Unit 9: Parametric Equations, Polar Coordinates, and Vector-Valued Functions
- Syllabus
- 2020
- Section
- —
- Level
- —

Parametric equations x=x(t) and y=y(t) assign both coordinates to the same parameter value. As t changes, the point (x(t),y(t)) traces a curve. Because both coordinates depend on t, their derivatives describe horizontal and vertical rates along that curve.
\frac{dy}{dt}=\frac{dy}{dx}\frac{dx}{dt}\quad\Longrightarrow\quad\frac{dy}{dx}=\frac{dy/dt}{dx/dt},\qquad \frac{dx}{dt}\ne0
Example: x=t2+1 and y=t3−3t. Then dx/dt=2t and dy/dt=3t2−3, so dy/dx=(3t2−3)/(2t). At t=2, the curve is at (5,2) and the slope is 9/4. The tangent line is therefore y−2=49(x−5).
Do not divide the coordinate functions themselves: dy/dx is the ratio of their derivatives. The quotient formula requires dx/dt=0 at the parameter value; when dx/dt=0, this calculation does not produce a finite tangent slope.
For a parametric curve, dy/dx is usually still written in terms of t. Differentiating it with respect to t gives change in slope per unit t, not per unit x. Divide by dx/dt once more to convert that rate into the second derivative with respect to x.
\frac{d^2y}{dx^2}=\frac{\dfrac{d}{dt}\left(\dfrac{dy}{dx}\right)}{dx/dt},\qquad \frac{dx}{dt}\ne0
Example: x=t2+1 and y=t3−3t. Then dy/dx=(3t2−3)/(2t)=23(t−1/t). Hence d(dy/dx)/dt=23(1+1/t2), and d2y/dx2=4t33(t2+1). At t=2, d2y/dx2=15/32>0, so the curve is concave up there.
Stopping after d(dy/dx)/dt leaves a rate with respect to t; it is not d2y/dx2. The conversion formula requires dx/dt=0, and any parameter values excluded while simplifying must remain excluded from the final expression.
For x=x(t) and y=y(t), a tiny parameter change produces horizontal change (dx/dt)dt and vertical change (dy/dt)dt. The Pythagorean magnitude of these components is the nonnegative rate at which length is traced.
L=\int_\alpha^\beta\sqrt{\left(\frac{dx}{dt}\right)^2+\left(\frac{dy}{dt}\right)^2},dt
Example: x=3cost and y=3sint for 0≤t≤π/2 trace one quarter of a circle of radius 3. Here dx/dt=−3sint and dy/dt=3cost, so the speed is 9sin2t+9cos2t=3. Thus L=∫0π/23dt=3π/2 units.
Use the parameter bounds, not automatically the curve's x-coordinates. The square root is the magnitude of both component rates, so neither sign can make length negative. If the parameterization retraces a segment, this integral counts the retraced distance again.
A planar vector-valued function r(t)=⟨x(t),y(t)⟩ gives both coordinates of a moving point at the same parameter value. Its derivative records the instantaneous horizontal and vertical rates together, so it points tangent to the path when it is nonzero.
\mathbf r'(t)=\left\langle x'(t),y'(t)\right\rangle
Example: r(t)=⟨t2−1,t3+2t⟩. Then r′(t)=⟨2t,3t2+2⟩. At t=1, the point is r(1)=⟨0,3⟩ and the derivative is r′(1)=⟨2,5⟩, a tangent direction and, for a position function, the instantaneous velocity vector.
The derivative is a vector, not the sum or magnitude of the component derivatives. Its magnitude ∥r′(t)∥=[x′(t)]2+[y′(t)]2 is a separate scalar quantity; for motion, that magnitude is speed.
If r′(t)=v(t)=⟨vx(t),vy(t)⟩, then the position vector is found by integrating the horizontal and vertical rates separately. An initial position fixes both constants and selects the one particular solution that describes the motion.
\mathbf r(t)=\mathbf r(t_0)+\int_{t_0}^{t}\mathbf v(u),du=\mathbf r(t_0)+\left\langle\int_{t_0}^{t}v_x(u),du,\int_{t_0}^{t}v_y(u),du\right\rangle
Example: suppose v(t)=⟨2t,3⟩ and r(1)=⟨4,−2⟩. Then
r(t)=⟨4,−2⟩+⟨∫1t2udu,∫1t3du⟩=⟨4,−2⟩+⟨t2−1,3t−3⟩=⟨t2+3,3t−5⟩.
Indeed, r′(t)=⟨2t,3⟩ and r(1)=⟨4,−2⟩, so both the rate vector and initial condition are satisfied.
Do not integrate the speed ∥v(t)∥ when the question asks for position: speed is scalar and its integral gives distance traveled. Integrating the velocity vector gives vector displacement. With indefinite integrals, remember that the two components can have different constants before the initial position determines them.
For planar motion, r(t)=⟨x(t),y(t)⟩ is position. Differentiate once for velocity and twice for acceleration; take the magnitude of velocity for speed. Integrate velocity to obtain displacement, or integrate speed to obtain total distance traveled.
\mathbf v(t)=\mathbf r'(t)=\langle x'(t),y'(t)\rangle,\qquad \text{speed}=|\mathbf v(t)|=\sqrt{[x'(t)]^2+[y'(t)]^2},\qquad \mathbf a(t)=\mathbf v'(t)
\text{displacement on }[a,b]=\int_a^b\mathbf v(t),dt=\mathbf r(b)-\mathbf r(a),\qquad \text{distance}=\int_a^b|\mathbf v(t)|,dt
Example: a particle's position in meters is r(t)=⟨t2+1,t3−3t⟩, with t in seconds. Then v(t)=⟨2t,3t2−3⟩ m/s and a(t)=⟨2,6t⟩ m/s2. At t=1, its position is ⟨2,−2⟩ m, velocity is ⟨2,0⟩ m/s, speed is 22+02=2 m/s, and acceleration is ⟨2,6⟩ m/s2. From t=1 to t=2, displacement is r(2)−r(1)=⟨5,2⟩−⟨2,−2⟩=⟨3,4⟩ m.
Displacement is the net position change, so opposite motions can cancel; total distance cannot cancel because speed is nonnegative. In the example, ∥⟨3,4⟩∥=5 m is the magnitude of displacement, not automatically the distance traveled along the curved path.
For a polar curve r=f(θ), write x=rcosθ and y=rsinθ. Both coordinates depend on θ, so differentiate each with the product rule and then use the parametric derivative rule.
\frac{dx}{d\theta}=\frac{dr}{d\theta}\cos\theta-r\sin\theta,\qquad \frac{dy}{d\theta}=\frac{dr}{d\theta}\sin\theta+r\cos\theta
\frac{dy}{dx}=\frac{dy/d\theta}{dx/d\theta},\qquad \frac{d^2y}{dx^2}=\frac{d}{d\theta}\left(\frac{dy}{dx}\right)\bigg/\frac{dx}{d\theta}
Example: let r=2cosθ. Then dr/dθ=−2sinθ, dx/dθ=−2sin(2θ), and dy/dθ=2cos(2θ). Therefore dy/dx=−cot(2θ). At θ=π/4, the slope is 0. Also,
dx2d2y=−2sin(2θ)2csc2(2θ)=−csc3(2θ),
so at θ=π/4 the second derivative is −1, indicating local concave-down behavior with respect to x.
Do not use dr/dθ as the Cartesian slope: r measures radial distance, not vertical position. The quotient for dy/dx requires dx/dθ=0; when dx/dθ=0, analyze the component derivatives separately because the tangent may be vertical or the point may require further investigation.
A small change dθ sweeps out a thin sector of radius r=f(θ). Because a sector with angle dθ has area approximately 21r2dθ, adding all sectors gives the area bounded by one polar curve.
A=\frac12\int_{\alpha}^{\beta}[r(\theta)]^2,d\theta
Example: the cardioid r=1+cosθ is traced once for 0≤θ≤2π. Its enclosed area is
A=21∫02π(1+cosθ)2dθ=21∫02π(1+2cosθ+cos2θ)dθ.
Over a full period, ∫02π1dθ=2π, ∫02π2cosθdθ=0, and ∫02πcos2θdθ=π. Therefore A=21(3π)=23π square units.
Do not compute ∫rdθ: polar area depends on r2 and includes the factor 1/2. Also, bounds describe a traversal, not merely the visible width of a sketch; if the interval traces the same region twice, the integral double-counts its area.
For a fixed angle θ, the region between two polar curves runs radially from an inner radius to an outer radius. Its thin-slice area is therefore the outer sector minus the inner sector, which produces a difference of squared radii.
A=\frac12\int_{\alpha}^{\beta}\left([r_{\text{outer}}(\theta)]^2-[r_{\text{inner}}(\theta)]^2\right),d\theta
Example: find the area inside r=2cosθ but outside r=1. Intersections satisfy 2cosθ=1, giving θ=±π/3. On [−π/3,π/3], 2cosθ is the outer radius. Thus
A=21∫−π/3π/3(4cos2θ−1)dθ=3π+23
square units. The integrand is nonnegative on this interval, consistent with the chosen outer and inner curves.
Do not subtract the radii first and then square: (router−rinner)2 is not a sector-area difference. Also, one curve may not remain outer across the whole region; an unsplit integral can create negative contributions or cancel genuine area.