in environmental amenity. A rather different approach, known as ‘hedonic’ analysis, attempts to account for the price or asset value of a complex good, by disaggregation into contributions from different attributes (Pearce 1993; Herriges and Kling 1999). Thus real estate values in otherwise similar areas may reveal an implicit valuation for (avoiding) sulfur dioxide pollution from a nearby plant. Sometimes the valuation modeling is more elaborated, as in the ExternE project of the European Commission (1995). Other tech- niques include environmental cost accounting, environmental accounting and life cycle costing (LCC). All these may be included under the rubric of ‘environmental economics’.
In life cycle assessment (see Chapter 12.5) and its subprocedure life cycle impact assess- ment (LCIA), the object of study is a product or service. The goal of an LCA may vary but, mostly, the LCIA is intended to be used – sooner or later – in a choice between two products or processes. The comparative element in LCIA requires a comprehensive approach, in which the focus is different from risk assessment and EIA. Besides studying each impact type separately, the weighting of various impacts becomes an issue. The LCIA procedure is standardized by the International Standards Organization (ISO) and described in the ISO 14042 standard. A technical report (ISO TR 14047) is at present being worked out with examples on how the standard may be implemented (ISO 2000).
Engineering science and natural science make different demands on LCIA. In engineer- ing science the product’s overall performance is the focus. The product is intended to func- tion as well as possible in a number of situations. In natural science, the theory is central.
The theory is intended to function as well as possible in a number of situations. The inclu- sion of uncertain models or data in a natural science-oriented context may be objection- able, whereas omission of it would be objectionable to the engineering scientist, as it would be tantamount to neglecting a likely problem. Experience, in particular from the LCA area, has revealed many such methodological conflicts.
From a system analysis point of view, all impact evaluation techniques may be seen to deal with the technical, natural and the social subsystems (Figure 13.1). The technical system may be further divided into a foreground and a background system. The fore- ground system is the one you know and can specify in detail. The background system includes, for instance, market behavior and infrastructure.
Social system
Technical system Natural system
Figure 13.1 An impact evaluation combining scenarios for technique, environment and human attitudes
In industrial ecology (IE) you will need a toolbox for different types of impact evalua- tions. But instead of describing the different tools as they mostly are used, one by one, some procedural steps, which are common to all tools, will be used to structure the text of this chapter:
● formulation of goal and scope;
● selection of impact indicators;
● modeling or recognizing interactions between technical system indicators and impact indicators;
● comparing different types of impacts and evaluation of total impact;
● analysis of uncertainty and sensitivity;
● data documentation and reporting.
FORMULATION OF GOAL AND SCOPE
The choice of goal and scope has a very significant influence on the outcome of an impact evaluation. Experience shows that this rather obvious statement needs to be repeated often. There are numerous examples of misunderstandings arising when telling an impact evaluation story without adequately specifying the goal and scope.
The choice of goal and scope is an ethical or normative issue that is normally left out of discussion in a scientific context. However, in the ISO documentation on LCA stan- dards (ISO 2000) it is recommended that the impact evaluation should include all signifi- cant impacts on human health, ecosystems and natural resources. Generally speaking, there are three questions that need to be addressed when setting out a goal and scope:
What is to be included in the study? How to deal with trade-offs? How to handle uncer- tainty?
When deciding upon what to include in the study, there are many dimensions to keep in mind. One is the qualitative dimension. In general terms, one may think of things to include as belonging to ‘safeguard subjects’ or ‘areas of protection’, such as human health or natural resources. In LCA the concept of impact categories exists, which is more focused but still not a quantitative indicator. The quantitative indicators, called ‘category indicators’ in LCA and ‘impact indicators’ in many other methodologies, define the qual- itative system borders of the ‘environment’ we study.
Another dimension where system borders need to be set is time. The consequences of an emission or impact may never end, even if our possibilities of following and modeling them decrease as time elapses after the intervention. It is particularly important to recog- nize the depreciation of future impacts achieved by narrowing system borders or (as econ- omists do), by discounting, when dealing with global warming effects or depletion of natural resources (Azar and Sterner 1996). Yet another dimension is space. There are many examples of local environmental issues having been ‘solved’ by shifting the impact to another scale or a wider region.
If we choose to use global system borders, we must face the problem of trade-offs between local and global impacts. In impact evaluation, trade-offproblems are ubiqui- tous, even if they are not always explicitly identified. For instance, when deciding to include an impact indicator in the study, there has to be some kind of weighting of its
significance compared to other indicators or compared to some reference. In impact eval- uation, as in many other types of evaluation, there are two ways of handling trade-offs.
One is to try to minimize or maximize an objective function of some sort. This may be called a ‘utilitarian’ approach. (In modern politics it is often associated with the right wing.) Another is to try to achieve some type of justice, that is to deal with each indica- tor separately and try to reach an acceptable compromise. This notion is close to Herbert Simon’s ideas of ‘satisficing’ in contrast to ‘optimizing’. (In modern politics this world- view is mainly associated with the left wing.) In LCIA used for design purposes the utili- tarian approach is often used (that is, the overall best option is sought) while in RA and EIA the latter approach is more often used. Of course, in practice, combinations of the two tradeofftypes are common.
The way of handling uncertainty depends on the study context, but also on the practi- tioner’s general attitudes. A common way is to let the degree of uncertainty decide whether an issue or figure should be included or not in the evaluation. Another, some- times more fruitful, way would be to accept uncertainty as a part of reality and try to describe its consequences. Instead of focusing on what is ‘correct’, or not, one may ask what our present knowledge, in terms of data and models, tells us. The ‘precautionary principle’ is often used in impact evaluation and it works well with the ‘justice’ type of trade-offapproach, but, for a utilitarian approach, safety margins in one impact type tend to decrease the appreciation of other impacts.
In IE we look for strategies to decrease the environmental impact from technical systems. Because normative aspects, such as choice of system borders, are of such an importance, these must be identified, handled in a systematic way and reported to the reader/decision maker.