Choosing a Strategy Before Calculating
Flux problems can have several possible routes to an answer. A closed surface may invite the divergence theorem, but a difficult volume integral can make direct surface integration preferable. An open surface may be completed with a convenient cap. And if a vector field is singular inside a proposed solid, the divergence theorem may not apply at all, even when its divergence is simple away from the singularity.
The useful question is not simply, “Which formula do I know?” It is, “Which method applies to this field and boundary, and which produces the clearest calculation?” The answer depends on three things: whether the surface is closed, whether the field is sufficiently regular throughout the solid, and whether the resulting surface or volume integrals are manageable.
The Decision: Boundary, Field, and Work
The divergence theorem relates flux through a closed boundary to an integral over the solid it encloses. In its standard form, let \(E\) be a bounded solid with a piecewise-smooth boundary \(\partial E\), oriented outward. If \(\mathbf{F}\) has continuous first partial derivatives on an open region containing \(E\), then
Here, if \(\mathbf{F}=\langle P,Q,R\rangle\), its divergence is \(\nabla\cdot\mathbf{F}=P_x+Q_y+R_z\). The theorem requires both a closed boundary and regularity of the field throughout the solid, not merely on the surface. It can be especially useful when the divergence is simple and the volume has an easy description.
If the target is an open surface, a cap can sometimes complete it to a closed boundary. Apply the theorem to that boundary, then subtract the cap’s flux with its correct orientation. If neither a theorem-based volume integral nor a useful cap simplifies the problem, direct surface integration may be the most reliable route. Direct integration is also necessary when a condition for the divergence theorem fails and no valid alternative solid removes the issue.
| Situation | Strategy to consider | Check before calculating |
|---|---|---|
| Closed surface and regular field | Divergence theorem or direct integration | Which of the volume and surface integrals is simpler? |
| Open surface with an easy cap | Complete the boundary, then subtract cap flux | Does the completed boundary enclose a valid solid? |
| Field singular inside the proposed solid | Direct integration, or a different valid region if available | Does the field actually satisfy the theorem’s regularity condition? |
Why Completing a Surface Works
Flux is additive over surface pieces: if a boundary is divided into pieces with their boundary orientations, its total flux is the sum of the fluxes through those pieces. Suppose a closed boundary consists of the target open surface \(S\) and a cap \(C\), with both normals pointing outward from the completed solid. Then
Rearranging gives the flux through \(S\) as the total boundary flux minus the cap flux. The minus sign is not a separate rule about caps; it follows by isolating the target term. The cap’s normal must still be the one pointing outward from the completed solid.
Worked Example: A Closed Cylinder with a Simple Divergence
Worked Example: A Closed Cylinder with a Simple Divergence
Find the outward flux of \(\mathbf{F}=\langle xz,yz,z^2\rangle\) through the boundary of the cylinder \(E=\{(x,y,z):x^2+y^2\leq 4,\ 0\leq z\leq 3\}\).
The surface is closed, made of the curved side and two disks. The field is polynomial, so its first partial derivatives are continuous throughout the solid. The divergence theorem applies. Compute the divergence:
In cylindrical coordinates, \(0\leq r\leq 2\), \(0\leq\theta\leq2\pi\), and \(0\leq z\leq3\), with \(dV=r\,dz\,dr\,d\theta\). Therefore
Check by summing the fluxes on the three boundary pieces. On the side, \(r=2\), an outward normal-area vector is \(\langle\cos\theta,\sin\theta,0\rangle r\,d\theta\,dz\), which here equals \(\langle2\cos\theta,2\sin\theta,0\rangle d\theta\,dz\). The dot product with \(\mathbf{F}\) is \(4z\), so the side flux is \(\int_0^{2\pi}\int_0^3 4z\,dz\,d\theta=36\pi\). The top disk has \(z=3\), \(F_z=9\), and area \(4\pi\), so its flux is \(36\pi\). On the bottom disk \(z=0\), \(F_z=0\), so its flux is zero. The sum is \(36\pi+36\pi+0=72\pi\), agreeing with the theorem.
Worked Example: Completing a Paraboloid with a Disk
Worked Example: Completing a Paraboloid with a Disk
Find the upward-oriented flux of \(\mathbf{F}=\langle x,y,z\rangle\) through the paraboloid patch \(S\) given by \(z=4-x^2-y^2\), \(x^2+y^2\leq1\). Its boundary circle lies in \(z=3\). The patch is open, but the paraboloid and disk \(C\) in \(z=3\) enclose the solid between them. The outward direction on the paraboloid is upward; on the disk, it is downward.
Since \(\nabla\cdot\mathbf{F}=1+1+1=3\), and the field is smooth throughout the solid, the divergence theorem applies to the completed boundary. In cylindrical coordinates the solid has \(0\leq r\leq1\), \(0\leq\theta\leq2\pi\), and \(3\leq z\leq4-r^2\). Its volume is
Thus the total outward flux through \(S\cup C\) is \(3(\pi/2)=3\pi/2\). On \(C\), the outward normal is \(\langle0,0,-1\rangle\), so \(\mathbf{F}\cdot\mathbf{n}=-3\). The disk has area \(\pi\), giving cap flux \(-3\pi\). Subtract it from the total:
Check directly on the graph. For \(f(x,y)=4-x^2-y^2\), the upward normal-area vector is \(\langle-f_x,-f_y,1\rangle dA=\langle2x,2y,1\rangle dA\). On the graph, the dot product is \(2x^2+2y^2+(4-x^2-y^2)=4+r^2\). Hence
The direct calculation confirms both the magnitude and the orientation used in the cap method.
Worked Example: A Singularity Makes Direct Integration Necessary
Worked Example: A Singularity Makes Direct Integration Necessary
Find the outward flux of \(\mathbf{F}=\langle x,y,z\rangle/(x^2+y^2+z^2)^{3/2}\) through the sphere \(x^2+y^2+z^2=4\). The sphere is closed, but the field is undefined at the origin, which lies inside the ball. Therefore the field is not continuously differentiable throughout the enclosed solid, and the divergence theorem cannot be applied to the ball.
Use direct integration. On the sphere of radius \(2\), the outward unit normal is \(\mathbf{n}=\langle x,y,z\rangle/2\). Since \(x^2+y^2+z^2=4\), the field is \(\mathbf{F}=\langle x,y,z\rangle/8\). Thus
The sphere has area \(4\pi(2^2)=16\pi\), so its flux is \((1/4)(16\pi)=4\pi\). As a check using spherical parameters, the area element is \(4\sin\phi\,d\phi\,d\theta\), and the integrand remains \(1/4\). Therefore the flux is
A Common Mistake: Using Divergence Without Checking the Solid
For the field in the sphere example, direct differentiation gives \(\nabla\cdot\mathbf{F}=0\) wherever \(x^2+y^2+z^2>0\). A tempting but invalid calculation is to integrate this value over the ball and conclude that the sphere’s flux is zero. That calculation ignores the point where the field is undefined. The divergence theorem requires the stated regularity throughout the solid, and it fails here. Direct integration gives \(4\pi\), not zero.
In general, a simple divergence does not by itself justify the theorem. Check the field on the entire solid, including its interior. If there is an interior singularity, do not treat a formula valid away from that point as though it held throughout the region. Choose a valid method for the actual domain and field.
A Practical Strategy
Determine whether it is closed. If it is open, look for a simple cap that closes it.
Look for points where the field or its first partial derivatives are undefined. A singularity inside the proposed solid blocks direct use of the divergence theorem.
Estimate whether the divergence and volume bounds are simpler than the surface parameterizations. A theorem is a tool, not an obligation.
For a completed boundary, identify the outward normal on every piece, especially the cap. For a direct calculation, match the chosen normal to the requested orientation.
Where practical, compare with a direct piecewise flux, verify a cap subtraction, or check the surface integrand and normal independently.
Check Your Understanding
For each question, identify the key condition or calculation that determines the strategy.
- A smooth vector field is given on a solid with a closed, piecewise-smooth boundary. What two integrals should you compare before choosing the divergence theorem or direct surface integration?
- An open surface is completed by a disk whose outward flux is \(-5\), while the total flux across the completed boundary is \(12\). What is the flux through the original surface?
- Why is a field’s being defined and differentiable on the boundary alone insufficient to apply the divergence theorem to the enclosed solid?
- For the paraboloid example, why does subtracting a negative cap flux increase the flux through the target surface?
- A field has zero divergence away from a point singularity inside a ball. Is it valid to conclude that the flux through the sphere is zero by integrating the divergence over the ball? Explain.