Artificial intelligence (AI) is playing a bigger role in weather forecasting, often competing with physical models. However, combining AI models with data assimilation, a process that improves weather forecasts by incorporating observation data, is still relatively unexplored. This study explored the coupling of ensemble data assimilation with an AI weather prediction model, ClimaX, which succeeded in employing weather forecasts stably by applying techniques conventionally used for physical models.

Studies on plastic pollution have emerged as a rapidly growing field of research. Modelling microplastic transport in estuaries stems from their complex hydrodynamics and diverse particle behaviours affecting the dispersion and retention of microplastics. We review key modelling approaches applied in estuaries analysing their set-ups and parameterizations. We provide recommendations and future directions to improve the accuracy and modelling strategies for estuarine microplastic research.

Ozone (O₃) pollution harms rice production and threatens food security. To understand these impacts, we calibrated a crop model using unique data from experiments where rice was grown in open fields under controlled O₃ exposure (free air). This is the first time such data have been used to improve a model's ability to predict how rice responds to O₃ pollution. Our work provides a more accurate tool to study O₃'s effects and guide strategies to protect agriculture.

The increasing availability of oceanographic physical and chemical data necessitates accompanying methods for optimizing use of these data. This project produced algorithms (PyESPERs) for estimating biogeochemical seawater properties in Python, a freely available coding language. These algorithms were based on empirical seawater property estimation routines (ESPERs), which were originally written in the proprietary MATLAB coding language and can be used in studies of marine carbonate chemistry.

This paper presents GEOSPACE and its 1D implementation: an open-source tool for simulating soil–plant–atmosphere continuum (SPAC) interactions. Using object-oriented programming, GEOSPACE modularizes SPAC processes, focusing on infiltration, evapotranspiration, and root water uptake. The 1D deployment integrates plant transpiration, soil evaporation, and root growth, providing a flexible and validated framework for ecohydrological modeling and applications.

This work uses machine learning to predict saturation vapor pressures of atmospherically-relevant organic compounds, crucial for partitioning of secondary organic aerosol (SOA). We introduce a new method using graph convolutional neural networks, in which molecular graphs enable the model to capture molecular connectivity better than with non-structural embeddings. The method shows strong agreement with experimentally determined vapor pressures, and outperforms existing estimation methods.