Medas M.(a,b), Cheshire D.(a), Cripps A.(a), Connaughton J.(b) and Peters M.(b)
a) AECOM, Building Engineering, London, UK
b) Technologies for Sustainable Built Environments Centre, University of Reading, Reading, UK
Keywords: BIM; embodied carbon; resource efficiency; building services.
Abstract: Recent research has highlighted the importance of increasing the resource efficiency of building services. In this paper, a literature review of life cycle assessment and embodied carbon studies clearly signposts the need for more research on the embodied impacts of building services. Theoretical and practical challenges are identified in the use of LCA methods in construction along with possible solutions compatible with BIM-led design. A case study of horizontal, ducted fan coil units demonstrates that variations in embodied carbon of similar units could be associated with variations in total cooling capacity. This raises the possibility that metrics could subsequently be devised to predict embodied carbon for a generic class of fan coil units. When combined with data on operational carbon emissions and product service life this could enable optimisation of lifetime carbon impacts. Implications of these results are explored and recommendations made for future studies.
If resource efficiency is defined as making ‘best use of materials, water and energy over the lifecycle of built assets to minimise embodied and operational carbon’ and other environmental impacts (WRAP, 2015), its implementation requires measurement and mitigation of these impacts. Building on previous research into resource efficiency of building services (CIBSE, 2014), this study considers how better calculation methods for the embodied impacts of these services might support building information modelling (BIM) tools to enable resource-efficient choices of services. Before this, we appraise current gaps in knowledge about the resource impacts of building services. A case study into embodied impacts of fan-coil units is then described and its implications discussed, after which recommendations are set out for future studies.
Published research into resource efficiency in the built environment has prioritised operational energy efficiency, driven by greenhouse gas (GHG) emission targets and the building sector’s responsibility for a third of global anthropogenic emissions (Ibn-Mohammed et al, 2012). The main theoretical approach used in these studies has been life cycle assessment (LCA), an approach supported by general and construction-specific ISO standards (BSI, 2006a, 2006b, 2011, 2013).
The importance of embodied impacts
Recent studies indicate that improved operational energy efficiency of buildings will raise the embodied proportion of lifetime building environmental impacts (RICS, 2010, HM Government, 2010). The term ‘operational carbon’ (OC) hereafter refers to CO2 equivalent emissions from a building or its components during its use phase, while ‘embodied carbon’ (EC) describes non-operational emissions from all life-cycle phases. Estimates from published research of the EC proportion of lifetime building carbon emissions vary owing to their sensitivity to parameters such as building lifetime and the lack of agreed calculation methods (Ng, Chen & Wong, 2013). Consequently studies are needed that can measure embodied impacts as robustly as has been done with operational impacts. While recent LCA studies have considered whole buildings as well as specific building elements (Ortiz, Castells & Sonnemann, 2009), few have measured life cycle impacts of building services, especially their embodied impacts (Passer, Kreiner & Maydl, 2012).
LCA studies of building services
The few published LCA studies into building services produced since 2004 have yielded relevant insights: (a) When comparing systems or products, lower embodied impacts are often associated with higher operational impacts (Hekkila, 2004, Chen, Zhang & Setunge, 2012) or higher reusability of an alternative (Franklin Associates, 2008); (b) Embodied impacts of service components relative to mass can exceed those of building fabric (Chau et al, 2007); and (c) Obtaining reliable data on raw materials within composite building services components presents challenges (Chau et al, 2007, Whitehead et al 2012). These points support a need for accurate measurement of embodied impacts of building services.
Embodied carbon based studies
Following a UK government report that recommended ‘whole life carbon appraisal’ of construction products (HM Government, 2010), recent UK studies have measured the EC intensity of buildings and their components in CO2 equivalent, using global warming potential (GWP) to represent multiple environmental impacts addressed by LCA, while meeting LCA standards (Moncaster & Symons, 2013). Results show that building services can represent 3-15% of initial EC of a typical office building, worth 20-126 kg CO2e/m2 (CIBSE, 2014), while recurring EC of building services from 30 years of maintenance and replacement may be six times the value of initial EC (Franklin & Andrews, 2010).
In seeking reliable measurements for embodied impacts of building services, LCA is challenged theoretically by the uniqueness and complexity of buildings and resulting variation of method between LCA studies (Buyle et al, 2013, Cabeza et al, 2014). This can be redressed (a) by using LCA standards specific to buildings and building products (BSI, 2011,2013); (b) by considering only GWP (RICS, 2010), thus avoiding weighting between impacts; and (c) by studying comparable building components rather than hard-to-compare entire buildings. The latter approach underlies Type III Environmental Product Declarations (EPDs) (BSI, 2013, EPD International, 2015).
LCA studies of building services are also challenged by practical gaps in data, tools and policy. Life cycle inventory (LCI) databases in construction typically contain simple materials and not composite building services components (Hammond & Jones, 2011, Ecoinvent, 2015). EPDs can fill this data gap, but are limited by being voluntary and costly, while UK building regulations omit embodied emissions. The data gap might be resolved by developing parametric methods to predict embodied life cycle impacts of generic building services components (Moncaster & Symons, 2013).
LCA tools and BIM integration
Three types of tools can measure life cycle impacts of building components (Cabeza et al, 2013). These are (a) product comparison tools such as Simapro (2015) and GaBi (2015), (b) whole-building decision tools such as Athena Eco-calculator (2015); and (c) whole-building assessment frameworks such as BREEAM (BRE Global, 2014) and LEED (USGBC, 2009). The first two measure impacts quantitatively, while the third combines quantitative and qualitative methods. Product comparison tools are complex and ill-suited to lay users, while other tools sometimes include embodied impacts of building services (EtoolLCD, 2015) but still face data gaps.
A BIM approach to design may improve whole building support tools either by interfacing a 3D model with an LCI database (Capper et al, 2014, Tah et al, 2012) or by adding ‘EC’ fields to BIM files representing building components, as in CIBSE’s ‘product data templates’ (CIBSE, 2015). Figure 1 illustrates how this promises integrated calculation of carbon and 4D-design data for rapid decision-making. However, for this to happen the theory, metrics and data informing EC values must be robust.
Case study into fan coil units
Given the need for suitable metrics and data on embodied impacts of building services, the hypothesis to be tested is whether a parametric approach can reliably predict the EC value of composite building services components. This requires measurement of the EC value of a ‘base component’ representing a component class. This value is then multiplied by one or more coefficients such as mass or power rating to predict the EC value of components of different sizes. For empirical investigation, the UK office sector was chosen, due to its high value within mechanical and electrical building services work (BSRIA, 2014) and commercial property stock (BCO, 2013). The most common item of installed central air-conditioning plant in the UK is the fan coil unit (FCU) (BSRIA, 2014), for which selection is typically based on operational performance and cost (CIBSE, 2008a, ACR News, 2009).
The study focuses on water-side FCUs, (see Figure 2) in which heating or cooling is modulated by the flow rate of hot or chilled water, and which make up 75% of the UK market (CIBSE, 2008a). The main components and materials within horizontal, ducted, water- side FCUs, are (a) galvanised steel casing lined with polyurethane acoustic foam, (b) fans and motors made mainly of steel, aluminium and copper and (c) aluminium and copper heating and cooling coils. Details of the full range of FCUs produced by a UK based manufacturer were obtained, including masses of main materials, yield loss in manufacturing and cradle-to-gate transport data. This was combined with reference data on the carbon intensity of materials (Hammond & Jones, 2011) and energy used in manufacturing and transportation (DEFRA/DECC, 2015) in order to calculate carbon coefficients and EC values. Product details were obtained from four other UK manufacturers of comparable FCUs that had been tested against similar standard operating conditions. A comparison was made of the ratio between relative mass and total cooling capacity of each FCU by size and by manufacturer. Assuming consistency of raw materials between manufacturers, the EC intensity of FCUs in relation to total cooling capacity was compared by manufacturer. Explanations for variations between manufacturers were considered and a sensitivity check done to explore uncertainty in the results. Results were then considered vis-a-vis the relationship between operational and embodied carbon.
Table 1 shows a bill of materials averaged from a range of FCUs of between 2-8 kW in total cooling capacity. Comparing columns 3 and 4, steel represents over 77% of FCU mass but only 56.2% of total EC of the product, as the greater carbon intensities of other materials give them larger shares of total EC. The total EC of the FCU is 193 kg CO2 e for a total mass of 69 kg, giving an EC coefficient of 2.8 kg CO2 e/kg. The total value of EC also exceeds the EC of raw materials within the finished product, as 24.3% of total EC is from scrap steel and polyurethane foam generated in making the FCU’s casing.
How do parameter values vary at different levels of total cooling capacity (TCC)? While the material breakdown of each FCU is broadly consistent across the size range, total mass and EC both rise proportionately with TCC as shown in Figure 3.
Figure 4 shows that mass and EC of each FCU relative to TCC varies slightly downwards as FCU size and TCC increase, except for a spike at the second model in the range. The spike is explained by an increase in FCU mass between points 1 and 2 on the X-axis that exceeds the increase in TCC over this range. Generally, as FCU size increases, mass and EC rise absolutely but fall relatively in relation to TCC. This suggests that TCC could act as a coefficient to transform the EC values of a base FCU into values for an FCU of any particular size, provided that anomalies in FCU mass can be explained.
Figure 5 plots mass against TCC of FCUs made by five UK manufacturers, demonstrating that TCC varies directly with mass for all manufacturers, with the rate of variation differing between manufacturers.
Figure 6 plots the relative mass against TCC for the five manufacturers, showing an inverse relationship. Downward pointing trend lines for each series of coordinates indicate that larger models are evidently more efficient on a power to weight basis. However, this trend varies in strength between manufacturers with an r2 value of between 0.47 and 0.93.
Based on the detailed analysis of FCUs of one manufacturer, we assume that all five manufacturers use consistent proportions of materials across their size range, and that these are associated with consistent proportions of EC. If the variations between manufacturers in mass and mass/power ratios at each level of TCC in Figures 5 and 6 can then be explained, a base FCU could subsequently be defined, from which EC values of other units might be predicted by varying selected parameters.
Possible explanations for variations in mass and mass/power ratios between manufacturers might include (a) the use of materials of different densities to make similar parts; (b) differences in dimensions of the FCU casing, likely to affect relative mass of the FCU; and (c) differences in design quality giving rise to differences in material density. With limited data available drawn from product literature, these issues may only be explored tentatively via an exploratory sensitivity check.
To consider the first explanation, if galvanised steel casing represents the majority of FCU mass and UK-produced galvanised steel is of standard density, varying steel thickness should significantly vary FCU mass. Manufacturers A and C are known to use 1.2mm gauge steel and manufacturer E uses 1mm gauge steel, with typical densities of 9.083 kg/m2 and 7.888 kg/m2 respectively (Custompartnet, 2015). If the FCU casing represents 70% of average FCU mass, as is the case in our empirical case study, then manufacturer E’s models should have 1-(0.7- (7.888/9.083*0.7)) = 9.08% less mass than those of manufacturer A. In Figure 6, the effect of reducing the mass of FCU ‘A’ by 9.08% would shift its mass / power curve closer towards the curve of FCU ‘E’. This would explain approximately 30% of the difference in mass between manufacturers A and E at each level of TCC. While this cannot explain the mass discrepancy between manufacturers A and C, who both use 1.2mm gauge steel, it shows the importance of material density in explaining variations in mass.
The case study shows that variation in EC of FCUs of similar type at different rated levels of total cooling capacity are in theory predictable as long as differences between FCUs by manufacturer can be explained. The aim would be to compare FCUs of similar function but varying embodied content, and eventually compare alternative types of cooling systems, given sufficient data.
But what do these results mean in relation to the balance of embodied versus operational impacts?
The answer requires several calculations. The difference in initial EC between FCUs made by different manufacturers in this case study could be estimated by assuming that EC content per raw material was consistent between manufacturers, while material quantities may vary between manufacturers. Values for operational energy consumption and carbon emissions could also be estimated for each FCU given assumptions about daily and seasonal loads. However, to balance the EC and OC of a particular FCU would need two additional calculations.
Firstly, lifetime OC emissions of a building or building component vary with length of its service life and future changes in carbon intensity of electricity grids. A typical service life for an FCU based on its economic and technical lifespan is 15 years (CIBSE, 2008b), but new commercial leases last on average 5 years (BCO, 2013), at which point tenant fit-outs may involve scrapping FCUs. The calculation of FCU operational emissions must therefore be based on robust assumptions. Secondly, the recurring EC associated with maintenance, replacement and disposal of an FCU must be calculated and added to values of initial EC within the present case study. These calculations must precede estimation of lifetime carbon as each can increase the EC/OC ratio for an FCU. Product service life and embodied and operational carbon intensity are all therefore variables that influence the lifetime resource efficiency of a choice of building service product.
Conclusions and recommendations
This study has highlighted the need for more research on the embodied impacts of building services. It shows that BIM design tools can include assessment of these impacts if suitable metrics and data are first developed. A case study of horizontal, ducted coil fan coil units showed that variations in embodied carbon of similar units could be associated with variations in total cooling capacity. This clearly indicates that if differences between FCUs by manufacturer could be explained, metrics could be devised to predict the embodied carbon of a generic class of FCUs. Calculation of lifetime carbon impacts of FCUs was observed to depend on additional calculations, assumptions and variables including product service life.
In order to test fully the method of embodied carbon analysis used in this study, the preliminary findings would need to be validated using additional data. This should cover material composition and initial and recurring embodied carbon content of FCUs produced by a representative sample of manufacturers. The results should then be subjected to a detailed sensitivity analysis. In principle, if such a method can robustly predict embodied carbon impacts of a class of component, other environmental impacts linked to material composition might be predicted. More widely it is recommended that further studies are carried out on the embodied impacts of building components in order to raise awareness on the need for the design and application of robust methods to balance embodied and operational impacts of buildings.
ACR News (2009) Masterclass: Fan coil units – Part 37. Retrieved 28.2.15, from: http://www.acr- news.com/masterclass-fan-coil-units-part-37-
Athena EcoCalculator [computer software] (2015) Retrieved 28.02.15, from: http://www.athenasmi.org/our-software- data/ecocalculator/
BRE Global. (2014). BREEAM UK New construction: Non -domestic buildings, Technical Manual SD 5076: 0.1 (DRAFT). Watford, UK: BRE Global
British Council of Offices (BCO). (2013). Property Data Report 2013. London, UK: BCO.
British Standards Institution (BSI). (2006a). BS EN 14040: Environmental management – Life cycle assessment – Principles and framework. Retrieved from: https://bsol.bsigroup.com
British Standards Institution (BSI). (2006b). BS EN 14044: Environmental management – Life cycle assessment – Requirements and guidelines. Retrieved from https://bsol.bsigroup.com
British Standards Institution (BSI). (2011). BS EN 15978: Sustainability of construction works — Assessment of environmental performance of buildings — Calculation method. Retrieved from: https://bsol.bsigroup.com
British Standards Institution (BSI). (2013). BS EN 15084: Sustainability of construction works – Environmental product declarations – Core rules for the product category of construction products. Retrieved from https://bsol.bsigroup.com
Building Services Research and Information Association (BSRIA). (2013). The illustrated guide to mechanical cooling. Bracknell, UK: BSRIA
Building Services Research and Information Association (BSRIA). (2014). The BSRIA Bue Book. Bracknell, UK: BSRIA
Buyle, M., Braet, J., & Audenaert, A. (2013) Life cycle assessment in the construction sector: A review. Renewable and Sustainable Energy Reviews, 26, 379–388.
Cabeza, L., Rincon, L., Vilarion, V., Perez, G., & Castell, A. (2014). Life cycle assessment (LCA) and life cycle energy analysis (LCEA) of buildings and the building sector: A review – Renewable and Sustainable Energy Reviews. 29, 394–416.
Capper, G., Matthews, J, & Lockley, S. (2012, April). Incorporating embodied energy in the BIM process. In CIBSE Technical Symposium, London, UK.
Chartered Institution of Building Services Engineers. (2008a). CIBSE TM43, Fan coil units. London, UK: CIBSE.
Chartered Institution of Building Services Engineers. (2008b). Maintenance engineering and management – CIBSE Guide M. London, UK: CIBSE
Chartered Institution of Building Services Engineers. (2014). CIBSE TM56, Resource efficiency of building services. London, UK: CIBSE.
Chartered Institution of Building Services Engineers. (2015). Product Data Templates. Retrieved 28.02.15 from: http://www.cibse.org/knowledge/building- information-modelling/product-data-templates
Chau, C.K., Yik, F.W.H., Hui, W.K., Liu, H.C., & Yu, H.K. (2007). Environmental impacts of building materials and building services components for commercial buildings in Hong Kong. Journal of Cleaner Production.15,1840 -1851.
Chen, S., Zhang, K., & Setunge, S. (2012, July). Comparison of three HVAC systems in an office building from a life cycle perspective. In ICSBE (International Conference on Sustainable Built Environment’), 2nd International Conference on Sustainable Built Environment. Yogyakarta, Indonesia. Retrieved 28.2.15 from: http://www.civil.mrt.ac.lk/conference/ICSBE2012/ic sbe2012-12-259.pdf
Custompartnet. (2015). Sheet metal gauge size chart. Retrieved on 28.2.15 from: www.custompartnet.com/sheet-metal-gauge
Department for Environment, Food and Rural Affairs / Department of Energy and Climate Change. (2015) ‘UK government conversion factors for company reporting’’ (spreadsheet) Retrieved on 1.4.15 from: http://www.ukconversionfactorscarbonsmart.co.uk/
Econinvent [online database] (2015) Retrieved on 28.2.15 from: http://www.ecoinvent.org
EPD International AB. (2015). Retrieved on 28.2.15 from: http://www.environdec.com
EtoolLCD. [computer software] (2015). Retrieved on 28.2.15 from: < http://etoolglobal.com/software/>
Franklin + Andrews (2010) Hutchins UK Building Blackbook: The Capital Cost and Embodied CO2 Guide. Volume 2: Major Works. London, UK: Franklin + Andrews.
Franklin Associates. (2008). Life Cycle inventory of the production of plastic and metal pipes for use in three piping applications. Kansas: Eastern Research Group Inc. Retrieved on 28.2.15 from: http://www.lubrizol.com/CPVC/Documents/Sustain ability/Life-Cycle-Inventory-of-Plastic-Pipe.pdf
GaBi. [computer software] (2015). Retrieved on 28.2.15 from: http://www.gabi-software.com/uk- ireland/index/
Hammond, G., & Jones, C. (2011). Inventory of Carbon and Energy (ICE). Bath, UK: Bath University. Retrieved on 28.2.15 from: http://www.circularecology.com/embodied-energy- and-carbon-footprint- database.html#.VGEY0b9OXIY
Heikkila, K. (2004). Environmental impact assessment using a weighting method for alternative air-conditioning systems – Building and Environment, 39, 1133–1140.
HM Government. (2010). Low Carbon Construction Innovation and Growth Team: final report. Retrieved on 28.2.15 from: https://www.gov.uk/government/uploads/system/u ploads/attachment_data/file/31773/10-1266-low- carbon-construction-IGT-final-report.pdf
Ibn-Mohammed, T., Greenough, R., Taylor, S., Ozawa-Meida, L., & Acquaye, A. (2012). Operational vs. Embodied Emissions in Buildings – A Review of Current Trends. Energy and Buildings, 66, 232-245.
Moncaster, A., & Symons, K. (2013). A method and tool for ‘cradle to grave’ embodied carbon and energy impacts of UK buildings in compliance with the new TC350 standards – Energy and Buildings, 66, 514–523.
Ng, S.T., Chen, Y., & Wong, J.M.W. (2013).
Variability of building environmental assessment tools on evaluating carbon emissions – Environmental Impact Assessment Review, 38, 131-141.
Ortiz, O., Castells, F., & Sonnemann, S. (2009). Sustainability in the construction industry: A review of recent developments based on LCA. Construction and Building Materials, 23, 28–39
Passer, A., Kreiner, H., & Maydl, P. (2012). Assessment of the environmental performance of buildings: a critical evaluation of the influence of technical building equipment on residential buildings. International Journal of Life Cycle Assessment, 17 (9):1116-1130
Royal Institute of Chartered Surveyors (RICS). (2010). Redefining zero: Carbon profiling as a solution to whole life carbon emission measurement in buildings, London, UK: RICS.
SimaPro. [computer software]. (2015). Retrieved on 28.2.15 from: http://www.simapro.co.uk
Tah, J., Zhou, W., Abanda, F.H., & Cheung, F.K.T. (2012, June-July). Towards a holistic modeling framework for embodied carbon and waste in the building lifecycle. In Tizani, W. (ed.), Proceedings of the International Conference on Computing in Civil and Building Engineering. Nottingham, UK: Nottingham University.
US Green Building Council (USGBC). (2009). LEED 2009 for New Construction and Major Renovations Rating System. Retrieved on 28.2.15 from: http://www.usgbc.org/sites/default/files/LEED%202 009%20RS_NC_10-2013_1b.pdf
Waste Resources Action Plan (WRAP). (2015). Definition of resource efficiency in construction. [Fact sheet] – Retrieved on 28.2.15 from: http://www.wrap.org.uk/sites/files/wrap/WRAP%20 RE%20construction%20definition.pdf
Whitehead, B., Tozer, R., Shah, A., Kosik, B., Maidment, G., Andrews, D., & Dunn, A. (2012, April). The Environmental Burden of Data Centres – A Screening LCA Methodology. In CIBSE ASHRAE Technical Symposium. London: CIBSE.