Integral theorems · Definition
Stokes' Theorem
Before you start you need to know
It lets you compute a circulation in space as a flux of the curl, choosing the most convenient surface that has the curve as its boundary.
Definition
Stokes' theorem relates the circulation of a field \(\mathbf F\) along the closed curve \(C\) bounding an oriented surface \(S\) to the flux of the curl of \(\mathbf F\) through \(S\).
Orientation is coupled by the right-hand rule: if the thumb points along the normal of \(S\), the curled fingers give the direction of travel along \(C\). One needs \(S\) piecewise smooth and \(\mathbf F\) with continuous partial derivatives on an open set containing \(S\). Reversing the normal flips the sign of the flux, just as reversing \(C\) flips the circulation.
A practical consequence is that the flux of the curl depends only on the boundary: two surfaces with the same curve \(C\) give the same value, so a flat disk can replace a hemisphere. A guiding example: for \(\langle -y,x,z\rangle\) around a horizontal circle of radius \(a\), the curl is \(\langle0,0,2\rangle\) and the circulation is \(2\pi a^2\), regardless of the circle's height.
When \(S\) lies in the \(xy\)-plane with normal \(\mathbf k\), the integrand reduces to \(Q_x-P_y\) and Green's theorem is recovered. If \(\nabla\times\mathbf F=\mathbf 0\) on a region without holes, the circulation around every closed curve vanishes and \(\mathbf F\) is conservative. Typical mistakes: forgetting that \(S\) and \(C\) need compatible orientations, or using a surface whose boundary is not \(C\).
Stokes' theorem links a curve to a surface, and the divergence theorem, which comes next, links a closed surface to the solid it encloses.
Think about it
If two different surfaces \(S_1\) and \(S_2\) have the same oriented boundary \(C\), how do \(\iint_{S_1}(\nabla\times\mathbf F)\cdot d\mathbf S\) and \(\iint_{S_2}(\nabla\times\mathbf F)\cdot d\mathbf S\) compare?
They are equal: both equal \(\oint_C\mathbf F\cdot d\mathbf r\).
What is the circulation of \(\langle -y,x,z\rangle\) around a horizontal circle of radius \(1\)?
\(2\pi\): the curl is \(\langle0,0,2\rangle\) and the disk area is \(\pi\).
Exercises 10
- Let \(S\) be the disk \(x^2+y^2\le1\) in the plane \(z=0\), with normal \(\mathbf k\). Compute \(\iint_S(\nabla\times\mathbf F)\cdot d\mathbf S\) for \(\mathbf F=\langle -y,x,0\rangle\).
- Let \(\mathbf F=\langle z,x,y\rangle\) and \(C\) the boundary of the triangle of the plane \(x+y+z=1\) in the first octant, counterclockwise seen from above. Compute \(\oint_C\mathbf F\cdot d\mathbf r\).
- With the same curve \(C\) as in the previous exercise, compute \(\oint_C\mathbf F\cdot d\mathbf r\) for \(\mathbf F=\langle y,z,x\rangle\).
- Let \(\mathbf F=\langle 2xyz,\;x^2z,\;x^2y\rangle\). What is \(\oint_C\mathbf F\cdot d\mathbf r\) over any smooth closed curve \(C\) in space?
- If \(S\) lies in the \(xy\)-plane with normal \(\mathbf k\) and \(\mathbf F=\langle P,Q,0\rangle\), what does \((\nabla\times\mathbf F)\cdot\mathbf k\) reduce to?
- Let \(S\) be the hemisphere \(x^2+y^2+z^2=1\), \(z\ge0\), with upward normal, and \(\mathbf F=\langle y,-x,0\rangle\). Compute \(\iint_S(\nabla\times\mathbf F)\cdot d\mathbf S\).
- If \(\iint_S(\nabla\times\mathbf F)\cdot d\mathbf S=4\) with \(S\) oriented upward, what is it after reversing the orientation of \(S\) (downward normal)?
- Let \(C\) be the circle \(x^2+y^2=4\) in the plane \(z=3\), counterclockwise seen from above, and \(\mathbf F=\langle -y,x,z\rangle\). Compute \(\oint_C\mathbf F\cdot d\mathbf r\).
- If \(\nabla\times\mathbf F=\mathbf 0\) on all of \(\mathbb R^3\), which statement is correct?
- Let \(\mathbf F=\langle 0,0,xy\rangle\) and \(C\) the unit circle in the \(xy\)-plane, counterclockwise. What is \(\oint_C\mathbf F\cdot d\mathbf r\)?
References
- James Stewart, Daniel K. Clegg, Saleem Watson, Calculus: Early Transcendentals, 9th edition, Cengage, 2021. §16.8
- Joel Hass, Christopher Heil, Przemyslaw Bogacki, Maurice D. Weir, George B. Thomas Jr., Thomas' Calculus, 15th edition, Pearson, 2023. §16.7
- Jerrold E. Marsden, Anthony Tromba, Vector Calculus, 6th edition, W. H. Freeman, 2012. §8.2