Assessments of the energy performance of green roofs have traditionally been performed with experimental and modelling procedures. Experimental green roof energy studies have significant disadvantages (expensive, time-consuming, cannot be generalised, etc.). On the other hand, modelling studies are often done with over-simplified models (for example, by ignoring the effect of moisture flows in the green roof layers) or with detailed models that require inputs which are often difficult to be obtained in practice and may involve a certain degree of uncertainty. This paper describes the methodologies for the development of databases that include data inputs that could be used within building simulation programs for green roof energy performance appraisals. The methods described are based on the data model used by a new green roof model within the ESP-r simulation program; however several of these data could also be used by other detailed models. Where applicable, data are used from the literature and measurements are also taken for input model variables that affect the heat and moisture flows in the plant, air canopy and soil layers. Measurement methods are also described in the paper. The availability of such databases could facilitate the appraisals of green roofs within whole building integrated energy simulations and the methodologies described in this paper could be replicated by researchers in order to include additional green roof data for modelling alternative green roof types.