Nowadays, more and more systems are coordinated and controlled to compose globally-distributed, highly-complex systems of systems (SoS). Existing testing methods focus on single devices and on performance characteristics from the network architecture upwards, but ignore the innate complexity caused by the spatial and temporal separation of these systems over vastly differing scales. As a result, quality assurance of today's SoS is a major challenge, especially where there is a need to consistently and robustly account for the spatial and temporal context of systems. In this paper, we propose the development of a methodology that tests the rich spatial and temporal properties specified in the spatio-temporal aware models. We illustrate our approach based on a spatio-temporal modeling and reasoning language, namely BeSpaceD. Our research highlights the opportunity of developing a comprehensive tool framework for supporting the quality maintenance and improvement of today's spatio-temporal aware SoS.