This document discusses developing a system for real-time monitoring and forecasting of Arctic ice floes. The system would use satellite imagery, sonar, radar, and other data to continuously track the location, speed, and geometry of individual ice floes. It would also forecast the trajectory of each ice floe and quantify the uncertainty in those forecasts. The goals are to help plan oil and gas operations in Arctic waters by visualizing ice floe movement and predicting future positions and routes. Such a system could save hundreds of millions or billions of dollars on Arctic projects by extending drilling seasons and improving platform design. It presents challenges from limited Arctic infrastructure and harsh working conditions but stream computing is seen as enabling effective real-time analytics of the