We have a new website go to gov.scot

Scottish Energy Study Volume 5: Energy and Carbon Dioxide Projections for Scotland

Listen

6. Sensitivity Analysis

The previous sections have developed two projections for Scotland's energy use and CO 2 emissions to 2020, which were based on the corresponding projections for the UK. These projections were based on the Central Fuel/Central EWP ( CC) and High Fuel/Central EWP ( HC) scenarios developed by BERR (see Introduction), which were then adjusted to take account of Scotland's current pattern of energy supply and demand, and the Scottish Government's targets for increased economic growth and population growth. Of course scenario projections are only indicative of a broad range of possible future trajectories, and the adoption of the CC and HC scenarios should not be taken to infer that these represent the most likely outcome for Scotland. Uncertainties over key drivers affecting both energy demand and the balance of supply side options mean that these projections should be regarded as a starting point or illustrative baseline from which to investigate other potential outcomes. To further scope the possible trends in Scottish energy and CO 2 pathways to 2020 this section assesses how these projections may be altered by a set of factors specific to Scotland.

6.1. Choice of Sensitivity Parameters

The UK's energy and CO 2 projections use scenarios designed to examine three key drivers affecting the energy sector, namely economic growth, fossil fuel prices and policy measures. Clearly these parameters are also important to Scotland's energy sector, and accordingly this sensitivity analysis focuses on how divergence from the baseline UK scenarios may alter Scotland's energy and CO 2 emission patterns. Particular emphasis has been placed on examining the impact of parameters that are more specific to Scotland.

The parameters affecting energy demand that have been examined are:

Gross Domestic Product - The level of economic activity, as measured by the rate of growth in Scotland's GDP, affects demand for energy services in the industry and services sectors, and also in freight transport.

Population - Population affects the demand for energy services in the domestic sector as it leads to an increase/decrease in the number of households. It also affects energy demand for public and private transport through greater demand for mobility and increased car ownership.

Car ownership - Scotland's car ownership is presently only about 90% of the UK average. An increase in ownership will affect the demand for road transport fuels.

Air travel - The more people travel by air the greater consumption of aviation fuel.

Clearly there is some interaction between these factors. For example GDP could be increased through an increase in population, and hence work-force. Similarly a higher population could increase demand for air travel. The aim herein is to assess the direct impact of each of these variables, and to avoid double counting.

Factors affecting energy supply and its related emissions of CO 2 that have been considered are:

Fuel mix in electricity generation - The balance of fossil, nuclear and renewable energy sources used to generate electricity can have a considerable impact on fossil energy consumption and CO 2 emissions.

Increased renewable heat utilisation - In addition to power generation renewable energy sources including biomass, solar thermal and ground source heat pumps may be used to supply heat to the domestic, industry and services sectors.

The price of energy clearly will have an impact on demand from both business and the private consumer, and may also impact indirectly through its influence on economic growth. This factor is considered separately.

The remainder of this section examines how variations to these individual parameters may affect Scotland's energy sector and its associated CO 2 emissions.

6.2. Demand Sensitivity Analysis

The sensitivity analysis of the key parameters affecting Scottish energy demand has assessed the variations listed in Table 17. For comparison purposes the table also gives the historic trend and the assumptions used either explicitly or implicitly in the baseline scenario projections.

Table 17 Factors affecting energy demand

Parameter

Historic trend

Central scenario assumption

Sensitivity analysis assumption

GDP

Average annual growth rate (1975 - 2005): 1.8%

Average annual growth rate of 2.5% between 2007-2020

Continuation of historic annual growth rate (1.8% per year)

Population

Average annual change (1995-2005): -0.2%

Population increases between 2007 and 2020 at an average annual rate of 0.4%

Population increases in line with the projections of the General Register Office for Scotland (0.27% per year) 64

Car ownership

Scotland has 42 cars per 100 people compared to 46 per 100 UK wide.

Scottish car ownership increases at the UK average rate (and therefore remains below UK average in 2020)

Scottish car ownership increases faster than the UK average to match the UK average by 2020

Air Travel

57% increase in passengers between 1995 and 2002 to reach 20 million per annum

40 million passengers per annum by 2020

34 and 46 million passengers per annum by 2020

Lower growth rate for Gross Domestic Product ( GDP)

Energy demands in the services and industry sectors, and also in freight transport, generally increase with the level of economic activity. It could be argued that energy demand in the domestic sector would also increase with GDP as increasing wealth drives demand for home comfort. However, the view taken herein is that home heating is an essential requirement, the demand for which has very limited market elasticity, and therefore is insensitive to relatively small variations in GDP. Personal transport may also increase with GDP through greater car ownership. However, this parameter has been assessed separately and is not considered as part of the analysis of GDP.

The sensitivity analysis has investigated the potential impact of Scotland's economy growing at a slower rate than assumed in the initial scenario assumptions, which were based on the Scottish Government's targets for increasing economic development. The alternative lower growth rate investigated was an average of 1.8% per year compared to the growth rate used in the scenario, which averaged 2.5% per year over the period 2007 to 2020 (Table 1). This lower rate broadly corresponds to the long term average rate of growth of Scotland's economy over the period 1975-2005. Adjustments to the baseline energy projections were made by assuming a linear relationship between GDP and energy demand for those sectors affected.

Table 18 lists the impact of the lower GDP assumption on energy demand by sector relative to the CC baseline scenario. As discussed above the greatest impact is on the industry and commercial sectors with reductions of 7.4% and 6.9% respectively by 2020. The impact is less with transport because GDP is assumed to only affect commercial operations. Overall Scottish final energy consumption is 3.6% less by 2020 with this slower GDP growth rate than it would have been with the baseline CC and HC scenario assumption for GDP growth.

Table 18 also shows that a lower GDP growth assumption reduces CO 2 emissions but the impact is less than for energy (2.1%) due to the mix of fuels affected.

Table 18 Impact of the lower GDP growth assumption on Scottish final energy demand (results indexed to CC scenario projections = 100.0)

2002

2005

2010

2015

2020

Domestic

100.0

100.0

100.0

100.0

100.0

Services & Agriculture

100.0

100.0

99.4

97.0

93.1

Industry

100.0

100.0

99.4

96.8

92.6

Transport

100.0

100.0

99.9

99.3

98.3

Total energy

100.0

100.0

99.7

98.6

96.4

CO 2 emissions

100.0

100.0

99.8

99.2

97.9

Lower population growth rate

Population has a direct effect on domestic energy demand because it determines the number of households in a country. It also has a direct affect on the demand for mobility both through public and private transport. Population also has an indirect affect of industry and services, and commercial transport by affecting the number of economically active people in the economy ( i.e. assuming constant GDP per capita).

The sensitivity analysis has investigated the potential impact of Scotland's population growing at a slower rate than assumed in the scenario assumptions, which were based on the Scottish Government's targets for increasing population. The alternative population growth rate investigated was based on the projections of the General Register for Scotland, which gave a smaller average annual increase of 0.27% per annum from 2007 to 2020 (Table 17).

The impact of this lower population assumption on energy demand and CO 2 emissions is shown in Table 19. Note the impact of population on transport energy demand is based on a contraction of the existing demand pattern, it does not include a change in car ownership per capita, which is examined later in a separate sensitivity assessments. For the purpose of this assessment population change is assumed not to have a direct affect on passenger aviation, which is also investigated separately below.

Table 19 shows that the impact of the lower population assumption is less than the GDP assumption examined previously with total final energy demand decreasing by just over 1%. This is mirrored by the trend in CO 2 emissions.

Table 19 Impact of different population trends on Scottish final energy demand (results indexed to CC scenario projections = 100.0)

2002

2005

2010

2015

2020

Domestic

100.0

100.0

99.9

99.2

98.3

Services & Agriculture

100.0

100.0

99.9

99.3

98.6

Industry

100.0

100.0

99.9

99.3

98.5

Transport

100.0

100.0

99.9

99.6

98.8

Total energy

100.0

100.0

99.9

99.4

98.6

CO 2 emissions

100.0

100.0

99.9

99.6

99.1

Increased car ownership

The projections presented in Section 2 showed transport to be a fast growing energy consumption sector with demand increasing by 11% to 15% between 2005 and 2020. Road transport and aviation accounted for most of this growth, and therefore it is desirable to investigate the sensitivity of these projections to variations in key factors. In addition to the GDP and population variations examined above, another important factor is the level of car ownership.

In 2005 Scotland's car ownership was lower than the UK average at 42 cars per hundred people compared to 46 per hundred for the UK overall 65,66. In the baseline projections Scottish car ownership was implicitly assumed to increase at the same rate as the UK overall, which meant that the difference in the absolute level of ownership between Scotland and the UK remained. This sensitivity assessment examines the impact of Scottish car ownership growing to match the UK average by 2020.

Table 20 shows the impact of this higher growth in car ownership. By 2020 this would increase road transport energy demand by over 6%, equivalent to about 4% of total transport energy demand but only 1.5% of total Scottish final energy demand. Total CO 2 emissions are also increased by 1.5% by 2020.

Table 20 Impact of increased car ownership on Scottish final energy demand (results indexed to CC scenario projections = 100.0)

2002

2005

2010

2015

2020

Transport - Road

100.0

100.0

101.3

104.0

106.2

Transport - Total

100.0

100.0

100.9

102.8

104.2

Total Energy

100.0

100.0

100.3

101.0

101.5

CO 2 emissions

100.0

100.0

100.3

101.0

101.5

Higher and lower growth in air travel

The baseline projections presented in Section 3 indicated an 18%-28% increase in aviation fuel consumption between 2005 and 2020. This growth in demand was expected to be driven by continued growth in air passenger travel, which has increased by 57% (measured in terms of passenger movements) between 1995 and 2002 67. Current estimates expect passenger movements to reach 40 million per year by 2020 from a 2002 level of just under 20 million. This sensitivity assessment examines the impact of higher and lower passenger growth to 46 million and 34 million by 2020.

Results presented in Table 21 show that these passenger trends would increase or decrease the baseline transport energy demand by about 3% by 2020, but the impact on overall Scottish final energy demand is only about +/-1%. Overall CO 2 emissions are also affected by about +/-1%.

Table 21 Impact of different air passenger trends on Scottish energy demand (results indexed to the CC scenario projections = 100.0)

2002

2005

2010

2015

2020

High Case - increase to 46 million per year by 2020

Transport - Air

100.0

106.0

110.0

112.9

115.0

Transport - Total

100.0

100.0

101.9

102.7

103.4

Total energy

100.0

100.0

100.6

100.9

101.2

CO 2 emissions

100.0

100.0

100.6

100.9

101.2

Low Case - increased to 34 million per year by 2020

Transport - Air

100.0

100.0

93.3

87.1

85.0

Transport - Total

100.0

100.0

98.7

97.3

96.6

Total energy

100.0

100.0

99.6

99.1

98.8

CO 2 emissions

100.0

100.0

99.6

99.1

98.8

6.3. Overall Impact of Demand Sensitivity Analysis

The above analysis has considered a range of factors that could either increase or decrease Scottish final energy demand relative to the baseline CC and HC projections presented in Section 2. Those factors driving an increase in energy demand ( i.e. higher car ownership and air travel) could in total increase overall demand by about 3.5% by 2020. Those factors causing a decrease in energy demand ( i.e. lower growth in GDP, population and air travel) could in total reduce demand by about 6% by 2020. The factor having the greatest effect was the lower GDP growth assumption that reduced energy consumption by around 4% compared to the baseline CC scenario projection in 2020. The overall trend in final energy demand for these high and low cases compared to the baseline CC projection is shown in Figure 11.

The impact of these factors on CO 2 emissions parallels their effect on energy demand. The factors driving a reduction in energy demand result in a reduction in CO 2 emissions of about 4% by 2020, and the factors driving an increase lead to CO 2 emissions about 3% above the CC baseline projections. The overall trends in CO 2 emissions compared to the CC baseline projection are shown in Figure 12. [ NB The changes in CO 2 emissions are slightly different to the energy demand variations because the variables assessed do not affect emissions from power generation or oil refineries.]

Figure 11 Overall impact of the sensitivity analysis variables on Scottish final energy demand compared to the baseline CC scenario projection.

Figure 11 Overall impact of the sensitivity analysis variables on Scottish final energy demand compared to the baseline CC scenario projection.

Figure 12 Overall impact of the demand side sensitivity analysis variables on Scottish energy related CO 2 emissions compared to the baseline CC scenario projection

Figure 12 Overall impact of the demand side sensitivity analysis variables on Scottish energy related CO2 emissions compared to the baseline CC scenario projection

6.4. Supply Side Sensitivity Analysis

Supply side variations do not affect Scottish final energy demand, but do affect the mix of primary energy sources used to meet this demand, the level of electricity exports and also the level of CO 2 emissions. This section concentrates on the effect of possible supply side variations on Scottish CO 2 emissions. No analysis has been made of the cost effectiveness, or additional cost of these supply changes, the analysis is concerned only with their impact on CO 2.

Fuel mix in electricity generation

Electricity generation accounts for a significant proportion of Scotland's primary energy consumption and CO 2 emissions, amounting to 34% of CO 2 emissions in 2002, and, in the baseline scenario projections, was still 21% in the CC scenario and 28% in the HC scenario by 2020. There is much uncertainty concerning the mix of power generation capacity in Scotland out to 2020 that will be affected by decisions including:

  • Scottish Power is considering replacement options for thermal based generation at both Longannet and Cockenzie.
  • There is a possibility that the two nuclear plants in Scotland may get life time extensions that permit their operation up to or beyond 2020.
  • The baseline projection indicated some new gas fired generation being built by 2015, but this deployment could be sensitive to the continued operation of nuclear plant, the expansion of renewable energy generation, future plans for coal fired generation and limits on transmission capacity to England and Northern Ireland.
  • Scotland has ambitious plans to expand the supply of energy from renewable sources that are likely to present a number of challenges.

To illustrate the impact of these uncertainties, three alternative generation mixes representing possible variations in renewable, coal and gas generation in 2020 are listed in Table 22. It should be stressed that these sensitivity factors were selected purely to investigate the potential variation in CO 2 emissions from power generation. No consideration has been given to their technical viability in terms of load following, transmission constraints and security of supply.

These sensitivity variations considered three situations in which Scotland's electricity supply sector (a) generates more power from fossil fuels (b) generates less power from fossil fuels and (c) falls short of its renewable electricity targets. These variations have been applied to the baseline CC scenario projection for total generation. To illustrate these assumptions:

  • (a) Increasing coal fired generation from 9% to 22% in the CC scenario is likely to involve building a new ~1.6 GW coal fired power station by about 2015 and operating it at a load factor of around 65% in 2020.
  • (b) Generating less from fossil fuels involves building no replacement gas or coal fired power plant in Scotland to 2020 and reducing the load factor of the one remaining coal fired power plant, Longannet, from 20% to 10% by 2020.
  • (c) Scotland falls short of its renewable energy target generating only 15 TWh from new renewable sources instead of the 17 TWh projected using the CC scenario. This gives a total 90% of what is needed to supply 50% of Scotland's gross consumption from all renewable sources (new renewables plus large hydro see Table 13).

Table 22 Factors affecting electricity generation

Factor

Historic position (2002)

CC Scenario assumption (2020)

Sensitivity Analysis (2020)

More Fossil

Share of generation:

Coal - 30%

Gas - 18%

RE - 12% 68

Share of generation in 2020:

Coal - 9%

Gas - 18%

RE - 43% 69,70

Additional 9 TWh coal generation increasing share to 22% by 2020.

Less Fossil

No new gas generation and coal generation down by 50% reducing share to 4% by 2020

Less Renewables

New renewable sources provide 15 TWh of electricity by 2020 71.

The results are presented in Figure 13, which also contains the result from the baseline CC scenario projection to aid comparison. This shows the potential for electricity generation to cause significant variations in Scotland's total CO 2 emissions with at one extreme total CO 2 emissions 17% above the CC scenario baseline and at the other extreme emissions 11% below by 2020.

CO 2 emissions were not sensitive to a reduction in generation from renewable energy sources because small changes in the level of renewable generation are not assumed to affect generation from other sources. This is because Scotland is assumed to generate a surplus of electricity over the full period to 2020. The lower level of renewables generation considered as part of the sensitivity analysis only reduced the level of Scotland's exports to England and Northern Ireland. As the reduction is not assumed to be made up by additional generation from sources using fossil fuels there is no increase in Scottish CO 2 emissions under this sensitivity test.

Figure 13 Influence of variations in the fuel mix used for electricity generation on total Scottish CO 2 emissions in the CC scenario

Figure 13 Influence of variations in the fuel mix used for electricity generation on total Scottish CO2 emissions in the CC scenario

Carbon dioxide capture and storage ( CCS) gives the option of continuing to use fossil fuels while reducing the associated CO 2 emissions by up to 85%-90%. Presently the UK government is running a competition to support a full-scale demonstration of this technology, which is scheduled to be operational by 2014-2015. Should this project be located in Scotland, replacing some of the existing coal fired capacity, it would reduce emissions by between 0.9 and 3.7MtCO 272 per year, equivalent to a 2-10% reduction in Scotland's overall CO 2 emissions in 2020. One purpose of the CCS demonstration is to facilitate further deployment when the EUETS allowance price makes this commercially viable. Current estimates of the EUETS price suggest that commercial deployment of CCS may not occur until after 2020 73.

Renewable heat utilisation

In addition to power generation renewable energy sources including biomass, solar thermal and ground source heat pumps may be used to supply heat to the domestic, industry and services sectors. The Scottish Energy Study Volume 1 indicated that renewable sources only made a modest contribution to heat supply amounting to ~1% of services and 1.3% of domestic demand, and, in line with BERR's projections for the UK, these shares did not change significantly in the baseline projections. This sensitivity assessment has examined the impact of an increased renewable heat contribution on Scottish CO 2 emissions. The assumptions used in the assessment were:

  • Renewable energy sources can replace heat demands in domestic, services and industry.
  • 72% of fossil energy supplies and 36% of electricity are used for space heating in Scottish domestic dwellings 74.
  • 75% of fossil energy supplies and 38% of electricity are used for space heating in Scottish commercial buildings 64.
  • 44% of fossil energy use in industry is for space heating and low grade process heat 75.
  • Industry increases renewable heat utilisation incrementally to 6% in 2010 and 10% in 2020.
  • Services increases renewable heat utilisation from 2.5% in 2002, to 5% in 2010 and 10% in 2020.
  • Domestic increases renewable heat utilisation from 1.3% in 2002, to 2.5% in 2010 and 10% in 2020.

The impact of these assumptions on sectoral CO 2 emissions is shown in Table 23. Substitution of fossil and electrical heating with 10% renewable sources reduced CO 2 emissions by 5.4%, 4.5% and 1.6% by 2020 for domestic, services and industry respectively. Total Scottish CO 2 emissions were reduced by 1.6%.

Table 23 Reduction in Scottish CO 2 emissions resulting from the deployment of renewable heat reaching 10% in 2020 (indexed to Baseline Project = 100.0)

2002

2005

2010

2015

2020

Domestic

100.0

100.0

99.1

96.8

94.6

Services & Agriculture

100.0

100.0

97.9

96.7

95.5

Industry

100.0

100.0

99.7

99.2

98.4

Total Emissions

100.0

100.0

99.6

99.0

98.4

Implications of different assumptions for fossil fuel prices

It was noted in Section 2 that the fossil fuel price assumptions used by BERR in its projections are significantly lower than recent peaks in market prices, particularly for oil and natural gas. Clearly these assumptions carry through to the projections presented here for Scotland.

History shows that in some periods fossil energy prices can be volatile, therefore current prices are not necessarily a sound indication for the future. Nonetheless it is reasonable to consider how these projections may be affected by higher price assumptions. Table 24 compares the price and demand differences for oil, gas, coal and total fossil primary energy between the CC and HC scenarios in 2020. The table shows that the higher prices for oil and gas in the HC scenario results in reductions in final demand for these fuels, but in contrast demand for coal is slightly higher in the HC scenario despite its higher price. Moreover, the total demand for fossil energy in primary consumption is actually 3% higher in the HC scenario (Table 14). This seemingly perverse outcome is due to the absolute price differentials causing significant levels of fuel switching, particularly in power generation, while having a much more limited impact on overall demand for energy services. Therefore the impact of higher price assumptions may be more limited than might otherwise be expected.

BERR is currently developing a new set of energy and CO 2 projections that will investigate higher fossil fuel prices. The Scottish Government will consider up-dating these projections for Scotland when the new results from BERR are available.

Table 24 Impact of fuel price assumptions on final demand for fossil fuels.

Fuel

Price difference of HC vs. CC scenario (%)

Demand change in HC vs. CC scenario (%)

Oil

51%

-3%

Gas

38%

-5%

Coal

41%

1%

6.5. Overall Impact of Supply Side Sensitivity

Figure 14 shows the overall impact of the supply side sensitivity analyses on Scottish CO 2 emissions. Together these could cause an increase in CO 2 of about 17% above the central CC scenario projection or a reduction of about 13% by 2020. This illustrates the powerful influence of the electricity generation sector, and in particular decisions on future fossil generation, on Scotland's total CO 2 emissions. Heat generation contributes a small proportion of the total Scottish CO 2 emissions projected for 2020, in comparison to the 21% of emissions from electricity generation (see Table 15). The impact on CO 2 emissions from the introduction of 10% of heat supplies from renewable sources is small in comparison to the changes made to fossil electricity generation.

Figure 14 Overall impact of the supply side sensitivity analysis variables on Scottish energy related CO 2 emissions compared to the baseline CC scenario projection

Figure 14 Overall impact of the supply side sensitivity analysis variables on Scottish energy related CO2 emissions compared to the baseline CC scenario projection

6.6. Overall Impact of both Supply and Demand Side Sensitivity

The combined impact of all the sensitivity factors investigated, covering both supply and demand is shown in Figure 15. Together these could cause an increase in CO 2 emissions of about 19% above the central CC scenario projection or a reduction of about 16% both by 2020. These variations are more significant than the change in emissions cause by the difference in price assumptions between the CC and HC scenarios, which was around 6%. Changes in the supply side have a greater impact upon emissions than changes on the demand side.

Figure 15 Overall impact of sensitivity analysis variable on energy related CO 2 emissions in the CC scenario

Figure 15 Overall impact of sensitivity analysis variable on energy related CO2 emissions in the CC scenario

6.7. Projection Uncertainty

As indicated previously, projections into the future are inherently uncertain. This uncertainty arises from a number of sources including:

  • Uncertainty in the baseline (2002) data on which the projections are based (linked to statistical margins and estimating uncertainties).
  • Modelling approximations.
  • Impact of policy measures ( i.e. measures in the UK Climate Change Programme, and EWP).
  • Variations in future parameters ( e.g.GDP, population, etc.)

The uncertainty over future parameters has been covered in this work through the analysis of two scenarios, and a set of sensitivity analyses. The sensitivity analysis has shown that this area of uncertainty can cause variations of about 20% around the central projection by 2020.

The other factors above are analytical uncertainties that are likely to have less impact than the uncertainty over future parameters. For example BERR has estimated the uncertainty around its central projection, in terms of CO 2 emissions, to be about plus or minus 10%.

These observations suggest that, for a given scenario ( i.e. setting aside the uncertainties in the scenario itself), the uncertainty applying to a particular projection may be of the order of 10-15%. However, since these uncertainties are the same for all projections, they will largely cancel out when considering differences between scenarios. Therefore the relatively small trends between scenarios reported herein are significant, and should not be dismissed as "within the projection error".