Change of variables theorem and the Jacobian determinant
Statement
Let Φ:U→Φ(U)⊆R2 be a C1 diffeomorphism with Jacobian matrix DΦ(u,v)=(xuyuxvyv). For any integrable function f on Φ(U), we have ∬Φ(U)f(x,y)dxdy=∬Uf(x(u,v),y(u,v))∣detDΦ(u,v)∣dudv, where detDΦ=xuyv−xvyu.
Why is it true?
Just as dx=g′(u)du rescales length in one-variable substitution, ∣detDΦ(u,v)∣ measures the local area-stretching ratio when a tiny (u,v)-rectangle is mapped to an (x,y)-parallelogram.
Proof sketch
Consider a small rectangle Rij=[ui,ui+Δu]×[vj,vj+Δv] in U. By first-order Taylor expansion around (ui,vj), the edges (Δu,0) and (0,Δv) map approximately to the tangent vectors a=Φu(ui,vj)Δu=(xu,yu)Δu and b=Φv(ui,vj)Δv=(xv,yv)Δv. The area of the parallelogram spanned by a and b in R2 equals ∣xuyv−xvyu∣ΔuΔv=∣detDΦ(ui,vj)∣ΔuΔv up to higher-order error o(ΔuΔv).
Substituting ΔAij≈∣detDΦ(ui,vj)∣ΔuΔv into the Riemann sum ∑i,jf(Φ(ui,vj))ΔAij yields ∑i,jf(Φ(ui,vj))∣detDΦ(ui,vj)∣ΔuΔv. In particular, for polar coordinates x=rcosθ, y=rsinθ, we have detDΦ=(cosθ)(rcosθ)−(−rsinθ)(sinθ)=r(cos2θ+sin2θ)=r, giving dxdy=rdrdθ.