Heavy Asset Intelligence (HAI)
Stakeholders throughout the energy industry are incentivized to shift towards renewable resources, as environmentally-driven policy is making them affordable and profitable.
Utility-scale solar farms are being built faster than can be reliably recorded, especially in world’s fastest growing economies. Astraea has built an interactive web application that allows users to visualize and chart the growth of utility-scale solar farms.
Using publicly available satellite imagery and advanced computer vision techniques to locate solar farms, the SEIP provides a cost-effective way to track the growth of large-scale solar arrays in every country in the world, at any given time.
Astraea is accelerating the rapid adoption of renewable energy by providing actionable information at each layer of the value chain, from installers to policy makers. We recognize that remote-sensing imagery is uniquely positioned to answer complex questions about our planet. With our proprietary algorithms, we can reliably cross-reference this information with property, weather, and policy data, enabling our clients to gain a competitive advantage in this fast-moving space.
Case Study: Roof Material Classification
- Sun Tribe Solar engaged Astraea to generate new business leads by identifying commercial buildings with newly installed Thermoplastic Polyolefin rooftops, indicating solar opportunities for building owners.
- Astraea’s team built a custom solution, powered by artificial intelligence. Using a semi-supervised model, Astraea was able to identify the roof material of over six thousand buildings across 12 counties in Virginia.
- Combining machine learning analysis and high-resolution imagery, Astraea was able to save the Sun Tribe Solar team thousands of hours in manual rooftop labeling.
Case Study: Global Pollution Tracker
- As the world’s population grows and becomes more densely concentrated in urban environments, air pollution from industrial processes poses a considerate threat to public health.
- Satellites that measure aerosols and other trace gases offer a border free, cost effective means for urban planners and local governments to monitor and track air quality around the world.
- Partnering with the Children’s Investment Fund Foundation, Astraea is building a web app that shows a time lapse of air pollutants in the atmosphere by aggregating satellite air pollution data and combining it with ancillary data listing known forest fires and industrial processes that output pollutants.
Supporting sustainable management of our natural resources, Astraea provides organizations access to valuable current and historical satellite imagery through a powerful analytics platform with flexible compute to scale to the problem at hand. We are currently working with clients on a diverse set of projects, from land use planning and understanding the built environment to deforestation/reforestation and agriculture.
Case Study: Deforestation Monitoring
- Monitoring changes in forests across large, inaccessible regions over extended time periods is difficult but essential for understanding the impacts forest loss and degradation have on climate change.
- Earth observation data provides a valuable source of information to conduct global, independent estimates of regional and national forest cover and deforestation.
- Astraea’s data science team used over 2,000 global data points from public satellite data and built a pixel-based model to visualize year-over-year forest loss for the state of Mato Grosso, Brazil.
Case Study: Controlled Burn Analysis
- The oak and pine forests in the Appalachian Mountains are specialized ecosystems whose health is managed through careful, controlled burns by forest monitors like The Nature Conservancy.
- The variance of these burns in tree mortality and regrowth success leads to difficult burn preparation and analysis.
- The Astraea team worked with TNC to collect and process historical satellite data on each burn site to derive phenology indicators immediately before the burn events.
- Our EarthAI platform provided the flexibility to query and extract data from multiple data products, and to rapidly aggregate the imagery data across the burn sites through RasterFrames operations.