32
$\begingroup$

After reading Wikipedia articles on advection and convection, I still cannot determine whether there is a consensus on a difference between these two terms.

Sometimes, the term convection seems to include advection and diffusion, sometimes not. After all, equations exhibit two different terms for advection and diffusion, and we don't call the sum of them a convective term. When we talk about a convective term, it is indeed the advective term!

It seems:

  • transport of pollutants in a river by bulk water flow downstream relates to advection

  • boiling water for pasta relates to convection

Maybe, a conserved property such as energy can be advected, but we only use convection for transport of "ensembles of molecules".

So, is there a difference? Could we figure it out with examples?

$\endgroup$
2

3 Answers 3

21
$\begingroup$

Convection is the movement of a fluid, typically in response to heat.

Advection is the movement of some material dissolved or suspended in the fluid.

So if you have pure water and you heat it you will get convection of the water. You can't have advection because there is nothing dissolved or suspended in the fluid to advect.

If you have silt suspended in the water and heat it then you will get convection of the water and advection of the silt.

If you have silt suspended in the water and the water is just flowing in a river you will get advection of the silt, but you would not normally describe the water movement as convection.

$\endgroup$
4
  • 12
    $\begingroup$ It can be pointed out that the mechanism of convection is the advection of the velocity field along with the fluid. This is proper usage of the term, even when there isn't a different material solute, just a velocity field. $\endgroup$
    – Ron Maimon
    Commented Apr 27, 2012 at 18:25
  • 4
    $\begingroup$ Note that the "material" being advected does not necessarily have to be a physical material (such as silt) but can in fact be something more abstract such as a property of the fluid; for example, it is possible to advect vorticity and/or density in an isothermal fluid (i.e without any convection). $\endgroup$
    – Biggsy
    Commented Dec 10, 2018 at 13:56
  • $\begingroup$ @RonMaimon What do you mean by "the advection of the velocity field"? Yes, the advection of the velocity field is one type of convection, more specifically, it is convection of the velocity field, that is, how the value of the velocity field in one part of the fluid moves or spreads to other parts of the fluid. This however doesn't say anything about the convection of other properties, such as density or temperature. $\endgroup$ Commented Sep 21, 2022 at 11:47
  • $\begingroup$ @RonMaimon Besides, advection isn't the only thing that causes convection; diffusion also causes convection. $\endgroup$ Commented Sep 21, 2022 at 11:48
5
$\begingroup$

The terms advection and convection have different meanings depending on in which context they are used.

Fluid dynamics

In fluid dynamics, advection is a type of convection, with the other type of convection being diffusion. Both advection and diffusion act to move around various intensive properties of a fluid.

In advection, properties are transported by bulk motion of the fluid. In other words, as soon as the velocity field of the fluid is non-zero, you have convection of the various intensive properties of the fluid, because they are moving along the velocity field.

For example, if you pour some dye into a river so that a colored blob is formed in the river, that blob is going to follow along with the river; the movement of the blob along the river is a form of advection.

In diffusion, things are spread out from points of high concentration or high property values to points of low concentration or low property values, as a result of particle random walk, which causes nearby regions of the fluid to blend together and has the effect of "blurring" the the various properties of the fluid. Mathematically, this means that high frequency components of the properties tend to decay over time, with components with higher frequencies decaying faster than components with lower frequencies.

When modelling turbulent flow, a common practical strategy is to ignore the small-scale vortices, or eddies, and to instead model their large-scale effects on various properties as a type of effective diffusion.

For example, thermal conduction is a form of diffusion, where the diffused property is the temperature. The effect of viscosity is another example of diffusion, where the diffused property is the velocity. For turbulent flow, the mixing of the velocity field caused by the eddies is sometimes modeled as an effective diffusion called the "eddy viscosity."

And to take an example that uses dye to explain diffusion, if you add a drop of dye carefully in a glass of water, you may form a very small blob of dye on the surface of the water, with a clear boundary towards the water. As time progresses, though, even if the water seems to perfectly still, that dye is going bleed out into the water, and the crisp boundary between the dye and the water is going to become blurry end eventually disapear. This is because even if the water is still on the macroscopic level, on the microscopic level the molecules move around and mix with each other, and dye molecules will become mixed with nearby water molecules. This movement of the particles on the microscopic level causes a blurring effect on the macroscopic level and is a form of diffusion.

Meterlogy

In meterology, as pointed out in Kobus' answer, advection means horizontal winds, and convection means vertical winds. Convection is typically caused by a steep lapse rate, which gives rise to a Rayleigh–Taylor instability in the atmosphere.

$\endgroup$
-4
$\begingroup$

Advection is the horizontal movement of air Convection is the vertical movement of air

$\endgroup$
2
  • $\begingroup$ Interesting approach. But what happens if we do not consider air but cells for instance? $\endgroup$
    – Wok
    Commented Jun 11, 2014 at 12:40
  • $\begingroup$ Need to add that: There are also vertical advection phenomenon of air too. IMHO that in this paper written in 1981, the author developed the physical model to depict the vertical advection and it continued to use in today's model. $\endgroup$ Commented Mar 27, 2016 at 14:43

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