Since the Industrial Revolution, human polluting activity known as anthropogenic interference has already caused 1.0°C of global warming (IPCC, 2019). A further increase to 1.5°C will be reached between 2030 and 2052 if emissions increases continue at the current rate (IPCC, 2018). However, scientists calculate that reaching and sustaining net zero global anthropogenic carbon dioxide (CO2) emissions by mid-century, will halt global warming on a multi-decadal scale and temperature gains will begin to peak (IPCC, 2018). To achieve this goal, it is calculated that the world cannot rely solely on key greenhouse gas abatement strategies, such as vehicle electrification and renewable energy transition (Dft, 2019a & b, IEA, 2019, UNEP, 2019). This is because evidence indicates that the rapidity of adoption and associated abatement will not be sufficient to bridge the projected 32GtCO2e annual emissions gap forecast for 2030 (UNEP, 2019). As an alternative, scientists and governments agree that all aspects of human pollutant activity must be examined and low carbon alternatives researched and diffused during the next decade to compensate for this limitation (IPCC, 2018). Specifically, the United Nations Environmental Programme (UNEP) suggests that to bridge the gap, the world must combine existing technology with innovation to drive behavioural changes capable of reducing societal emissions (UNEP, 2019).
Considering the criteria, this research proposes personal computing as a candidate technology for participation in this alternate strategy. The rational being that as a mature technology, end user computing (EUC) generates in excess 1% of global GHG annual emissions (Andraea and Edler, 2015; Bekaroo et al., 2014; Belkhir and Elmeligi, 2017; GeSI, 2008, 2012, 2015, 2019; Malmodin et al., 2013) and represents a rich source of pollution abatement. This pollution is caused by the yearly manufacturing of 460 million devices (Gartner, 2021; Statistica, 2020 and 2021) and the associated energy consumed by 4.2bn active users (Datareportal, 2021) (IEA, 2021a). In context this is equal to annual pre pandemic aviation emissions (IEA, 2021b). The emissions are generated by four phases of the EUC product lifecycle including embodiment, transportation, use phase energy (UPE) consumption and disposal. Transport and waste emissions contributions are consistent, producing 5% and 1% respectively (Apple, 2021; Dell, 2021; HP, 2021; Lenovo, 2021; Microsoft, 2021). Consequently, life cycle assessment (LCA) research indicates that the total carbon footprint of EUC devices is predominantly generated by the embodied and use phase emissions (Andrae and Andersen, 2010; Andre et al., 2019; Arushanyan et al., 2014; Subramanian and Yung, 2016). Whilst this is agreed, the proportionate representation of each value varies considerably between findings. As an example., the contribution of the embodied phase ranges from 12–97% and conversely use phase emissions from 3–88% (Atlantic Consulting and IPU, 1998; Choi et al, 2006; Duan et al., 2008; Hart, 2016; IVF, 2007; Kemna et al., 2005; Kim et al., 2001; Lu et al., 2005; PE International, 2008; Sahni, S. et al., 2010; Socolof et al., 2005 and 2017; Tekawa et al., 1997; Teehan and Kandliker, 2012; Williams, 2004).
The incongruity is caused by four key variables, one attributed to the embodied emissions and three to the use phase emissions. The first being that whilst the lifecycle inventory (LCI) process is governed by international standard ISO2006 (Haque, 2020), embodied results variations are caused by differences in the way LCI data sources are calculated. As an example, values for metal, a key component of EUC device cases, fixings, wiring, chipsets, fans and hard drives, are subject to differing LCI quantification methods that either include or exclude the environmental influence of post-sales values such as exergy (Sonderegger et al., 2017; Steen, 2006). As such, depending on which database is accessed during calculation, the embodied value may change in prominence whilst remaining theoretically accurate (Finnveden et al., 2016; Peters and Weil, 2016; Rigamonti et al., 2016; Rorbech et al, 2014).
The second variable is caused by the annual kilowatt hour (kWh) value attributed to the device. Unless specifically measured in the field with a watt metre, researchers and manufacturers predominantly rely upon two sources of data to determine the kWh value. If either source is inaccurate or not appropriate, then the proportionate representation of the UPE value will be affected. The first source is existing secondary field measured data. Whilst legacy sources from the late twentieth century exist in relative abundance, contemporary sources are recognised as highly limited (Greenblatt et al, 2013; Karpagam and Yung, 2017; Malmodin et al, 2010). As such, due to energy efficiency improvements (GeSI, 2008; McManus, 2002) EUC device UPE consumption quantification that relies upon legacy data will generate findings that are subject to a lack of specificity, obsolescence and extrapolation (Malmodin et al., 2010) thus reducing accuracy and validity. The recent limitation of available field data is due to 84% of devices now being mobile (Gartner, 2021; Statistica, 2020 and 2021) meaning that as people work from multiple locations daily (Gartner 2019,, traditional static watt metre measurement is unfeasible. Greenblatt et al. (2013) emphasise that consequently, widespread UPE field measurement is now avoided due to scale and mobility creating unsurmountable logistical complexities. Such is the limited availability of contemporary field data, Karpagam and Yung (2017) note that whilst conducting EUC device LCA comparison analysis their work was made all the more difficult by what is described as a field that is ‘data starved’. Belkhir and Emeligi, (2018) concur, conceding that EUC UPE consumption findings are subject to error as validity of use profile variations is sought from sources predominantly tied to the desktop era between 1988 to 2002 (Norford et al, 1988; Koomey et al. 1995, Kunz, 1997, Komor, 1997, Hosni, Jones, and Xu, 1999, Roth et al., 2002). Intellect (2016) consequently echo Malmodin’s (2010) concerns concluding that using legacy source data to calculate modern day EUC emissions is unreliable due to data being obsolete.
To compensate for the recent limitation, a second source of UPE consumption data offers a contemporary baseline value in the form of pre-sale energy efficiency Energy Star benchmarks (Energy Star, 2021). Conducted under strict test set up and conduct regulations, the programme accurately measures newly manufactured EUC devices for power draw in no-user present operational modes such off, sleep and idle (Energy Star, 2017). The results are published online (Energy Star, 2021) and include a typical energy consumption (TEC) value to represent an anticipated annual kilo-watt hours (kWh) value. Whilst used as the basis for manufacturer carbon footprint publications the values are ultimately without validity in the context of an LCA as they do not include the active operation mode when a user is interacting with the device. Research determines (Sutton-Parker, 2020) that this causes the TEC value to be inappropriate as a substitute for EUC UPE consumption field measurements as the additional power required as the device carries out useful work is excluded from any calculations (Sutton-Parker, 2020). Specifically, the inaccuracy ranges from − 48% to + 107% (Sutton-Parker, 2020) consequently causing LCI calculations reliant upon the TEC method to under estimate the proportionate representation of EUC UPE consumption by an average of 30%.
The third variable is arguably simple, in the fact that the incongruity relates to the number of years of energy consumption included within the final LCA. The rationale being that including five years as opposed to three years of energy consumption will affect the UPE contribution by 40%. As an example, when preparing product carbon footprint reports the world’s five largest EUC device suppliers (Gartner, 2021) include various duration including 3-years (Microsoft, 2021), 4-years (Apple, 2021; Dell, 2021; HP, 2021) and 5-years (Lenovo, 2021). The lack of uniformity is driven by consensus of opinion not delivering specificity. As such, prevailing research determines that the initial ‘first use’ retention period has an average duration of between three and five years (Hart, 2016; Prakash et al., 2016; Thiébaud et al., 2017; Teehan and Kandliker, 2012; Williams and Hatanaka, 2005). This is predominantly influenced by factors such as company asset management and depreciation accounting and refreshes forced by a necessity to keep pace with new applications (Boyd, 2012). Where a ‘second use’ exists, if the device is sold or repurposed rather than disposed of, then this additional retention period is between two and three years (Prakash et al., 2016; Thiébaud et al., 2017). Consequently, it is reasonable to conclude that the lifetime input for the use profile ranges from three years to eight years before disposal. However, extending the lifecycle beyond five years is subject to research opinions that suggest the diminishing efficiency performance undermines the sustainability case for displacement of new devices (Bakker et al., 2014; Boyd, 2012, Cooper and Gutowski, 2017; Deng et al., 2011; Prakash et al., 2016; Schiscke et al., 2003; Vadenbo et al., 2017; Wolf et al., 2010). Consequently, similar to the embodied methodologies, selecting one duration over another remains theoretically accurate yet highly inconsistent.
The final variable affecting the GHG emissions generated by the determined UPE consumption value is the location in which the electricity is consumed. CO2e is the accounting unit that represents a unified value for all of the greenhouse gases (WBCSD and WRI, 2004). UPE consumption CO2e GHG emissions are calculated by multiplying the electricity consumed value (kWh) by the GHG conversion factor published annually by each government where the energy is consumed (DoBEIS, 2021). The factor is created to reflect the carbon intensity of the electricity supply grid. EUC device manufacturers supply goods into regions with three different volts alternating current (V ac) electricity supplies including North America and Taiwan (115 V ac), Europe, Australia and New Zealand (230 V ac) and Japan (100 V ac) (Energy Star, 2017). As the 115 and 230 V ac are the largest, the relevant product carbon footprint reports are predominantly produced for one or the other depending on the brand (Apple, 2021; Dell, 2021; HP, 2021; Lenovo, 2021; Microsoft, 2021). The reality is that each country within a region will publish a different conversion factor based upon the carbon intensity included within the national supply grid (Carbon Footprint, 2020). The factors differ because all countries adopt renewable energy at different rates. Therefore, a country with a higher percentage of renewable energy supply will have a lower conversion value as it is producing less emissions per energy unit consumed. As an example the USA, which has been slow to transition to solar, wind and water source energy, has a conversion factor of 0.45322 (Carbon Footprint, 2020) compared to the UK factor of 0.21233 (DoBEIS, 2021). The difference being that for 10 kWh of electricity consumed in the former will create 4.5 kgCO2e compared to the latter of 2.1 kgCO2e. Consequently, any LCA research or quantification relying directly upon manufacturer supplied UPE GHG data sources may over or under emphasise the value due to the impact of location.
Whilst the issue of embodied emissions incongruity is beyond the scope of this research, previous research designed to address key issues such as scale and mobility affecting the accuracy of EUC UPE consumption values have been undertaken. Notably, in response to increasing legislation and policy to reduce scope 2 emissions in the public sector, Cartledge (2008) and Hopkison et al (2009) produced the SustIT/JISC tool. Essentially an EUC device specific version of the UPE consumption input tables from Kenma et al (2005) LCA energy consumption calculator, the tool enables any organisation wishing to complete EUC use phase emissions quantification to do so following a few simple steps. First the organisation simply conducts an asset profile exercise and then inputs the high level results (e.g. 20 x notebooks) into the tool. Using the TEC use profile formula, the researchers create a type (e.g. notebook or desktop) relevant use profile based upon energy measurements conducted within the relevant universities where the research was conducted. The resulting value is then multiplied by the relevant carbon emissions factor and a kgCO2e unit value is produced. Whilst logical, again the limitation of the imposed use profile based upon a fixed seventy active hours per week may address the inclusion of an active value, it does introduce error of non-specificity raised by Malmodin et al (2010). The issue lies within the uniform UPE consumption value applied to device types (e.g. notebooks) rather than the specific notebook used by an organisation. As an example, the field measured annual electricity consumption in the workplace for a Chromebook illustrated previously is 11.93 kWh (Sutton-Parker, 2020) generating 2.53 kgCO2e of annual scope 2 emissions if used in the UK (DoBEIS, 2021). Conducting the same calculation using the estimation tool (Hopkison et al, 2009) as the Acer device is categorised as a notebook, an average electricity consumed value of 30 kWh is applied (JISC, 2019). This value is translated to scope 2 GHG emissions of 6.37 kgCO2e per device. As such, the inaccuracy introduced is equal to + 152%. To overcome both non-specificity and mobility barriers software has previously been trialled to achieve the similar action of a watt metre based upon the simple rationale that in this form the ‘measuring component’ can simply move with the EUC device. The approach was called Joulemeter (Kansell, 2010) and was capable of measuring and reporting real time energy consumption of both physical IT hardware, virtual machines (VM) and software applications. Whilst the idea of moving to software based measurement would have offered scope for wide scale EUC device UPE consumption data to be generated, the tool suffered a setback for two reasons. Firstly, it required a watt meter for a calibration phase, thus re-introducing the issue it was designed to overcome plus upon scrutiny (Bekaroo et al, 2014) it proved to only achieve 59% accuracy. Subsequently, the software failed to progress and is noted only by Microsoft as no longer publicly available and deprecated. Consequently, to advance the concept of capturing EUC device UPE consumption values regardless of scale, mobility and location, and ultimately improve the availability of associated field data, this research uses analytics to capture both asset and use profile data. As such the following sections describe the methodology used to conduct the field experiment and the results and discussion generated by the undertaking.