Why I am sceptical about Geothermal Energy

Friends, some people are of the opinion that geothermal energy offers practically unlimited opportunities for the generation of electricity with low carbon dioxide emissions. For example,

  • The US Government’s GeovisionReport is an extensive and overwhelmingly positive assessment of the potential for geothermal energy generation in the US.
  • This IEA video describes new drilling technologies and possible applications in Germany.

For a more general overview, take a look at this video on the ‘Just have a think‘ channel, or read these 10 pages on How Stuff Works.

And there are indeed real possibilities. However I am sceptical that such opportunities are ‘practically unlimited’, particularly in the UK.

And since we are in an emergency situation and need to act urgently, I am doubly sceptical of novel technologies which might steal investment from known solutions.

Allow me to explain….

General Situation

Below the surface, the temperature of Earth increases at typically 25 °C to 30 °C per kilometre of depth. So a few kilometres below the surface, rocks are typically at several hundred degrees Celsius. And this thermal energy represents a potential energy resource.

The US Government’s Geovision Report outlines a number of ways in which this geothermal energy might be extracted. These are summarised in the graphic below.

Click on image for a larger version. The Geovision summary of geothermal potential.

The large-scale projects typically involve four parts.

  1. The injection of cool water down a drilled well to some hot rocks.
  2. Percolation of water through the hot rocks.
  3. Extraction of hotter water from a second well.
  4. Generation of steam and electricity.

Despite the very large volume of hot rocks available under our feet, there are two basic difficulties that all these schemes face.

  • Firstly, rocks of all kinds have a poor thermal conductivity. So if we extract heat from the surface of a rock, that surface will cool down. However heat will only flow back into the cooled rock slowly. This limits the rate at which heat can be extracted.
  • Secondly, except in a few geologically exceptional places, the upwards geothermal heat flow from the deep Earth is very low – typically less than 0.1 W/m^2.

Given these difficulties, there are a lot of technological tricks that can be exploited to optimise extraction of heat in particular circumstances.

For example in Enhanced Geothermal Systems (EGS), engineers ‘frack’ a volume of rock which increases the surface area over which water can flow. As water percolates through the rock it is then able to extract heat more efficiently.

However, no technological advance can overcome the basic fact that the heat flux from the interior of the Earth is – in most places – very low. This means that the technology is essentially ‘one-shot’ i.e. the heat extracted from the rock will be only slowly replaced by heat from the interior of the Earth.

This means that once a geothermal plant has ‘harvested’ the geothermal energy from a block of rock, it will need to move on to a new block. So the big question is: “How long will that ‘one-shot’ last?”

Simplified Model 

I have constructed a spreadsheet model to assess how much electricity could be generated from one cubic kilometre of rock, and how long that extraction could go on for. The model is illustrated in the graphic below.

Click on image for a larger version. A simple model considering how much energy could be extracted from a cubic kilometre of rock, a few cubic kilometres under the Earth. I have illustrated a cubic kilometre at a depth of 3 kilometres, but in most places the rocks would don’t reach that temperature until nearer 10 km.

The model is very crude. For example, it does not capture the dynamics of the extraction process. But it does represent a simple way to assess the resources available. I’ll discuss the shortcomings of the model below.

Details of the model are given at the end of this article, and here I just describe the results.

One cubic kilometre of rock

The model considers one cubic kilometre of ‘fracked’ rock through which water can easily percolate, and be warmed by 100 °C.

The total heat available to be extracted is ~1.6 x 10^17 joules or 45 billion kWh_th. The “_th” suffix indicates thermal energy.

If we suppose that we wish to generate 100 MW_e of electricity, then using a typical generating efficiency of 33%, we need to extract heat at a rate of 300 MW_th. The “_e” suffix indicates electrical energy. This corresponds to a very high flow rate of around 0.7 cubic metres of water per second.

Based on a uniform extraction rate, the thermal resource would be exhausted after ~17 years, a lifetime which could be extended by extracting energy at a lower rate.

But once exhausted the time to restore this thermal resource is tens of thousands of years.

Once the thermal resource has been extracted, the cubic kilometre of rock will have shrunk sufficiently that the land above it will subside by roughly 0.6 metres.

One might also usefully compare the 100 MW_e generation from 1 cubic kilometre of rock with the Solar PV generation from 1 square kilometre of the Earth’s surface.

Assuming that 1 square metre of solar panels generates 1 kWh/day in summer, then a square kilometre of solar panels will produce 1,000 MWh_e/day which can be compared with the 24 x100 = 2,400 MWh_e produced from the rock below.

If one reduced the rate of extraction by a factor 2.4 to 42 MW_e to match the solar PV generation, then the lifetime would be extended to about 40 years – similar to a normal power plant.

What might a geothermal plant look like?

The geothermal resource can be used alongside the  solar PV and wind generation to produce year-round, reliable low-carbon electricity.

If one cubic kilometre of rock could generate 42 MW_e for 40 years, then to generate say 3.5 GW_e – similar to the output of Hinkley C –  we would need roughly 3,500/42 ~ 82 square kilometres – or an area ~ 9 km x 9 km – which is a large amount of land, but small compared to our total land area of the UK.

The geothermal surface plant need not occupy much of this area, but all the area would be potentially affected by subsidence, and/or small earthquakes during the fracking process.

Additionally, the geothermal energy produces a large amount of waste heat, which might also be available for district heating or warming crops.

Discussion of the Model

I have used figures from the report  which suggest that injecting water at 200 °C and withdrawing at 300 °C represent useful operating parameters. The water would be used in a so-called ‘binary’ plant to heat a secondary fluid which would then power a generator.

The idea of warming water by 100 °C is pretty optimistic since the thermal gradient within the Earth is only 25 °C/km. And in most of the UK one would need to drill to ~10 km to reach rocks at this temperature, something which is extremely challenging. So I have just assumed that somewhere in the UK, hotter rocks are available nearer the surface.

I have also not considered any impacts on groundwater flow.

The most obvious criticism of the model is that extracting heat energy at a uniform rate is unrealistic. In practice, the heat would be extracted easily at first, and over years the rate of heat extraction would fall exponentially. This could be used to boost initial power output as the expense of a shorter lifetime. But the calculated amount of heat is the maximum energy that could be removed over the lifetime of the well.

One could of course remove heat from a greater vertical depth, but this would leader to greater subsidence, and in the UK we do not have large areas over which subsidence is acceptable. In old mining areas, subsidence is a considerable blight, and so this could only really be considered in the most isolated of places.


Drilling deep wells is hard work and so expensive. One figure from the ‘How stuff works‘ article suggests $20 million dollars per 10 km well.

So to extract heat from a 9 km x 9 km area would require (I guess) around 81 boreholes and so the capital cost would be in the range £1 billion to £2 billion. Even with the additional costs of generating plant, the cost would likely be less than a nuclear station (£20 billion). And the plant could be enlarged slowly with new blocks being drilled as the first blocks came on line.

So geothermal energy is a large but still finite resource from which heat is harvested and then wells are abandoned and the plant moves on to new areas., after perhaps a decade or so. And so we could imagine plants slowly ‘grazing’ across the country under areas which were insensitive to subsidence or earthquakes.

This is a rather different proposition to that which is marketed by geothermal advocates in which geothermal energy is considered as practically infinite and having negligible impact on the land. It may make sense in the western US, but I am sceptical that it makes sense at scale in the UK, especially given the potential risks.

And as I mentioned at the start we are in a Climate Emergency and need to act urgently. And this makes me doubly sceptical of novel technologies which might steal investment from known solutions.

In contrast, investments in wind and solar and batteries will definitely bring electricity costs down, reduce carbon emissions and reduce – but not yet completely eliminate – the use of gas.



Details of the spreadsheet model

The spreadsheet model can be downloaded here and is described below. I apologise in advance for any errors I have made.

Click on image for larger version. These are the basic parameters used in the model

Click on image for larger version. This section of the spreadsheet calculates the thermal properties of one cubic kilometre of rock.

Click on image for larger version. This section of the spreadsheet calculates the operational parameters of a geothermal power plant assuming a constant rate of heat extraction.

Obviously the properties of rocks vary greatly, and the hottest rocks often occur under very hard rock – very different from the rocks through which oil and gas companies normally drill.

However, I think the spreadsheet model captures the correct order of magnitude of the geothermal resource. It’s large, but finite.

2 Responses to “Why I am sceptical about Geothermal Energy”

  1. Enrico Says:

    Michael, what is your take on the 100% geoEnergy sustaining Iceland?

    Your old time friend, Enrico

    • protonsforbreakfast Says:


      Lovely to hear from you. I trust you and your family are well.

      In the article I said:

      Secondly, except in a few geologically exceptional places, the upwards geothermal heat flow from the deep Earth is very low – typically less than 0.1 W/m^2.

      Iceland is one of those exceptional places. They have a very small population (0.4 million people) and very large geothermal resources, so geothermal energy makes sense. But in most places it doesn’t and it’s a ‘one-shot process.

      Additionally in most places, the fact that Geothermal extraction causes earthquakes would be considered a disadvantage. In Iceland, the tiny population experiences a plethora of earthquakes and volcano but probably don’t even notice. It is however a serious problem in Northern California (where Earthquakes are also not uncommon!) because there are more people around.

      Just compare the global average of solar irradiation (240 watts per square metre) with the heat flow from the Earth (<0.1 watts per square metre). That's a factor 2,400. In general it's much easier (i.e. cheaper) to get access to genuinely renewable energy from solar than from geothermal.

      Best wishes


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: