Calculus III: Multivariate Functions, Partial Differentiation, and Multiple Integration

Section 15.1: Functions of Several Variables

  • Multivariate Functions: Unlike vector-valued functions (which have one independent variable and several dependent variables), multivariate or multivariable functions possess one dependent variable and several independent variables.

  • Functions of Two Independent Variables: Most properties are developed for functions with two independent variables, xx and yy, and a dependent variable zz.

    • Explicit Form: z=f(x,y)z = f(x, y).
    • Implicit Form: F(x,y,z)=0F(x, y, z) = 0.
  • Domains and Ranges:

    • The Domain is the collection of all ordered pairs (x,y)(x, y) in the plane that produce a real output. This differs from single-variable calculus where the domain is an interval of single values.
    • The Range is the set of resulting real outputs.
  • Graphs of Functions of Two Independent Variables: The graph is the set of all points (x,y,z)(x, y, z) satisfying z=f(x,y)z = f(x, y).

    • Vertical Line Test: A relation F(x,y,z)=0F(x, y, z) = 0 is a function if every line parallel to the zz-axis intersects the graph at most once.
  • Level Curves:

    • A contour curve is a path on a surface where the elevation (the zz-value) is a constant z0z_0.
    • When a contour curve is projected onto the xyxy-plane, it produces the equation f(x,y)=z0f(x, y) = z_0, known as a level curve.
    • A set of level curves provides a 2D representation of a 3D surface.
    • Proximity Rule: Closely spaced level curves indicate that the surface is changing more rapidly (steeper), while spread-out curves indicate a slower change.
  • Functions of Three or More Independent Variables:

    • Algebraic properties and definitions of domain/range extend naturally.
    • Visualization: A function of three independent variables has a graph in four dimensions, making it difficult to visualize. Graph work is generally limited to two independent variables, though algebra and calculus calculations apply to any number of variables.

Section 15.2: Limits and Continuity

  • Definition of a Limit: A function of two independent variables has a limit LL if, as P(x,y)P(x, y) approaches a fixed point P0(a,b)P_0(a, b), the distance ∣f(x,y)−L∣|f(x, y) - L| can be made arbitrarily small for all points PP in the domain sufficiently close to P0P_0.

    • Notation: lim⁡(x,y)→(a,b)f(x,y)=L\lim_{(x, y) \to (a, b)} f(x, y) = L
  • Single Variable vs. Two-Variable Limits:

    • In single-variable calculus (y=f(x)y = f(x)), a limit exists if the left-hand and right-hand limits match (only two directions).
    • In multivariable calculus, a point on a surface can be approached from infinitely many paths.
    • Existence Requirement: For a limit to exist, the function must approach the same value LL from every possible path.
  • Basic Limit Laws:

    • Limits can be evaluated using substitution if the function is continuous.
    • If substitution results in an indeterminate form, algebraic techniques (factoring, canceling, rationalization) should be employed.
  • Showing a Limit Does Not Exist:

    • To prove a limit does not exist, find two different paths in the domain that lead to different values being approached.
    • Critical Warning: Finding two paths that approach the same value is not sufficient to prove a limit exists, as two paths do not represent all paths.
  • Continuity:

    • A function is continuous at (a,b)(a, b) if lim⁡(x,y)→(a,b)f(x,y)=f(a,b)\lim_{(x, y) \to (a, b)} f(x, y) = f(a, b).
    • This core definition and the rules for composite functions remain the same as single-variable calculus.
  • Topological Definitions:

    • Interior Point: A point where there exists an open disk centered at the point contained entirely within the set.
    • Boundary Point: A point where every open disk centered at the point contains both points inside and outside the set.
    • Open Set: A set consisting entirely of interior points.
    • Closed Set: A set that contains all its boundary points.

Section 15.3: Partial Derivatives

  • Concept: If standing on a surface at point P(0,0)P(0, 0), the slope depends on the direction of travel (e.g., walking East/up-hill vs. North/down-hill).

    • Partial Derivative with respect to xx (fxf_x): Fix yy as a constant and differentiate with respect to xx. This represents the slope of the surface in the xx-direction.
    • Partial Derivative with respect to yy (fyf_y): Fix xx as a constant and differentiate with respect to yy. This represents the slope in the yy-direction.
  • Limit Definitions:

    • fx(a,b)=lim⁡h→0f(a+h,b)−f(a,b)hf_x(a, b) = \lim_{h \to 0} \frac{f(a+h, b) - f(a, b)}{h}
    • fy(a,b)=lim⁡h→0f(a,b+h)−f(a,b)hf_y(a, b) = \lim_{h \to 0} \frac{f(a, b+h) - f(a, b)}{h}
  • Notation:

    • Leibniz notation uses the ∂\partial (del) symbol: ∂z∂x\frac{\partial z}{\partial x} or ∂f∂y\frac{\partial f}{\partial y}.
    • Subscript notation: fx(a,b)f_x(a, b) or zyz_y.
  • Higher-Order Partial Derivatives:

    • Pure Partials: Differentiating with respect to the same variable repeatedly (e.g., fxx,fyyf_{xx}, f_{yy}).
    • Mixed Partials: Differentiating with respect to different variables in sequence (e.g., fxyf_{xy} or fyxf_{yx}).
  • Mixed Partial Notation & Ordering:

    • Subscript Notation: Differentiate from left to right (e.g., fxyf_{xy} means differentiate by xx first, then yy).
    • Leibniz Notation: Differentiate from right to left (e.g., ∂2f∂x∂y\frac{\partial^2 f}{\partial x ∂ y} means differentiate by yy first, then xx).
  • Equality of Mixed Partials (Clairaut's Theorem): If fxyf_{xy} and fyxf_{yx} are continuous on an open set, then they are equal (fxy=fyxf_{xy} = f_{yx}). This applies to higher orders, such as fxxy=fxyx=fyxxf_{xxy} = f_{xyx} = f_{yxx}, provided continuity is maintained.

  • Differentiability:

    • A function is differentiable at (a,b)(a, b) if partial derivatives exist in a neighborhood and are continuous.
    • Theorem: If a function is differentiable at a point, it is also continuous at that point.

Section 15.4: Chain Rules

  • Chain Rule (One Independent Variable): If z=f(x,y)z = f(x, y) where x=x(t)x = x(t) and y=y(t)y = y(t), then zz is ultimately a function of the single independent variable tt.

    • The Tree Diagram: zz branches to xx and yy (partial); xx and yy branch to tt (ordinary).
    • Formula: dzdt=∂z∂xdxdt+∂z∂ydydt\frac{dz}{dt} = \frac{\partial z}{\partial x} \frac{dx}{dt} + \frac{\partial z}{\partial y} \frac{dy}{dt}
    • Definitions: zz is the dependent variable, tt is the independent variable, and x,yx, y are intermediate variables.
  • Chain Rule (Several Independent Variables): If z=f(x,y)z = f(x, y) where x=x(s,t)x = x(s, t) and y=y(s,t)y = y(s, t).

    • Formula for ss: ∂z∂s=∂z∂x∂x∂s+∂z∂y∂y∂s\frac{\partial z}{\partial s} = \frac{\partial z}{\partial x} \frac{\partial x}{\partial s} + \frac{\partial z}{\partial y} \frac{\partial y}{\partial s}
    • Formula for tt: ∂z∂t=∂z∂x∂x∂t+∂z∂y∂y∂t\frac{\partial z}{\partial t} = \frac{\partial z}{\partial x} \frac{\partial x}{\partial t} + \frac{\partial z}{\partial y} \frac{\partial y}{\partial t}
  • Implicit Differentiation (Partial Style): Given an implicit relation F(x,y)=0F(x, y) = 0, where yy is a function of xx:

    • ∂F∂xdxdx+∂F∂ydydx=0\frac{\partial F}{\partial x} \frac{dx}{dx} + \frac{\partial F}{\partial y} \frac{dy}{dx} = 0
    • Resulting formula: dydx=−FxFy\frac{dy}{dx} = -\frac{F_x}{F_y}

Section 15.5: Directional Derivatives and the Gradient

  • Directional Derivatives: Partial derivatives only quantify change parallel to the axes. The directional derivative calculates the rate of change in any direction specified by a unit vector u=⟨u1,u2⟩u = \langle u_1, u_2 \rangle.

    • Limit Definition: Duf(a,b)=lim⁡h→0f(a+hu1,b+hu2)−f(a,b)hD_u f(a, b) = \lim_{h \to 0} \frac{f(a + h u_1, b + h u_2) - f(a, b)}{h}
    • Calculation Theorem: If ff is differentiable, then Duf(a,b)=fx(a,b)u1+fy(a,b)u2D_u f(a, b) = f_x(a, b) u_1 + f_y(a, b) u_2.
  • The Gradient Vector:

    • In 2D: ∇f(x,y)=⟨fx,fy⟩\nabla f(x, y) = \langle f_x, f_y \rangle
    • In 3D: ∇f(x,y,z)=⟨fx,fy,fz⟩=fxi+fyj+fzk\nabla f(x, y, z) = \langle f_x, f_y, f_z \rangle = f_x \mathbf{i} + f_y \mathbf{j} + f_z \mathbf{k}
    • Relationship to Directional Derivative: Duf=∇f⋅uD_u f = \nabla f \cdot \mathbf{u}.
  • Interpretations of the Gradient:

    • The direction of maximum rate of increase is ∇f\nabla f. The magnitude of this increase is ∣∇f∣|\nabla f|.
    • The direction of maximum rate of decrease is −∇f-\nabla f. The magnitude of this rate is −∣∇f∣-|\nabla f|.
    • The directional derivative is zero in any direction orthogonal to ∇f\nabla f.
  • Gradients and Level Curves/Surfaces:

    • The gradient vector ∇f(a,b)\nabla f(a, b) is always orthogonal to the level curve f(x,y)=kf(x, y) = k at point (a,b)(a, b).
    • In 3D, the gradient is orthogonal to level surfaces.

Section 15.6: Tangent Planes, Linear Approximations, and Differentials

  • Tangent Plane for Implicit Relations: For a surface F(x,y,z)=0F(x, y, z) = 0, the gradient ∇F(a,b,c)\nabla F(a, b, c) serves as the normal vector to the tangent plane at point (a,b,c)(a, b, c).

    • Equation: Fx(a,b,c)(x−a)+Fy(a,b,c)(y−b)+Fz(a,b,c)(z−c)=0F_x(a, b, c)(x - a) + F_y(a, b, c)(y - b) + F_z(a, b, c)(z - c) = 0
  • Tangent Plane for Explicit Surfaces: For z=f(x,y)z = f(x, y), the gradient of the surface rewritten as F(x,y,z)=z−f(x,y)=0F(x, y, z) = z - f(x, y) = 0 gives properties Fx=−fxF_x = -f_x, Fy=−fyF_y = -f_y, and Fz=1F_z = 1.

    • Formula: z−f(a,b)=fx(a,b)(x−a)+fy(a,b)(y−b)z - f(a, b) = f_x(a, b)(x - a) + f_y(a, b)(y - b)
  • Linear Approximation: The tangent plane can linearly approximate the value of a function near a point (a,b)(a, b).

    • Equation: L(x,y)=f(a,b)+fx(a,b)(x−a)+fy(a,b)(y−b)L(x, y) = f(a, b) + f_x(a, b)(x - a) + f_y(a, b)(y - b)
  • The Differential (dzdz): Approximates the change in the dependent variable Δz\Delta z when moving from point (x,y)(x, y) to (x+dx,y+dy)(x+dx, y+dy).

    • Equation: dz=fx(x,y)dx+fy(x,y)dydz = f_x(x, y) dx + f_y(x, y) dy

Section 15.7: Local Maxima and Minima

  • Critical Points: A point (a,b)(a, b) is a critical point if fx(a,b)=0f_x(a, b) = 0 and fy(a,b)=0f_y(a, b) = 0, or if either partial derivative is undefined.

  • Saddle Points: A point (a,b)(a, b) is a saddle point if in every open disk centered at (a,b)(a, b), there exist points where f(x,y)>f(a,b)f(x, y) > f(a, b) and points where f(x,y)<f(a,b)f(x, y) < f(a, b).

  • Second Derivative Test: Let (a,b)(a, b) be a critical point and define the determinant D=fxx(a,b)fyy(a,b)−[fxy(a,b)]2D = f_{xx}(a, b) f_{yy}(a, b) - [f_{xy}(a, b)]^2.

    • Local Minimum: If D>0D > 0 and fxx(a,b)>0f_{xx}(a, b) > 0.
    • Local Maximum: If D>0D > 0 and fxx(a,b)<0f_{xx}(a, b) < 0.
    • Saddle Point: If D<0D < 0.
    • Inconclusive: If D=0D = 0.

Section 16.1: Double Integrals over Rectangular Regions

  • Riemann Sum Formulation: A rectangular region RR in the xyxy-plane is partitioned into nn subrectangles with area ΔAk=ΔxkΔyk\Delta A_k = \Delta x_k \Delta y_k. The volume of boxes under surface z=f(x,y)z = f(x, y) approximates the volume under the surface.

    • Double Integral Definition: ∬Rf(x,y) dA=lim⁡Δ→0∑k=1nf(xk∗,yk∗)ΔAk\iint_R f(x, y) \, dA = \lim_{\Delta \to 0} \sum_{k=1}^n f(x_k^*, y_k^*) \Delta A_k
  • Iterated Integrals: Double integrals are evaluated by performing partial integration repeatedly.

    1. Evaluate the inner integral while fixing outer variables.
    2. Evaluate the outer integral with respect to the remaining variable.
  • Fubini's Theorem: For a rectangular region, the order of integration does not matter:   ∬Rf(x,y) dA=∫ab∫cdf(x,y) dy dx=∫cd∫abf(x,y) dx dy\iint_R f(x, y) \, dA = \int_a^b \int_c^d f(x, y) \, dy \, dx = \int_c^d \int_a^b f(x, y) \, dx \, dy

  • Average Value: The average value of ff over a rectangular region RR is given by:   Avg(f)=1Area(R)∬Rf(x,y) dA\text{Avg}(f) = \frac{1}{\text{Area}(R)} \iint_R f(x, y) \, dA

Section 16.2: Double Integrals over General Regions

  • Type I Regions: Bounded above and below by continuous functions of xx: g(x)≤y≤h(x)g(x) \leq y \leq h(x) for a≤x≤ba \leq x \leq b.

    • Setup: ∫ab∫g(x)h(x)f(x,y) dy dx\int_a^b \int_{g(x)}^{h(x)} f(x, y) \, dy \, dx
  • Type II Regions: Bounded left and right by continuous functions of yy: g(y)≤x≤h(y)g(y) \leq x \leq h(y) for c≤y≤dc \leq y \leq d.

    • Setup: ∫cd∫g(y)h(y)f(x,y) dx dy\int_c^d \int_{g(y)}^{h(y)} f(x, y) \, dx \, dy
  • Critical Rules for Non-Rectangular Regions:

    1. Variables in Limits: The variable of integration cannot appear in the limits of the same integral (e.g., if integrating dydy, limits cannot contain yy), nor can they appear in any outer limits.
    2. Switching Order: You cannot simply swap the differentials and limits. You must sketch the region and redefine the bounding functions (from xx-perspective to yy-perspective or vice-versa).
  • Volume Between Two Surfaces: If f(x,y)≥g(x,y)f(x, y) ≥ g(x, y), the volume is:   V=∬R(f(x,y)−g(x,y)) dAV = \iint_R (f(x, y) - g(x, y)) \, dA

  • Area Calculated by Double Integration: If the integrand is set to f(x,y)=1f(x, y) = 1, the numerical result of the double integral equals the area of region RR.

Section 16.3: Double Integrals in Polar Coordinates

  • Polar Rectangle: A region defined by a≤r≤ba \leq r \leq b and α≤θ≤β\alpha \leq \theta \leq \beta.

  • Conversion to Polar Integration: When converting from Cartesian to polar, the differential dAdA transforms to r dr dθr \, dr \, d\theta.

    • Formula: ∬Rf(x,y) dA=∫αβ∫abf(rcos⁡(θ),rsin⁡(θ))r dr dθ\iint_R f(x, y) \, dA = \int_{\alpha}^{\beta} \int_a^b f(r \cos(\theta), r \sin(\theta)) r \, dr \, d\theta
    • Critical Note: Do not forget the extra factor of rr in the integrand.
  • Coordinate Relations:

    • x=rcos⁡(θ)x = r \cos(\theta)
    • y=rsin⁡(θ)y = r \sin(\theta)
    • x2+y2=r2x^2 + y^2 = r^2