The different ways to operationalise the social in applied models and simulations of sustainability science: A contribution for the enhancement of good modelling practices

• A sociologically reflected synthesis of the modelling process shows dependencies and feedback loops between the different intertwined decision levels and provides orientation over the different stages. • We bring together and evaluate the various strands of discussion in which the social is considered in applied modelling in sustainability science. • A scheme is used to show how suitable components and processes of social and ecological aspects can be taken into account in sufficient detail and linked with each other via processes. • The extent to which and the quality with which social issues are taken into account is an important issue for the quality and the acceptance of applied models. There are several concepts out there, which describe modelling as a circular process with several rounds of iteration. The aim of these concepts is to structure the process and gain insights into the various steps carried out during a modelling project. It is common to all these concepts that they are created by modellers themselves, hence the focus is on generating a good modelling practice (GMP) in standardising technical aspect. However, modelling must also be considered as a social process, which have so far only been marginally considered in these GMP protocols. Therefore, we explored the different approaches to incorporate the social aspects into GMP of interdisciplinary applied modelling projects in sustainability science with reference to sociological knowledge. While discussing the social dimensions in the modelling process, we found that two perspectives need to be distinguished, namely a structural and a procedural one. From a structural perspective, various approaches have developed to model interactions and feedbacks between ecological and social aspects of a sustainability problem. Among the best known is the concept of the social-ecological system, which makes it possible to grasp the complexity of reality in ecological and social subsystems that are intertwined within each other. From a procedural perspective, several components describing decision points and feedback processes along the modelling pathway have been identified. This forms a new GMP scheme different from the so far published ones as it incorporates all the feedback loops active during the modelling process. It therefore breaks the so far common circular approach. The new scheme emphasises the fact that every model formation is a social, communicative process at all phases of a modelling project. Additionally, we have gained new insights into the path dependency of model structures and identified an extended hierarchical structure of social modelling steps. Most importantly, we describe how the iterative application of these two perspectives should be used to improve the GMP of an active modelling project.