What Is the Jacobian Determinant and How Do You Calculate It?

The Jacobian determinant summarizes how a differentiable change of coordinates locally stretches, shrinks, or reverses area or volume. To calculate it, form the matrix of first partial derivatives and take its determinant. In a multiple integral, the area or volume scaling factor is the absolute value of that determinant.

The title can refer either to calculating this determinant or understanding what it means. The method below covers both, with a worked transformation and the connection to integration.

Define the Jacobian matrix

Suppose a map takes input coordinates $(u,v)$ to output coordinates $(x,y)$:

$$(x,y) = (x(u,v), y(u,v)).$$

Its Jacobian matrix is

$$\frac{\partial(x,y)}{\partial(u,v)} = \begin{pmatrix} \frac{\partial x}{\partial u} & \frac{\partial x}{\partial v} \\ \frac{\partial y}{\partial u} & \frac{\partial y}{\partial v} \end{pmatrix}.$$

The Jacobian determinant is the determinant of that matrix:

$$J = \frac{\partial(x,y)}{\partial(u,v)} = \frac{\partial x}{\partial u}\frac{\partial y}{\partial v} - \frac{\partial x}{\partial v}\frac{\partial y}{\partial u}.$$

The notation $\partial(x,y)/\partial(u,v)$ denotes a determinant, not an ordinary fraction. For a map from $n$ input variables to $n$ output variables, the Jacobian matrix is the $n\times n$ matrix of partial derivatives, and its determinant is defined only when the input and output dimensions match.

Calculate a Jacobian determinant step by step

For a two-variable map:

1. Identify the input variables and output functions.
2. Differentiate each output function with respect to each input variable.
3. Arrange the derivatives so each row corresponds to one output and each column to one input.
4. Evaluate the determinant.

Some books put inputs in rows and outputs in columns instead. That convention transposes the matrix, but does not change its determinant.

Worked example

Consider the linear transformation

$$x=u+v, \qquad y=u-v.$$

The partial derivatives are $x_u=1$, $x_v=1$, $y_u=1$, and $y_v=-1$. Thus

$$J = \det\begin{pmatrix}1&1\\1&-1\end{pmatrix} = (1)(-1)-(1)(1)=-2.$$

The determinant is $-2$. Its magnitude, $|J|=2$, means that small areas in the $(u,v)$ plane are scaled by a factor of 2 in the $(x,y)$ plane. The negative sign indicates that the transformation reverses orientation.

For example, the unit square $0\leq u\leq1$, $0\leq v\leq1$ has area 1. Its image is a parallelogram with area 2, agreeing with the scale factor. This linear example has a constant Jacobian; for a nonlinear map, the determinant can vary from point to point.

Why the absolute value appears in integrals

Under the usual change-of-variables hypotheses—for example, when the coordinate map is one-to-one and continuously differentiable with nonzero Jacobian on the region—a small area element changes according to

$$dx\,dy = |J|\,du\,dv.$$

So the change-of-variables formula is

$$\iint_R f(x,y)\,dx\,dy = \iint_S f(x(u,v),y(u,v))\,|J|\,du\,dv,$$

where $S$ is the region in input coordinates and $R$ is its image. The absolute value is necessary because area is nonnegative, even when the map reverses orientation.

A familiar example is polar coordinates, $x=r\cos\theta$ and $y=r\sin\theta$. Their Jacobian determinant is

$$\det\begin{pmatrix} \cos\theta & -r\sin\theta \\ \sin\theta & r\cos\theta \end{pmatrix} = r\cos^2\theta+r\sin^2\theta=r.$$

For the usual polar-coordinate region, $r\geq0$, so $|J|=r$ and $dx\,dy=r\,dr\,d\theta$. This is why the factor $r$ belongs in a double integral written in polar coordinates.

What the determinant tells you

When $J\ne0$ at a point, the derivative map there is invertible, so the map has a nonsingular first-order linear approximation at that point. If the map is continuously differentiable near that point, the inverse function theorem also guarantees that it is locally invertible. When $J=0$, the derivative collapses at least one direction, so the derivative is singular at that point. A zero determinant does not by itself prove that the whole transformation is non-invertible everywhere; it describes the local derivative at the point where it vanishes.

In higher dimensions, the same idea applies. The absolute value of the determinant gives the local volume scale. For example, a determinant with magnitude 3 means a tiny volume near that point is scaled by a factor of 3.

Common mistakes

• Forgetting the absolute value in an integral. The determinant can be negative, but area and volume scaling use $|J|$.
• Using the wrong variables. Make sure the denominator in the Jacobian notation matches the input coordinates you differentiate with respect to.
• Mixing up rows and columns. Choose a consistent convention. Transposing the full matrix does not change the determinant, but selectively rearranging entries does.
• Assuming a zero determinant settles global invertibility. It shows local singularity of the derivative at that point; global behavior requires more information.
• Leaving out the Jacobian factor in a coordinate change. Changing the integrand and the region is not enough; the area or volume element must also be transformed.