Start with the Quantity You Need
A vector field can be integrated in different ways, and the right method depends first on what the problem asks for. Flux measures the component of a vector field passing through a surface. Circulation measures the component of a vector field along a curve. These quantities use different integrals, so one theorem cannot be substituted for another just because the same vector field appears.
Direct integration is available when the curve or surface and its orientation can be described explicitly. Stokes’ theorem relates circulation around the boundary of an oriented surface to the flux of the curl through that surface. The divergence theorem relates outward flux through a closed surface to a volume integral of the divergence. Choosing well means matching the requested quantity and the boundary type to the theorem’s hypotheses, then checking which resulting integrand is simplest.
Three Methods, Three Roles
Direct integration evaluates the integral on the curve or surface that appears in the question. For a parameterized surface \(\mathbf{r}(u,v)\), with orientation chosen so that \(\mathbf{r}_u\times\mathbf{r}_v\) points in the desired normal direction, flux can be computed as an integral of \(\mathbf{F}(\mathbf{r}(u,v))\cdot(\mathbf{r}_u\times\mathbf{r}_v)\) over the parameter region. For a parameterized curve \(\mathbf{r}(t)\), circulation is the integral of \(\mathbf{F}(\mathbf{r}(t))\cdot\mathbf{r}'(t)\).
Stokes’ theorem is useful when the requested quantity is circulation around a boundary curve and the curl is easier to integrate over a convenient spanning surface than the original field is to integrate around the curve. It does not turn an arbitrary flux integral of \(\mathbf{F}\) into a circulation integral. Its surface integrand is \(\nabla\times\mathbf{F}\), not \(\mathbf{F}\).
The divergence theorem is a method for finding total outward flux through a closed surface. It replaces a surface integral by a volume integral; it does not directly evaluate flux through an open surface or circulation around a curve. If a theorem’s conditions are not met, direct integration may still be available, but the theorem cannot simply be applied.
A Practical Choice
Before calculating, identify the requested integral and inspect its boundary. Then compare the work required by direct integration with the work required by the applicable theorem. A useful decision guide is:
| Requested quantity and boundary | Method to consider | What to compare |
|---|---|---|
| Flux through an open surface | Direct surface integration | Whether a parameterization and normal make the integrand manageable |
| Circulation around a curve that bounds an oriented surface | Direct line integration or Stokes’ theorem | Whether the line integral or the curl-flux integral is simpler |
| Outward flux through a closed surface | Direct surface integration or the divergence theorem | Whether the face integrals or the divergence over the solid is simpler |
“Consider” is important: a theorem’s availability does not guarantee that it saves work. For example, a surface may have many pieces that are awkward to parameterize, while the divergence is a simple constant. Conversely, a volume integral may be unnecessarily complicated when the surface flux is immediate. The goal is not to use a theorem at every opportunity; it is to choose a valid calculation whose geometry and integrand suit the problem.
Worked Example: Direct Flux Through an Open Surface
Worked Example: Direct Flux Through an Open Surface
Find the upward flux of \(\mathbf{F}(x,y,z)=(0,0,1)\) through the paraboloid \(z=x^2+y^2\) above the unit disk \(x^2+y^2\leq 1\).
This is flux through an open surface, so the divergence theorem does not apply to the given surface. Parameterize it by \(\mathbf{r}(x,y)=(x,y,x^2+y^2)\) over the unit disk \(D\). Then $$ \mathbf{r}_x=(1,0,2x),\qquad \mathbf{r}_y=(0,1,2y),\qquad \mathbf{r}_x\times\mathbf{r}_y=(-2x,-2y,1). $$ The third component is positive, so this cross product has the upward orientation. Since \(\mathbf{F}(\mathbf{r}(x,y))=(0,0,1)\), $$ \mathbf{F}(\mathbf{r}(x,y))\cdot(\mathbf{r}_x\times\mathbf{r}_y) =(0,0,1)\cdot(-2x,-2y,1)=1. $$ Therefore the flux is $$ \iint_D 1\,dA = \int_0^{2\pi}\int_0^1 r\,dr\,d\theta = \int_0^{2\pi}\left[\frac{r^2}{2}\right]_0^1d\theta = \int_0^{2\pi}\frac12\,d\theta = \pi. $$ As a check, the parameterized flux integrand is \(1\), and the unit disk has area \(\pi\), giving the same value. The upward choice matters: reversing the normal would change the flux to \(-\pi\).
Why Stokes’ Theorem Matches Circulation
The relation in Stokes’ theorem is not a coincidence of notation. For a smooth parameterization \(\mathbf{r}(u,v)\) over a parameter region \(D\), define \(A=\mathbf{F}(\mathbf{r})\cdot\mathbf{r}_u\) and \(B=\mathbf{F}(\mathbf{r})\cdot\mathbf{r}_v\). Along the parameterized boundary, the circulation becomes the line integral of \(A\,du+B\,dv\). Green’s theorem converts that boundary integral into \(\iint_D(B_u-A_v)\,du\,dv\).
Applying the chain rule, the terms involving \(\mathbf{r}_{uv}\) cancel because the mixed partial derivatives agree. The remaining terms satisfy $$ B_u-A_v = (D\mathbf{F}(\mathbf{r})\,\mathbf{r}_u)\cdot\mathbf{r}_v - (D\mathbf{F}(\mathbf{r})\,\mathbf{r}_v)\cdot\mathbf{r}_u = (\nabla\times\mathbf{F})(\mathbf{r})\cdot(\mathbf{r}_u\times\mathbf{r}_v). $$ Thus the boundary circulation equals the curl flux through the parameterized surface, with the boundary orientation induced by the selected normal. This calculation explains both why Stokes’ theorem relates these two quantities and why its orientation condition is essential.
Worked Example: Stokes’ Theorem for Circulation
Worked Example: Stokes’ Theorem for Circulation
Let \(C\) be the circle \(x^2+y^2=4\), \(z=0\), traversed counterclockwise as viewed from above. Find the circulation of \(\mathbf{F}(x,y,z)=(-y/2,x/2,0)\) around \(C\).
The curve bounds the disk \(S\) of radius \(2\) in the plane \(z=0\). Its upward normal is consistent with the stated counterclockwise boundary direction. The curl is $$ \nabla\times\mathbf{F} = \left( \frac{\partial F_3}{\partial y}-\frac{\partial F_2}{\partial z}, \frac{\partial F_1}{\partial z}-\frac{\partial F_3}{\partial x}, \frac{\partial F_2}{\partial x}-\frac{\partial F_1}{\partial y} \right) = (0,0,\tfrac12-(-\tfrac12)) = (0,0,1). $$ Stokes’ theorem gives $$ \oint_C\mathbf{F}\cdot d\mathbf{r} = \iint_S(0,0,1)\cdot(0,0,1)\,dS = \iint_S1\,dS = \pi(2)^2 = 4\pi. $$ Check by direct line integration. Use \(\mathbf{r}(t)=(2\cos t,2\sin t,0)\), \(0\leq t\leq2\pi\). Then \(\mathbf{r}'(t)=(-2\sin t,2\cos t,0)\), and \(\mathbf{F}(\mathbf{r}(t))=(-\sin t,\cos t,0)\). Hence $$ \mathbf{F}(\mathbf{r}(t))\cdot\mathbf{r}'(t) = 2\sin^2t+2\cos^2t = 2, \qquad \int_0^{2\pi}2\,dt=4\pi. $$ The direct check agrees. Here Stokes’ theorem is especially clear because the curl is constant across the disk.
Worked Example: The Divergence Theorem for Closed-Surface Flux
Worked Example: The Divergence Theorem for Closed-Surface Flux
Find the outward flux of \(\mathbf{F}(x,y,z)=(x^3,y^3,z^3)\) across the boundary of the cube \(E=[-1,1]^3\).
The boundary is closed, and the field is continuously differentiable throughout the cube, so the divergence theorem applies. Compute $$ \nabla\cdot\mathbf{F}=3x^2+3y^2+3z^2. $$ The volume integral separates into three terms. For example, $$ \iiint_E 3x^2\,dV = 3\left(\int_{-1}^1x^2\,dx\right) \left(\int_{-1}^1dy\right) \left(\int_{-1}^1dz\right) = 3\cdot\frac23\cdot2\cdot2 = 8. $$ The \(3y^2\) and \(3z^2\) terms each give \(8\) by the same explicit integrals: the squared coordinate integrates to \(2/3\), and the other two coordinates each integrate to \(2\). Thus $$ \iint_{\partial E}\mathbf{F}\cdot\mathbf{n}\,dS = \iiint_E(3x^2+3y^2+3z^2)\,dV = 8+8+8 = 24. $$ Check directly on the six faces. On \(x=1\), \(\mathbf{F}\cdot\mathbf{n}=1\) and the face has area \(4\), so the flux is \(4\). On \(x=-1\), \(F_1=-1\) and the outward normal is \((-1,0,0)\), so the dot product is again \(1\) and the flux is \(4\). The two \(x\)-faces contribute \(8\). The two \(y\)-faces contribute \(8\) by the same calculation, and the two \(z\)-faces contribute \(8\). The total is \(8+8+8=24\), as predicted.
A Common Mix-Up: Curl Is Not Divergence
A frequent error is to choose a theorem by recognizing a familiar vector field, rather than by checking the quantity being requested. For instance, consider the outward flux of \(\mathbf{F}(x,y,z)=(x,y,z)\) through the unit sphere. A mistaken calculation might use \(\nabla\times\mathbf{F}=\mathbf{0}\) and conclude that the flux is zero. That uses the curl where the divergence theorem requires the divergence, so it does not calculate the requested quantity.
The correction is to compute \(\nabla\cdot\mathbf{F}=1+1+1=3\) and apply the divergence theorem to the unit ball: $$ \iint_S\mathbf{F}\cdot\mathbf{n}\,dS = \iiint_E3\,dV = 3\left(\frac{4\pi}{3}\right) = 4\pi. $$ Indeed, on the unit sphere the outward unit normal is \(\mathbf{n}=(x,y,z)\), so \(\mathbf{F}\cdot\mathbf{n}=x^2+y^2+z^2=1\); the direct flux is the sphere’s area, \(4\pi\). The check confirms the corrected result.
A Final Strategy Check
Decide whether the problem asks for surface flux or curve circulation.
Check whether it is an open surface, a curve bounding a surface, or a closed surface.
Confirm the required smoothness, field regularity, and normal or curve direction.
Choose direct integration or a valid theorem based on which calculation is simpler.
Use a direct calculation, a sign check, or a geometric check when practical.
The boundary type narrows the options, but the desired quantity determines which theorem could apply. Stokes’ theorem exchanges boundary circulation for curl flux; the divergence theorem exchanges closed-surface flux for a volume integral of divergence. Direct integration remains the baseline method, particularly when the surface is open or a theorem offers no simplification.
Check Your Understanding
For each question, identify the requested quantity and the method or methods worth considering.
- A vector field’s outward flux is requested across a closed, piecewise-smooth surface. Which theorem may replace the surface integral, and what volume integrand does it use?
- A curve bounds an oriented surface, and circulation around the curve is requested. What integral does Stokes’ theorem use on the surface?
- Why does the divergence theorem not directly evaluate flux through the open paraboloid in the first worked example?
- What orientation relationship must hold between the normal on a spanning surface and the boundary direction in Stokes’ theorem?
- A theorem applies, but its resulting integrand is more complicated than the original integral. What should you compare before choosing a method?