The math optional, made finite. Daily Practice

Jacobian

At a Glance

Why This Chapter Matters

The Jacobian appears in three distinct roles in UPSC exams, each with its own algorithm. The simplest (compute a Jacobian determinant, 7 marks) is purely mechanical. The hardest (use a Jacobian substitution to evaluate a double integral, 15 marks) requires recognising which substitution transforms the integration region into a rectangle and separates the integrand into Beta functions. The middle archetype (functional dependence via vanishing Jacobian) is conceptual but short. Know all three roles.

Minimum Theory

Definition. For functions u=u(x,y)u = u(x,y) and v=v(x,y)v = v(x,y), the Jacobian is (u,v)(x,y)=u/xu/yv/xv/y=uxvyuyvx.\frac{\partial(u,v)}{\partial(x,y)} = \begin{vmatrix}\partial u/\partial x & \partial u/\partial y\\ \partial v/\partial x & \partial v/\partial y\end{vmatrix} = \frac{\partial u}{\partial x}\frac{\partial v}{\partial y} - \frac{\partial u}{\partial y}\frac{\partial v}{\partial x}.

Change of variables. If (u,v)=T(x,y)(u,v) = T(x,y) and TT is one-to-one, then f(x,y)dxdy=f(x(u,v),y(u,v))(x,y)(u,v)dudv.\iint f(x,y)\,dx\,dy = \iint f(x(u,v),y(u,v))\left|\frac{\partial(x,y)}{\partial(u,v)}\right|du\,dv.

Jacobian roles: change of variables (left) and functional dependence (right)

Functional dependence. If (u,v)/(x,y)0\partial(u,v)/\partial(x,y) \equiv 0 on an open region, then uu and vv are functionally dependent — there exists a function Φ\Phi such that u=Φ(v)u = \Phi(v) (or v=Φ(u)v = \Phi(u)). To identify Φ\Phi, evaluate at a convenient special point.

Question Archetypes

ArchetypeRecognition
compute-jacobian”Find (u,v)/(r,θ)\partial(u,v)/\partial(r,\theta)” — direct computation of the 2×22\times2 determinant
jacobian-change-of-variable”Using the transformation … evaluate \iint\ldots” — change of variables in a double integral
functional-dependence”Show u,vu,v are functionally related; find the relation” — Jacobian =0=0 implies dependence

compute-jacobian (1 question(s); 2021)

Compute a Jacobian determinant directly

Recognition Cues

Solution Template

  1. Express uu and vv in terms of the independent variables (simplify using trig identities if possible).
  2. Compute all four partial derivatives.
  3. Evaluate the 2×22\times2 determinant.

Worked Example

2021 Paper 1, 2021-P1-Q3a-i (7 marks)

u=x2+y2u=x^2+y^2, v=x2y2v=x^2-y^2, x=rcosθx=r\cos\theta, y=rsinθy=r\sin\theta. Find (u,v)/(r,θ)\partial(u,v)/\partial(r,\theta).

Step 1. u=r2u = r^2 (depends only on rr); v=r2cos2θv = r^2\cos2\theta (using cos2θsin2θ=cos2θ\cos^2\theta-\sin^2\theta=\cos2\theta).

Step 2. u/r=2r\partial u/\partial r=2r, u/θ=0\partial u/\partial\theta=0; v/r=2rcos2θ\partial v/\partial r=2r\cos2\theta, v/θ=2r2sin2θ\partial v/\partial\theta=-2r^2\sin2\theta.

Step 3. Determinant: (2r)(2r2sin2θ)(0)(2rcos2θ)=4r3sin2θ.(2r)(-2r^2\sin2\theta)-(0)(2r\cos2\theta) = \boxed{-4r^3\sin2\theta.}

Common Traps

jacobian-change-of-variable (1 question(s); 2014)

Use a Jacobian change of variables to transform a double integral

Recognition Cues

Solution Template

  1. Invert the substitution to express (x,y)(x,y) in terms of (u,v)(u,v).
  2. Compute the Jacobian (x,y)/(u,v)\partial(x,y)/\partial(u,v).
  3. Transform the limits: find the image of the original region in (u,v)(u,v) coordinates.
  4. Rewrite the integrand in (u,v)(u,v); check for factorisation.
  5. Evaluate using Beta functions if the integrand has the form ua(1u)bu^a(1-u)^b.

Worked Example

2014 Paper 1, 2014-P1-Q2c (15 marks)

Using x+y=ux+y=u, y=uvy=uv, evaluate xy(1xy)dxdy\displaystyle\iint\sqrt{xy(1-x-y)}\,dx\,dy over the triangle x0x\ge0, y0y\ge0, x+y1x+y\le1.

Step 1. Inversion: x=u(1v)x=u(1-v), y=uvy=uv.

Step 2. Jacobian: (x,y)/(u,v)=1vuvu=u\partial(x,y)/\partial(u,v) = \begin{vmatrix}1-v&-u\\v&u\end{vmatrix} = u.

Step 3. Limits: the triangle maps to the unit square 0u10\le u\le1, 0v10\le v\le1.

Step 4. Transform: xy=u2v(1v),1xy=1u,xy(1xy)=uv(1v)(1u).xy = u^2v(1-v),\quad 1-x-y=1-u,\quad \sqrt{xy(1-x-y)}=u\sqrt{v(1-v)(1-u)}. Integral: I=0101uv(1v)(1u)ududv=01u21udu01v(1v)dv.I=\int_0^1\int_0^1 u\sqrt{v(1-v)(1-u)}\cdot u\,du\,dv=\int_0^1 u^2\sqrt{1-u}\,du\cdot\int_0^1\sqrt{v(1-v)}\,dv.

Step 5. Beta functions: B(p,q)=01tp1(1t)q1dt=Γ(p)Γ(q)/Γ(p+q)B(p,q)=\int_0^1t^{p-1}(1-t)^{q-1}dt = \Gamma(p)\Gamma(q)/\Gamma(p+q). 01u2(1u)1/2du=B(3,3/2)=Γ(3)Γ(3/2)Γ(9/2)=2(π/2)105π/16=16105.\int_0^1 u^2(1-u)^{1/2}\,du = B(3,3/2) = \frac{\Gamma(3)\Gamma(3/2)}{\Gamma(9/2)} = \frac{2\cdot(\sqrt\pi/2)}{105\sqrt\pi/16} = \frac{16}{105}. 01v1/2(1v)1/2dv=B(3/2,3/2)=(Γ(3/2))2Γ(3)=π/42=π8.\int_0^1 v^{1/2}(1-v)^{1/2}\,dv = B(3/2,3/2) = \frac{(\Gamma(3/2))^2}{\Gamma(3)} = \frac{\pi/4}{2} = \frac{\pi}{8}. I=16105π8=2π105.\boxed{I = \frac{16}{105}\cdot\frac{\pi}{8} = \frac{2\pi}{105}.}

Common Traps

functional-dependence (2 question(s); 2019, 2024)

Use a vanishing Jacobian to detect and find a functional relation between uu and vv

Recognition Cues

Solution Template

  1. Compute (u,v)/(x,y)\partial(u,v)/\partial(x,y) directly.
  2. Show the Jacobian is identically zero — this proves functional dependence.
  3. Identify Φ\Phi: set one variable to a convenient value (e.g. y=0y=0) to pin down u=Φ(v)u=\Phi(v).

Worked Example 1

2024 Paper 1, 2024-P1-Q2b (15 marks)

u=(x+y)/(1xy)u=(x+y)/(1-xy), v=arctanx+arctanyv=\arctan x+\arctan y. Find (u,v)/(x,y)\partial(u,v)/\partial(x,y). Are u,vu,v functionally related?

Step 1. ux=(1+y2)/(1xy)2u_x = (1+y^2)/(1-xy)^2, uy=(1+x2)/(1xy)2u_y = (1+x^2)/(1-xy)^2; vx=1/(1+x2)v_x = 1/(1+x^2), vy=1/(1+y2)v_y = 1/(1+y^2).

Step 2. Jacobian: uxvyuyvx=1+y2(1xy)211+y21+x2(1xy)211+x2=1(1xy)21(1xy)2=0.u_xv_y - u_yv_x = \dfrac{1+y^2}{(1-xy)^2}\cdot\dfrac{1}{1+y^2} - \dfrac{1+x^2}{(1-xy)^2}\cdot\dfrac{1}{1+x^2} = \dfrac{1}{(1-xy)^2}-\dfrac{1}{(1-xy)^2} = \boxed{0.}

Step 3. Functionally dependent. By the arctan\arctan addition formula (valid for xy<1xy<1): v=arctanuv = \arctan u, i.e. u=tanv.\boxed{u = \tan v.}

Worked Example 2

2019 Paper 1, 2019-P1-Q4c-ii (8 marks)

If f(x)=1/(1+x2)f'(x)=1/(1+x^2) and f(0)=0f(0)=0, show f(x)+f(y)=f ⁣(x+y1xy)f(x)+f(y)=f\!\left(\dfrac{x+y}{1-xy}\right) using the Jacobian.

Step 1. Let u=f(x)+f(y)u=f(x)+f(y) and w=(x+y)/(1xy)w=(x+y)/(1-xy). Then ux=1/(1+x2)u_x=1/(1+x^2), wx=(1+y2)/(1xy)2w_x=(1+y^2)/(1-xy)^2.

Step 2. Jacobian (u,w)/(x,y)=11+x21+x2(1xy)211+y21+y2(1xy)2=0\partial(u,w)/\partial(x,y) = \frac{1}{1+x^2}\cdot\frac{1+x^2}{(1-xy)^2} - \frac{1}{1+y^2}\cdot\frac{1+y^2}{(1-xy)^2} = 0.

Step 3. Functional dependence: u=Φ(w)u=\Phi(w). Set y=0y=0: w=xw=x and u=f(x)+f(0)=f(x)u=f(x)+f(0)=f(x). So Φ=f\Phi=f, giving f(x)+f(y)=f ⁣(x+y1xy)f(x)+f(y)=f\!\left(\dfrac{x+y}{1-xy}\right). \blacksquare

Common Traps

Marks-Aware Writing

7-mark Jacobian computation: show each partial derivative computed, display the 2×22\times2 determinant, then the evaluation. Three lines of algebra suffice.

15-mark change-of-variables: the examiner awards marks for (a) the Jacobian correctly computed, (b) the transformed limits (unit square), (c) the separation of the double integral, (d) the Beta function identification, and (e) the numerical answer. Missing the Jacobian factor (forgetting the uu) or mis-identifying the Beta parameters each costs 3 marks.

8-mark functional dependence: the vanishing of the Jacobian (shown by explicit calculation) and the identification of Φ\Phi via y=0y=0 are both required. A proof with only the Jacobian step and no identification of Φ\Phi earns half marks.

Practice Set

We've mapped all 13 years of this exam. Get new chapters, tools, and solutions as we release them — free.