10. The Economics of the Geothermal Well
We now move on to the economics of the geothermal well itself. The key question we seek to answer is: what is the likely unit cost of the heat produced by such a well? If this is too high it will be uncompetitive against alternative future heat sources for the Banchory heat network, such as natural gas or (especially) continuing biomass heat.
Production costs: assumptions about the geothermal well
To model the financial viability of a geothermal heat generating station at Hill of Banchory we must first make some reasonable - and clearly defined - assumptions about the characteristics of the geothermal system itself. These will include:
- system lifetime;
- operation & maintenance costs;
- heat production capacity and (crucially)
- the likely capital cost of drilling the boreholes
System lifetime and operation and maintenance costs
Deep geothermal heat systems are known for having exceptionally long lifespans - several decades is quite normal. From the point of view of financial modelling this could actually be problematic, as it raises difficult questions over how to value income far in the future. In practice, energy sector investors will not make decisions based on a time frame greater than 20-25 years. Furthermore, RHI income is only available to a project for 20 years from commissioning.
While we would expect the major physical elements of the system (especially the boreholes) to last for much longer, we have therefore based the financial modelling of the Banchory system on a twenty year lifetime.
For operations and maintenance costs, we have assumed a figure of 0.5p/kWh, which is a normal (and quite conservative) assumption across the geothermal heat sector.
The heat capacity of the geothermal well: flow rates and temperature
Chapters 4 to 7 set out the Consortium's analysis and conclusions about the likely geothermal potential of the Hill of Fare Pluton. These suggest that, on a reasonably conservative estimate, water temperatures of approximately 75°C could be obtained at a depth of 2.5 km. It is entirely reasonable that an efficient and effective heat network could operate using this input temperature; we might also assume a maximum achievable 'ΔT' (i.e. the difference between input and output temperatures for the heat network) of around 30°C, with the network operating at between 70-80°C flow temperature and a 40-50°C return temperature.
We address the challenges of matching well output and network input temperatures below - the current HOB network operates on a flow/return of 85/60°C, with the return temperature sometimes being as high as 65°C.
Figure 29: Biomass fuel at the Hill of Banchory Biomass Energy Centre
Image: Cluff Geothermal / Hobesco
The available heat output from the system will be heavily dependent on the maximum sustainable flow rate (in litres per second) from the geothermal well. The geological members of the Consortium recommended that three representative flow scenarios ('favourable', 'modest', and 'unfavourable') should be used when modelling the geothermal potential as set out in Table 9.
Table 9: Estimated flow rates for the Hill of Fare granite at Banchory
Flow rate (l/s)
Table 10 below illustrates how flow rate and output temperature influences the heat capacity of the well. We vary the achievable ΔT from 5-30°C and vary the flow rates across the 5 l/s to 50 l/s range suggested by our geological analysis. As will be seen, there are many combinations of flow rate and output temperature that imply a well heat capacity of more than about 1.5 megawatts, the level at which the project becomes worthwhile from a presentational point of view. If conditions turn out to be favourable, a capacity of 5 MW plus can be expected.
Table 10: Estimated heat available from geothermal resources for given flow rates and temperature drops
It should be noted that the values in Table 10 are for the heat capacity of the well at the wellhead. There will be some heat loss as the heat is transported to the network and processed though the heat exchanger - though these should only be a few percent at most.
As the primary goal of this project was to look at the geothermal integration with the existing scheme, for the purposes of the financial modelling, the current network's operating temperatures were used as a benchmark for the resource temperature requirements in the later stages of system design. In other words, we use as a benchmark a source temperature of 85°C, a minimum expected ΔT of 20°C and thus a return temperature of 65°C. This will allow the geothermal well to interface directly with the HOB network.
The capital costs of the boreholes
We have used £1.5 million as the approximate unit cost of the boreholes. Every borehole is different and the cost of each one is subject to uncertainty. However, we are aware of costs for similar geothermal boreholes in the recent past. We would also note that drill rig hire costs will depend, inevitably, on the level of demand in the UK onshore drilling market.
It must be noted that there will always be some uncertainty over the physical parameters of any proposed geothermal project. Ultimately, the output temperature and flow rate from the well will be known with 100% confidence only once a borehole has been drilled and the geothermal potential measured.
Observations on the economics of deep geothermal heat systems
For almost all deep geothermal energy systems drilling costs will have a dominant impact on the cost per unit of heat generated. In this case, significant outlay will also be required for the pipeline connecting the Hill of Banchory Energy Centre to the likely site of the wellhead (which we assume will be not far from the B977 road).
Two other factors are key determinants of the final unit cost (here we have used pence per kilowatt-hour) of the geothermal heat produced. These are:
- volume of heat sold per year; and
- availability of subsidy income, in particular the Renewable Heat Incentive ( RHI)
Volume of heat sold. This is important because the costs of building a deep geothermal system are only weakly proportional to the final heat capacity. Drilling the borehole(s) will inevitably involve a multi-million pound investment, and this must be made before any heat is produced. While there will be some continuing production costs (powering the pump, maintenance etc.) this up-front capital expenditure dominates the unit cost of heat, even in the medium to longer term.
Consider a 3 MW geothermal system that has cost, say, £3 million to build. The cost of financing an investment of this size will be high, certainly in the hundreds of thousands of pounds per year. For the sake of arithmetic simplicity, let us imagine that the full cost of operating the system and meeting the financing costs is £500,000 per year. If the system sells only 100 megawatt-hours (MWh) of heat per year, this would represent a very expensive way to produce heat: £5,000 per MWh. However, if the same system operates at a 40% load, it will sell 10,512 MWh per year at very little extra cost equating to a production cost of £48/MWh. (Given the long system lifetime we would expect unit heat cost to drop as the system reached financial maturity.)
In other words, the financial viability of a deep geothermal heat system is strongly dependent on good heat sales. The presence of the already existing Hill of Banchory heat network thus represents a great advantage to this project: subject to a suitable water temperature being achieved, it offers a ready-made large scale heat customer - and one with considerable scope for expansion.
Renewable heat subsidies. At present geothermal heat is eligible for support under the RHI at a rate of £50 per MWh produced. This is payable only on 'useful heat' i.e. heat that has been consumed by customers who meet defined energy efficiency standards i.e. it is not payable simply on heat produced. The RHI tariff is index linked and payable for 20 years from a system's commissioning date. DECC is currently considering a review of the structure of the RHI, though a commitment has been made to its future funding. The deep geothermal tariff is generally seen as unlikely to be adversely affected by the RHI review, though there can be no guarantee of this.
As a consequence of EU State Aid rules, projects in receipt of a public sector grant for the heat-production facilities (i.e. not including the cost of pipes, civils etc.) are not eligible for the RHI. However, OFGEM (who administer the RHI) have put in place a recognised and officially sanctioned process under which projects that pay back any such grants can regain RHI eligibility. In many cases the cumulative RHI payments will be more valuable than the grant, and this will normally be the case for deep geothermal projects. In modelling the financial viability of the Hill of Banchory project we have assumed that the grant will be paid back, using bank finance, in order to regain RHI eligibility.
Results of the production costs modelling
The parameters set out above were input into an adaptation of Cluff Geothermal's financial model for deep geothermal heat systems. We also made some assumptions that are not directly dependent on the geothermal outcome:
- the RHI in the first year of operation is available at a rate of £52/MWh;
- the cost of the connecting pipeline between the wellhead and the heat network is £1,000,000;
- fixed annual costs for the geothermal well (site rental, salaries, are £20,000 per year);
- the relevant part of the grant can be repaid through a loan at 6% over 15 years; and
- heat prices rise with RPI, which we take to be 2.5%
We also need to decide how to compare the different variables. We have adopted an approach based on well capacity, desired rate of investment return, the associated price for heat sold, and - as the output - annual heat sales. For example, if the well has a capacity of 3 MW, and heat is sold at 1p/kWh, how much heat must be sold to meet an investor's required rate of return of - say - 10%. (These rates of IRR are not unreasonable given the high risk factor involved in this type of exploratory drilling)
To phrase this another way, given a particular combination of well capacity and required investment return, reviewing how the other variables scale tells us how much the geothermal operator needs to charge for heat for a given level of heat sales.
Matching supply and demand
One complication for this part of the analysis is that the geothermal well will generally not be able to supply 100% of the heat demand from the network. Geothermal wells produce heat at a constant rate, while heat demand varies considerably both seasonably and diurnally. When demand is too high the well cannot simply be 'turned up' - instead, the backup heat generators (either gas or biomass) will need to cover the peaks. This problem is mitigated by having heat stores installed (essentially large water tanks; Kyriakis and Younger 2016). Such heat stores are already installed at Hill of Banchory.
This does not mean that heat customers cannot be supplied - it should always be possible to meet a given level of heat demand using the back-up heat sources. However, it does mean that some of the time the geothermal well is 'missing out' on supplying heat (and also receiving RHI income). Producing an exact figure for the percentage of 'missed sales' is not possible without knowing the exact future heat demand profiles. The model incorporates reasonable estimates for the 'missed heat' factor, for each geothermal well capacity being considered.
Outputs of the financial model
We set out this relatively complex set of output data in Table 11 below, and also in graphical form in figure 31. The 'heat sales required' figure refers to the overall sales across the heat network; the actual heat sold by the geothermal well to the network will be slightly less.
Table 11: Heat sales required to reach 10% and 20% IRR for wells of differing capacities
Summary of downstream financial model outputs
Given reasonable assumptions about the capacity of the well, a deep geothermal heat system at Banchory would be commercially viable if it could sell heat at 2p/kWh at a volume approaching 10,000 MWh. This would make it competitive against natural gas heating, but not biomass at the lower end of the price range (as analysis in the following chapter shows).
Note - Output/input temperature: a key challenge for geothermal heat at Banchory
Before moving on to consider the economics of the Banchory heat network itself, we consider one specific issue that has major implications for the system's overall viability. This is the output temperature of the geothermal well.
Many heat networks operate at a relatively high water input temperature - say 85-90 °C. This is the temperature of the water as it leaves the Energy Centre to begin its journey around the network. The existing Hill of Banchory network uses a flow temperature of 85°C. Given the typical thermal gradient in the UK, and the constraints imposed by drilling costs and likely levels of permeability, we would expect the output temperature of a geothermal well to be lower than this - perhaps in the range 60-85 °C.
If the geothermal well's output water temperature proved to be too low to be compatible with the existing heat network, there would be a requirement to boost it, either by fossil fuel or renewable means. That would mean extra cost, and it is possible this would increase unit heat costs to the point where they became uncompetitive. However, there is no reason that water at lower temperatures - 70 °C, say - cannot be used to provide space and water heating. However, doing so requires installing heat delivery equipment that is designed specifically to utilise lower temperatures (this can be underfloor heating or larger radiators). Retro-fitting existing residential heat customers to allow them to use lower temperature water in this way is generally expensive and unlikely to be economic - though the cost of conversion will be lower for larger scale heat customers.
Our assumption is that powering the existing Hill of Banchory heat network using the water from the geothermal well (without boosting the temperature) is unlikely to be possible. However, it would be entirely possible to exploit the geothermal heat available if future extensions to the existing heat network are designed to be compatible with lower temperatures.
There is real scope for this development route, building on cooperation with Aberdeenshire Council, which has several large properties and future projects in the areas most likely to see organic network expansion. These include a group of properties on Raemoir Road: the Banchory Academy and Sports Centre, Banchory Primary School and Nursery, Dalvenie Gardens Sheltered Housing and around 150 council-owned residential dwellings. The heat demand for these buildings was assessed in a report produced by Jigsaw Energy in 2013 as 5,300 MWh/yr, and peak load estimated at 3MW. 
A heat network large enough to make a 'mid-case', 2.5 MW geothermal well economically viable - for example, able to sell heat at 2p/kWh - would, under our analysis, require annual heat sales of 7,800 MWh. This group of properties would appear to form a highly attractive anchor heat load to form the core of such a network.