Jacobians in Calculus: Definition, Formula and Examples

When you change variables in a double or triple integral — for example switching from x and y to polar coordinates r and θ — the little area element dx dy does not simply become dr dθ. The grid of the new coordinates is stretched and squeezed relative to the old one, and you need a correction factor that measures exactly how much area is distorted at each point.

That correction factor is the Jacobian. For a transformation u = u(x, y), v = v(x, y), the Jacobian ∂(u, v)/∂(x, y) is the determinant of the matrix of all first partial derivatives. Its absolute value tells you how much a tiny patch of area expands or shrinks under the transformation — and in integrals you multiply by |J| so that areas and volumes come out right.

Key Concepts

• Jacobian matrix: the 2×2 matrix of first partials, rows = new variables, columns = old variables: [[ux, uy], [vx, vy]].
• Jacobian determinant: J = ux vy − uy vx, written ∂(u, v)/∂(x, y).
• Geometric meaning: |J| > 1 means local expansion of area; |J| < 1 means contraction; J = 0 means the transformation collapses area and is not invertible there.
• Three variables: the same idea with a 3×3 determinant ∂(u, v, w)/∂(x, y, z).

Formulas

• ∂(u, v)/∂(x, y) = ux vy − uy vx
• Change of variables: &iint;R f(x, y) dA = &iint;S f(u, v) |∂(x, y)/∂(u, v)| du dv — always the absolute value.
• Polar coordinates: x = r cos θ, y = r sin θ gives J = r, so dA = r dr dθ.
• Inverse relation: ∂(x, y)/∂(u, v) = 1 / ∂(u, v)/∂(x, y) wherever the Jacobian is nonzero.

Worked Example

Compute the Jacobian of u = x² − y², v = 2xy, and verify the polar Jacobian.

1. Partials. ux = 2x, uy = −2y; vx = 2y, vy = 2x.
2. Determinant. J = (2x)(2x) − (−2y)(2y) = 4x² + 4y² = 4(x² + y²).
3. At (1, 1): J = 8 — a tiny square near (1, 1) is stretched to 8 times its area by this map.
4. Polar check. x = r cos θ, y = r sin θ: xr = cos θ, xθ = −r sin θ, yr = sin θ, yθ = r cos θ. J = cos θ · r cos θ − (−r sin θ)(sin θ) = r(cos²θ + sin²θ) = r.
5. Bonus — area of a quarter-disk of radius 2: ∫0π/2 ∫02 r dr dθ = (π/2) × 2 = π, which matches πr²/4 = π. The Jacobian made the integral trivial.

Common Misconceptions

• “The Jacobian is the determinant of any matrix.” It must be the determinant of the partial-derivative matrix of the transformation, with the variables in the right order.
• “You can drop the absolute value.” Area is positive — always use |J| inside integrals, or regions where J is negative will subtract instead of add.
• “dx dy becomes du dv directly.” Without the Jacobian factor the integral is wrong whenever the new grid is stretched relative to the old one.

Real-World Applications

• Physics and engineering: polar, cylindrical, and spherical coordinates (Jacobians r, r, and ρ² sin φ) simplify problems with circular symmetry.
• Probability: transforming joint density functions when random variables are redefined.
• Robotics and computer vision: how small motions in joint space map to motion in the workspace.
• Economics: change of variables in multi-dimensional models, e.g. switching between price and quantity coordinates.

The Jacobian is the exchange rate between coordinate systems: it tells you how many units of new area each unit of old area buys. Compute it as the determinant of the partial-derivative matrix, take the absolute value, and multiply it into the integrand — then even the most awkward region can become a simple rectangle.