Q5: LCA Tools and Data Collection

Does it matter if I use LCA data from different sources? Can I compare different LCAs? What are the impacts of geography on LCA?  Why do I see negative GWP numbers for timber buildings in some LCA studies and not others?

Author: Adam Slivers, Kate Simonen; Contributors: Frances Yang, Terry McDonnell

Yes, it does matter which LCA data you use, due to the source of the data, the level of detail and scope of the life cycle inventory, and the data’s applicability.  LCA tools (datasets and software) vary in scope, audience and flexibility.  Tools like the Athena Impact Estimator are developed using the ‘best available’ life cycle inventory data for specific regions in North America, integrating regional energy sources and typical transportation modes and distances to reflect average industry materials and typical U.S. projects.   If one wants to refine transportation or energy sources for a project, for example, tools such as SimaPro (from Switzerland) or GaBi (from Germany) enable users to customize the LCA based on user input of LCA variables. 

Variations are significant orders of magnitude.  As an example, a simple, one-room building modeled in SimaPro, Sustainable Minds, and Athena Impact Estimator resulted in a calculated embodied Global Warming Potential of 4,000 kg CO₂e, 18,000 kg CO₂e, and 6,000 kg CO₂e from the respective software (Moore 2012). 

In order for LCAs to be comparable, the methodology and data should be comparable.  Different users of these tools can make different assumptions, capturing a different scope of the life cycle processes, resulting in different conclusions.

Variables in an LCA that can influence results include:

Energy source:  Life cycle inventories conducted in different geographical areas will be different due to the different fuel types and proportions used in electricity production.  A region with a higher proportion of hydroelectric power will result in different types and magnitudes of emissions than areas with more nuclear or coal power plants.  This is particularly relevant for materials that require large amounts of electricity to manufacture, such as aluminum (Fernandez 2008). 

Transportation:  Transportation can also play a factor if the region requires import of construction materials.  Life cycle assessments report results based on assumptions of an average distance and mode (truck, train, barge, etc.) of material transport, which are particular to the geographic region studied.  Modes of transport have different impacts.  For instance, diesel trucking requires 10 times more energy per mile than shipping over water using marine heavy fuel oil (Athena 1993).

Regional manufacturing and construction methods: Different ways to manufacture and construct may lead to significantly different emissions.  Newer manufacturing plants and construction equipment are generally more energy-efficient.  Different regions of the U.S. and the world have different mixes of new and old technology and different efficiency standards.

Level of detail and scope of the LCA: Differences can relate to system boundaries and number of data sources.  Some tools bound system scopes to exclude the impacts of extraction and distribution of raw fuels (including sources of electricity), or embodied impacts of byproducts. Other tools include these “upstream” scopes, but often it is approximated from aggregated sources or industry averages, such as economic data (financial transactions) rather than specific material processes (energy and mass flows).  According to ISO 14044 (ISO 2006), the scope and level of detail of an LCA depends on the subject and the intended use of the study, and thus they can differ considerably while still meeting the standard.

Treatment of biogenic carbon:  Carbon dioxide is absorbed during the photosynthesis that results in growth of trees and other bio-based material resources.  Wood thus physically contains (or sequesters) carbon that can be released again through combustion or decomposition.  The embodied emission or absorption of carbon dioxide (‘biogenic carbon’) in the lifecycle of a tree varies over time through its growth and end-of-life.  LCA results for wood can appear to have negative carbon emissions by starting and stopping at different points in time in the life-cycle of wood, as the various Life Cycle Inventory (LCI) datasets demonstrate in Figure 1.

Figure 5-1: Embodied CO₂ in Wood Over Time, As Modeled by Various LCI Data

The World Resource Institute/World Business Council for Sustainable Development Greenhouse Gas Protocol (WRI 2011) recommends that biogenic carbon absorption and emissions be reported as a separate inventory item.  LCAs that report biogenic carbon in this manner are most transparent.  The LCI data for U.S. wood processes and products reported in the National Renewable Energy Laboratory’s LCI database and in CORRIM Reports does not include biogenic carbon, however, the carbon sequestered in a piece of lumber can be estimated based on wood chemistry.  If forests are sustainably managed (trees planted to replace those cut) over the life of a plant the carbon absorbed through growth is considered equal to the amount emitted at end of life.  If the forest is clear cut and not replanted, the environmental impacts of land use change should be accounted for.   Methods to consistently report land use change impacts are still evolving however a method of reporting carbon impacts from land use change is outlined in the Appendix B of the Greenhouse Gas Protocol Product Standard. 

Recycled content methodology:  In materials made of high recycled content or of high recyclability, the method used to account for impacts of recycled material and recycling make a significant impact on the LCA results.   Notably the U.S. and European methods for reporting the impact of steel recycling are different.   The U.S. uses the ‘recycled content method’ where recycled steel comes with no ‘burden’ of impacts from the original manufacturing.   This method promotes the use of recycled materials Another method, termed the ‘end of life approach’ gives a credit to processes in the end of life stage that result in recyclable steel at the end of life.  This method promotes the recycling of materials. The Worldsteel/European method uses a combination method to develop LCI emissions to attribute to scrap steel and apply burdens to the use of scrap and credits to its recycling.   Worldsteel has published a comprehensive LCA document that includes details of this recycling methodology discussion in an appendix (Worldsteel 2011).

Treatment of end of life impacts: The end-of-life assumptions can vary significantly as it is difficult to predict in the distant future.   Research and their resulting data may not address current understanding of impacts such as sequestration of carbon (absorption of CO₂ via photosynthesis) in wood products (Lippke et al. 2004), carbonation of recycled concrete (formation of calcium carbonate from calcium hydroxide and carbon dioxide), or recycling of steel.  Additionally some regions landfill most of their waste and others burn it and co-generate energy and heat.   The difference in emissions and impacts from these different treatment methods can be significant.

Limited scope of studied environmental impacts:  Tools commonly focus (sometimes exclusively) on CO₂ or greenhouse gas emissions (CO₂e) and their impacts to climate change.  More complete life cycle assessments compliant with ISO 14044 should address other environmental impacts such as ozone depletion, acidification, eutrophication, tropospheric ozone (smog) formation, ecotoxicity, human particulate effects, and human carcinogenic effects.  (See Q8 for more detail.)

Complexity of the tool vs. sophistication of the user:  Some LCA tools are more transparent in explaining the assumptions and limitations of the research when used for material and building assemblies, while others are less transparent.  Most tools incorporate some publicly disclosed datasets, such as the National Renewable Energy Laboratory’s LCI database, while also incorporating proprietary data that is not disclosed to the user.  This means that the user could substitute a similar material or assembly for another where data is unavailable, and unknowingly substitute data that is inappropriate.  Or the user may misinterpret the specifications or assumptions of a product because they are unaware of the details in the data collected and LCA results reported.  SimaPro and GaBi require the user to understand and use judgment when inputting data, while Athena removes much of the judgment from the user.

As researchers continue to develop methodologies for measuring, fine-tuning, and publishing life-cycle inventory data, using LCA tools should come with some trepidation.  That should not prevent one from conducting a life cycle assessment. Just understand these caveats:

• Depending on the underlying methodology, different tools can present impacts at different orders of magnitude. 
• Do not mix results from different tools unless you can verify that the methods, scopes and results are consistent. 
• Beware of claims that are not supported by peer-reviewed research, or are not verified by a third party.
• Follow ISO 14040 and 14044 standards for life cycle assessment procedures.
• State the major assumptions made during the study and the methodology used. 
• Use data as specific to the region and products being studied as possible. 
• Address the uncertainty of data on a component. If the quantity of a component with insufficient data makes a substantial difference on the measured impacts of the entire project under study, provide the range of uncertainty in your results. 
• Conduct sensitivity analysis to focus on major impact contributors.  Focus on the sources of the largest impacts and aim to reduce or substitute those sources with a material or process that makes a substantial improvement. 

Because different LCA datasets and software can produce results that may not be compatible, just using the one tool to focus on a partial or simplified impact analysis may provide enough information to weigh the relative benefits during structural material selection.

References

Alcorn, A. (2003). “Embodied Energy and CO2 Coefficients for NZ Building Materials.” Centre for Building Performance Research, Victoria University of Wellington, Wellington, New Zealand.

Athena Sustainable Materials Institute, Canada Centre for Mineral and Energy Technology, & Radian Canada Inc. (1993) “Raw Material Balances, Energy Profiles and Environmental Unit Factor Estimates: Cement and Structural Concrete Products” Ottawa, Canada., 16

Fernandez, N. P. (2008). “The Influence of Construction Materials on Life-Cycle Energy Use and Carbon Dioxide Emissions of Medium Size Commercial Buildings” Victoria University of Wellington, Wellington, New Zealand.

International Organization for Standardization (2006) “14044:2006 Environmental Management - Life Cycle Assessment - Requirements and Guidelines.” ISO, Geneva, Switzerland.

Lippke, B., Perez-Garcia, J., & Comnick, J. (2004) “The Role of Northwest Forests and Forest Management on Carbon Storage.” CORRIM Fact Sheet in care of College of Forest Resources, University of Washington.

Moore, E. E., & Peterson, E (2012). "Comparative Case Study in Life Cycle Assessment Modeling Software for Buildings." LCA for WA Stakeholder Workshop. Integrated Design Lab, Seattle. May 7, <http://courses.washington.edu/lcaforwa/Video_12-05-07/06_Moore/index.html>, 22-23. (Jan. 20, 2012).

World Resources Institute, World Business Council for Sustainable Development. (2011). “Greenhouse Gas Protocol Product Life Cycle Accounting and Reporting Standard.”  <http://www.ghgprotocol.org/files/ghgp/Product%20Life%20Cycle%20Accounting%20and%20Reporting%20Standard.pdf> (Feb. 26, 2013).

World Steel Association (2011). ”Life Cycle Assessment Methodology Report.” <http://www.worldsteel.org/dms/internetDocumentList/bookshop/LCA-Methodology-Report/document/LCA%20Methodology%20Report.pdf> (Feb. 26, 2012).