Table 4. Improvements needed in different sustainability aspects and indicators for shale gas to be the most sustainable option.
Individual sustainability indicators and aspects
The results of the analysis when considering improvements in one indicator at a time together with its related aspect are shown in Table 5. As can be seen, the environmental impacts would need to be reduced by 9–329 times and their importance would have to be 10,000 times higher than of the other indicators, together with a 100 times greater importance of the environmental aspect relative to the other two. For the economic indicators, the levelised cost of electricity (LCOE) needs to be reduced by 32%, with its importance increasing by 100 times, together with a similar increase in the importance of the economic aspect.
As mentioned in the previous section, the capital cost does not need to be reduced, but if it is, then its importance must be increased by up to 10,000 times relative to the other indicators, along with a 100–10,000 times higher importance of the economic aspect (Table 5). The social indicators would need improvements similar in magnitude to those needed for the environmental indicators: direct employment by 16.4 times and public support by 13.6 times. Worker injuries do not need to be reduced for the ranking to change, but this indicator must be considered 1000–10,000 times more important than the others, together with a similar increase in the importance of the social aspect over the other two.
Table 5. Improvements needed in each indicator for shale gas to be the most sustainable option (changing one indicator at a time)a.
Therefore, the above results suggest that shale gas is unlikely to be the best option in comparison to the other alternatives considered in this work as large improvements and considerable, sometimes extreme, increase in the importance of indicators and aspects would be needed.
Changes needed for shale gas to be comparable to different electricity options
The results in the previous section demonstrate that it is all but impossible for shale gas to be considered the most sustainable option. While this is informative, arguably, it is not necessary for shale gas to be the most sustainable option and it could still potentially be viable if it can compete with some of the other established electricity options. Therefore, this section considers what would be needed to achieve that, starting with the focus on other fossil fuels (conventional gas and LNG), followed by nuclear and finally the renewable options (hydroelectricity and biomass).
Wind and solar PV are not considered as they are the most sustainable options based on the results discussed in 3.1 Sustainability of shale gas compared to other electricity options, 3.2 Sensitivity analysis so that improvements needed for shale gas to compete with these two would be similar to those considered in the previous section. Coal is not considered either as shale gas is a more sustainable option for most scenarios discussed in the previous sections. For brevity, only an overview of the results is provided here; for the detailed discussion, see Section S4 in SI.
Comparison of shale gas with conventional gas and LNG
As conventional gas and LNG rank higher than shale gas in the base case (Section 3.1), large reductions in the impacts of shale gas are needed, along with changes in the importance of the sustainability aspects. As indicated in Table 6, the environmental and social aspects need significant improvements while only a moderate reduction is needed for the economic costs. For example, for the LVF, to be comparable with conventional gas, a 20% reduction in environmental impacts from shale gas is necessary and the environmental aspect must be 13 times more important than the other two. Alternatively, 80% reduction in impacts should be achieved if all three aspects are considered equally important. For the EVF, a 100-fold reduction in environmental impacts is needed and the aspect must be three times more important. Similar results are found for LNG assuming the EVF.
Table 6. Improvements needed for each sustainability aspect for shale gas to become comparable to conventional gas and LNG.
When the individual indicators are considered, reductions are needed in nine out of 11 environmental impacts for shale gas to compete with conventional gas and in eight relative to LNG. Improvements are also needed in social and economic indicators: 18%–31% for LCOE and fuel cost and 32% to 6.5-fold for direct employment and public support (Table S2), along with large increases in the importance of these indicators (100–1000) and their related sustainability aspects (10–100 times). For the remaining indicators, no improvements are needed, but unlike the environmental indicators, large increases in aspect/indicator importance are needed for fuel cost and diversity of fuel supply (100–10,000 times).
Comparison of shale gas with nuclear power
As nuclear power ranks significantly better than shale gas in the base case (Fig. 2), significant improvements and increases in the importance of sustainability aspects and indicators are needed if shale gas is to be comparable. The magnitude of the improvements and increases in importance are similar to those needed for it to compete with conventional gas as nuclear has a similar ranking to it (Fig. 2). As shown in Table 7, the environmental and social aspects need the largest improvements (up to 100 times) while the needed reductions in costs are smaller (25%–40%).
Table 7. Improvements needed for each sustainability aspect for shale gas to become comparable to nuclear power.
When each indicator is targeted individually (Table S3), improvements are needed in seven out of the 11 environmental indicators. For these, 89% to 91-fold reductions are needed along with large increases in aspect/indicator importance (100–10,000 times). For the economic and social indicators, improvements are needed in the levelised cost of electricity, fuel cost, direct employment and public support (Table S3).
The levelised costs of electricity need a 21% reduction and 10–100 times increase in aspect/indicator importance, while the fuel cost must be reduced 16-fold and the importance of the aspect and the indicator should increase by 100–1000 times. A 72%–79% increase in direct employment and public support are required along with a 100-fold increase in the aspect and indicator importance (Table S3).
Comparison of shale gas with hydro and biomass electricity
As both hydro and biomass electricity are closer in ranking to shale gas than conventional gas, LNG and nuclear, smaller improvements and increases in the importance of the aspects and indicators are needed, as shown in Table 8. The social aspect needs the largest improvement (8–10 times), followed by the environmental (20%–50%) and economic (20%) aspects.
Table 8. Improvements needed for each sustainability aspect for shale gas to become comparable to hydro and biomass electricity.
However, significant improvements (9–329 times) are needed in all environmental indicators for shale gas to compete with hydroelectricity (Table S4). A 100–10,000 times increase in the importance of the aspect and the indicators is also needed. For the economic indicators, shale gas has lower levelised and capital cost than both hydro and biomass electricity, but its fuel cost is higher. As a result, no reductions in levelised and capital cost are needed but an increase in aspect/indicator importance of up to 1000 times is required (Table S4). On the other hand, fuel cost must be reduced to zero and the importance of the aspect/indicator increase 10,000-fold for it to compete with hydroelectricity while a 20% reduction and a 100-fold increase in the importance is needed for it to compete with biomass.
No improvement in worker injuries is needed but up to 50-fold increase in aspect/indicator importance is required. Direct employment should be improved by 16.4 times and 13 times higher public support is required for shale gas to compete with hydropower, along with a 100-fold increase in aspect/indicator importance (Table S4). To compete with biomass, an eight-fold increase in direct employment and 10.4 times greater public support are needed, together with 100–1000 times increase in aspect/indicator importance. For the diversity of fuel supply, biomass scores lower than shale gas and hence no improvements are needed, but a five to 100 times increase in aspect/indicator importance is necessary.
Effect of shale gas on the sustainability of electricity generation
The results in Table 9 suggest that, assuming equal importance of all the sustainability aspects and indicators, the electricity mix with low penetration of shale gas (1% on the grid) is considerably more sustainable than for the higher contribution (8%), with the respective sustainability scores of 0.74 and 0.44. This is to be expected because, as discussed in the previous sections, shale gas generally scores poorly in various impacts, including global warming potential, fuel cost and public support (see Table 1). Only when a 10,000 lower importance is placed on the environmental aspect do the two electricity mixes become comparable (Table 9).
Table 9. Sustainability scores for the low and high penetration of shale gas into the 2030 electricity mix in comparison with the current mix, assuming differing importance of sustainability aspects.
When the individual indicators are considered, for the high-penetration electricity mix to become comparable with the low, improvements are necessary in all but two environmental impacts as well as in capital and fuel costs, employment and public support. As can be seen in Table 10, the improvements needed range from 2%–16%. However, no changes to the importance of any aspect are required, except for the social when considering diversity of fuel supply; however, the importance assigned to each indicator must increase five to 80-fold.
Table 10. Improvements needed for each indicator for the 2030 high shale contribution mix and present mix to become comparable to the low shale contribution mixa.
It can also be seen in Table 9 that both 2030 electricity mixes are more sustainable than the present mix, assuming equal importance of all three sustainability aspects. This is not because of shale gas but due to a large drop in the contribution from coal and growth in renewables. The current mix is only better if the economic aspect is 3.9 times more important than the other two. This is due to the average levelised cost of fossil fuels being lower than that of renewables, making 2030 electricity more expensive. For the current electricity to be comparable to the 2030 mixes, improvements must be made to all social indicators (6% to 2.2 times), eight environmental impacts (36% to 7.9-times) and fuel cost (26%); see Table 10. An increase in the importance of the indicators and their related aspects (up to 1000-fold) is also required.
To assess the robustness of the results with respect to the MCDA method and the weightings used, the same analysis has been performed using direct weighting (DW) (Mustajoki and Hamalainen, 2000) as an alternative. This method is similar to SMART except that the weightings are inputted directly into the model, while in SMART they are calculated based on the assigned scores (see Section 2.1 and Section S1 in the SI). The rankings obtained through DW remained the same for all the weightings considered in SMART and discussed in the previous sections, thus validating the robustness of the results. It is possible that the rankings would change with other MCDA methods but their exploration is out of the scope of this paper.
As discussed in Section 2.3, the quality of the data underlying the sustainability assessment has been evaluated according to the six criteria in the pedigree matrix (see Table S2 in the SI). Overall, the data quality is estimated to be ‘medium’ for the environmental and economic assessments, ‘high’ for the social sustainability assessment and ‘medium’ for the 2030 electricity mix (Table 11). This would suggest that the results are valid, but further improvements to the data used would increase their robustness.
Table 11. Data quality assessment using a pedigree matrix.
Some data sources were of poor quality, in particular the sample size for the LCC data and geographical correlation for the LCA (Table S3 in SI). This is due to the data used to estimate the cost of producing shale gas in the UK being based on reports which estimate the cost of establishing a UK shale industry. Similarly, as the UK has no shale gas industry but only exploration wells, US data for material and process requirements have been used to model shale gas wells. Despite this, the overall data quality is ‘medium’ to ‘high’. Also, the quality of the literature data scored well in comparison to the Ecoinvent data used (Table S3 in SI).
The results of this study show that, assuming equal importance of the environmental, economic and social aspects, shale gas ranks seventh out of the nine electricity options considered for both values functions. In that case, wind and solar PV are the most sustainable and coal is the worst option. If the environmental impacts are the most important, hydropower becomes the best option, with shale gas ranking fourth to seventh, depending on the value function used.
For high importance of the economic aspect, solar PV is the best option while coal and hydropower represent the least sustainable options; shale gas ranks sixth or seventh. Finally, if social aspect is the most important, solar PV is the most sustainable option with coal and LNG being the worst options; shale gas is in the sixth to eighth place. Therefore, while overall not the worst, shale gas is not one of the better options either.
Despite this, it is possible to arrive at an outcome where shale gas is the best option by altering the importance placed on the indicators and aspects, as well as by improving its performance in different indicators. However, these are very significant and unrealistic. For example, if the importance of the capital cost and the economic aspect is 10,000 higher, shale gas becomes the best option (together with conventional gas and LNG). Similarly, when the importance of worker injuries and the social aspect is increased 10,000-fold, shale gas emerges as the most sustainable option, along with conventional gas. However, for the other indicators, large improvements would be needed in combination with very significant increases in the importance placed on the sustainability aspects and indicators.
For the environmental aspect, improvements in impacts can lead to shale gas becoming the best option (jointly with hydro) but only at a 9–329 fold reduction and in combination with significant increases in their importance (100–10,000 times). For the economic aspect and indicators, the levelised cost must be reduced by a minimum of 32% and their importance must be increased by 10–100 times. Alternatively the fuel cost must be zero and its importance increased by 10,000-fold for shale gas to be the most sustainable option, together with hydro, wind and solar PV.
Large increases in the importance (100–1000 times) are also required for public support and employment, together with improvements in their values (13.6 and 16.4 times, respectively). No improvements are necessary for diversity of fuel supply but the importance of this indicator and the social aspect must increase 10,000-fold and even then it is level with conventional gas, hydro, wind and solar PV as the most sustainable options.
To be comparable with conventional gas, LNG and nuclear power, large improvements in the performance of shale gas are needed, along with significant increases in the importance of the sustainability aspects and indicators. For example, to compete with nuclear power, an 89% to 91-fold reduction in environmental impacts is needed and their importance must be increased by 100–1000 times. However, this is only applicable to seven out of the 11 indicators. For the remaining four, no improvements are needed as shale gas has lower impacts, but a 5–1000 times increase in their importance is necessary. In some scenarios, shale gas is already more sustainable than hydro and biomass, but in others, large improvements to environmental and social impacts would be needed.
The results also suggest that a future electricity mix with a lower penetration of shale gas is more sustainable than the one with higher contribution, assuming the sustainability aspects are of equal importance. If higher importance is placed on the economic or social aspect, the high shale gas mix outranks the low due to the relatively low cost of shale gas compared to renewables.
Although the quality of the data used in this study is considered ‘medium’ to ‘high’, some data are derived from non-UK sources, which is one of the limitations of this work. If or when the exploitation of shale gas starts in the UK, using actual field data would help to refine the findings of this research. A further limitation is the limited number of economic and social indicators considered and future work should consider others, such as tax revenue, contribution to gross domestic product, community benefits, local employment, noise and traffic, to name a few.
Another limitation is a lack of stakeholder input into the decision analysis, particularly their preferences for different sustainability aspects and indicators. Despite the study showing that the overall conclusions are robust to changes in preferences, future work should consider involving relevant stakeholders and exploring the effect of their actual preferences on the outcomes of the sustainability assessment. Future work could also explore the effect on the sustainability of shale gas of different technological solutions. These include techniques for disposal of drilling waste and wastewater treatment, ‘green completion’ and carbon capture, storage and/or utilisation.
While this study focused on the UK, the methodological approach is generic enough to be applicable to sustainability evaluations of shale gas in other countries and regions – this is recommended as part of future research to help decision-makers make more informed decisions. It would also help to find out if the results for different regions (e.g. Europe) can be generalised to guide future policy development related to shale gas.
This work was funded by the UK Engineering and Physical Sciences Research Council (grant no. EP/K011820/1) and The University of Manchester’s Alumni Donor Society. This funding is gratefully acknowledged.
Appendix A. Supplementary data
Althaus et al., 2007 H.-J. Althaus, G. Doka, R. Dones, T. Heck, S. Hellweg, Roland Hischier, T. Nemecek, G. Rebitzer, M. Spielmann, G. Wernet Overview and Methodology: Data v2.0 (2007) The Ecoinvent Centre, Zurich, CH (2007) Retrieved from http://www.ecoinvent.org/database/older-versions/ecoinvent-version-2/methodology-of-ecoinvent-2/methodology-of-ecoinvent-2.html
Cooper, 2017 J. Cooper Life Cycle Sustainability Assessment of Shale Gas in the UK. PhD Thesis The University of Manchester (2017)Cooper et al., 2014 J. Cooper, L. Stamford, A. Azapagic Environmental impacts of shale gas in the UK: current situation and future scenarios Energ. Technol., 2 (2014), pp. 1012-1026
Cooper et al., 2016 J. Cooper, L. Stamford, A. Azapagic Shale gas: a review of the economic, environmental, and social sustainability Energ. Technol., 4 (2016), pp. 772-792
Edwards, 1977 W. Edwards How to use multiattribute utility measurement for social decisionmaking IEEE Trans. Syst. Man Cybern., SMC-7 (1977), pp. 326-339
Gosden, 2017 E. Gosden Cuadrilla starts drilling at Lancashire fracking site The Times, 18 August 2017 (2017) Available from: https://www.thetimes.co.uk/article/cuadrilla-starts-drilling-at-lancashire-fracking-site-qn85dgtkf
House of Lords, 2014 House of Lords The Economic Impact on UK Energy Policy of Shale Gas and Oil. May, 2014 House of Lords: Economic Affairs Committee, London, UK (2014) Retrieved from: http://www.publications.parliament.uk/pa/ld201314/ldselect/ldeconaf/172/172.pdf
Johnston, 2017 I. Johnston Election 2017: conservatives back fracking ‘revolution’ in the party manifesto The Independent 18 May 2017 (2017) Available from: http://www.independent.co.uk/news/uk/politics/election-2017-conservatives-fracking-party-manifesto-tory-gas-shale-domestic-enivronment-a7742496.html
Lewis et al., 2014 C. Lewis, J. Speirs, R. MacSweeney Getting ready for UK shale gas: Supply chain and skills requirements and opportunities United Kingdon Onshore Oil and Gas and Ernst and Young (2014) (London. Retrieved from)
Moore et al., 2014 V. Moore, A. Bereford, B. Gove, R. Underhill, S. Parnham, H. Crow, R. Cunningham, H. Huyton, J. Sutton, T. Melling, P. Billings, M. Salter Hydraulic Fracturing for Shale Gas in the UK: Examining the Evidence for Potential Environmental Impacts The Royal Society for the Protection of Birds (RSPB), London, UK (2014) Retrieved from: https://www.rspb.org.uk/Images/shale_gas_report_evidence_tcm9-365779.pdf
Mustajoki and Hamalainen, 2000 J. Mustajoki, R.P. Hamalainen Web-HIPRE: global decision support by value tree and AHP analysis Information Systems and Operational Research (INFOR), vol. 38 (2000), pp. 208-220
Ward, 2017 A. Ward Ineos wins injunction against shale protesters Financial Times (FT), 31 July 2017 (2017) Available from: https://www.ft.com/content/352de042-7607-11e7-a3e8-60495fe6ca71
Weidema et al., 2013 B.P. Weidema, C. Bauer, R. Hischier, C. Mutel, T. Nemecek, J. Reinhard, C.O. Vadenbo, G. Wernet Overview and Methodology: Data Quality Guideline for Ecoinvent Database Version 3 (Final) The ecoinvent Centre, Zurich, CH (2013) Retrieved from: https://www.ecoinvent.org/files/dataqualityguideline_ecoinvent_3_20130506.pdf
E-mail address: [email protected] (A. Azapagic).
0048-9697/© 2018 The Authors. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).