Tutorials › Calculus III › Computing Curl Flux on Parametrized Surfaces

Home  /  Tutorials  /  Stokes’ Theorem

Stokes’ Theorem · Tutorial 928 of 999

Computing Curl Flux on Parametrized Surfaces

Build the oriented area vector from a surface parametrization, then use it to compute curl flux accurately over the parameter domain.

Advanced 11 min read

What You'll Learn

  • Set up the curl-flux integral from a regular surface parametrization
  • Compute the two parameter tangent vectors and their cross product
  • Match the cross-product direction to the required surface orientation
  • Evaluate the curl field on the parametrized surface
  • Integrate the resulting scalar function over the parameter domain
  • Detect and correct sign errors caused by reversing the tangent-vector order

From a Parametrized Surface to Curl Flux

Stokes’ Theorem relates the circulation of a vector field around the boundary of an oriented surface to the flux of the field’s curl through that surface. When the surface is parametrized, the surface integral can be evaluated directly over a two-dimensional parameter domain.

The central calculation is to combine three ingredients: the curl evaluated at points on the surface, the two tangent vectors supplied by the parametrization, and the cross product that gives the oriented area element. Keeping these ingredients separate until the final dot product helps make both the geometry and the sign of the answer clear.

Definition: Let \(S\) be parametrized by \(\mathbf{r}(u,v)\), where \((u,v)\) ranges over a region \(D\) in the parameter plane. If \(\mathbf{r}_u\) and \(\mathbf{r}_v\) are linearly independent on the interior of \(D\), the parametrization is regular there. The oriented surface element determined by the order \(u,v\) is \(\mathbf{r}_u\times\mathbf{r}_v\,du\,dv\). Thus, for a vector field \(\mathbf{G}\), \[ \iint_S \mathbf{G}\cdot\mathbf{n}\,dS = \iint_D \mathbf{G}(\mathbf{r}(u,v))\cdot \bigl(\mathbf{r}_u\times\mathbf{r}_v\bigr)\,du\,dv. \]

For curl flux, take \(\mathbf{G}=\nabla\times\mathbf{F}\). The resulting formula is

$$ \iint_S(\nabla\times\mathbf{F})\cdot\mathbf{n}\,dS = \iint_D (\nabla\times\mathbf{F})(\mathbf{r}(u,v)) \cdot\bigl(\mathbf{r}_u\times\mathbf{r}_v\bigr)\,du\,dv. $$

This formula assumes a regular parametrization that covers the surface once, apart from possible overlaps along its boundary. If the surface is not regular at isolated points, the integral may still be evaluated by treating those points as limits, provided the integral exists. The order of the tangent vectors determines the orientation: switching them changes the sign of the cross product and reverses the oriented surface.

Why the Cross Product Gives the Surface Element

A small rectangle in the parameter domain, with side lengths \(du\) and \(dv\), maps under \(\mathbf{r}\) to a small patch on the surface. To first order, its two side vectors are \(\mathbf{r}_u\,du\) and \(\mathbf{r}_v\,dv\). The cross product of these vectors is

$$ (\mathbf{r}_u\,du)\times(\mathbf{r}_v\,dv) = (\mathbf{r}_u\times\mathbf{r}_v)\,du\,dv. $$

The magnitude of this cross product is the area of the approximating parallelogram, and its direction is perpendicular to that patch. Adding the field’s dot product with these oriented area vectors over small patches, then taking the limit, gives the surface integral formula. This explains why the formula uses the cross product itself—not just its magnitude—and why reversing its order reverses the flux.

In Cartesian coordinates, the curl used in the calculation is

$$ \nabla\times\mathbf{F} = \left( \frac{\partial F_3}{\partial y}-\frac{\partial F_2}{\partial z}, \frac{\partial F_1}{\partial z}-\frac{\partial F_3}{\partial x}, \frac{\partial F_2}{\partial x}-\frac{\partial F_1}{\partial y} \right), \qquad \mathbf{F}=(F_1,F_2,F_3). $$

For the Stokes’ Theorem interpretation, the field \(\mathbf{F}\) should have continuous first partial derivatives on an open region containing the oriented, piecewise-smooth surface and its boundary. The direction of \(\mathbf{n}\) must agree with the chosen surface orientation; the corresponding boundary orientation is the one specified by the right-hand rule.

A Reliable Calculation Sequence

For a direct computation, write down the parameter domain and the parametrization first. Then carry out the vector and integral steps in order.

1
Set the orientation and parameter domain.
Identify the ranges of \(u\) and \(v\), and determine which normal direction the problem requires.
2
Find the tangent vectors.
Differentiate \(\mathbf{r}(u,v)\) with respect to each parameter to obtain \(\mathbf{r}_u\) and \(\mathbf{r}_v\).
3
Form and check the oriented area vector.
Compute \(\mathbf{r}_u\times\mathbf{r}_v\), then check that it points in the required direction. Reverse its sign if the orientation requires it.
4
Compute and evaluate the curl.
Find \(\nabla\times\mathbf{F}\) in space, then substitute \(\mathbf{r}(u,v)\) into that vector field.
5
Dot and integrate.
Take the dot product with the oriented area vector and integrate the resulting scalar function over \(D\).

Worked Examples

Worked Example: Outward Curl Flux Through a Cylindrical Patch

Let \(\mathbf{F}(x,y,z)=(0,0,x)\). Compute the curl flux through the quarter-cylinder of radius \(2\), with \(0\leq u\leq\pi/2\) and \(0\leq v\leq1\), parametrized by \(\mathbf{r}(u,v)=(2\cos u,2\sin u,v)\). Use the outward orientation.

First compute the tangent vectors:

$$ \mathbf{r}_u=(-2\sin u,2\cos u,0), \qquad \mathbf{r}_v=(0,0,1). $$

Their cross product is

$$ \mathbf{r}_u\times\mathbf{r}_v = (2\cos u,2\sin u,0). $$

This points radially outward, as required. For \(\mathbf{F}=(F_1,F_2,F_3)=(0,0,x)\), the curl is

$$ \nabla\times\mathbf{F} = \left( \frac{\partial x}{\partial y}-\frac{\partial 0}{\partial z}, \frac{\partial 0}{\partial z}-\frac{\partial x}{\partial x}, \frac{\partial 0}{\partial x}-\frac{\partial 0}{\partial y} \right) =(0,-1,0). $$

The curl is constant, so evaluating it on the surface does not change it. The integrand is

$$ (0,-1,0)\cdot(2\cos u,2\sin u,0) = -2\sin u. $$

Therefore the flux is

$$ \int_0^1\int_0^{\pi/2}-2\sin u\,du\,dv = \int_0^1\left[2\cos u\right]_0^{\pi/2}dv = \int_0^1(-2)\,dv = -2. $$

As a check, the outward area vector has a positive \(y\)-component on this quarter-cylinder, while the curl points in the negative \(y\)-direction. Their dot product must therefore be negative, consistent with the computed flux.

Worked Example: Curl Flux Through a Helicoid Patch

Let \(\mathbf{F}(x,y,z)=(-y,x,0)\), and parametrize a helicoid patch by \(\mathbf{r}(u,v)=(u\cos v,u\sin v,v)\), for \(0\leq u\leq1\) and \(0\leq v\leq\pi/2\). Use the orientation given by \(\mathbf{r}_u\times\mathbf{r}_v\).

The tangent vectors are

$$ \mathbf{r}_u=(\cos v,\sin v,0), \qquad \mathbf{r}_v=(-u\sin v,u\cos v,1). $$

Thus,

$$ \mathbf{r}_u\times\mathbf{r}_v = (\sin v,-\cos v,u). $$

Now compute the curl:

$$ \nabla\times\mathbf{F} = \left( 0-0,\;0-0,\;\frac{\partial x}{\partial x} -\frac{\partial(-y)}{\partial y} \right) =(0,0,2). $$

The curl is constant, so its value at \(\mathbf{r}(u,v)\) is also \((0,0,2)\). The dot product with the oriented area vector is

$$ (0,0,2)\cdot(\sin v,-\cos v,u)=2u. $$

Integrating over the parameter rectangle gives

$$ \int_0^{\pi/2}\int_0^1 2u\,du\,dv = \int_0^{\pi/2}\left[u^2\right]_0^1\,dv = \int_0^{\pi/2}1\,dv = \frac{\pi}{2}. $$

The result is positive, consistent with the positive \(z\)-component \(u\) of the oriented area vector for \(u>0\), since the curl points in the positive \(z\)-direction.

Worked Example: Curl Flux Through a Spherical Patch

Let \(\mathbf{F}(x,y,z)=(-y/2,x/2,0)\). Consider the unit-sphere patch parametrized by \(\mathbf{r}(u,v)=(\cos v\cos u,\cos v\sin u,\sin v)\), where \(0\leq u\leq\pi/2\) and \(0\leq v\leq\pi/3\). Use the orientation given by \(\mathbf{r}_u\times\mathbf{r}_v\).

Differentiate with respect to each parameter:

$$ \mathbf{r}_u=(-\cos v\sin u,\cos v\cos u,0), \qquad \mathbf{r}_v=(-\sin v\cos u,-\sin v\sin u,\cos v). $$

The cross product is

$$ \mathbf{r}_u\times\mathbf{r}_v = (\cos^2v\cos u,\cos^2v\sin u,\cos v\sin v). $$

For the given field,

$$ \nabla\times\mathbf{F} = \left(0,0,\frac{\partial(x/2)}{\partial x} -\frac{\partial(-y/2)}{\partial y}\right) =(0,0,1). $$

Consequently, the parameter-domain integrand is

$$ (0,0,1)\cdot (\cos^2v\cos u,\cos^2v\sin u,\cos v\sin v) = \cos v\sin v. $$

The flux is

$$ \int_0^{\pi/2}\int_0^{\pi/3}\cos v\sin v\,dv\,du = \int_0^{\pi/2} \left[\frac{1}{2}\sin^2v\right]_0^{\pi/3}du = \int_0^{\pi/2}\frac{3}{8}\,du = \frac{3\pi}{16}. $$

As a geometric check, the cross product is \(\cos v\) times the outward unit radial vector \((\cos v\cos u,\cos v\sin u,\sin v)\). The curl’s dot product with that unit vector is \(\sin v\), so the flux density is \(\cos v\sin v\), exactly the integrand above. The computed value is therefore consistent with the outward orientation.

Orientation and a Common Sign Error

The cross product order is not interchangeable. In the cylindrical example, the outward area vector was \(\mathbf{r}_u\times\mathbf{r}_v=(2\cos u,2\sin u,0)\). If the order is mistakenly reversed, then

$$ \mathbf{r}_v\times\mathbf{r}_u = -(\mathbf{r}_u\times\mathbf{r}_v) = (-2\cos u,-2\sin u,0). $$

Using this reversed vector with the same curl gives the wrong integrand:

$$ (0,-1,0)\cdot(-2\cos u,-2\sin u,0) = 2\sin u, \qquad \int_0^1\int_0^{\pi/2}2\sin u\,du\,dv=2. $$

The value \(2\) is the flux for the inward orientation, not the requested outward orientation. The correction is to check the cross product against the required normal direction before integrating. A correct magnitude with the wrong direction still gives the wrong oriented flux.

Using the Integral in Stokes’ Theorem

If \(\mathbf{F}\) has continuous first partial derivatives on a region containing the oriented surface \(S\), Stokes’ Theorem equates the curl flux with the circulation around the boundary curve \(C\) oriented consistently with \(S\):

$$ \oint_C \mathbf{F}\cdot d\mathbf{r} = \iint_S(\nabla\times\mathbf{F})\cdot\mathbf{n}\,dS. $$

The parameter-domain calculation evaluates the right-hand side directly. It does not remove the need to check the orientation: reversing the surface orientation reverses the cross product and the flux sign, and it also reverses the compatible boundary direction. For a surface consisting of several smooth patches, the integral can be computed patch by patch, provided their orientations agree across the pieces.

A useful final audit is to ask whether the tangent vectors span the intended surface, whether their cross product points in the required direction, whether the curl has been evaluated on the surface, and whether the bounds cover the parameter domain exactly once. These checks address the most common sources of error before any integration is done.

Check Your Understanding

Use the parametrized-surface formula and explain the role of orientation in each response.

  1. For a parametrization \(\mathbf{r}(u,v)\), what vectors must be dotted with the curl to form the parameter-domain integrand?
  2. What happens to the oriented surface integral if the tangent-vector order in the cross product is reversed?
  3. For the cylinder in the first worked example, why must the outward-oriented integrand be negative?
  4. What regularity condition on \(\mathbf{F}\) is sufficient for the standard Stokes’ Theorem relation?
  5. In the spherical-patch example, how does the cross-product magnitude help check the computed flux density?

Previous

Next

Back to Tutorials