Partial derivative including a dependent parameter

68 Views Asked by At

This may be a simple partial derivative formula but I'm having a hard time to understand it. $$ \left( \frac{\partial }{\partial t} \right)_x f(t,\vec{r}(t,\vec{x})) = \left( \frac{\partial }{\partial t} \right)_r f(t,\vec{r}) + \dot{r} \cdot \left( \frac{\partial}{\partial r} \right)_t f(t,\vec{r}) $$

Since the function "f" explicitly depends on "t", isn't just the partial derivative by "t" enough?

I already know the basic concept of the chain rule, but this formula is tricky for me to understand. What kinds of steps are going on?

1

There are 1 best solutions below

0
On

The subscripting is used to distinguish the two $\tfrac{\partial f}{\partial t}$ notations by clearly indicating which 'independent variable' are being kept 'constant for purpose of partial differentiation'.

$\left(\dfrac{\partial f(t, r(x,t))}{\partial t}\right)_x$ is the partial derivative of the composite function $f(t,r(x,t))$ with respect to $t$ when $x$ is the independent variable. If we define $g(t,x)=f(t,r(x,t))$ then it is the differentiation of $g$ with respect to its first argument.

$\left(\dfrac{\partial f(t, r)}{\partial t}\right)_r$ is the partial derivative of the function $f(t,r)$ with respect to $t$ and 'independent variable' $r$ . So it is just the differentiation of $f$ with respect to its first argument.


So we often first define $g(t,x)= f(t,r(x,t))$ and present the chain rule as $$\dfrac{\partial g(t,x)}{\partial t}=\left[\dfrac{\partial f(t,y)}{\partial t}+\dfrac{\partial r(t,x)}{\partial t}\cdot\dfrac{\partial f(t,y)}{\partial y}\right]_{y=r(x,t)}$$

Which you will see most commonly shorthanded to: $$\dfrac{\partial g}{\partial t}=\dfrac{\partial f}{\partial t}+\dfrac{\partial r}{\partial t}\cdot\dfrac{\partial f}{\partial r}$$

Alternatively:

$$g^{(1,0)}(t,x) = f^{(1,0)}(t,r(x,t)) + r^{(0,1)}(x,t)\cdot f^{(0,1)}(t,r(x,t))$$