The U.S. Navy has awarded Charles River Analytics a follow-on contract to help the Office of Naval Research develop a tool for the military branch to generate probability forecasts of weather.
Charles River Analytics said Tuesday it aims to produce big data-based machine learning software for an extended-range weather prediction tool through the Climatological Observations for Maritime Prediction and Analysis Support Service initiative.
The company will collaborate with Benjamin Kirtman, a professor at the University of Miami Rosenstiel School of Marine and Atmospheric Science, Clear Science founder Bruce Ford on the COMPASS project.
The program seeks to integrate the North American Multi-Model Ensemble and several other weather forecasting systems into a single platform that will work to help users visualize climate and weather predictions through mission-specific forecast displays.
Charles River Analytics also intends to utilize its Figaro open-source probabilistic programming language and machine learning techniques as part of efforts to identify a method to combine various models for prediction of temperature, cloud cover and wind stress.