3

I've had a tough time trying to interpret line integrals such as this one, which may be found on Wikipedia's Line Integral article:

$$\int_C f(\mathbf r) \,ds$$

More specifically, I'm trying to understand what $\mathbf r$ is in this integral. I understand that:

  • $C$ is a curve and $ds$ is the differential arc length function of $C$, which depends on the parameterization variable.
  • $f$ is a scalar field, which takes in a vector and returns a real number.

But, what is $\mathbf r$? Wikipedia says that $\mathbf r$ is a parameterization of $C$. Does this mean $\mathbf r$ is a vector-valued function and not a vector? Wikipedia certainly treats it like a function of the parameterization variable. If so, how is the expression $f(\mathbf r)$ valid if $f$ takes in a vector? This is like trying to fit a square peg into a round hole.

Then, I see this equivalent expression, which makes sense because we are getting the value of $\mathbf r$ at $t$:

$$\int_a^b f(\mathbf r(t))|\mathbf r'(t)| \,dt$$

However, this makes matters worse because now I'm more convinced that $\mathbf r$ is a function. How is the expression $f(\mathbf r)$ in the first integral possible? Is this just another way to write a composite function like $f \circ \mathbf r$?

As I gain more knowledge in mathematics, I'm discovering that I appear to be a perfectionist regarding syntax, and things like this make me go crazy. I have a background in object oriented programming, so when I see notation like this and fail to find logical consistency, I freak out a little bit. This is my first post here so I'm excited to see what people will say.

3 Answers3

3

If you want to be strict with the notation, then you simply write $\int_Cf\,ds$, and state appropriate regularity conditions on $f$ and $C$, and state that $ds$ denotes the arclength measure along $C$. Note that the symbol $dl$ might often be used in place of $ds$. Then, given a parametrization $\mathbf{r}:[a,b]\to\Bbb{R}^n$ with image $C$, one defines \begin{align} \int_Cf\,ds&:=\int_a^bf(\mathbf{r}(t))\,\|\mathbf{r}'(t)\|\,dt. \end{align} The LHS is just a symbol, and the RHS is the meaning of the LHS.

As for why we write $f(\mathbf{r})$, well, this sort of thing is the usual abuse of language where people refer to functions not by the name of the function, but rather by the function values (and here often one overloads the symbol $\mathbf{r}$ to mean an element of $\Bbb{R}^n$ and also a mapping $[a,b]\to\Bbb{R}^n$). For instance, we might often speak of "the function $f(x)=x^2$", when in reality, what we should say is "the function $f:A\subset\Bbb{R}\to\Bbb{R}$, defined on some set $A$, having the 'rule' $f(x)=x^2$".

peek-a-boo
  • 55,725
  • 2
  • 45
  • 89
  • So in other words, it's another example of functional notation abuse. I've seen plenty of this in high school and it's been throwing me off as I'm trying to re-learn calculus. – Alejandro Miller Aug 06 '22 at 07:00
  • 1
    @AlejandroMiller yes, it is an abuse (kind of like having an unclosed bracket), but it's a convenient one at times (once you understand the abuse of course). For example, suppose you want to talk about the function $f:\Bbb{R}^2\to\Bbb{R}$, $f(\xi,\eta)=\xi^2+\eta$, and you want to consider the integral $\int_Cf,ds$. Clearly, this is a lot to write down, compared to "the arclength integral $\int_C(\xi^2+\eta),ds$ in the plane"; by talking about the function values you can by-pass the step of explicitly writing out the definition of the function and so on. – peek-a-boo Aug 06 '22 at 07:06
  • 1
    also, it's like how in everyday language, we don't always speak with fully grammatically correct sentences/ we text with informal language and punctuation, but still the message gets across. – peek-a-boo Aug 06 '22 at 07:09
0

A parameterization of a curve $C$ $\mathbf r:[a,b]\to\Bbb R^n$, is a vector function, for which a $t\in [a,b]$ gives the vector $\mathbf r(t)$ of a point of the curve and conversely for a point of the curve there is a parameter $t\in [a,b]$ which gives the position vector $\mathbf r(t)$ of the point.

SK_
  • 575
-1

Is $r$ a vector-valued function? Yes, it has a one-dimensional input and a multi-dimensional output.

An example is in order.

Suppose, the contour we want to evaluate is a circle. Then we can parameterize the circle as:

$x = \cos t\\ y = \sin t$

or $r(t) = (\cos t, \sin t)$

$f(r) = f(\cos t, \sin t)$

$r'(t) = (-\sin t, \cos t)$
$\|r'(t)\| = (-\sin t)^2 + (\cos t)^2 = 1$

Which means $ds = dt$

If our contour was a square with coordinates $(0,0),(1,0),(1,1),(0,1)$ we can parameterize each line segment as $r(t)=(t,0), r(t)=(1,t),r=(1-t, 1),r=(0,1-t)$ respectively.

I hope this helps.

user317176
  • 11,017
  • Would you care to explain your downvote? I am happy to ammend my answer. – user317176 Aug 06 '22 at 07:53
  • 2
    I didn't downvote, but I didn't mark it accepted either because it doesn't address my problem with the expression $f(\mathbf{r})$. $f$ is a function of a vector, and $\mathbf{r}$ is a function, so I asked why $f(\mathbf{r})$ is commonly written. I do want to add that $ds$ in my context denotes the magnitude of the curve differential and not $dt$. – Alejandro Miller Aug 06 '22 at 08:34
  • $f(u)$ where $u$ is a variable in relation to $x$ is standard notation in calculus. $u$ might be a function of $x$ but it might not meet the standards of a function. – user317176 Aug 06 '22 at 09:10