The left part of Navier-Stokes equation is: $\dfrac{D\vec{v}}{D t}= \dfrac{\partial\vec{v}}{\partial t}+ \vec{v} \cdot\nabla \vec{v}$

Let's take $\vec{v}$ as a two dimentional vector: $(u,v)$. Then: $\dfrac{\partial\vec{v}}{\partial t}$ is $(\dfrac{\partial u}{\partial t}, \dfrac{\partial v}{\partial t})$, which is a vector. However, $\vec{v} \cdot\nabla \vec{v}$ will be $(u,v)\cdot(\dfrac{\partial u}{\partial x},\dfrac{\partial v}{\partial y}) = u\dfrac{\partial u}{\partial x}+v\dfrac{\partial v}{\partial y}$, which is a scalar. How a "vector" can be summed with a "scalar"?

Of course my thought is wrong somewhere. Please help me.

  • $\begingroup$ directional derivative $\endgroup$ – timur Dec 10 '18 at 7:40
  • 3
    $\begingroup$ $\nabla\vec{v}$ is not $(\dfrac{\partial u}{\partial x},\dfrac{\partial v}{\partial y})$, it is the rank-2 tensor with entries $\dfrac{\partial u}{\partial x},\dfrac{\partial u}{\partial y},\dfrac{\partial v}{\partial x},\dfrac{\partial v}{\partial y}$ $\endgroup$ – user10354138 Dec 10 '18 at 7:46
  • 1
    $\begingroup$ @user10354138 -- Then the dot must be interpreted as tensor contraction. I guess that produces the same results, but it's probably not what was meant by the equation. $\endgroup$ – mr_e_man Dec 10 '18 at 7:51
  • 2
    $\begingroup$ @mr_e_man Yes it is what is meant by the equation, in the definition of $\frac{D}{Dt}=\frac{\partial}{\partial t}+\mathbf{v}\cdot\nabla$ as the material derivative. $\endgroup$ – user10354138 Dec 10 '18 at 8:16
  • $\begingroup$ @user10354138 -- I mean it's probably supposed to be $(v\cdot\nabla)v$ instead of $v\cdot(\nabla v)$. $\endgroup$ – mr_e_man Dec 10 '18 at 19:18

As stated in the comments, this is the material derivative. If $\mathbf{v} = (v_1,v_2)$ then

$$ (\mathbf{v}\cdot \nabla) \mathbf{v} = (\mathbf{v}\cdot \nabla v_1, \mathbf{v}\cdot \nabla v_2) = \left(v_1 \frac{\partial v_1}{\partial x} + v_2\frac{\partial v_1}{\partial y}, v_1\frac{\partial v_2}{\partial x} + v_2\frac{\partial v_2}{\partial y}\right) $$

You can also intepret it as a matrix product

$$ \mathbf{v}\cdot \nabla \mathbf{v} = \begin{bmatrix} v_1 & v_2 \end{bmatrix}\begin{bmatrix} \dfrac{\partial v_1}{\partial x} & \dfrac{\partial v_2}{\partial x} \\ \dfrac{\partial v_1}{\partial y} & \dfrac{\partial v_2}{\partial y} \end{bmatrix} $$

where $\nabla \mathbf{v}$ represents a rank-2 tensor


You're right in that scalars and vectors don't add, but have unfortunately - and understandably - got muddled up with how the operators are put together. If you put brackets around (v dot del) then it will make sense. This will be a scalar operator just like partial_t.

  • $\begingroup$ Scalars and vectors can be added in geometric algebra, though that's not happening here. $\endgroup$ – mr_e_man Dec 10 '18 at 7:46
  • $\begingroup$ That's interesting. It occured to me after I said it that there was likely some abstract spaces where mixing could occur. $\endgroup$ – Paul Childs Dec 10 '18 at 7:51

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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