Substance flow analysis can be used in many applications. So far, there is hardly any standardization of the tool: systems definition, quantification and interpretation of the results can be, and in fact are, performed in many ways. Nevertheless, the steps to be taken and the choices to be made can be specified, as has been attempted in the above. A three-step framework has been defined: goal and systems definition, quantification of the overview of flows and stocks, and interpretation of the results.
In general, the outcome of the methodological choices will depend on the goal of the study and the questions to be answered. Whether or not to include environmental flows, in what detail to define the system's nodes, which sectors to include and whether or not to include stocks must be decided case-by-case. Something more can be said about the choice of the type of quantification, the second step of the general framework. Three options are presented: accounting, static modeling and dynamic modeling. It is important to realize that there is no 'best' choice: each type of modeling is useful and each has different functions for supporting environmental policy. The display below summarizes the possibilities for application of the three modeling types.
Type of quantification application Accounting Static modeling Dynamic modeling signaling +
spotting trends +
evaluation ex post +
origins analysis +
comparing regimes +
evaluation ex ante + +
extrapolating trends +
scenario analysis +
For the third step of the framework, the interpretation of the results, it is important to realize the limitations of the SFA tool. The conclusions to be drawn should fall into the boundary conditions of SFA and therefore necessarily be restricted to the life cycle of the investigated substance. Many other variables, such as impacts on the life cycle of other substances, costs or rebound effects, are also relevant. Some attempts have been made to combine different aspects within one model (for example, Kandelaars and van den Bergh 1997), some even successfully, but so far limited to very small systems. In the future it may become clearer whether this is a fruitful road to travel, or whether it makes more sense to leave the individual tools small and simple but use them together.
Was this article helpful?