Non-energy energy policies and the use of time: Explaining shifts in UK electricity demand from 1974 to 2014

Transcript

1. Non-energy energy policies and the use of time: Explaining shifts

   in UK electricity demand from 1974 to 2014 Ben Anderson (@dataknut) School of Engineering University of Southampton Centre for Sustainability University of Otago

2. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut and 2012 coal prices for British power stations were level whilst gas prices rose 50% ( ). DECC, 2016b As seen in , an increasing share of electricity is being imported Fig. 2 through interconnectors to France and the Netherlands.1 This elec- tricity is treated as zero carbon when calculating national emissions inventories, which conveniently neglects the fact Britain now exports “ ” around 3% of its power sector emissions abroad. This accounting • weather-dependent renewables su taneous demand; • wholesale prices falling negative i winter (due to wind); • solar output forcing minimum ne run output (nuclear and CHP). These will no doubt test the capa the system, and may necessitate n deployment of greater interconnectio The aim of this paper is to disen on the British electricity system to analyse their impacts on the elec whether they can continue contrib the coming years. outlines Section 2 tions employed. explore Section 3 demand, supply, carbon emissions a contributions that individual techno made towards decarbonisation, th supplementary material provides fur In an e ort to promote transpar ff behind all the gures in this paper ar fi The processed data can also be electricinsights.co.uk. 2. Methods and data 2.1. Conventions The electricity sector has many ac of. Power ow is measured in kW / fl measured in MWh / GWh / TWh. year will yield 8.76 TWh, which is en according to the common media ana The chemical energy contained sured in MJ / GJ / TJ, or as million =41,868 TJ =11.63 TWh), or million Fig. 1. The historic and required future carbon content of British electricity, highlighting the average year-on-year change during each decade. Data from (CCC, 2015a; MacLeay et al., 2016). Fig. 2. Biannual generation by type and carbon emissions from the electricity sector. What’s the UK problem? § A de-carbonisation story… • Staffell (2018) https://doi.org/10.1016/j.enpol.2016.12.037 2

3. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut The menu § Peak demand in the UK: – What’s the problem? – What’s changing § Introducing the data – Multinational Time Use Study 1974-2000 – UK Time Use Survey 2014 § What creates ‘peak’? – And is it stable? § What does it all mean? 3

4. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut of 25% in December 2015, averaging 19% over the whole of 2015. The second half of December 2015 saw wind overtake coal to supply 20% vs. 13% of electricity. May 2016 then saw coal supply less than wind, biomass or solar for the rst time. Imports have grown steadily from fi being net-zero in 2009 10 to 7% of British demand by 2015, as the – price di erential between Britain and its neighbours grows. Biomass ff jumped from 2.5% to 5% share in July 2015, producing more energy than wind for two weeks of the year. During the rst half of 2016, fi Fig. 7. The average daily pro le of generation at half-hourly resolution in June and December, comparing 2009 and 2015. fi I. Sta ell ff Energy Policy 102 (2017) 463–475 What’s the UK problem? § A de-carbonisation peak reduction story…? • Staffell (2018) https://doi.org/10.1016/j.enpol.2016.12.037 4

5. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut Peak demand is shifting… 5 Mean GW Overall reduction Industrial decline Efficient appliances Efficient lighting Higher winter temperatures Embedded PV/wind Some subtle shifts

6. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut Peak demand is shifting… 6 Normalised Mean GW Relatively more ‘peaky’ Shifting later Shifting later PV effects?

7. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut Why is ‘peak’ a problem? • ‘Dirty’ energy Carbon problems: • Higher priced energy Cost problems: • Inefficient use of resources; • ‘Local’ overload; Infrastructure problems: 7 UK Housing Energy Fact File Graph 7a: HES average 24-hour electricity use profile for owner-occupied homes, England 2010-11 Gas consumption The amount of gas consumed in the UK varies dramatically between households. The top 10% of households consume at least four times as much gas as the bottom 10%.60 Modelling  to  predict  households’  energy   consumption – based on the property, household income and tenure – has so far been able to explain less than 40% of this variation. Gas use varies enormously from household to household, and the variation has more to do with behaviour than how dwellings are built. 0 100 200 300 400 500 600 700 800 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 Heating Water heating Electric showers Washing/drying Cooking Lighting Cold appliances ICT Audiovisual Other Unknown Watts Filling the trough Peak load

8. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut What to do? Storage •Just reducing it per se Demand Reduction •Shifting it somewhere else in time (or space and time) Demand Response 8 What makes up peak demand? How did we get here? What might be changed? Who might adapt?

9. Non-energy energy policies and the use of time, CfS, August

   2018 @dataknut The menu § Peak demand in the UK: – What’s the problem? – What’s changing § Introducing the data – Multinational Time Use Study 1974-2000 – UK Time Use Survey 2014 § What creates ‘peak’? – And is it stable? § What does it all mean? 9

10. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Source: https://www.flickr.com/photos/stevendepolo/3761877701 Source: https://upload.wikimedia.org/wikipedia/commons/9/96/R ush_Hour_on_London_Bridge.jpg "Drip Coffee Bangkok" by Takeaway - Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons - https://commons.wikimedia.org/wiki/File:Drip_Coffee_Bangkok.jp g https://upload.wikimedia.org/wikipedia/commons/f/fa/Laundry_room _%28tv%C3%A4ttstuga%29.JPG Timings matter… 10 UK Housing Energy Fact File Graph 7a: HES average 24-hour electricity use profile for owner-occupied homes, England 2010-11 Gas consumption The amount of gas consumed in the UK varies dramatically between households. The top 10% of households consume at least four times as much gas as the bottom 10%.60 Modelling  to  predict  households’  energy   consumption – based on the property, household income and tenure – has so far been able to explain less than 40% of this variation. Households with especially high or low consumption do not have particular behaviours that make them easy to identify. Instead they tend to have a cluster of very ordinary behaviours that happen to culminate in high or low Gas use varies enormously from household to household, and the variation has more to do with behaviour than how dwellings are built. 0 100 200 300 400 500 600 700 800 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 Heating Water heating Electric showers Washing/drying Cooking Lighting Cold appliances ICT Audiovisual Other Unknown Watts

11. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Digression: Practices, proxies & traces 11 (an empiricist
    s view) Image: Anthony B. Wooldridge Image: Eric Shipton “The recurrent enactment of specific practices leaves all sorts of “marks” – diet shows up in statistics on obesity; heating and cooling practices have effect on energy demand, and habits of laundry matter for water consumption. Identifying relevant “proxies” represents one way to go.” ESRC Sustainable Practices Working Group (SPRG) Discussion Paper, 2011

12. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Time Use: Traces of practices… 12 BBC 1961 ONS 2005 • 1998/1999 • 2010/2011

13. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut How often have we had them? 13 Year Sample Size & season Time interval Format 1974 All 5+ in representative household sample 2,598 February, March, August, September 30 minutes 7 diary days, primary & secondary activities (73 codes), location known, co- presence unknown 1983 Representative sample 14+ 1,350 January, February, September, November, December 15 minutes 7 diary days, primary & secondary activities (188 codes), location known, co- presence of others known 1987 Representative sample 14+ 1,586 March - June 15 minutes 7 diary days, primary & secondary activities (190 codes), location known, co- presence of others known 1995 Representative sample 16+ 1,962 May 15 minutes 1 diary day, primary activities only (31 codes), location & co-presence of others unknown 2001 All 8+ in representative household sample 8,688 All months 10 minutes 2 diary days (1 weekday & weekend), primary & secondary activities (265 codes), location known, co-presence of others known 2005 Representative sample 16+ 4,854 March, June, September, November 10 minutes 1 diary day, primary & secondary activities (30 codes), location known, co- presence of others unknown 2014 All 8+ in representative household sample 9,388 All months 10 minutes 2 diary days (1 weekday & weekend), primary & secondary activities (265 codes), location known, co-presence of others known ‘1985’ Not used (small, 1 month) Not used (small, limited coding) Most recent! 1960s… Source: Multinational Time Use Study & ONS

14. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Survey Time interval ‘Laundry’ Original source code MTUS Definition Notes 1974 30 minutes 50 Other essential domestic work 21: Laundry, ironing, clothing repair i.e. NOT 'routine housework' (defined as cleaning); NOT 'Prepare meals or snacks' ‘1985’ 15 minutes 0701 Wash clothes, hang out / bring in washing; 0702 Iron clothes; 0801 Repair, upkeep of clothes 21: Laundry, ironing, clothing repair Note bundled clothing related activities. 2000 10 minutes 3300 Unspecified making and care for textiles; 3310 Laundry; 3320 Ironing; 3390 Other specified making and care for textiles 21: Laundry, ironing, clothing repair Note bundled clothing related activities. 2014 10 minutes 3300 Unspecified making and care for textiles; 3310 Laundry; 3320 Ironing; 3390 Other specified making and care for textiles 21: Laundry, ironing, clothing repair Note bundled clothing related activities. Measuring change over time 14 1974 ‘1985’ 2000/1 2014 Harmonisation Applied MTUS coding Source: Multinational Time Use Study & ONS Synthetic MTUS 1974 - 2014 Trace: “Any X in a half- hour”

15. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut The menu § Peak demand in the UK: – What’s the problem? – What’s changing § Introducing the data – Multinational Time Use Study 1974-2000 – UK Time Use Survey 2014 § What creates ‘peak’? – And is it stable? § What does it all mean? 15

16. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut 30 years of change… 16 % of sample by age group % of gender by work status (16-64 only) Source: Synthetic MTUS 1974-2014 (weighted)

17. @dataknut DEMAND 1974 - 2014 1. Sleep 2. Personal/home care

    3. Work (paid) 4. Travel 5. Food related (cooking & eating) 6. Media use (all) 7. Education related 8. Sport/exercise 9. Social/leisure 10. Shopping & service use 11. (Not recorded) 10 Activity ‘classes’

18. @dataknut DEMAND 1974 - 2014 1974 - 2014 1. Sleep

    2. Personal/home care 3. Work (paid) 4. Travel 5. Food related (cooking & eating) 6. Media use (all) 7. Education related 8. Sport/exercise 9. Social/leisure 10. Shopping & service use 11. (Not recorded) 10 Activity ‘classes’

19. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Work 1974 – 2014: 16-64 19 Reduction in shift working?

20. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Travel 1974 – 2014: 16 - 64 20 Multiple purposes Commuting Shopping Leisure Walking the dog School pick-up

21. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Personal/home care 1974 – 2014: 16 - 64 21 Remember: far fewer men are now in work!

22. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Food related 1974 – 2014: 16 - 64 22 Remember: far fewer men are now in work!

23. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Food related 1974 – 2014: All 23 Remember: ageing population!

24. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Media related 1974 – 2014: 16 - 64 24 Day time TV did not exist!

25. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut 40 years of weekday change… 25 1974 65+ small n => noisy data?

26. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut 40 years of Saturday change… 26 1974 65+ small n => noisy data?

27. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut 40 years of Sunday change… 27 1974 65+ small n => noisy data?

28. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Enough of the descriptives… 28 § Poisson models of the number of reported acts – 16:00 – 18:00 = ‘Early’ – 18:00 – 20:00 = ‘Late’ Early food: Model 1 Early food: Model 2 All All 1985 (contrast = 1974) 0.263*** (0.221, 0.305) 0.001 (-0.022, 0.024) 2000 0.196*** (0.147, 0.244) -0.062*** (-0.089, -0.034) 2014 0.082** (0.029, 0.134) -0.203*** (-0.233, -0.173) Female 0.717*** (0.679, 0.756) 0.366*** (0.334, 0.398) Female * 1985 -0.325*** (-0.375, -0.275) Female * 2000 -0.342*** (-0.400, -0.283) Female * 2014 -0.352*** (-0.416, -0.289) Job: Full-time (not in paid work) -0.295*** (-0.331, -0.259) Job: Missing -0.339*** (-0.521, -0.157) Job: Part time -0.172*** (-0.234, -0.109) Job: Unknown hours -0.273*** (-0.400, -0.147) Female * Full time -0.047* (-0.094, -0.0002) Female * Missing 0.001 (-0.235, 0.237) Female * Part time 0.137*** (0.069, 0.204) Female * Unknown hours 0.232** (0.079, 0.386) Constant -0.041* (-0.074, -0.007) 0.346*** (0.313, 0.379) Observations 33041 33021 -0.4 -0.2 0 0.2 0.4 0.6 0.8 2014 Full time Female * Full time Coefficient Factor Poisson models (selected results) Late food Early food These values are additive Basically the descriptive results are mostly robust -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 2014 Full time Female * Full time Coefficient Factor Poisson models (selected results) Late personal/home care Early personal/home care

29. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Does this explain the GW shift? 29 Mean GW (normalised) Remember this? 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Changing patterns of time use_R1_Manuscript_clean.docx Last saved on 7/16/2018 4:28:00 AM Total word count (Complete manuscript): 9764 Figure 6: Comparison of % point change in half hours reported and mean change in normalised GW demand (Source: MTUS 2000 – 2014, National Grid weekdays in July 2006 – 2016, loess curve with SE) • So (on this evidence) the relative shift in GW has: • a bit to do with personal/home care • quite a lot to do with food • but nothing to do with media use 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Changing patterns of time use_R1_Manuscript_clean.docx Last saved on 7/16/2018 Total word count (Complete manuscript): 9764 Figure 6: Comparison of % point change in half hours reported and mean change in normalised GW de (Source: MTUS 2000 – 2014, National Grid weekdays in July 2006 – 2016, loess curve with SE) laundry, cooking and media use accounted for 8 + 20 + 14 ~= 42% of evening peak demand in the UK Household Electricity Use Study (Jason Palmer et al., 2013a)

30. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Making (some) sense of it all… 30 Changing timing driven by Labour market transitions Gendered (domestic) practices Extended work hours (?) Extended commuting? These are not energy policies!

31. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut The menu § Peak demand in the UK: – What’s the problem? – What’s changing § Introducing the data – Multinational Time Use Study 1974-2000 – UK Time Use Survey 2014 § What creates ‘peak’? – And is it stable? § What does it all mean? 31

32. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Key message I: Change is endemic 32 Demanding Practices Change Variation Reconfiguration? ? Normality & Need Infrastructures Non-energy energy policy Labour market policies Working hours School hours (Sub)Urban planning Transport options

33. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Key message II: Sequences matter 33 We don’t have good levers for shifting this around ($s) don’t seem to cut it… & people are locked together

34. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut § Diffusing peak : – Work – Extended (randomized) commute § Intensifying peak: – Cramming in domestic tasks – Older people’s eating habits – Charging EVs § Matching PV – Solar citizens… Key message III: Power paradoxes 34 Might help Won’t help Win win?

35. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut What’s next? § Repeat J 35 • 1998/1999 • 2010/2011 • www.energy.soton.ac.uk/tag/spatialec • 2 year EU Global Fellowship @Otago CfS Watch this space

36. Non-energy energy policies and the use of time, CfS, August

    2018 @dataknut Thank you • SPRG • DEMAND With thanks to: 36 b.anderson@soton.ac.uk @dataknut Further reading: • https://www.southampton.ac.uk/engineering/about/staff/ba1e12.page?#publications • www.demand.ac.uk • Spurling (2018) Matters of time: Materiality and the changing temporal organisation of everyday energy consumption (https://doi.org/10.1177/1469540518773818) pixabay.com