used in “inverse models” to improve emissions reporting, but one of the key components, the simulations of atmospheric transport, struggle to scale to large datasets. We introduce the model GATES, an AI-driven emulator that outputs transport plumes 1000× faster than traditional models. Applied to Brazil’s methane emissions, GATES produces estimates consistent with physics-based methods, offering a scalable path for timely emissions monitoring.

We present a new deep learning framework designed to assess seasonal climate predictability by identifying the key predictors that influence climate variability across different regions. This tool enhances understanding of how remote areas are connected through climate interactions and providing accurate and explainable seasonal predictions. Our results demonstrate its potential to support more reliable and informed climate services at both regional and global scales.

We developed a new method to guide the simulated atmosphere in an Earth system model so it better reflects real-world weather. By adjusting temperature and humidity, it reduces unwanted side effects and improves the realism of rainfall, energy flows, land–surface conditions, and extreme storms such as cyclones and atmospheric rivers. This makes the model more useful for testing its performance, understanding high-impact weather events, and creating reliable training data for machine learning.

We developed a model to simulate seafloor biogeochemical processes across a wide range of marine environments, from shallow coastal zones to deep-sea sediments. From this model, we derived a set of simple equations that predict how carbon, oxygen, and alkalinity are exchanged between sediments and overlying waters. These equations provide an efficient way to improve how ocean models represent seafloor interactions, which are often missing or overly simplified.

The characterization of vegetation at two sites proved insufficient to adequately simulate the evapotranspiration. A dry surface layer was implemented in the land surface model SURFEX-ISBA (Externalized Surface-Interactions Soil-Biosphere-Atmosphere) v9.0. It is compared to simulations without a soil resistance. The application to an alfalfa site and a natural grass site in semiarid conditions results in an improvement in the estimation of the latent heat flux. The surface energy budget and the soil and vegetation characteristics are explored in detail.

The newly developed model AccRo (Accumulation-based Runoff and Pluvial Flood Estimation Tool) is a computationally efficient method to derive key parameters for estimating pluvial flood hazards. Here, we compare results of AccRo with the data of two hydrodynamic models for different cases. We find that AccRo is able to represent the simulations of the hydrodynamic models in high quality, but with much lower computational effort, making it a valuable tool for assessing pluvial flood hazards.

Regional coupled human–natural systems models are essential for regional sustainability. We developed a new model, CHANS-SD-YRB, using System Dynamics for the Yellow River Basin in China, which faces severe human-water conflicts. The model links 10 components, including Population, Economy, Energy, Food, Water, Sediment, Land, Carbon, and Climate to simulate basin's key human-natural interactions. The model is applicable for sustainable development through scenario analyses and predictions.

Wildfires can strongly affect local weather by heating the land surface and changing wind and turbulence near the ground. This study used computer simulations together with field observations from a mountain wildfire in China to examine how different schemes represent fire–weather interactions. The results show that one scheme performs better, helping improve wildfire prediction in complex terrain.

We assess the implementation of additional organic carbon pathways into a global setup of a numerical model, which simulates the ocean circulation, sea ice, and biogeochemical processes. With a focus on the Arctic Ocean, this model tracks the temporal and spatial dynamics of phytoplankton, exudation of organic carbon, and its aggregation to so-called transparent exopolymer particles. We evaluate the simulation using measurements from ship-based and remote-sensing campaigns in the Arctic Ocean.

We present the development of a global chemistry-climate coupled model BCC-GEOS-Chem v2.0, with improved representation of comprehensive troposphere-stratosphere chemistry and new capability to account for radiative-cloud feedbacks from short-lived climate forcers. The development of the BCC-GEOS-Chem v2.0 provides a useful framework for investigating climate–chemistry interactions and for future projections of global atmospheric chemistry

This paper describes the ocean BiogeochemicAl Model for Hypoxic and Benthic Influenced areas (BAMHBI). BAMHBI is a moderate complexity marine biogeochemical model that describes the cycling of carbon, nitrogen, phosphorus, silicon and oxygen through the marine foodweb. BAMHBI is a stand-alone biogeochemical model that can be coupled to any hydrodynamical model and is particularly appropriate for modelling low oxygen environments and the generation of sulfidic waters.

Tropical land surface processes shape the Earth's climate, but models often lack accuracy in the tropics due to limited data for validation. We improved the Noah with Multi-Parameterizations (Noah-MP) land surface model for the tropics using data from forests in Panama and Malaysia. Calibration enhanced simulations of energy and water fluxes, and revealed key vegetation and soil parameters, as well as future directions for model improvement in tropical regions.

Geochemical anomalies associated with mineralization represent one of the most significant types of geo-anomalies for mineral exploration.This study develops a AI method that combines geological knowledge with a flexible deep learning model. It helps identify geochemical anomaly patterns more accurately and reliably by focusing on key features like ore-controlling faults. The model's decisions are easier to understand through visual explanations, increasing transparency and trust in the results.

We develop a novel framework using the convection-permitting Weather Research and Forecasting (WRF) model to assess how stratospheric aerosol injection, a solar climate intervention strategy, affects future convective weather over the contiguous U.S. Results demonstrate the feasibility and scientific potential of this approach for evaluating weather-scale impacts and suggest that such intervention may mitigate changes in temperature, precipitation, and convective activity due to warming.