A Practitioner’s Introduction to LCA Databases: EPiC and ICE

By Amalka Nawarathna, Ghada Karaki, Francisco Sierra, Alireza Moghayedi & Alice Moncaster (all at University of the West of England)

This short comparative review of two construction material databases explains their potential use for assessing embodied carbon to designers and practitioners not yet expert in the field. It introduces and examines the Australian Environmental Performance in Construction (EPiC) database (updated in 2024) and the UK Inventory of Carbon and Energy (ICE) which was significantly updated in 2019.

Introduction

There is increasing interest across the construction and property industries in the whole life environmental performance of buildings (IEA EBC 2023). In order to conduct whole life assessments, as well as modelling the operational energy use, designers need to understand the environmental impacts of construction materials and products over the life of the building, usually through Life Cycle Assessment (LCA).  Full LCA considers a raft of environmental indicators. However, the imperative to reduce energy and greenhouse gas (GHG) emissions has led to these often being the dominant focus. The indicator for GHG emissions is also known as Global Warming Potential (GWP), or ‘carbon’ in common parlance across many countries, and is measured in kg of carbon dioxide equivalent (kg CO₂e). Even limiting analysis to one or two indicators, however, leaves a significant complexity (see for example Cabeza et al. 2021).

One approach to the complexity is to focus on the material coefficients for the most common construction materials. Such data can be sourced from a rising number of databases, increasingly published open access.  One of the first open access databases to focus on construction materials was the UK Inventory of Carbon and Energy (ICE), first released in 2006 and since used across the world. Other databases have also been around for a while such as Ökobaudat, also construction specific and first produced in 2009, and the cross-industry Ecoinvent, since 2003.  A full review of databases and their application to buildings would be an extensive undertaking (see for example Teng et al. 2023, who compare 20 databases).  Instead this article offers a short comparative review of two open access databases.

Four specific perspectives are used to examine the utility of these databases for reducing embodied carbon:

• At the early design stage, when material choices are made, with the choice of structural materials commonly responsible for a significant proportion of the total impacts up to the end of construction (Robati et al. 2021).
• The embodied impact of services components.
• The impact of materials at the end of life of a building and beyond are now becoming a critical concern. 
• The potential of these two databases in contexts where national databases do not exist.

Introduction to the databases

EPiC is an Australian-based database first published under this name in 2019 and updated in 2024 (Crawford et al. 2019, 2024). It provides embodied coefficients of environmental flows, including energy and water as well as GHG emissions (carbon), for a wide range of commonly used construction materials and products. The scope covers the cradle-to-gate system boundary, encompassing the processes involved in producing these materials and products, such as raw material supply, transport, and manufacturing (known as module A1-A3 in EN 15978:2011). Additionally, the database includes appendices with information on calculating construction stage impacts (A4 and A5 within EN 15978:2011) for various construction project types, including residential, non-residential, and heavy civil engineering construction. Coefficients are derived using a hybrid approach known as Path Exchange hybrid analysis (Lenzen & Crawford 2009). The database adheres to International (ISO) and European (CEN) standards for building sustainability assessments.

The ICE database was first released by the University of Bath in the UK as a Beta version in 2006 (Hammond & Jones 2006) and has been periodically updated. The latest version, ICE V3.0, was released in 2019 (Circular Ecology 2019). This database provides embodied coefficients of carbon for a large number of common construction materials. Similar to EPiC, the scope of the ICE Database includes the cradle-to-gate boundary (module A1-A3). ICE V3.0 has seen significant improvements in terms of boundaries, methodologies, and implemented data quality indicators to assess the quality of data. In the current version, Environmental Product Declarations (EPD) conforming with EN 15804 have been the source of data for almost 90% of data, providing a standardised methodology.

The two databases clearly differ in world region. As material coefficients will vary with differences in manufacturing processes, variation in production efficiency, quantity and treatment of waste, regional differences in fuel mix and transport distances (RICS 2023; Mohebbi et al. 2021), country of origin can make a significant difference. EPiC utilises the Australian Life Cycle Inventory (AusLCI 2024) and Australian input-output data from 2014-15 as its primary data sources and so has internal consistency. The ICE coefficients on the other hand draw on EPD data from around the globe. While this means that there is less consistency within the data sources, the ICE database explains where figures come from, and also includes values for certain materials (for example, aluminium) by continent and for a few specific countries, as well as a “world average production with world average recycled content”.

While ICE provides all figures in kgCO₂e per kg of bulk material, the EPiC Database often provides values for manufactured material products, in different units. For the example of aluminium again, it offers figures for aluminium extruded round tube (25 mm dia.), or aluminium sheet (1.6 mm) rather than for bulk aluminium. These differences can each be a benefit for different applications.

Currency of data is also an important factor in accurately reflecting technological advancements and increasing efficiency in processes. Both databases have some data which is several years old.  For the most detailed and accurate calculations, designers should always use product-specific and up to date data such as provided by manufacturer’s EPD.

The ICE Database is available in Excel file format, making it easily accessible for various applications. EPiC is available in various digital formats, including Excel files, a Grasshopper plug-in, and a Python package, catering to a wide range of applications.

The perspective of structural engineers

The project engineer or designer must consider the design stage, stakeholders' requirements, and relevant standards to determine what database to choose, although usability and relevance to purpose are likely to be the key deciding factors, alongside project location and supply chain.

The sources for the EPiC environmental coefficients are clearly described and accessible.  These coefficients comprehensively estimate environmental flows, including direct emissions (associated with main production processes), and indirect emissions (associated with the entire supply chain, processes and activities upstream of and supporting the main processes). In the Grasshopper version EPiC additionally documents the service life (enabling the calculation of recurrent environmental flows) and wastage coefficients for most materials.

ICE provides data from diverse EPD, but also offers a data quality matrix based on the statistical analysis of collected data, assessing method, assurance, temporal factors, geography, transparency and sample size, all combined in a Data Quality Indicator (DQI Total). This usefully informs the designer about the reliability of the calculated embodied GHG emissions. An Excel-based tool is provided within ICE database for cement, mortar and concrete (V1.1 Beta 2019), allowing designers to input cement replacements with different percentages. The tool then analyses the contributions to GHG emissions. In previous versions of the ICE database, the embodied energy was also provided for the documented materials.

Examples of some results for typical structural materials are given in Table 1 from the two databases.  As is known, input-output and hybrid methods such as used in EPiC produce higher figures due to the wider system boundaries considered. However EPiC also provides percentages of process data representing the hybrid environmental flow as fractions of 1. This feature can help identify energy- or resource-intensive materials and facilitate the comparison of EPiC coefficients with those from process-based databases (e.g. ICE), helping designers in assessing differences and variations when used in embodied GHG calculations.

Table 1: Examples of typical results from EPiC and ICE.

* assuming density 2400 kg/m³
** densities vary

Use of the databases for services components

Studies have suggested that the embodied carbon of the mechanical, electrical and public health/plumbing (MEP) equipment of a building may comprise between 2% and 27% of the total embodied carbon in new construction projects (Bagenal et al. 2019). This is due to their material composition (metals, plastics and electronics) and processes involved in their manufacturing. For whole life calculations which include component replacements over the building lifetime, this percentage can be even higher (Rodriguez et al. 2020). CIBSE (2021) published “Embodied Carbon in Building Services: A Calculation Methodology (TM65)”, which includes a table of embodied carbon coefficients of key materials used in MEP products. The table is primarily based on data obtained from the ICE database, although cast iron and silicon were sourced from the EPiC Database, as their coefficients are not available in the ICE database.

Embodied carbon assessments for building services are often performed at the product level, using average embodied carbon coefficients, since the bill of quantities is often expressed in raw materials. EPiC’s focus on material products, such as aluminium sheet, rather than raw materials such as aluminium, may in some instances make it more difficult to use in the calculation of the footprint of MEP products, which are often manufactured from raw material. However, it is crucial to use data that reflects current manufacturing processes. Therefore, the most accurate estimate would use both databases to calculate the A1 module embodied carbon of MEP products, using ICE as the default database, and EPiC for specific manufactured material products.

End-of-life and circular economy considerations

Neither EPiC nor ICE offer detailed data on end-of-life (EoL) scenarios (module C in EN 15978:2011) and beyond (module D in EN 15978:2011).  

While EPiC provides input-output data that can be used to assess end-of-life processes, it lacks detailed information on material circularity and the impacts of various disposal methods, such as landfilling, incineration, recycling, and re-use.  This omission limits the database's utility for thorough EoL analyses and strategic planning aimed at circular economy and Net-Zero targets. Consequently, users who need extensive EoL data must look to additional resources to supplement EPiC's offerings.

Similarly, the ICE database also focuses on the product stage (module A1-A3) and does not provide comprehensive data on EoL or beyond. This limitation reduces its effectiveness for in-depth end-of-life and circular economy evaluations. While the ICE database is suitable for quick assessments of embodied carbon, its simplified approach and limited scope necessitate additional resources for whole life cycle assessments (LCA) and sustainability planning.

This gap in both databases reduces their utility for contemporary methods of carbon accounting and tracking, such as material passports and blockchain-based carbon tracing at end-of-life projects, which will need to use other data sources.

Use in countries lacking a national database

EPiC and ICE, both open access databases, serve as viable proxies for countries without their own environmental performance databases. Although each has limitations in providing accurate results, both databases offer valuable representations of embodied environmental flows to support informed design decision-making. They both present embodied coefficients in common functional units and summary tables, streamlining the assessment process and facilitating easier adaptation for countries without specialised expertise in environmental assessments.

The choice between the two depends on a clear understanding of the intended use. Factors influencing this decision include the specific requirements of environmental assessments, geographical location, data availability, and the desired level of detail and comprehensiveness.

EPiC is particularly advantageous for creating comprehensive environmental performance profiles for construction projects due to its coverage of various impacts such as water, energy, and GHGs, supporting a standard lifecycle assessment approach. Its availability in multiple digital formats is an added benefit. However, its development using Australian-specific data may necessitate adjustments for use in other regions, and it is important to recognise the uplift in impacts due to the wider boundaries in hybrid analyses for countries considering targets and benchmarks, or when comparing results with process-based calculations.

In contrast, ICE focuses specifically on embodied carbon coefficients (and previously energy), drawing data from a global pool, and using process-based EPD sources. This extensive data coverage may make it a more versatile proxy for countries without national databases.

Recommendations

Both EPiC and ICE are valuable tools for initial and high-level lifecycle assessments.  Their open access makes it easy for practitioners to use them either as raw data or through incorporation into in-house building LCA tools. Indeed, the development of ICE as an open access database almost 20 years ago arguably increased understanding of LCA across the UK construction industry and wider (Moncaster & Malmqvist 2020). The availability of EPiC, with its additional digital formats, is also likely to increase use yet further, and reportedly has already done so in Australia.

Both databases have limitations. Because both compile average data, rather than for individual specific materials and products from a particular source, they are most useful for informing decisions at the early design stage; average data doesn’t support more detailed design choices or offer high accuracy for final calculations. EPiC and ICE also primarily focus on cradle-to-gate assessments, although EPiC also includes information on durability, therefore providing some of the data needed for the in use life cycle stage assessments.  Both lack detailed data for end-of-life scenarios and beyond lifecycle stages, necessary for comprehensive lifecycle assessments and strategic planning aimed at circular economy and Net-Zero targets. For more detailed calculations of building environmental impacts, it is advisable to consider the use of specific EPDs and material passports. However these two databases both provide an excellent free resource which will support the further development and increased understanding of this important field.

References

AusLCI. (2024). The Australian Life Cycle Inventory Database Initiative. https://www.auslci.com.au

Bagenal, C., Hamot, L. & Levey, R. (2019). Understanding the importance of whole life carbon in the selection of heat-generation equipment. CIBSE Technical Symposium: Transforming Built Environments: driving change with engineering. University of Sheffield. 25–26 April 2019. London: The Chartered Institution of Building Services Engineers.

Cabeza, L.F., Boquera, L., Chàfer, M. & Vérez, D. (2021). Embodied energy and embodied carbon of structural building materials: Worldwide progress and barriers through literature map analysis. Energy and Buildings, 231. 110612. https://doi.org/10.1016/j.enbuild.2020.110612

Circular Ecology. (2019). ICE version 3.0. https://circularecology.com/embodied-carbon-footprint-database.html  

Crawford, R.H., Stephan, A. & Prideaux, F. (2019). Environmental Performance in Construction (EPiC) Database, The University of Melbourne, Melbourne. https://www.researchgate.net/profile/Andre-Stephan/publication/337730690_Environmental_Performance_in_Construction_EPiC_Database/links/5e5cfd67a6fdccbeba12cb87/Environmental-Performance-in-Construction-EPiC-Database.pdf

Crawford, R.H., Stephan, A. & Prideaux, F. (2024). Environmental Performance in Construction (EPiC) Database, The University of Melbourne, Melbourne. https://doi.org/10.26188/10257728.v11

Hammond, G. & Jones, C. (2006). Inventory of Carbon & Energy (ICE) version 1.5 Beta. University of Bath, Department of Mechanical Engineering. https://www.inference.org.uk/sustainable/LCA/ICE%20Version%201.5%20Beta.pdf

IEA EBC. (2023). Annex 89: Ways to Implement Net-zero Whole Life Carbon Buildings.  International Energy Agency, Energy in Buildings and Communities Programme. https://annex89.iea-ebc.org

Lenzen, M. & Crawford, R. (2009). The path exchange method for hybrid LCA. Environmental Science & Technology, 43.21: 8251-8256.

Mohebbi, G., Bahadori-Jahromi, A., Ferri, M. & Mylona, A. (2021). The role of embodied carbon databases in the accuracy of life cycle assessment (LCA) calculations for the embodied carbon of buildings. Sustainability, 13(14), 7988. https://doi.org/10.3390/su13147988

Moncaster, A. & Malmqvist, T. (2020). Reducing embodied impacts of buildings – insights from a social power analysis of the UK and Sweden. IOP Conference Series: Earth and Environmental Science. Vol. 588. No. 3. IOP Publishing. https://doi.org/10.1088/1755-1315/588/3/032047

RICS. (2023). Whole life carbon assessment for the built environment. 2nd edition. https://www.rics.org/content/dam/ricsglobal/documents/standards/Whole_life_carbon_assessment_PS_Sept23.pdf

Robati, M., Oldfield, P., Akbar Nezhad, A., Carmichael, D.G. & Kuru, A. (2021). Carbon value engineering: A framework for integrating embodied carbon and cost reduction strategies in building design. Building and Environment,192, 107620. https://doi.org/10.1016/j.buildenv.2021.107620

Rodriguez, B. X., Huang, M., Lee, H. W., Simonen, K. & Ditto, J. (2020). Mechanical, electrical, plumbing and tenant improvements over the building lifetime: estimating material quantities and embodied carbon for climate change mitigation.Energy and Buildings,226, 110324.https://doi.org/10.1016/j.enbuild.2020.110324

Teng, Y., Zhengdao Li, C., Shen, G.Q.P.,  Yang, Q. & Peng, Z. (2023). The impact of life cycle assessment database selection on embodied carbon estimation of buildings. Building and Environment, 243, 110648. https://doi.org/10.1016/j.buildenv.2023.110648

Share:

Previous article

Why Convergence Research is Needed

Next article

Community-based Monitoring of Urban Environmental Data