A new study published in Advances in Atmospheric Sciences explores how machine learning and statistical techniques can refine forecasts of photovoltaic by correcting errors in weather models.

A research team led by Prof. WANG Min from the Institute of Vertebrate Paleontology and Paleoanthropology of the Chinese Academy of Sciences has discovered two bird fossils in Jurassic-era rocks from Fujian Province in southeast China. These rocks date back approximately 149 million years. The fossils fill a spatiotemporal gap in the early evolutionary history of birds and provide the evidence yet that birds were diversified by the end of the Jurassic period.

Scientists from Prof. XIE Qi's lab at the Institute of Genetics and Developmental Biology of the Chinese Academy of Sciences identified two key genes and the strategy in sorghum's resistance to Striga.

A comprehensive seven-year study across 20 Chinese provinces has revealed a concerning rise in antimicrobial resistance (AMR) among diarrheagenic Escherichia coli (E. coli), a major foodborne pathogen.

Researchers explored Mendelian randomization to identify novel drug targets for metabolic dysfunction-associated steatotic liver disease (MASLD). This approach could revolutionize therapies for a condition affecting over one-third of adults globally.

Data-driven networks influence convective-scale ensemble weather forecasts

Spatial-Temporal Forecasting, as an important research direction in Geospatial Artificial Intelligence (GeoAI), plays a central role in integrating surveying, mapping, geographic information technologies, and artificial intelligence. It promotes intelligent innovation and enables the deployment of spatial intelligence technologies across diverse real-world application scenarios. In January 2025, a research team from the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, reviews the history of GeoAI-driven spatiotemporal forecasting, and offers a comprehensive survey of spatial-temporal prediction models including statistical learning, deep learning, and generative large models. This study critically evaluates existing forecasting research and identifies five key challenges in the field. Finally, four future trends and research directions were proposed to advance geospatial intelligent prediction technologies.

A new study has uncovered a novel P-type PPR protein, BoYgl-2, which plays a crucial role in chloroplast RNA editing and chlorophyll biosynthesis in cabbage. 

Research team led by Chuliang Weng introduces D2-GCN, a groundbreaking disentangled graph convolutional network that dynamically adjusts feature channels for enhanced node representation, outperforming traditional methods in node classification tasks.

Research team led by Chuliang Weng introduces D2-GCN, a groundbreaking disentangled graph convolutional network that dynamically adjusts feature channels for enhanced node representation, outperforming traditional methods in node classification tasks.