Talk:Material derivative

Thank you
Probably not the correct place for this, but the swimmer description of this derivative wrt lake temperature was wonderful and *really* clarified this for me. Thank you! 133.87.57.32 (talk) 13:06, 23 June 2012 (UTC)

Notation consistancy
Am I confused, or does this article change notation between the introduction and the proof? Is D/Dt the same operator as d/dt? Tom Duff 19:19, 30 January 2006 (UTC)
 * Yes I have fixed it. d/dt is used in many texts but I think D/Dt makes it clear that we are discussing a property in a vector field.  Rex the first  talk 01:00, 27 February 2007 (UTC)

Also, what does the hat on the B in the "proof" section refer to? A hat can have many meanings, there should be a line saying "where B-hat indicates that [whatever it indicates]" 140.184.21.115 13:35, 19 September 2007 (UTC)

Two of a kind
Looks to me like this article ought to be rolled in with substantive derivative. Linuxlad 14:19, 22 June 2006 (UTC)

Identity
The identity given for taking the material derivative of an integral is the Reynolds Transport Theorem, though written in a form that is dissimilar to the one listed in the article concerning that theorem. This is also a varient of the Liebnitz Rule.

Parentheses
Is there any special reason for the parentheses used on the RHS in the definitions? —DIV (128.250.204.118 09:04, 6 April 2007 (UTC))

Same as "total" derivative
Assume that


 * $$\phi = \phi (x,y,z,t)$$

By the chain rule


 * $$d\phi = \frac{\partial \phi}{\partial x} dx + \frac{\partial \phi}{\partial y} dy + \frac{\partial \phi}{\partial z} dz + \frac{\partial \phi}{\partial t} dt $$

dividing both side by $$dt$$, we get


 * $$\frac{d\phi}{dt} = \frac{\partial \phi}{\partial x} \frac{dx}{dt} + \frac{\partial \phi}{\partial y} \frac{dy}{dt} + \frac{\partial \phi}{\partial z} \frac{dz}{dt} + \frac{\partial \phi}{\partial t}$$

since $$\frac{dx}{dt}=u$$, $$\frac{dy}{dt}=v$$ and $$\frac{dz}{dt}=w$$, the above equation becomes


 * $$\frac{d\phi}{dt} = \frac{\partial \phi}{\partial t} + u \frac{\partial \phi}{\partial x} + v \frac{\partial \phi}{\partial y} + w \frac{\partial \phi}{\partial z} = \frac{\partial \phi}{\partial t} + (\mathbf{u}\cdot\nabla)\phi$$

Hence, we see that $$\frac{d\phi}{dt}$$ and $$\frac{D\phi}{Dt}$$ are one and the same. Therefore, the substantial derivative is nothing more than a total derivative with respect to time. The only advantage of the substantial derivative notation is that it higlights more of the physical significance (time rate of change following a moving fluid element).

I think that the terminology "substantial derivative" and "total derivative" are unnecessarilly confusing (As far as I know, this terminology is mainly prevalent in fluid dynamics) The wikipedia article should explain that they are different way to express the same thing.

199.212.17.130 13:47, 31 August 2007 (UTC)


 * In transport phenomena the partial derivative and the material derivative (the latter also called the substantial derivative) are both special cases of the total derivative. There are three cases to consider with respect to the terms $$dx/dt$$, $$dy/dt$$, and $$dz/dt$$ which describe the motion of the observer and which appear in the definition of the total derivative:


 * In general, the motion of the observer may be an arbitrary function of time, defined by $$d\boldsymbol{r}(t)/dt=[dx(t)/dt,dy(t)/dt,dz(t)/dt]$$. Note that this motion of the observer $$d\boldsymbol{r}(t)/dt$$ is independent of the motion of the fluid and does not necessarily establish an inertial frame of reference.  In this general case, the rate of change in $$\phi$$ as observed by the arbitrarily moving observer is given by the total derivative $$d\phi/dt = \partial\phi/\partial t + (\partial\phi/\partial x)(dx/dt) + (\partial\phi/\partial y)(dy/dt) + (\partial\phi/\partial z)(dz/dt)$$, which may be interpreted as the stationary rate of change (given by the partial derivative; see the next case below) plus corrections due to the motion of the observer.


 * If the observer is stationary, $$dx/dt=0$$, $$dy/dt=0$$, and $$dz/dt=0$$. The rate of change in $$\phi$$ as observed by the stationary observer is then given simply by the partial derivative $$\partial\phi/\partial t$$.


 * If the motion of the observer follows the motion of the fluid, then $$dx/dt=v_x$$, $$dy/dt=v_y$$, and $$dz/dt=v_z$$, where the vector $$\boldsymbol{v}(t,x,y,z)=[v_x(t,x,y,z),v_y(t,x,y,z),v_z(t,x,y,z)]$$ describes the motion of the fluid. The rate of change in $$\phi$$ as observed by the observer drifting along with the moving fluid is given by the substantial derivative $$D\phi/Dt = \partial\phi/\partial t + v_x(\partial\phi/\partial x) + v_y(\partial\phi/\partial y) + v_z(\partial\phi/\partial z)$$.


 * It appears that the commenter above mistook $$dx/dt$$, $$dy/dt$$, and $$dz/dt$$ to mean always the velocity of the fluid, and arrived at the incorrect conclusion that the total derivative and the material (substantial) derivative are the same.


 * Ydw (talk) 06:56, 24 November 2010 (UTC)

What has been proven?
The first section of this article claims to define the convective derivative. The next section offers a proof. How can a definition be proven? I am confused. Is the proof intended to show that the convective derivative is the partial derivative with respect to time in a frame that moves with material particles? That requires some reasoning, I think, not just direct application of the chain rule.

155.37.79.216 14:27, 7 September 2007 (UTC)

A subtlety

 * There is a subtlety that the proof has missed. Consider the point $$\mathbf{r}=\mathbf{r}(t)$$ and $$\phi(\mathbf{x},t)$$  for $$\mathbf{x}$$ a position coordinate independent of $$t$$, then the total derivative $$d\phi(\mathbf{r}(t), t) / dt$$ may be found


 * $$\frac{d\phi(\mathbf{r}(t), t)}{dt} = \left. \frac{\partial \phi(\mathbf{x},t)}{\partial t} \right|_{\mathbf{x}=\mathbf{r}} +  \frac{d \mathbf{r}}{dt}  \cdot  \left[\nabla \phi(\mathbf{x},t) \right]_{\mathbf{x}=\mathbf{r}} $$

If and only if $$\mathbf{r}$$ is a Lagrangian point, so $$d \mathbf{r}(t) / dt = \mathbf{u}$$, does the total time derivative equal the convective derivative.

139.80.48.19 (talk) 22:29, 19 March 2008 (UTC)

Rename to Material derivative
I propose to rename this article to Material derivative, because:
 * 1) Convective derivative is ambiguous, since it is also used to denote only the spatial part, v•&nabla;.
 * 2) Material derivative is more commonly used. A Google Books search for the exact phrase gives:
 * {| class="wikitable" style="text-align:left"

! phrase !! hits Crowsnest (talk) 22:26, 23 June 2008 (UTC)
 * material derivative
 * 837
 * substantial derivative
 * 693
 * convective derivative
 * 650 (incl. uses for only the spatial part)
 * Lagrangian derivative
 * 490
 * substantive derivative
 * 407
 * derivative following the motion
 * 321
 * particle derivative
 * 177
 * hydrodynamic derivative
 * 162
 * Stokes derivative
 * 106
 * advective derivative
 * 77
 * }
 * hydrodynamic derivative
 * 162
 * Stokes derivative
 * 106
 * advective derivative
 * 77
 * }
 * advective derivative
 * 77
 * }

TOC right
According to Help:Section, the TOC is floated "when it is beneficial to the layout of the article, or when the default TOC gets in the way of other elements." In what way is the template here beneficial to the layout of the article? More than 99% of our mathematics articles have never used this template; many of them have identical layouts of a lede section, a TOC, and then several lower sections, without images or other floating elements. &mdash; Carl (CBM · talk) 18:13, 21 May 2010 (UTC)

I agree. I've read a lot of science, engineering and maths articles, and never seen this layout before. I didn't even see the t.o.c. until reading this talk page entry. Michael Hodgson (talk) 07:30, 25 October 2021 (UTC)


 * The TOC to the right is benificial for this article, to my opinion, because it directly reveals the alternate names (most of them redirecting to here) given in the section "Names", which otherwise are separated by the TOC. So the TOC on the right is used to minimise in the lay-out the distance between the lead and the "Names" section. This article is special in the sense that a lot of alternative names are used for the same operator, and the TOC on the right helps to directly have an overview (on the same screen) without need of scrolling. -- Crowsnest (talk) 17:50, 11 June 2010 (UTC)

Formula wrong for vector fields?
I want to ask if the formula for the material derivative for vector fields is wrong: it does not seem to define an objective quantity. I would have strongly expected an additional corotational term $+\vec{\omega} \times \vec{A}$, where $\vec{\omega}$ is the local vorticity of the fluid, because a rotating flow should rotate any local vectorial quantity $\vec{A}$ associated with the material. In the literature, this material derivative for vectorial quantities (including the additional corrotation term) is sometimes referred to as Jaumann corotational derivative if I understand correctly.

I would like to change the main article, but want to consult with the community first. Benjamin.friedrich (talk) — Preceding undated comment added 09:09, 16 January 2021 (UTC)


 * Dear Benjamin.friedrich, thanks for pointing this out. Please go ahead to improve the article. I did not look deeply into it, but a quick search on material derivative for vector fields seems to support your assessment. Please add some references endorsing the changed or new material, see WP:REF. -- Crowsnest (talk) 20:09, 24 January 2021 (UTC)
 * An important point regarding this co-rotational term seems to be whether the observation is made in a rotating frame of reference or not. -- Crowsnest (talk) 20:36, 24 January 2021 (UTC)


 * See Upper-convected time derivative for the derivative in fluids with rotation. That article says that this is identical to the Lie derivative, which would then tell you about the general case of coordinate systems with non-trivial metric tensors (e.g. hydrodynamics near a black hole). Anyway, this article should clarify such questions/issues. 67.198.37.16 (talk) 20:53, 21 March 2024 (UTC)