We use a decision-support system for agriculture based on a wireless sensor network, in order to understand how scientists solve a typical problem that requires environmental data (such as soil moisture, temperature, etc.). Subjects use an existing interface to solve simple problems, and a qualitative study of the users’ interactions with the system is performed to better understand their requirements. The study is performed from the point-of-view of distributed cognition, the goal being to understand how a new tool fits in the mental model that agriculture scientists have of their own field of work, and how it may influence or enhance their mental representations and work-processes.