So I read somewhere that the inner core temperature of Earth might be as high as 7000 °C.

Depending on the composition of the different layers towards the surface, the temperature should decline until the surface temperature of the air, at about average 14 °C, is reached.

This does however not seem to be the case as the temperature at first gets colder if one starts digging?

How can this be explained?

And please excuse any naivety on my behalf as I´m no scientist.

  • 2
    $\begingroup$ Welcome to the site, and thank you for a good question. I'm sure somebody with more knowledge of this area will be along, but my quick answer would be that the temperature on the very surface is an anomaly, subject to weather, seasons, etc., and that once you go down a short way - into mines or caves - it's fairly constant. So perhaps that shallow depth is what should be considered the "surface" for the purposes of what you are asking. $\endgroup$ Commented May 18, 2017 at 10:30
  • $\begingroup$ The just below the surface effect is because the sun is also a source of heat. $\endgroup$
    – John
    Commented May 23, 2017 at 20:46

1 Answer 1


The temperature does indeed increase with depth, something that is a problem in deep mines or deep drilling, but a benefit for geotermal heating. The heat originates mostly from radioactive decay, but there is also a fraction of primordial heat from the formation of the Earth.

As a rule of thumb, geologists often assume 25°C/km temperature increasements within continental plates, but steeper increasements in oceanic plates or volcanic regions. The temperature curve is not constant, however. Due to the convection, the temperature increases slower with depth in the mantle and outer core. In that sense, the crust is an isolating cover. The temperature also varies due to subductions, the dynamics of the mantle and radioactivity within the crust.

The shallow soil temperature and the temperature in the air depend mostly on the insolation, the radiation heat from the sun. The average annual solar isolation is 1361 $W/m^2$ but only 20-500 $mW/m^2$ heat flux comes from Earth's internal heat production. However, this tiny fraction can sometime sbe important, e.g. to understand basal melting of ice sheets and glaciers.

 Geothermal gradient adapted from Boehler, R. (1996). Melting temperature of the Earth's mantle and core: Earth's thermal structure. Annual Review of Earth and Planetary Sciences, 24(1), 15–40. CC Temperature gradient from crust to core. Recent research have improved the accuracy, but the principle in the figure remains (Annual Review of Earth and Planetary Sciences, 24(1), 15–40)

So why do you experience the temperature colder underground when you are digging?

The temperature at depth is the average temperature for day and night or even the year. The energy from the sun reaches only the uppermost soil layer that absorbs heat but the thermal conductivity is relatively low and it will not transport the heat down. The ground is reluctant to change its temperature regardless of variations on the surface.

How strongly you experience the temperature difference depends on where in the world you are and the local conditions. In some areas, the ground remains frozen, known as permafrost, even as the temperatures in the air can be pleasant during the summer and the surface layer melts. The depth of the suggests that it's been formed over a long time. It reaches down to the depth where the mean temperature over hundreds or thousands of years is 0°C. At larger depths, the heat from the Earth's interior takes over and melts the frost.

CC Original contributor, German Wikipedia user, "HylgeriaK", translation by User:HopsonRoad. Wikimedia Temperature in the active layer and permafrost. The thickness of active layer is less than a couple of meters and the thickness of the permafrost depends on its age, the bedrock and the distance to the coast. (By German Wikipedia contributor, "HylgeriaK")

See the plot, linked by @JeopardyTempest in the comments to better understand the variations. Plotted on another scale and depth, annual variations would probably be seen as well.

  • 1
    $\begingroup$ So if I understand this informative answer correct, because the surface temperature used to be much colder and the heat conductivity of soil is low, it will take a very long time until the heat exchange between surface and core will have heated up the layer between them. And could the speed of this process be measured by how fast the base of the permafrost is changeing? $\endgroup$
    – HeLi8
    Commented May 18, 2017 at 11:58
  • 2
    $\begingroup$ Heat from the core is lost at the surface, and if more heat is lost than added, the temperature goes down, as we see in the crust of the Earth. For shallow soil, the energy from the sun is much more important than the geothermal energy, but the variations in air temperature are much faster than the soil due to its thermal inertia and limited conductivity. $\endgroup$
    – user2821
    Commented May 18, 2017 at 12:02
  • 2
    $\begingroup$ @HeLi8 There's two layers sources of heat, and one sink. The Sun heats from the outside, and the Earth from the inside. The sink is only on the outside (Earth's thermal radiation). The average daily temperature is constant (ignoring seasonal changes, the atmosphere and oceans etc.), which means that on average the amount of heat radiated by the Earth's surface is the same as the amount of incident radiation (from the Sun) and the heat escaping from the Earth source. If you removed the Sun, the surface temperature would get much lower, while the deeper crust would stay the same. $\endgroup$
    – Luaan
    Commented May 18, 2017 at 12:41
  • 2
    $\begingroup$ @gerrit depth of active layer varies in a large range, depending on a lot of factors. Not only temperature but from my experience humidity have been very important. Near glaciers, the active layer can be almost nothing, and in some valleys, with warm summers it can be a couple of meters. The thickness of permafrost varies over an even larger range, depending on its age and the bedrock. It's just a principle sketch, the only CC I could find. $\endgroup$
    – user2821
    Commented May 18, 2017 at 13:23
  • 2
    $\begingroup$ @Tbb Good. Could you put a source attribution immediately below each figure, just to avoid confusion? I assume it's from one of the linked articles but it's better to be clear. $\endgroup$
    – gerrit
    Commented May 18, 2017 at 13:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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