Wood in Our Future: The Role of Life-Cycle Analysis: Proceedings of a Symposium (1997)

Effective implementation of international standards for environmental management, auditing, performance evaluation, and labeling depends in part on the availability of a quantitative, cradle-to-grave environmental assessment tool. Within the ISO-14000 series, life-cycle assessment is the one methodology that has demonstrated the greatest potential for addressing this need.

The life-cycle impact assessment (LCIA) standard is intended to guide practitioners and users in assessing potential environmental effects associated with specific industrial systems. The standard outlines a general ''inventory interpretation'' approach to LCIA in which potential system impacts are predicted on the basis of inventory analysis data. (Inventory analysis data are input-output data that have been aggregated and allocated without consideration of their spatial, temporal, threshold, and dose-response characteristics, and without regard to the magnitude of the actual environmental effects.)

As described in the standard, life-cycle analysis users must be cautioned that such an approach cannot, in and of itself, provide an accurate portrayal, or even an approximation, of the actual effects caused by the industrial system being studied. The standard further advises life-cycle analysis users that, to understand the actual environmental effects of a system, the findings should be considered in conjunction with environmental data typically generated by site-oriented environmental assessments, such as Environmental Impact Assessments (EIA).

This annex describes an approach to LCIA that merges the cradle-to-grave accountability of life-cycle analysis with the assessment of stressor-effects networks and consideration of relevant spatial, temporal, threshold, and dose-response characteristics common to other environmental assessment disciplines. Specifically, the life-cycle stressor effects assessment (LCSEA) approach is de-

signed to perform the iterative calculations required to determine the significance and relative contribution of an industrial system's inputs, outputs, and activities to actual measurable or observable environmental effects. Its purpose is to provide practitioners, industrial users, and other interested parties with a practical application of the proposed LCIA framework, free from the arbitrary allocation and aggregation procedures associated with inventory analysis data.

Within the life-cycle inventory, the term "unit operation" refers to individual physical processes or groups of processes that produce a single product or service and their associated inputs and outputs. The input-output data for each unit operation are typically normalized with respect to mass units of production, then averaged over a selected period (for example, 12 months) to account for fluctuations in industrial processes. Once normalized and averaged, the input and output data for all unit operations of the system are aggregated and allocated to produce an overall mass and energy balance.

Just as life-cycle inventory models link unit operations, the LCSEA architecture is based on LCSEA unit operations linked by process. In addition, LCSEA unit operations are linked by effect. The LCSEA unit operation is distinguished from the standard life-cycle inventory unit operation in several respects, as described below. Because the objective of the LCSEA approach is to quantitatively assess the significance and contribution of an industrial system's effects on the environment, each LCSEA unit operation is defined in terms of its relevant stressor-effects networks.

As described in Figure A-1, the physical boundaries of the LCSEA unit operation generally correspond with the boundaries of a standard LCI unit operation. Appropriately aggregated input and output data associated with the standard LCI unit operation provide the initial set of useful data for the LCSEA unit operation. These input-output data are classified into corresponding stressor effects networks.

To determine the full array of stressor effects networks relevant to the system being studied, however, it is necessary to look beyond the standard life-cycle inventory input-output data. Some stressor effects networks do not have direct input or output causes, but are nevertheless related to the activities of a system and can be measured through quantifiable effects indicators. For example, mechanical disruptions that occur at mining sites can cause effects, even though inputs or outputs are measured under a standard life-cycle inventory.

Figure A-1 Useful components of a life-cycle inventory.

Effects indicators are calculated only from environmental data, which are not currently being collected under for life-cycle inventories but that are routinely collected for other environmental assessments. The LCSEA approach is the first life-cycle analysis approach specifically designed to integrate such data into the architecture. It is the inclusion of these environmental data that transforms life-cycle analysis from a tool used simply to model potential impacts into a tool that assesses the actual effects of a system.

Environmental data describe the spatial, temporal, threshold, and dose-response characteristics of input-output values and effects indicator values with respect to their relevant stressor effects networks. For example, for the eutrophication stressor effects network, environmental data include the baseline percentage of dissolved oxygen (2 percent) in a lake and the reduced percentage (0.2 percent) of dissolved oxygen in that same lake at a later date. These data, in turn, are used to characterize the effects indicator for the LCSEA unit operation—in this case, a 90 percent decrease in dissolved oxygen.

Environmental data also are used to normalize input-output values and effects indicator values to quantify their significance and contribution to a given effect. In the eutrophication stressor effects network example, external input-output data, such as the amount of phosphate emissions from other sources into the same lake, are needed to determine the relative contribution of the system's phosphate emissions to the total effect.

In some instances, life-cycle inventory models have aggregated input-output data from multiple unit operations under one unit operation. In other instances, specific unit operations with no measured inputs or outputs have been excluded from life-cycle inventory models. In such cases, environmental data provide the sole route for identifying relevant stressor effects networks, thereby triggering the need to create new LCSEA unit operations.

The new components of the LCSEA unit operation described above make possible, for the first time, the ability to

• determine the significance and contribution of each raw input-output data point to actual effects,
• identify specific nodes in the stressor effects networks associated with a given unit operation and quantify the magnitudes of their respective effects indicators,
• establish quantitative equivalency factors and characterization weighting factors with established levels of certainty,
• calculate the "environmental loadings" for each LCSEA unit operation per stressor effects network on a functional unit basis, and
• calculate the cumulative environmental loadings for the system, product, or service being studied

Because life-cycle analysis as historically practiced has been confined to inventory analysis and the assessment of potential impacts, users have been limited in their ability to establish meaningful equivalency and weighting factors. The LCSEA approach, in which a system's actual contribution to effects is assessed, allows new equivalency and weighting factors to be established:

• The relative aggregated magnitude of a specific input-output to a related group of inputs-outputs within a specific stressor effects network (e.g., molar acid equivalencies for outputs associated with acid rain).
• The relative magnitude of the effects in a specific stressor effects network compared with effects in other stressor effects networks of the same type.
• The relative magnitude of the effects of a specific stressor effects network compared with effects in similar types of networks (e.g., networks with the same endpoint effect).

An essential architectural feature of standard life-cycle inventory engineering models is that each unit operation is a stand-alone module. An individual unit

operation can be linked through the functional unit to any other unit operation to calculate the mass and energy balance for a given system, product, or service. In life-cycle inventory engineering studies, the aggregated inputs and outputs are typically allocated on a direct proportional basis by mass to the functional unit—per kilogram, MJ, kWh, or mile driven, for example. However, from an actual effects perspective, such a linear proportional relationship most often does not exist.

Like its life-cycle inventory counterpart, the LCSEA functional unit allows for the same critical linking of unit operations in a system. Accordingly, LCSEA functional units are also divided through on a per kilogram, MJ, kWh, or mile driven basis. The essential difference is that LCSEA unit operations are linked not only by process but by the significance of the various measurable effects. The environmental loadings (the normalized inputs–outputs and effects indicators) are allocated by effect to the functional unit.

Thus, the LCSEA is constructed as follows (WF is the weighing factor; EF is the equivalency factor):

Once the normalized input-output and effects indicator loadings per unit operation have been established, the cumulative environmental loadings can be calculated for the entire system or for the specific material, product, or service being studied. The calculation of overall environmental loadings is essential for such LCA applications as Type III labeling and the establishment of quantified environmental performance indexes.

The amount of material or energy used or the amount of releases into air, water, or ground from a given unit operation, without regard to any specific environmental effects.

A physical, chemical, or biologic measure of a specific node within a recognized stressor effect network.

Raw input-output data related to operations that are used to determine the contribution of the system inputs-outputs to effects in a given stressor effects network, but that are unrelated to the system defined by a given study.

A physical activity, or physical or chemical input or output that can trigger a subsequent environmental effect or network of effects.

The data needed to classify, characterize, or normalize input-output data, or to quantify effects indicators, for each stressor effects network associated with a given unit operation.

The explicit measurable or observable effect in the environment which identifies the assessment endpoint as relevant and meaningful and allows the significance of the impact assessment results to be evaluated.

The sequential physical, chemical, or biologic mechanisms involved in linking a specific stressor to specific environmental effects.

Characterization factor based on a recognized, well-defined stressor effects network and well-established properties of the stressors involved.

Subjective factor that translates individual stressors into a relative ranking or weighting scheme. It may then be possible to mathematically represent the effective loadings of a stressor effects network.

A value that represents the relative significance of input or output from a unit operation for the relevant stressor effects network for a given function.

Normalized input-output loadings and effects indicator loadings.

An industrial process within the scope of a given life-cycle study, and its associated stressor effects networks defined with respect to their spatial, temporal, threshold, and dose response characteristics.

A systematic process to define and evaluate the effect of variations of inventory and model data input on the impact assessment result.

A systematic process for defining and evaluating the sources of error and uncertainty in the impact assessment process, including the linkage of stressor effects networks and their inherent characteristics, such as space, time, dose response, and threshold.