The recovery of heat from power generation in Scotland: study

5 Core Results

This section sets out the core financial and emissions modelling outputs for each of the four power station sites. The core results assume current policy initiatives are in place.

The model results for each power station fit into five different scenarios:

• Scenario 1: An assumed 100% of anchor heat loads are picked up from year 1. In reality this will never be possible, but this result shows the financial returns under the most positive scenario possible, providing a useful yardstick.

• Scenarios 2 to 5: The other results use a 15 year built up of heating load, with heat sales increasing by 7% each year. This reflects the fact that most consumers will not consider connecting until their own boilers fall due for replacement. The timescales for this will depend on each site. In the past the period for heat load to build up has been very long, 20 years or more. With new pressures on customers (high energy costs, CRC etc.) we are assuming a typical build up period of 15 years. Scenarios 2 to 5 progressively test the impact of different levels of connection of heat load falling from 100% to 75% then 50% and finally 25%.

The results for each scenario are coded as P1, L4 etc, i.e. Peterhead scenario 1, Longannet scenario 4 etc. The following table shows the scenario assumptions and codes.

Table 6 Scenario Assumptions

	Scenario Code	Final heat load % of potential	Timescale of heat load build up
		%	years
Peterhead	P1	100%	1
	P2	100%	15
	P3	75%	15
	P4	50%	15
	P5	25%	15
Hunterston	H1	100%	1
	H2	100%	15
	H3	75%	15
	H4	50%	15
	H5	25%	15
Cockenzie	C1	100%	1
	C2	100%	15
	C3	75%	15
	C4	50%	15
	C5	25%	15
Longannet	L1	100%	1
	L2	100%	15
	L3	75%	15
	L4	50%	15
	L5	25%	15

5.1 Peterhead

With no additional support the screening process showed that only loads within Peterhead itself would be financially viable to connect to the power station. Around 90%, of the heat load would be non-domestic and it would not be viable to feed most new housing developments. The network would be as follows:

Figure 10 Potential pipe route for Peterhead

This network has two main legs:

• An eastern leg which runs through Peterhead town supplying existing loads.
• A western leg which largely supplies new development sites.

In practice, it may be desirable to complete the loop between the mixed development (M1) to the northwest and business park 1 to the Northeast. This could provide a level of redundancy in the network for sites to the north of the scheme when maintenance was to be carried out, but this option is not modelled below as it would be less cost effective.

Table 7 Peterhead Heat Sales

Scenario Code	Final heat load % of potential	Timescale of build up	Final Peak Load	Lifetime Heat Sales
	%	yrs	MW	MWh
P1	100%	1	51.606	3,341,323
P2	100%	15	51.606	2,756,591
P3	75%	15	38.704	2,067,443
P4	50%	15	25.803	1,378,296
P5	25%	15	12.901	689,148

Table 8 Peterhead Capital Costs

Scenario Code	Power Plant Conversion to CHP Capex	DH Capex Incl. distribution, HCUs and heat meters in dwellings less conventional domestic boilers	Total Capex
	£	£	£
P1	£10,788,177	£10,765,475	£21,553,652
P2	£10,788,177	£10,765,475	£21,553,652
P3	£10,788,177	£9,150,654	£19,938,831
P4	£10,788,177	£6,459,285	£17,247,462
P5	£10,788,177	£3,767,916	£14,556,094

Table 9 Peterhead Scenario Results

Scenario Code	NPV	IRR	Lifetime Carbon Saving
	£	%	Tonnes CO 2
P1	-£2,168,818	7.73%	347,287
P2	-£11,800,368	3.31%	286,512
P3	-£12,623,863	2.08%	214,884
P4	-£12,370,814	0.68%	143,256
P5	-£12,117,768	None	71,628

Probably the most likely scenario above is that 75% of identified heat loads would be connected over a 15 year period (P3). Assuming this is the case and the power station also develops the DH scheme then the NPV at DCR of 9% without subsidy is -£12.6M and the IRR to around 2% so a £12.6M net present value subsidy would be required to achieve an IRR of 9%. The capital cost would be around £19.9M (£10.8M to fit heat extraction capability to the power station and £9.1M for the DH network) and the required whole life subsidy as a percentage of capital cost would be around 63%.

Once the network has reached the peak of its development (75% of all identified potential heat load) it would be delivering 63,000 MWh/ Yr of heat to customers and saving 6,500TCO ₂/Yr.

The lifetime CO 2 savings would be around 215,000TCO ₂ so the lifetime subsidy required to make the scheme cost effective would be around £59/TCO ₂.

5.2 Hunterston

Two main pipe routes were initially considered:

• An eastern route, taking in loads in Dalry, Glengarnock and Kilbirnie.
• A southern route, taking in loads in Ardrossan, Saltcoats, Stevenson, Kilwinning, Irvine and the industrial sites south of Irvine.

The eastern route has several long sections without any connected heat loads. The initial logic was driven by the potential industrial anchor heat load in Dalry (Roche). This site already has a CHP system - so there is a high degree of uncertainty over the potential for connecting this load. Without this load the onwards legs to Glengarnock and Kilbirnie would not be financially viable.

As a result the analysis focused on the southern route:

Figure 11 Potential southern pipe route for Hunterston

Table 10 Hunterston Heat Sales

Scenario Code	Final heat load % of potential	Timescale of build up	Final Peak Load	Lifetime Heat Sales
	%	yrs	MW	MWh
H1	100%	1	107.453	18,617,060
H2	100%	15	107.453	15,359,074
H3	75%	15	80.590	11,519,306
H4	50%	15	53.726	7,679,537
H5	25%	15	26.863	3,839,769

Table 11 Hunterston Capital Costs

Scenario Code	Power Plant Conversion to CHP Capex	DH Capex Incl. distribution, HCUs and heat meters in dwellings less conventional domestic boilers	Total Capex
	£	£	£
H1	£27,000,000	£44,620,175	£71,620,175
H2	£27,000,000	£44,620,175	£71,620,175
H3	£27,000,000	£37,927,149	£64,927,149
H4	£27,000,000	£26,772,105	£53,772,105
H5	£27,000,000	£15,617,061	£42,617,061

Table 12 Hunterston Scenario Results

Scenario Code	NPV	IRR	Lifetime Carbon Saving
	£	%	Tonnes CO 2
H1	£20,249,722	12.72%	1,092,564
H2	-£27,254,200	5.01%	901,365
H3	-£31,652,504	3.63%	676,024
H4	-£31,588,900	2.17%	450,683
H5	-£31,525,404	-0.58%	225,341

In scenario H3 assuming this is the case and the power station also develops the DH scheme then the NPV at DCR of 9% without subsidy is -£32M and the IRR to around 3.6% so a £32 M net present value subsidy would be required to achieve an IRR of 9%. The capital cost would be around £65M (£27M to fit heat extraction capability to the power station and £38M for the DH network) and the required whole life subsidy as a percentage of capital cost would be around 49%.

Once the network has reached the peak of its development (75% of all identified potential heat load) it would be delivering around 350,000 MWh/year of heat to customers and saving around 20,500 TCO ₂/Yr.

The lifetime CO 2 savings would be around 680,000 TCO ₂ so the lifetime subsidy required to make the scheme cost effective would be around £47/ TCO ₂.

5.3 Cockenzie

This network is significantly larger than the preceding two examples, with 240 anchor heat loads.

Several main pipe routes were considered as shown below:

Figure 12 Potential pipe route for Cockenzie

This network is not only larger, but is likely to be more difficult than the others due to:

• The densely packed urban areas served with extensive networks of pipes and cables already in place.
• The potential competition from the proposed biomass plant at Leith.

It should be noted that the south western most leg extending to Penicuik rises a few hundred metres. This will necessitate high static pressures on most of the DH network but will not increase the pumping energy requirement which is purely due to friction as the static head will be recovered in the return leg.

Table 13 Cockenzie Heat Sales

Scenario Code	Final heat load % of potential	Timescale of build up	Final Peak Load	Lifetime Heat Sales
	%	yrs	MW	MWh
C1	100%	1	222.071	24,049,695
C2	100%	15	222.071	19,840,999
C3	75%	15	166.553	14,880,749
C4	50%	15	111.035	9,920,499
C5	25%	15	55.518	4,960,250

Table 14 Cockenzie Capital Costs

Scenario Code	Power Plant Conversion to CHP Capex	DH Capex Incl distribution, HCUs and heat meters in dwellings less conventional domestic boilers	Total Capex
	£	£	£
C1	£10,788,177	£158,166,950	£168,955,127
C2	£10,788,177	£158,166,950	£168,955,127
C3	£10,788,177	£134,441,908	£145,230,085
C4	£10,788,177	£94,900,170	£105,688,347
C5	£10,788,177	£55,358,433	£66,146,610

Table 15 Cockenzie Scenario Results

Scenario Code	NPV	IRR	Lifetime Carbon Saving
	£	%	Tonnes CO 2
C1	-£11,505,178	8.10%	2,499,655
C2	-£92,596,566	3.14%	2,062,216
C3	-£87,960,720	2.31%	1,546,662
C4	-£67,508,476	1.80%	1,031,108
C5	-£47,056,527	0.57%	515,554

In scenario C3 assuming this is the case and the power station also develops the DH scheme then the NPV at DCR of 9% without subsidy is -£88 M and the IRR to around 2.3% so a £88M net present value subsidy would be required to achieve an IRR of 9%. The capital cost would be around 145M (£11M to fit heat extraction capability to the power station and £134M for the DH network) and the required whole life subsidy as a percentage of capital cost would be around 61%.

Once the network has reached the peak of its development (75% of all identified potential heat load) it would be delivering around 450,000 MWh/year of heat to customers and saving around 47,000 TCO ₂/Yr.

The lifetime CO 2 savings would be around 1,500,000 TCO ₂ so the lifetime subsidy required to make the scheme cost effective would be around £57/ TCO ₂.

5.4 Longannet

Figure 13 Potential pipe route for Longannet

The network extends westward to Stirling and eastward to Dunfermline and beyond to the Exxon Mobile plant at Mossmorran.

The potential to cross the Kincardine Bridge was discounted due to:

• The heat recovery that could be provided from industrial sites in the Grangemouth area - which would not incur the cost of a pipe network to reach, and then cross the Kincardine Bridge and then cross the south bank of the Forth.
• The potential for additional competition from the proposed biomass plant on Grangemouth docks.

Table 16 Longannet Heat Sales

Scenario Code	Final heat load % of potential	Timescale of build up	Final Peak Load	Lifetime Heat Sales
	%	yrs	MW	MWh
L1	100%	1	162.272	21,999,211
L2	100%	15	162.272	18,149,349
L3	75%	15	121.704	13,612,012
L4	50%	15	81.136	9,074,674
L5	25%	15	40.568	4,537,337

Table 17 Longannet Capital Costs

Scenario Code	Power Plant Conversion to CHP Capex	DH Capex Incl distribution, HCUs and heat meters in dwellings less conventional domestic boilers	Total Capex
	£	£	£
L1	£27,000,000	£72,948,433	£99,948,433
L2	£27,000,000	£72,948,433	£99,948,433
L3	£27,000,000	£62,006,168	£89,006,168
L4	£27,000,000	£43,769,060	£70,769,060
L5	£27,000,000	£25,531,952	£52,531,952

Table 18 Longannet Scenario Results

Scenario Code	NPV	IRR	Lifetime Carbon Saving
	£	%	Tonnes CO 2
L1	£13,732,958	10.85%	1,117,719
L2	-£45,414,355	4.01%	922,118
L3	-£48,104,483	2.80%	691,589
L4	-£43,502,881	1.63%	461,059
L5	-£38,898,930	-0.74%	230,530

In scenario L3 assuming this is the case and the power station also develops the DH scheme then the NPV at DCR of 9% without subsidy is -£48 M and the IRR to around 2.8% so a £48M net present value subsidy would be required to achieve an IRR of 9%. The capital cost would be around 89M (£27M to fit heat extraction capability to the power station and £62M for the DH network) and the required whole life subsidy as a percentage of capital cost would be around 54%.

Once the network has reached the peak of its development (75% of all identified potential heat load) it would be delivering around 410,000 MWh/Yr of heat to customers and saving around 21,000 TCO2/Yr.

The lifetime CO 2 savings would be around 690,000 TCO ₂ so the lifetime subsidy required to make the scheme cost effective would be around £70/ TCO ₂.

5.5 Comparison of the Core Results for Each Station

The following table shows the key results for all four power station sites, showing the IRR under each of the five scenarios:

Table 19 Key Results

	Scenario Code	Final heat load % of potential	Timescale of build up	IRR
		%	years	%
Peterhead	P1	100%	1	7.73%
	P2	100%	15	3.31%
	P3	75%	15	2.08%
	P4	50%	15	0.68%
	P5	25%	15	None
Hunterston	H1	100%	1	12.72%
	H2	100%	15	5.01%
	H3	75%	15	3.63%
	H4	50%	15	2.17%
	H5	25%	15	-0.58%
Cockenzie	C1	100%	1	8.10%
	C2	100%	15	3.14%
	C3	75%	15	2.31%
	C4	50%	15	1.80%
	C5	25%	15	0.57%
Longannet	L1	100%	1	10.85%
	L2	100%	15	4.01%
	L3	75%	15	2.80%
	L4	50%	15	1.63%
	L5	25%	15	-0.74%

The key points from this are:

• Only two systems make a commercial rate of return of 9% or more are H1 and L1 - both are under the ideal circumstances of 100% load connected in year 1.
• The best rate of return for scenario 2 is for H2 at 5% - however this is far from financially viability.
• The best rate of return for scenario 3 is for H3 at 3.6% - however this is also far from financially viability.

Hence under current policy conditions none of the more realistic options is close to commercially viable. As there are no schemes of this scale of heat demand connected to any UK power station, this is not an unexpected result. However the model allows a range of policy options to be modelled and the core results give an indication of the key issues.

High capital cost is a well-recognised key issue for district heating. The following table shows the funding required enabling each scenario to reach a 9% IRR.

Table 20 Funding requirement levels

	Scenario Code	Final heat load % of potential	Timescale of build up	Funding required to Reach 9% IRR
		%	years	£ million
Peterhead	P1	100%	1	2.17
	P2	100%	15	11.80
	P3	75%	15	12.62
	P4	50%	15	12.37
	P5	25%	15	12.12
Hunterston	H1	100%	1	none
	H2	100%	15	27.25
	H3	75%	15	31.65
	H4	50%	15	31.59
	H5	25%	15	31.53
Cockenzie	C1	100%	1	11.51
	C2	100%	15	92.60
	C3	75%	15	87.96
	C4	50%	15	67.51
	C5	25%	15	47.06
Longannet	L1	100%	1	none
	L2	100%	15	45.41
	L3	75%	15	48.10
	L4	50%	15	43.50
	L5	25%	15	38.90

This shows that even with 100% load connected (scenario 2); the funding requirements are significant, ranging from £12 million to £92 million.

These are very large sums; however the amount of carbon saved is also significant. Hence the cost effectiveness of the potential funding is key.

The carbon savings are assessed over the 40 year lifetime of each scheme. As heat load builds up over 20 years the carbon savings in year 1 will be small, growing by 5% pa. Because the carbon savings are different in each year it is not possible to give a single figure for annual carbon savings, hence the table below shows the lifetime carbon saving and the funding needed per tonne of lifetime carbon savings.

Table 21 Capital Grant Cost Effectiveness

	Scenario Code	Lifetime Carbon Saved	Funding requirement to Reach 9% IRR	Funding per tonne CO 2
		ktonnes CO 2	£ million	£/tonne CO 2
Peterhead	P1	347	2.17	6.25
	P2	286	11.80	41.19
	P3	214	12.62	58.75
	P4	143	12.37	86.35
	P5	71	12.12	169.18
Hunterston	H1	1	0.00	0.00
	H2	901	27.25	30.24
	H3	676	31.65	46.82
	H4	450	31.59	70.09
	H5	225	31.53	139.90
Cockenzie	C1	2	11.51	4.60
	C2	2	92.60	44.90
	C3	1	87.96	56.87
	C4	1	67.51	65.47
	C5	515	47.06	91.27
Longannet	L1	1	0.00	0.00
	L2	922	45.41	49.25
	L3	691	48.10	69.56
	L4	461	43.50	94.35
	L5	230	38.90	168.74

This shows a range of cost effectiveness for the more realistic scenarios 2 to 5. This ranges from £30/tonne CO 2 to £168/tonne CO 2. Generally the higher the connection level the better the cost effectiveness.

This has an important lesson for the policy ideas, as policies that can increase connection rates will reduce the gap in capital funding needed and offer better cost effectiveness.

5.5.1 Comparison of Heat Loads

5.5.2 Breakdown between existing and proposed heat customers

The Peterhead network is dominated by proposed loads, hence this network is the most speculative, as there is significant risk that these heat uses are not developed. The Longannet network is dominated by existing users and as such this network is the least speculative.

Figure 14 the breakdown between existing and proposed heat customers for the four sites

5.5.3 Breakdown between the sectors served

There sectors served by the four stations may be categorised as domestic, public and business. Cockenzie is the closest of the four sites to a large conurbation; hence this has significant heat supply to all three sectors. Hunterston is dominated by heat supply to the business sites around Irvine.

Figure 15 The breakdown between the sectors served for each of the four sites:

5.6 Sensitivity Analysis

This section tests out how sensitive the core results are to changes in key variables and includes some basic assessment of how capital and revenue support might change investment returns.

Where the sensitivity test takes the IRR above 9%, the target rate, the result is in bold.

5.6.1 Sensitivity to Build out Rate

Tests have been run to show the impact of a more accelerated growth of the heat load.

With a large capital expenditure, increasing the revenue from heat sales at an earlier period has a potentially important impact. We have modelled three scenarios:

• Accelerated heat load build-up of 10 years.
• Accelerated heat load build-up of 5 years.

Table 22 IRR for the 2 sensitivity tests

	10 Years	5 Years
P2	4.56%	6.13%
P3	2.61%	3.74%
P4	0.23%	0.94%
P5	None	None
H2	6.93%	9.56%
H3	4.46%	6.40%
H4	1.47%	2.72%
H5	None	None
C2	4.45%	6.19%
C3	2.45%	3.70%
C4	0.02%	0.80%
C5	None	None
L2	5.78%	8.14%
L3	3.42%	5.16%
L4	0.53%	1.66%
L5	None	None

In no cases does a 10 year build out reach the 9% target IRR. In only one case, H2, does a 5 year build out reach the 9% IRR, however L2 is close to this rate.

As would be expected the test has the greatest impact on Scenario 2 as this has the highest heat loads and hence the highest revenue from heat sales.

In line with the IRR results the funding gap falls very significantly for the Scenario 2 cases.

Table 23 The funding gap for the 2 sensitivity tests:

Funding gap	10 Years	5 Years
P2	9.01	5.46
P3	12.14	9.48
P4	15.28	13.51
P5	18.42	17.53
H2	13.92	0.00
H3	28.34	15.26
H4	42.77	34.05
H5	57.19	52.83
C2	70.10	40.37
C3	94.81	72.51
C4	119.53	104.66
C5	144.24	136.81
L2	28.89	7.20
L3	46.65	30.39
L4	64.41	53.55
L5	82.18	76.75

5.6.2 Sensitivity to Capital and Revenue Support

The core results emphasis two key financial challenges for district heating:

• The large impact of the high upfront capital costs needed for the heat network. The impact is high as the sums of investment are large and they all occur in year one of the project's financial life.
• The modest impact of the income from heat sales. The impact is modest as the initial heat sales are low and the discounting process reduces the impact of the increasing heat sales in later years.

There are two obvious routes to change the status quo:

• Offer substantial capital funding support
• Offer revenue support on heat sales - a District Heating Incentive

Both of these routes would involve substantial sums of state aid. There are state aid guidelines for environmental support measures - however they tend to focus on renewable energy. The heat from these 4 power stations is predominantly derived from fossil fuel. Hence there will be a significant challenge to gain approval for significant aid. In addition the sums involved are likely to be substantial - raising value for money questions. For the District Heating Incentive this is the type of measure that would probably require UK level primary legislation.

The large scale financial support measures are unlikely to be practical or affordable policy ideas hence they have not been selected for the policy analysis stage of this report. To illustrate the scale of these funding routes and their impact this section provides a sensitivity analysis testing out these ideas.

Capital Funding

The capital costs vary from station to station, depending on the size of the heat network - which depends on the assumption on the level of connections. The following table shows the capital costs under the most extreme scenarios.

Table 24 Summary of Capital Costs

Power Station Site	Capital for Scenario 1 (100% heat load)	Capital for Scenario 5 (25% heat load)
Peterhead	£21,553,652	£14,556,094
Hunterston	£71,620,175	£42,617,061
Cockenzie	£168,955,127	£66,146,610
Longannet	£99,948,433	£52,531,952
Total	£362,077,387	£175,851,717

Hence 10% capital funding would be between £17 and £36 million for all four sites.

For this idea we propose testing the impact of 20% and 40% capital funding on Scenario 3 - 75% heat load within 15 years.

District Heating Incentive

District heating is a major capital investment. The financial returns are dominated by the value of heat sales vs. the high initial capital costs. In general fiscal support for low carbon technologies are moving towards revenue incentives, e.g. ROCs, FIT and the RHI.

This idea tests the impact of a District Heating Incentive - paid on the sales of heat. This would be similar in concept to the RHI. The proposals for the RHI included a possible uplift for renewable heat supplied via district heating. As currently implemented the RHI does not have this uplift.

This tests the impact of rewarding heat from district heating with the RHI level for biomass boilers over 1 MW, i.e. 2.6 p/ kWh and at half this level 1.3 p/ kWh.

Capital and Revenue Support Results

The table below shows that with high levels of capital or revenue support that in two cases Scenario 3 can improve from a low IRR of 2.08% to3 3.36% to levels above the 9% threshold.

Table 25 Results of Capital and Revenue Support Ideas

	Longannet		Hunterston		Cockenzie		Peterhead
Policy	NPV	IRR	NPV	IRR	NPV	IRR	NPV	IRR
No Support	-£48,104,483	2.80%	-£31,652,504	3.63%	-£87,960,720	2.31%	-£12,623,863	2.08%
Heat Incentive equivalent to RHI	£17,900,253	10.48%	£24,204,688	11.61%	-£15,803,867	8.14%	-£2,598,817	7.98%
Heat Incentive equivalent to RHI/2	-£15,102,115	7.55%	-£3,723,908	8.53%	-£51,882,293	5.79%	-£7,611,340	5.60%
40% Capital Grant	-£12,502,016	6.65%	-£5,681,644	7.60%	-£29,868,686	5.65%	-£4,648,331	5.26%
20% Capital Grant	-£30,303,250	4.39%	-£18,667,074	5.26%	-£58,914,703	3.68%	-£8,636,097	3.40%

The table below shows the financial cost of each of these ideas. In the case of capital support this is simply the value of the capital grant. In the case of the District Heating Incentive it is the NPV of the annual support.

The following charts show the impact of these ideas.

Figure 16 Sensitivity Results for Longannet

Figure 17 Sensitivity Results for Hunterston

Figure 18 Sensitivity Results for Cockenzie

Figure 19 Sensitivity Results for Peterhead

The following table shows the costs of these sensitivity tests.

Table 26 Cost of Capital and Revenue Sensitivity Tests

	Longannet	Hunterston	Cockenzie	Peterhead
	Policy Cost WLC	Policy Cost WLC	Policy Cost WLC	Policy Cost WLC
Heat Incentive equivalent to RHI	£66,004,737	£55,857,192	£72,156,853	£10,025,047
Heat Incentive equivalent to RHI/2	£33,002,368	£27,928,596	£36,078,427	£5,012,523
40% Capital Grant	£35,602,467	£25,970,860	£58,092,034	£7,975,532
20% Capital Grant	£17,801,234	£12,985,430	£29,046,017	£3,987,766

5.6.3 Conclusions on Sensitivity Tests

A small number of the sensitivity tests show that the target rate of a 9% IRR can be met. These are:

• Accelerating the build-up of the heat load from 15 years to 5 years.
• A high level of revenue support, similar in level to the Renewable Heat Incentive.

Of these two the revenue support is not a practical option. The NPV of the revenue support is higher than the funding gap. For Longannet the NPV of the revenue support is £66 million, whereas the funding gap is £45 million. In addition it would be difficult to present a State Aids cased to support heat from a fossil fuel source at the same support level as heat from renewable sources.

This suggests that:

• Policy measures to bring forward heat sales are a more affordable and practical option than revenue incentives.
• Policy measures that can reduce the costs of the heat network, e.g. by co-locating heat demand and heat supply, could play an important role.

Hence planning policy has an important role to play.