Mathematics
The Role of the Jacobian in Multivariable Change of Variables
Quick fact
The Jacobian determinant is the factor that tells you how much a small area or volume expands or shrinks when you change coordinates. For a transformation from (x,y) to (u,v), the Jacobian is the determinant of the matrix of partial derivatives, and it multiplies the differential element du dv to give the true physical size.
Why this is interesting
You've probably computed integrals in polar coordinates and noticed a mysterious extra 'r' multiplied into the area element. Why does that 'r' suddenly appear, and what does it have to do with stretching space?
Read the full explanation
Understanding The Role of the Jacobian in Multivariable Change of Variables
Imagine you have a map of a city, but the streets are drawn on a rubber sheet. When you stretch the rubber sheet, distances and areas change. Similarly, when you switch from one coordinate system to another, the shapes of infinitesimal areas get stretched or squashed. The Jacobian is like a magnifying glass that tells you exactly how much the local area is enlarged or reduced. Formally, if you have a change of variables x = x(u, v) and y = y(u, v), then a small rectangle in the uv-plane of size du by dv corresponds to a small parallelogram in the xy-plane. The area of this parallelogram is |J| du dv, where J is the determinant of the matrix of partial derivatives. This determinant is called the Jacobian. So when you integrate a function over a region, and you rewrite the integral in new coordinates, you must include this factor to account for the distortion. Without it, your integral would be off by an amount that depends on the local stretching.
A deeper explanation
Why does the Jacobian work? The key lies in linearization. Near any point, a smooth transformation is approximately linear. The derivative matrix (also called the Jacobian matrix) captures this linear approximation. When you change variables, you are effectively replacing a tiny rectangle du×dv with its image, which is a parallelogram spanned by the two vectors formed by the partial derivatives. The area of that parallelogram is exactly the absolute value of the determinant of the Jacobian matrix. Thus, the determinant measures the local scaling factor of the transformation. In higher dimensions, the same idea holds with volumes and higher-order determinants. This matters because the integral is defined as a limit of sums of function values times small area elements. If you change variables, you must compensate for the change in size of those area elements. The Jacobian is the bridge that lets you safely convert the integral, ensuring that the total accumulation is preserved. In practice, this is how we get the familiar rdrdθ in polar coordinates, and similar factors for cylindrical and spherical coordinates. The Jacobian also appears in probability when changing variables in joint distributions, and in physics when converting integrals to alternative coordinate systems to simplify computations.