Drones in general have become increasingly popular nowadays, mainly due to its cost attractive and popularization. At this stage, drones' technology is providing a new set of applications and services specially in the field of aquatic drones which can be used for numerous applications (e.g, sea life localization and environmental monitoring, bridges inspection, sea border patrolling...). Most of the Earth surface is covered by water, and therefore, aquatic drones can become great assets in aquatic environment monitoring tasks. Thus, it is of crucial importance to guarantee a persistent and reliable communication network to assure the drones' self-control, as well as to make sure that environmental data collected from sensors can reach the fixed gateway placed ashore. In this work we propose a mechanism to calculate and assess the link quality between aquatic drones based on the signal strength, stability of the connection and its bandwidth. This quality evaluation mechanism is then integrated in the routing process of a Delay Tolerant Network (DTN) for an efficient transportation of data among the aquatic drones. Having the quality mechanism as baseline, three routing strategies were proposed: i) routing through the first neighboring contact, ii) routing through the best quality contact, and iii) routing through the best quality contact to the gateway. To evaluate the performance of the proposed routing strategies, a set of evaluations were performed using real aquatic scenarios with movement. The results have shown that the strategy "best quality contact to the gateway" presents the highest delivery ratio when compared to other routing strategies.