Similar to any modelling technique, system dynamics (SD) modelling should start with the essential step of scoping and identifying the problem of interest before further analysis and modelling. In practice, this first step is a challenging task, especially when wicked issues such as water management are being addressed. There is still a vital need for modelling methods and tools that can support modellers to identify and assemble essential data to inform problem scoping and boundary setting. This article aims to narrow this gap by presenting a methodology for combining a series of conceptual modelling techniques (extending the usually linear Driver-Pressure-State-Impact-Response framework with causal loop diagrams, system archetypes, stock and flow diagrams) towards the development of a quantitative SD model. A case study of the Gorganroud-Gharesu Basin, in Iran, is used to illustrate the benefits of the methodology. Our experience shows that combining multiple conceptual models provides complementary insights into the problem boundaries and model structure, as a basis for developing the SD model.