A new AI-powered imaging technique that processes raw telescope data in a way, using "closure invariants" to automatically correct for common measurement errors.
Towards independent event horizon imaging of the supermassive black holes in M87 and the Milky Way · Nithyanandan Thyagarajan, Samuel Lai, Ivy Wong et al. · arXiv:2608.19675
A new physics-based model that uses satellite and weather data to precisely map how much water is available to plant roots in farm fields.
SoilWaterNow: Soil water nowcasting for mapping plant available water (PAW) across paddocks for improved on-farm decision-making · Yi Yu, Mikaela J. Tilse, Patrick Filippi et al. · arXiv:2608.19930
A novel AI model that can generate many different weather scenarios by effectively "rewinding" from an extreme event, such as a major heatwave or hurricane, to explore the diverse conditions that could have led to it.
Extremes on Rewind: Generating 1,000-Member Ensembles Initialized at a Final Condition · Jerry Lin, Mu-Ting Chien, Mansi Sakarvadia et al. · arXiv:2608.19008
A look at whether new AI-driven climate models can accurately predict how global climate zones, like deserts or rainforests, will shift as the Earth warms, comparing their predictions to established physics-based models.
How Do AI Climate Models Respond to Warming Across Climate Zones? · Charlotte C. Merchant, Milan Klöwer, Bradley Stanley-Clamp et al. · arXiv:2608.17986
How we prepare satellite data for machine learning models is key to accurately measuring things like water clarity and algae levels in the ocean. By optimizing this "data preparation," they significantly improve the accuracy of these measurements, which is vital for monitoring coastal water health and understanding environmental changes.
The impact of feature engineering and an optimisation framework for ocean colour machine learning · Edson Silva, Julien Brajard, Simon Cappe et al. · arXiv:2608.19899
This paper rigorously compares different radio technologies, like Wi-Fi, for their ability to monitor human activity and sleep from a ceiling-mounted device in a bedroom, all without physical contact.
A comparison between ceiling-mounted FMCW, IR-UWB and Wi-Fi radar for in-bedroom human activity monitoring and sleep interruption detection · Anton Lambrecht, Reda El Hail, Xianjun Jiao et al. · arXiv:2608.20322
A new instrument on India's Chandrayaan-3 mission that observes Earth from the Moon, measuring its light's color and how its waves are oriented. By studying our own planet as if it were a distant 'exoplanet,' scientists can learn what signs of life or habitability to look for.
Shaping SHAPE - A spectro-polarimeter onboard Chandrayaan-3 to observe Earth as an Exoplanet · Anuj Nandi, Swapnil Singh, Bhavesh Jaiswal et al. · arXiv:2608.19371
A new algorithm, MARA, designed to find Earth-like planets in the habitable zones of double-star systems. It works by analyzing incredibly precise measurements of how one star subtly shifts its position relative to its companion due to an orbiting planet's tiny gravitational tug.
Finding Habitable Exoplanets with Binary Relative Astrometry: Planet Detection and Characterization with the Microarcsecond Astrometric Retrieval Algorithm (MARA) · William Roberson, Eric L. Nielsen, Jessie L. Christiansen et al. · arXiv:2608.19444