Autonomous surveillance systems based on wireless sensor networks have brought many benefits for understanding, protecting, and preserving biodiversity thanks to the latest sensor and telecommunication technologies. For example, in archipelagoes with many rocks of various shapes and elevations interleaved with water, it is hard to deploy wireless sensing systems for covering all these given areas. In these sensing systems, coverages are defined as where information is accessible. In this article, a new approach is proposed, adopting cellular automata and massively parallel processing on GPUs. This work relates to the development of parallel algorithms and CAD tools to optimize coverage oriented to efficient deployment of wide-range wireless networks for various purposes such as environmental surveillance, early warning systems for natural hazards and risks, taking into account turbulence in topology. Some experiments on radio coverage were done in different complex terrain areas given positive results in terms of performance and functional requirements.