This question already has an answer here:

At what power output would we be using so much geothermal energy that we cool the core enough to endanger the Earth's magnetic field and have to stop using it? Is this a conceivable concern for a future energy crisis?


marked as duplicate by Mazura, Community Feb 8 '17 at 1:41

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


No, this seems highly unlikely.

According to online sources:

Average human power consumption in 2008 was estimated at 15 TW.

Total annual heat loss from the Earth due to the surface heat flux is estimated at 44.2 TW.

Estimates of the electricity generating potential of geothermal energy around the world are consistently less than 2 TW.

So although the average power consumption is a surprisingly large fraction of the total geothermal heat flux, the part that might possibly be extracted for power is quite small.

There are much more concerning environmental impacts to geothermal power development than freezing the core and destroying the Earth's Magnetic field. These include increased rates of release of carbon and sulfur into the atmosphere and land use issues.

  • 2
    $\begingroup$ However unlikely, commercially invalid, or otherwise infeasible with today's technology there is a number of watts that if removed from the earth's core will solidify it. +1 for other 'more concerning environmental impacts'. $\endgroup$ – Mazura Aug 1 '14 at 0:54
  • 1
    $\begingroup$ @Mazura, a key point though is you cannot draw an arbitrary wattage from the core with geothermal energy. You can only get some fraction of the heat flux that is already conducting up to the surface anyway. So you aren't really affecting the core cooling rate. $\endgroup$ – ZSG Nov 19 '14 at 5:56
  • $\begingroup$ Mark, not sure what you mean by increased rates of carbon and sulphur released, can you say more? $\endgroup$ – a different ben Nov 25 '14 at 22:45
  • $\begingroup$ Regarding land use issues, footprints per kW of current geothermal power plants are amongst the smallest of all electricity-producing industries. $\endgroup$ – a different ben Nov 25 '14 at 22:47
  • $\begingroup$ @a different ben - Because C & S gas species are in volcanic/geothermal emissions, there are measurable emissions of these from a geothermal power plant. Should this be regulated in the same way as a carbon emission from an coal-fired plant. It is measured at a 'stack'. On the other hand it may be 'natural' in the sense that it was going to be emitted from a fumarole anyway. My understanding is that this is treated by regulatory agencies as the same. Perhaps this is a confusion, but I believe this has in fact had an impact on the licensing of geothermal power plants in places such as Hawaii. $\endgroup$ – Mark Rovetta Nov 27 '14 at 2:00

This is a bit of a what-if kind of answer, with some rough estimates. Just considering the cooling aspect - let's pretend that Earth is solid for a momnet.

The time-dependent heat equation tells us how long it will take for a change in temperature to travel a certain distance by conduction (the 'characteristic timescale'). The distance to the outer core, where the magnetic field is generated, is about 3000 km.

Assume that the geothermal installation is at the surface, and that it loses its heat instantaneously. Then, the 'characteristic timescale' is given by $$ \tau = \frac{l^2}{\kappa} $$ Taking $ l=3 \times 10^{6} $m as the distance to travel, and $ \kappa = 1 \times 10^{-6} $m2/s as the mean thermal diffusivity of the rocks, then it would take $ 2.85 \times 10^{11} $ years for the change in temperature due to extracting geothermal heat at Earth's surface to propagate to the outer core. The sun would have transitioned to a red giant by then.

The problem being of course that it's not all solid. Convection in the mantle would obliterate the temperature signal. I don't know how you'd go about calculating that.

More practically, a deep geothermal project of the EGS flavour would be managed as a mining project. The reservoir would be used up within 20 years or so, then either abandoned or mothballed (until perhaps it reheated?). The maximum temperature drawdown for economic feasibility might be of the order of 50°C or less, depending on a boatload of factors.


The geothermal energy is produced in 3 main ways: When the Earth is formed eons ago, the kinetic energy of the mass that rammed into the surface of the earth turned into heat. Earth gravity force further compressed the materials that also generated some heat, element natural decay caused nuclear fission energy be produced.

Also, the Moon, the Sun’s gravity caused periodical (very slightly) distortions on Earth that also generate some heat though very little.

In these sources of heat: The Earth forming process stopped and the kinetic energy turned into heat no longer have supply. The gravity compress effect that generate heat no longer significant because matter already dense enough in Earth. Radioactive elements on Earth only can be consumed but no further supply from space; the heat they produced will be weaker and weaker when time elapse along. The tidal movement caused heat is too little on the Earth to compensate the loss of heat through natural dissipation.

The heat produced from the core (20×〖10〗^12W) cannot compensate the heat that being conducted to surface and dissipated out into space (44×〖10〗^12W).

So as an observed fact, the Earth’s total mass is cooling all the time. When there is no supply (or not enough supply) while there is a constant loss (natural energy dissipation) or accelerated loss (manmade energy dissipation by tapping into geothermal energy), it is a simple mathematical logic: the Earth’s geothermal energy is not sustainable --- don’t even mention renewable.

Someone argue: the Earth is so big and we are so small, we use a little bit of geothermal energy, it doesn’t matter. OK, if this proportional difference can allow human to use the geothermal energy long enough to the time the Sun’s life span is going to end and start to swell and scorch the Earth, then it makes sense. Otherwise it is non-sense.

Then how long it can last if we use geothermal energy?

Here is some rough calculation (because in this case we only need some qualitative analysis to decide sustainability issue on geothermal energy):

In 2006, human use 16 trillion w /s;
Earth total mass: 5.9763×〖10〗^24kg;
Mean density: 5517kg/M^3;
Estimated average heat capacity: 700 J/kg ℃;
Kilowatts/hour= 3.6×〖10〗^6Joule;
16 trillion w/s = 〖16×10〗^9kw×〖60〗^2s= 〖57.6×10〗^12kw/h;
1℃ down for Earth total mass:
5.9763×〖10〗^24kg×700 J/kg÷ (3.6 ×〖10〗^6Joule)

How many years human use can cause 1℃ down for Earth total mass?

1162×〖10〗^18kw/h÷〖57.6×10〗^12kw/h×24h×7d×52w = 2309year

How many years human use can cause 200℃ down for Earth total mass?

2309year×200℃ = 461,800year

How many years we taking off our descendants by using geothermal energy for one year by this rate?

2 billion year/461,800year≈4331year

The heat inside the Earth is the living force that keeps the Earth alive --- the heat inside the Earth resist water be sipped into the Earth, keep the molten core flowing that maintain the magnetic fields of the Earth that protect all lives on it and that also protect the atmosphere not been blown away by solar wind, keep continental plates moving that make mountains which keep the Earth uneven to maintain weather and prevent all Earth surface be submerged into 50 meters of water…

Earth with weakened geothermal energy will become another Mars.

  • $\begingroup$ You are assuming that we could extract all our power from geothermal energy. This is wrong for a whole multitude of reasons, some of which were mentioned in other answers. I suggest you read those first. $\endgroup$ – bon Feb 7 '17 at 18:23
  • $\begingroup$ you mean we can use some but not all, right? a compromise? $\endgroup$ – Ming Lou Feb 8 '17 at 6:46
  • $\begingroup$ what I mean is that use geothermal energy is fundamentally wrong, principally wrong. $\endgroup$ – Ming Lou Feb 8 '17 at 6:48

Not the answer you're looking for? Browse other questions tagged or ask your own question.