Some regions and integrands have circular symmetry that makes Cartesian coordinates cumbersome. Polar coordinates (r,θ) — where x=rcosθ and y=rsinθ — naturally describe disks, sectors, and annuli. This note shows how to set up and evaluate double integrals in polar coordinates.
1. The polar rectangle
Recall from Polar Coordinates that a point in the plane is described by (r,θ) with r≥0 and θ∈[0,2π). Just as Cartesian rectangles [a,b]×[c,d] are the building block for Cartesian integrals, polar rectangles are the natural building block for polar integrals.
A polar rectangle is the set
R={(r,θ):a≤r≤b,α≤θ≤β},
where 0≤a<b and 0≤α<β≤2π.
In Cartesian coordinates, R is the region between two arcs of circles (radii a and b) and two radial lines (angles α and β) — sometimes called an annular sector.
Use the sliders below to explore different polar rectangles.
Polar Rectangle: R = {(r,θ) : a ≤ r ≤ b, α ≤ θ ≤ β}
True area = ½(β−α)(b²−a²) = 4.1233
Naive ∫∫ dr dθ = (β−α)(b−a) = 2.3562(wrong!)
2. Why we cannot use drdθ naively
A natural first guess for the polar area element might be drdθ (by analogy with dxdy). Let us check this against a known area.
Take the full disk of radius R: 0≤r≤R, 0≤θ≤2π. The area of a disk is πR2.
Naive attempt:
∫02π∫0Rdrdθ=∫02πRdθ=2πR.
This gives 2πR, not πR2. The naive element drdθ is wrong.
The interactive graph above displays the "true area" versus the "naive area" computed without the r factor — you can see they differ whenever a=0 or the sector is not degenerate.
3. The polar area element: dA=rdrdθ
The correct area element in polar coordinates is rdrdθ. Here is where it comes from.
Geometry of a thin polar rectangle. Consider the polar rectangle with r∈[r0,r0+Δr] and θ∈[θ0,θ0+Δθ].
Its area is the area of the outer sector minus the area of the inner sector:
ΔA=21(r0+Δr)2Δθ−21r02Δθ=2Δθ[(r0+Δr)2−r02].
Expanding:
ΔA=2Δθ[2r0Δr+(Δr)2]≈r0ΔrΔθas Δr→0.
In the limit, ΔA→rdrdθ. The factor of r accounts for the fact that arcs at large r are much longer than arcs near the origin.
dA=rdrdθ.
Verification. For the disk of radius R:
∫02π∫0Rrdrdθ=∫02π2R2dθ=πR2.✓
4. The change-of-variables formula for polar coordinates
Let f(x,y) be continuous on the polar rectangle R={a≤r≤b,α≤θ≤β}. Then
∬Rf(x,y)dA=∫αβ∫abf(rcosθ,rsinθ)rdrdθ.
How to use this:
Replace x→rcosθ and y→rsinθ in the integrand.
Replace dA→rdrdθ.
Set the limits: r from a to b, θ from α to β.
Example. Evaluate ∬D(x2+y2)dA where D is the disk x2+y2≤4.
In polar: x2+y2=r2, D corresponds to 0≤r≤2, 0≤θ≤2π.
The polar formula is a special case of the general change-of-variables theorem for double integrals.
Suppose (x,y)=T(u,v) is a smooth, one-to-one transformation from a region S in the (u,v)-plane to a region D in the (x,y)-plane. The Jacobian of T is the determinant
J=∂(u,v)∂(x,y)=det(xuyuxvyv).
The change-of-variables formula is
∬Df(x,y)dA=∬Sf(T(u,v))∣J∣dudv.
For polar coordinates x=rcosθ, y=rsinθ:
J=det(cosθsinθ−rsinθrcosθ)=rcos2θ+rsin2θ=r.
So ∣J∣=r, giving dA=rdrdθ. This confirms the geometric argument above.
5. General polar domains
We are not limited to polar rectangles. A general polar domain is described by
D={(r,θ):α≤θ≤β,h1(θ)≤r≤h2(θ)},
where h1 and h2 are continuous and 0≤h1(θ)≤h2(θ).
This is the polar analogue of a Type I region (sweep θ from α to β; for each θ, r runs from h1(θ) to h2(θ)).