Two Routes to the Same Net Flux
A flux problem asks how much of a vector field passes outward through an oriented surface. If the surface is closed and bounds a solid, there are often two legitimate routes: integrate the field across the surface directly, or integrate its divergence throughout the enclosed volume.
Neither route is automatically shorter. A sphere may have a simple radial field that makes direct flux transparent, while a complicated collection of boundary pieces may be easier to handle through one volume integral. The useful skill is to recognize what each setup requires, compare the work before calculating, and check that the two methods agree when both are practical.
For direct integration, a smooth surface patch can be parametrized by \(\mathbf{r}(u,v)\). The tangent vectors \(\mathbf{r}_u\) and \(\mathbf{r}_v\) span a small tangent parallelogram. Its oriented area vector is \((\mathbf{r}_u\times\mathbf{r}_v)\,du\,dv\), or its negative if that cross product points opposite the required normal. Thus the direct flux over the patch is
with the sign chosen for the desired orientation. This formula explains why parametrization and orientation must be checked together: the cross product is not just a magnitude of area; its direction determines the sign of flux.
Compare the Setups Before Integrating
The direct route requires a parametrization for every boundary piece whose flux is needed, together with the correct orientation and parameter limits. The divergence route requires a closed boundary, a sufficiently smooth field on and inside it, and a manageable divergence and volume integral. A useful first comparison is therefore:
| Question | Direct surface integration | Divergence theorem |
|---|---|---|
| What is integrated? | The normal component of the field on each surface piece | The divergence throughout the enclosed solid |
| What setup is needed? | Surface parametrizations, area vectors, and limits | A closed solid, its volume limits, and the divergence |
| When can it be efficient? | When the field’s normal component or the surface geometry is simple | When the divergence or the volume description is simple |
| What must be checked? | Orientation and every boundary contribution | The theorem’s smoothness and closed-boundary conditions |
These are not competing definitions of flux. Under the theorem’s hypotheses, both calculations give the same total outward flux. The examples below work out both routes so that the source of the agreement, and the practical difference in setup, are visible.
Worked Example: Flux Through a Sphere by Both Methods
Worked Example: Flux Through a Sphere by Both Methods
Let \(\mathbf{F}(x,y,z)=(x,y,z)\), and let \(S\) be the sphere of radius \(2\) centered at the origin, oriented outward. Compute its flux directly and with the divergence theorem.
Direct calculation. Use the spherical parametrization \(\mathbf{r}(\phi,\theta)=(2\sin\phi\cos\theta,2\sin\phi\sin\theta,2\cos\phi)\), where \(0\leq\phi\leq\pi\) and \(0\leq\theta\leq2\pi\). Its tangent vectors are
The outward area vector per \(d\phi\,d\theta\) is
For example, at \(\phi=\pi/2,\theta=0\), this vector points in the positive \(x\)-direction, as it should at that point on the sphere. Since \(\mathbf{F}(\mathbf{r})=\mathbf{r}\), the dot product is
Therefore the direct flux is
Divergence-theorem calculation. The divergence is \(\nabla\cdot\mathbf{F}=1+1+1=3\). The sphere encloses a ball of volume \(\frac{4}{3}\pi(2)^3=\frac{32\pi}{3}\), so
Both methods give \(32\pi\). The direct calculation uses a simple field and an explicit spherical area vector; the theorem calculation uses a constant divergence and the familiar ball volume. Each is manageable here, and the agreement checks the parametrization and integral.
Worked Example: A Closed Cylinder with Three Boundary Pieces
Worked Example: A Closed Cylinder with Three Boundary Pieces
Let \(\mathbf{F}(x,y,z)=(x,y,2z)\). Let \(E\) be the cylinder \(x^2+y^2\leq1\), \(0\leq z\leq2\), with its side and both caps included. Find its outward flux by direct integration and by the divergence theorem.
Direct calculation on the side. Parametrize the curved side by \(\mathbf{r}(\theta,z)=(\cos\theta,\sin\theta,z)\), with \(0\leq\theta\leq2\pi\) and \(0\leq z\leq2\). Taking \(\mathbf{r}_\theta\times\mathbf{r}_z\) gives the outward area vector \((\cos\theta,\sin\theta,0)\). On the side, \(\mathbf{F}=(\cos\theta,\sin\theta,2z)\), so the flux density is
Thus the side contributes \(\int_0^{2\pi}\int_0^2 1\,dz\,d\theta=4\pi\). On the top cap, \(z=2\), the outward normal is \(\mathbf{k}\). The normal component is \(2z=4\), so the top contributes \(4\cdot\pi=4\pi\). On the bottom cap, \(z=0\), the outward normal is \(-\mathbf{k}\), but the field’s \(z\)-component is zero; its flux is \(0\). Adding all three pieces gives
Divergence-theorem calculation. Here \(\nabla\cdot\mathbf{F}=1+1+2=4\). The cylinder has volume \(\pi(1)^2(2)=2\pi\), so the theorem gives
The results agree. In this example, the divergence method is especially short because the divergence is constant and the volume is immediate. The direct method is also systematic, but requires separate calculations for the side and caps.
Worked Example: A Hemisphere and Its Base Disk
Worked Example: A Hemisphere and Its Base Disk
Let \(\mathbf{F}(x,y,z)=(x,y,z+1)\). Let \(E\) be the upper half of the unit ball, including its flat base disk at \(z=0\). Compute the outward flux directly and by the divergence theorem.
Direct calculation on the curved hemisphere. Use \(\mathbf{r}(\phi,\theta)=(\sin\phi\cos\theta,\sin\phi\sin\theta,\cos\phi)\), for \(0\leq\phi\leq\pi/2\), \(0\leq\theta\leq2\pi\). The outward area vector is
On the unit sphere, the outward unit normal is \(\mathbf{r}\) and the area element is \(\sin\phi\,d\phi\,d\theta\). Hence \(\mathbf{F}\cdot\mathbf{r}=x^2+y^2+z(z+1)=1+z=1+\cos\phi\). The curved flux is
On the base disk the outward normal points downward, \(-\mathbf{k}\). Since \(F_z=z+1=1\) at \(z=0\), the base flux is \(-1\) times the disk area, or \(-\pi\). The closed-surface flux is therefore \(3\pi-\pi=2\pi\).
Divergence-theorem calculation. The divergence is \(3\). The upper half-ball has half the volume of the unit ball, namely \(\frac12\cdot\frac43\pi=\frac{2\pi}{3}\). Therefore
Again, both methods agree. This example makes the piecewise orientation important: the disk is part of the closed boundary, and its outward normal is downward, not upward.
Why the Two Methods Agree
The direct formula measures the field’s outward component on each small surface patch and adds those contributions over the boundary. The divergence theorem says that, when its conditions hold, this total equals the integral of the local divergence throughout the enclosed solid. Divergence records the field’s net outward flow per unit volume; adding it over the solid accounts for the total boundary flow. The theorem therefore converts a boundary calculation into a volume calculation without changing the quantity being computed.
The examples also illustrate the theorem’s role in checking work. A discrepancy does not mean that the two notions of flux differ. It signals a setup or arithmetic issue to investigate: a missing boundary piece, an incorrect normal, an integration limit, or a computed divergence may be responsible.
A Practical Choice of Method
Determine whether the surface is closed, whether it bounds the specified solid, and whether the field has continuous first partial derivatives on an open set containing that solid.
Compute or inspect \(\nabla\cdot\mathbf{F}\). A simple divergence and familiar volume description often make the theorem efficient.
Look for boundary pieces where the normal component is zero, constant, or otherwise simple. Include every piece and orient each one outward.
Use the shorter reliable setup. If the other route is also manageable, compare the results; agreement is a strong check on signs and arithmetic.
A direct calculation is not necessarily difficult just because a surface is curved: a radial field on a sphere may have a simple normal component. Conversely, a closed piecewise surface is not automatically a reason to invoke the theorem if the divergence or volume integral is harder than the surface pieces. The geometry and the field must be considered together.
Common Pitfall: Reversing One Piece’s Normal
On the hemisphere example, the base disk’s outward normal is \(-\mathbf{k}\), because the solid lies above the disk. A common incorrect calculation uses \(+\mathbf{k}\), giving a base flux of \(+\pi\). Keeping the curved contribution \(3\pi\) and adding this mistaken cap contribution produces \(4\pi\), not the correct total \(2\pi\). The correction is to use the outward normal \(-\mathbf{k}\), which makes the base flux \(-\pi\).
This error is easy to miss if the curved surface is parametrized correctly but the flat piece is treated as though it had the same orientation as a graph viewed from above. For a closed boundary, determine the outward direction separately on each piece. Also, do not use the divergence theorem to avoid checking the boundary: its integral gives the total flux over the complete closed surface, not the flux through an arbitrarily chosen part alone.
Check Your Understanding
Use the definitions and comparisons above to answer these questions.
- What smoothness and boundary conditions are needed to apply the divergence theorem as stated here?
- For the sphere parametrization in the first worked example, why does the direction of \(\mathbf{r}_\phi\times\mathbf{r}_\theta\) need to be checked?
- In the cylinder example, what are the separate flux contributions from the side, top, and bottom?
- Why is the outward normal to the base disk of an upper hemisphere \(-\mathbf{k}\)?
- Give one reason the divergence-theorem calculation might be shorter than direct surface integration, and one reason direct integration might be preferable.