In the environmental sciences field, ecological system dynamics models are widely used as a tool for representing knowledge about ecosystems followed, in many cases, by simulation, prediction and decision making. The bases for model building are formed by human expertise and data, which can be collected on field experiments, extracted from the literature or guesstimated by modeler(s). In this talk we shall discuss the "very early" research ideas on the formalisation and seemingly possible partial automation of the model construction process supported by data. We shall present a simple ecological model example, discuss the human reasoning involved in designing the models, and some early ideas that have come up as a possible approach to conduct the research.