Building Better Forecasts with Atmospheric Data
WindBorne
Atmospheric data is the biggest missing piece to improving modern forecasts, and a new company is filling in that gap. The AI weather startup, WindBorne, fuses data from its long-duration balloons with state-of-the-art AI models to produce accurate weather forecasts, including over oceans and the Earth’s most remote places. Each balloon can control its own altitude, autonomously flying for weeks at a time. WindBorne is helping the U.S. and international governments, companies, and philanthropic organizations transform forecast data into actionable insights.
Helping Cities Make Investments in Climate Resilience
USC’s Tree Planning Tool
USC researchers have developed a new, free AI tool that could help cities better understand one of their best defenses against rising temperatures: trees. Using free aerial imagery and AI, the tool gives cities an affordable way to target tree planting, expand shade, and make investments in climate resilience. Unlike many expensive tree-mapping systems, the USC-developed tool works with free aerial photographs collected nationwide through the National Agriculture Imagery Program (NAIP). This reduces the cost of producing detailed tree canopy maps, making the technology practical for many cities. This AI technology is assisting communities make smarter, targeted investments in urban forests by giving them the fine-scale data necessary to know where to plant new trees.
Making Renewable Energy More Reliable
University of Houston’s Battery Decay Prediction Model
University of Houston researchers are deploying AI to help solve a critical vulnerability in the power grid: the unpredictable aging and decay of large-scale batteries. The AI-powered model predicts how batteries degrade under real-world conditions and accounts for various factors that affect battery health, which include temperature, charge rates, and overall usage patterns. To ensure the model is usable in real-time energy planning, the researchers created a simplified system that maintains accuracy while reducing demands, resulting in faster, more informed decisions about when to charge or discharge batteries.