This paper describes the theory and implementation of a trajectory optimization layer operating within an energy-aware, airborne, dynamic data-driven application system (EA-DDDAS). The work addresses wind field data estimation, trajectory optimization, and implementation of communication protocols to increase sUAS endurance in persistent sensing missions. The trajectory optimization layer acts as a middleware between a higher-level path planner and a lower-level guidance control layer that interfaces with the autopilot. The novel aspects of this work include a scalable trajectory optimization algorithm that generates unique dynamic soaring trajectories using assimilated wind field data from dual Doppler synthesis and online atmospheric planning. The optimization algorithm is used in a receding-horizon fashion to develop energy efficient trajectories between waypoints for missions in complex atmospheric phenomena. Mission trees are developed to form cost functions and constraints that guide the trajectory optimization layer to a suitable solution domain.