What if we told you that the secret to healthier soil, cleaner ecosystems, and smarter farming isn’t buried in a high-tech lab—but hidden in the data behind crop residues, wood chips, and food waste?

Meet the future of sustainable agriculture: a powerful new machine learning tool that can predict exactly how much biochar—a carbon-rich, soil-boosting material—can be made from any type of biomass, and how much nitrogen, phosphorus, and potassium it will contain. No crystal ball needed. Just smart science, powered by data.

Georgia State computer science students are building robotics to deliver tomorrow’s solutions — and learning as they do it.

A team of researchers used airborne radar to map out an unexplored region in the deep interior of East Antarctica in search of old ice that could record clues from Earth's ancient climate. Although they haven’t found a suitable ice sample yet, their survey revealed new insights about the geologic processes happening under the Antarctic ice sheet and how ice is moving around beneath it.

By combining ground-based direct imaging and radial velocity measurements with precise astrometric acceleration data from the Gaia space telescope, astronomers have discovered a brown dwarf companion orbiting a nearby red dwarf star about 55 light-years from Earth. The team accurately determined its mass (≈60 Jupiter masses) and orbital distance (≈4.3 AU) using Subaru’s Infrared Doppler instrument (IRD) under the IRD Strategic Program, Keck high-contrast imaging, and Gaia data. The object, classified as a late-T-type brown dwarf, shows roughly 30% variability in near-infrared brightness, offering a valuable benchmark for studying atmospheric clouds and circulation. This marks the first successful application of Gaia-only astrometric acceleration to precisely characterize a companion beyond Hipparcos’ detection limits.

WPI Assistant Professor Nitin Sanket has received a $704,908 National Science Foundation (NSF) Foundational Research in Robotics grant to develop sound-based navigation systems that enable tiny aerial robots to operate in environments where cameras and light sensors fail, such as smoke, dust, or darkness. Drawing inspiration from how bats use echolocation, Sanket’s project combines bio-inspired design, deep learning, and sensor fusion to create lightweight, energy-efficient drones capable of autonomous navigation in challenging conditions—advancing robotics for applications in disaster response, environmental monitoring, and search and rescue.