Life Cycle Assessment Explained

1. Goal and Scope Definition. During this phase, the client and LCA practitioner work closely to define the fundamental questions that the assessment will address and how they will carry out the study. First a study goal is set, including the purpose of the project and how the results will be used. A typical goal is to evaluate the environmental performance of a product or process, and/or to compare options. But how to compare? For each option, a common functional unit is defined, which describes the service being provided and establishes a quantitative basis of comparison for the entire study. Each option being compared is then scaled into a reference flow that delivers that common functional unit. For example, in comparing the impact of plastic bags to paper bags, it is not appropriate to simply compare one bag to another because they have different capacities (say 5 and 10 liters, respectively). So, first a common functional unit is defined, say 10 liters of capacity, and the reference flows would be 2 plastic bags versus 1 paper bag.

Once the study parameters are set, the LCA practitioner works with the client to decide the system boundary, essentially what types of inputs, processes, and emissions should be included in the study. This should be done in a way that serves the study goal, making sure that nothing is excluded that is likely to have a major influence on the results. Other aspects that are difficult to obtain data for and are likely to have a negligible influence can be excluded, as long as these assumptions are documented. For example, examining the life cycle of a cup of coffee could include transporting coffee beans but might exclude the emissions associated with building the roads that are used, as many other vehicles use these as well. Finally, the team decides on requirements for data quality and modeling approaches.

2. Life Cycle Inventory (LCI) Analysis. The LCI is a list of all of the flows of resource inputs and emissions outputs that occur in producing the reference flow of each product option. This is also the data collection phase. To compile the LCI, the LCA practitioner will first gather direct information from where the item is produced or used, such as how many tissues are used in a particular treatment. These direct quantities are called foreground data, but they only reflect the quantities being used directly and not the entire life cycle. In order to link the quantity of tissues with all of the emissions that occur throughout the supply chain of those tissues, LCA practitioners will link foreground data on usage with background data from LCI databases. These background data are based on decades’ worth of in-depth studies that have investigated the supply chains of thousands of materials and products and counted all of the inputs and emissions along the way. These investigations are extremely time-consuming, so using LCI databases is a way to build on what is already known. There are various LCI databases compiled by governments (e.g., the U.S. LCA Commons) and commercial entities (e.g., ecoinvent). In the STiCH project, conservators and engineers worked closely to define reference flows for each conservation item of interest, collected foreground data from manufacturers and practitioners, and then linked to background data from the ecoinvent database.  

Once all of the data have been linked together, an algorithm scales all of the flows in the product system according to the amounts needed to produce the final reference flow. The product system can be quite complex. Considering tissues, for example, the product system might include emissions from processing wood into pulp, from transporting logs to the pulping plant, from making the fuel that the trucks run on, from making the trucks themselves, and on and on up each branch of the supply chain. The final, scaled set of resource inputs and emissions is called the life cycle inventory. 

3. Life Cycle Impact Assessment (LCIA). This phase tells us how each of the flows in our life cycle inventory affects the environment. There are multiple ways that our activities are affecting the planet and our health, these are called impact categories. Examples include global warming, depletion of the ozone layer, and ecotoxicity. Each flow is classified according to whether it causes a certain impact. For example, nitrous oxide (N2O) causes both global warming and ozone depletion. Then it is assigned a value, or characterized, according to how potent a substance it is for each type of impact. This is called a characterization factor (CF) and is expressed relative to an indicator substance. In the case of global warming potential, we measure CFs relative to carbon dioxide, and so all greenhouse gases are characterized in units of carbon dioxide equivalents (CO2eq). 

Characterization relies on integrating models from environmental science, chemistry, physics, ecology, toxicology, and other fields in order to create a causal chain that links an emission in a particular location to the physical changes and damages that it causes all over the world. If some treatment chemical is disposed of in wastewater, we want to be able to understand how the chemical is transported through waterways, is transformed or degraded, comes into contact with aquatic organisms, and causes harm through ecotoxicity.  Most of these integrated models are developed by scientific research organizations within government labs and universities.  These organizations will compile a suite of models that correspond to each impact category that it deems important. Together, this collection is called an LCIA method.

In the United States, the Environmental Protection Agency developed the most commonly-used LCIA method, called Tools for Reduction and Assessment of Chemicals and other environmental Impacts (TRACI). This is the LCIA method used in all of the STiCH tools.  TRACI (v2.1) includes 10 impact categories (see the full list in the Appendix below), including global warming that is the primary focus of STiCH carbon calculator, while the case studies include the full list. Which impact categories are included in a study should be in line with the study Goal and can be decided by the LCA practitioner and client.

Because LCA compares options across multiple sustainability indicators, results typically have different units. It can be useful to normalize results across impact categories into relative units, for example as a proportion of a person’s average annual impacts, or further weight results into a single score so that product options can be compared more easily.


4. Interpretation. This last and most important phase of LCA is where results get translated into knowledge and recommendations. If Option A has lower impacts than Option B across all impact categories, then this supports a clear preference. But it frequently happens that one product option has higher impacts for some indicators, but lower for others. The LCA practitioner will discuss these different trade-offs in the context of the study Goal, leaving the final decisions to the client.

In many cases an LCA is motivated by a desire to improve designs and practices. These studies will include a contribution analysis to tell which processes or hot spots are driving results for different impact categories, in order to suggest where to focus efforts. If a particular material is driving impacts (such as foams in the Crates Case Study), then designers might investigate low-carbon alternatives or ways to reduce how much material is used. If a particular upstream process is the main culprit (such as electricity generation), then managers might work with suppliers to source a greener option (through renewable energy contracts for example).

LCA is not an exact science – there are various assumptions that are a part of every study, as well as inherent uncertainty in model parameters such as material quantities and emissions factors. It is critical to indicate the quality of the underlying data and their implication on the certainty of the results. For instance, if an LCA uses emissions factors that are 5 years old, then this must be clearly documented and its implications discussed, especially if in reality the emissions have improved. LCA studies often include a sensitivity analysis, where various assumptions are changed, such as the source of electricity or the quantity of materials required, in order to see how much the overall results respond. If the results are found to be very sensitive to a single parameter, then this parameter should be further investigated to ensure that the study uses the most accurate and up-to-date data possible. Even an alternate functional unit might be tested. In the Loans Case Study, shifting the functional unit from one loan to the number of views per object in each location had a dramatic influence on the results, changing the order of preference. LCA studies might also include a formal uncertainty analysis, which provides a statistical range of results.

Overall, the Interpretation Phase provides context to the study so that readers can fairly interpret the results and make educated choices.