A common layman explanation for why does it get colder to higher elevations (considering only the troposphere here) qualitatively boils down to The Sun heats the Earth's surface and the Earth's surface heats the atmosphere. I remember that, every time I heard or read this explanation as a child, I thought: then why is the Tibetan plateau still colder than lowlands at the same latitude? After all, the solar energy reaching the surface in Tibet is no less. Clearly there is more to it.

Considering a remark from a quantum physics lecturer that you only really understand something after you've taught the subject twice, I'm phrasing my question as such: How does one explain, in layman terms, why the Tibetan plateau is colder than lowlands at similar latitudes?


3 Answers 3


The answer is because the Earth is not a static system.

Due to the ideal gas law, air cools as it rises. This is referred to as the dry adiabatic lapse rate. However, you are curious why every location on earth is not the same temperature at the same latitude. We know this is not true. But why is not true? Weather.

The earth, as with most natural phenomenon, attempts to achieve static equilibrium. The differences in incoming and outgoing radiation leave the earth with net energy reserves that need to be balanced out. Weather is the mechanism that the Earth uses to attempt to achieve balance.

Okay, so we know that weather is disrupting static equilibrium on earth. But how does that work? Wind. Or better yet, advection. Advection is simply the transport of an atmospheric parameter (such as moisture, temperature, or rotation) from one location to another. So, wind can 'transport' temperature, moisture, and rotation.

You don't necessarily need a storm to advect any of these parameters. In fact, they are always in motion. Cold air is always moving away from the poles.

Still, that doesn't quite answer it, does it? Well, imagine you are hundreds or thousands of miles away from Tibet. You aren't in a mountainous region anymore but instead a flatlands at sea level. The environmental lapse rate is 6.5°C/1000m above this location (because we are not dealing with an air parcel). Tibet is 4500m - 8850m high. That would be around approximately 32.5°C of cooling at 5000m. The air that advects into Tibet from elsewhere is on average of 32.5-50°C cooler than at sea level.

So, it isn't so much that Tibet is receiving less radiation than other points at the same latitude (its not). Its more that, the air at other locations at the elevation of Tibet is much cooler than the air at sea level. Remember, we measure air temperature!

  • 1
    $\begingroup$ Right. The horizontal temperature gradient is instable. But a patch of sand in the sun in Tibet would probably still get really hot, I suppose. $\endgroup$
    – gerrit
    Commented Apr 16, 2014 at 3:19
  • 3
    $\begingroup$ A surface would struggle to store substantial amount of heat at that altitude. There are too many sources of heat loss surrounding any point that attempted to warm. In any direction above the surface, there would be cold, dry air sapping heat from said surface. In any direction to the sides or below an insolated surface, you would have colder earth also sapping any heat. Any kind of wind would only further accelerate the heat drain. $\endgroup$
    – DrewP84
    Commented Apr 16, 2014 at 3:39
  • $\begingroup$ Tibetan Soil Temperature at 4450m near Nagqu $\endgroup$
    – DrewP84
    Commented Apr 16, 2014 at 3:51
  • $\begingroup$ That soil temperature graph is interesting, where is it from, what is the context? $\endgroup$
    – gerrit
    Commented Apr 16, 2014 at 17:10
  • $\begingroup$ I merely wished to illustrate soil temperatures at 4500m in Tibet throughout a typical year. The soil is sub-freezing nearly half of the year, but does warm with longer-days. So, you are right that it still gets relatively 'hot'. Tibetan soil and moisture study PDF $\endgroup$
    – DrewP84
    Commented Apr 17, 2014 at 0:45

You are correct that solar input on the Tibetan plateau will be the same as a location at sea level at the same latitude. You are also correct that the Sun heats the Earths surface and that in turn heats the atmosphere. Now for the rest of the details.

  • Albedo

    Albedo is a measure of "whiteness" and gives us an idea of how solar irradiance interacts with the Earth. You probably know that dark colored things tend to get really hot when left in the sun, while light colored objects do not get as hot. This is because when a photon from the Sun hits the Earth, it is either absorbed or scattered. Without getting into the dynamics of radiation, I'll just leave it that dark colors tend to absorb and light colors tend toward scattering. Now we translate this to albedo -- a high value "white" means solar input is being reflected back into the atmosphere; a low value means solar input is being absorbed. A forest or ocean will have low albedo. Snowpack and ice will have high albedo.

    However, the albedo in Tibet on average is only slightly higher than average (~0.35 as cited in the comments below) and so this effect is limited with only a slight decrease in absorbed solar radiation than average.

  • Absorption

    The solar input that is absorbed in Tibet will excite the molecules in the ground and they will respond by warming up (and emitting longwave radiation). The emitted radiation upward does nothing to the air temperature, only the greenhouse gases are good at absorbing these wavelengths, regular dry air is not. It is notable that at these altitudes will also be less water vapor, which is probably the best greenhouse gas. If radiation isn't going to warm the air, that leaves conduction and convection.

  • Conduction

    The Earth heats the atmosphere via conduction with the lowest level of molecules in the atmosphere. For conduction to happen, molecules need to touch. At the lower pressures at high altitude, there are less molecules of air and conduction will be less efficient than at lower altitudes.

  • Atmosphere

    Air getting up to the Tibetan Plateau will have blown in, already at high altitude, or it will be blown upslope from lower elevations. In either case, this air will tend to be cold and dry. Upslope flow will tend to cool around 6C/km until it dries out (rain, clouds) and then cool at 10C/km. For the air already at altitude, the average elevation of Tibet is around 580 hPa and the standard temperature at that pressure is -14 C. Of these two effects, cool air advection of air already at this altitude dominates over upslope adiabatic flow . This air at altitude tends to move faster than at the surface and you will have a constant flow of new incoming, cold air blowing over the surface of Tibet.

Putting it all together:

The slightly higher albedo is our first suspect as to why Tibet is colder than lower elevations at the same latitude -- more of the sunlight is reflected to space and does not contribute to warming the ground. The next suspect is the less efficient heating of the atmosphere by the surface. This is followed up by the atmosphere that is already much colder than the surface at sea level (at similar latitudes). Take a much colder airmass than sea level, heat it up less than at sea level and you end up colder than you would be at sea level.

  • $\begingroup$ Some thoughts; according to Peixoto and Oort, Physics of Climate, Fig 6.10a, albedo is somewhat but not that much higher in Tibet (Sahara 0.3, Tibet 0.35, ice caps 0.8). As for the less molecules of air; yes, but it takes proportionally less energy to heat less molecules. I suspect the key issues are the wind and the lack of water vapour (which means even a wind-sheltered valley won't heat up much). $\endgroup$
    – gerrit
    Commented Apr 16, 2014 at 18:07
  • $\begingroup$ This is a clean, easy to follow explanation. Nice! I think it'd be the perfect answer if it emphasized that the bulk of the cooling is due to cool air advection versus adiabatic cooling (think of the volume of cool air transported by each), though adiabatic cooling is certainly relevant. $\endgroup$
    – DrewP84
    Commented Apr 18, 2014 at 19:46
  • $\begingroup$ @gerrit is correct, the albedo isn't as important up there as you might think. The plateau is rain (or snow) shadowed by the Himalayas for the most part. Satellite imagery of Tibet. $\endgroup$
    – DrewP84
    Commented Apr 18, 2014 at 19:53
  • 1
    $\begingroup$ @DrewP84 I made an edit to de-emphasize the albedo effect and replace them emphasis on cool air advection. $\endgroup$
    – casey
    Commented Apr 18, 2014 at 20:17

Temperature is a measure of kinetic energy which, in the atmosphere, is due to the number of molecules and their speed. There is less atmosphere at higher altitudes (e.g. lower pressure, less molecules), so a measure of temperature will be lower there because less molecules are moving around. That is the fundamental/basic reason that temperature is lower at higher altitudes in the troposphere. Furthermore, since there are a smaller number of greenhouse gas molecules trapping energy, more heat is lost to space at higher altitudes.

  • $\begingroup$ Do you have references for that? The other answers seem to give a different reason for the temperature difference. $\endgroup$
    – Gimelist
    Commented Nov 1, 2014 at 14:08
  • $\begingroup$ Temperature is the average kinetic energy of molecules. So less molecules does not equate to less temperature. In fact for the same amount of energy input, the result of having fewer molecules, with all other variables held the same, would be greater temperatures... as seen in the ideal gas law where n (the number of moles) and T (the temperature) are indirectly related. But pressure and volume both change markedly in the atmosphere as well. $\endgroup$ Commented May 8, 2018 at 13:57

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.