Calculus III: Double Integrals

Double integrals extend the powerful idea of integration into two dimensions, transforming how we calculate quantities like mass, charge, and heat flow over surfaces and volumes. For engineers, this tool is indispensable for analyzing stress distributions, fluid dynamics, and electromagnetic fields, moving beyond simple one-dimensional models to accurately describe complex, real-world systems.

From Summation to Integration: The Foundation

The concept of the double integral originates from the need to sum a quantity over a two-dimensional area. Consider a thin metal plate in the $xy$-plane with variable density $f(x,y)$ (in kg/m²). To find its total mass, we partition the plate's region $R$ into small rectangular subregions. For each subrectangle with area $\Delta A$, we choose a sample point $(x_{ij}^{\ast },y_{ij}^{\ast })$, approximate the mass as $f(x_{ij}^{\ast },y_{ij}^{\ast })\Delta A$, and sum over all rectangles. The double integral is defined as the limit of this Riemann sum as the number of subrectangles increases infinitely, shrinking their dimensions to zero. Formally, we write:

$${\iint }_{R}f(x,y)dA=\underset{m,n\to \infty }{\lim }\sum _{i=1}^m\sum _{j=1}^{n}f(x_{ij}^{\ast },y_{ij}^{\ast })\Delta A.$$

If $f(x,y)\ge 0$, this integral represents the volume of the solid between the surface $z=f(x,y)$ and the region $R$ in the $xy$-plane. This geometric interpretation is crucial for visualization: the double integral "slices" the solid into thin columns, sums their volumes, and takes the limit to get the exact volume.

Computation: Iterated Integrals and Fubini's Theorem

Actually computing the limit of a Riemann sum is impractical. Thankfully, Fubini's Theorem provides the essential computational bridge. It states that for a continuous function $f(x,y)$ over a rectangular region $R=[a,b]\times [c,d]$, the double integral can be computed as an iterated integral:

$${\iint }_{R}f(x,y)dA={\int }_{x=a}^b{\int }_{y=c}^{d}f(x,y)dydx={\int }_{y=c}^d{\int }_{x=a}^{b}f(x,y)dxdy.$$

You integrate with respect to one variable at a time, treating the other as a constant, and work from the inside out. The parentheses are implied: ${\int }_a^b\left({\int }_c^{d}f(x,y)dy\right)dx$. For example, to compute ${\iint }_R(x^2+y)dA$ over $R=[0,2]\times [1,3]$, we proceed step-by-step:

$$\begin{array}{rl}{\int }_{x=0}^2{\int }_{y=1}^3(x^2+y)dydx& ={\int }_0^2{\left[x^{2}y+\frac{1}{2}{y}^2\right]}_{y=1}^{3}dx\\ & ={\int }_0^2\left((3x^2+\frac{9}{2})-(x^2+\frac{1}{2})\right)dx\\ & ={\int }_0^2(2x^2+4)dx={\left[\frac{2}{3}{x}^3+4x\right]}_0^2=\frac{16}{3}+8=\frac{40}{3}.\end{array}$$

The order of integration ($dydx$ vs. $dxdy$) is interchangeable over a rectangle, a powerful feature of Fubini's Theorem.

Integrating Over General Regions

Real engineering problems rarely involve simple rectangles. We must integrate over general regions, which fall into two primary types. A Type I region is bounded vertically: it lies between two functions of $x$, so $a\le x\le b$ and $g_1(x)\le y\le g_2(x)$. A Type II region is bounded horizontally: $c\le y\le d$ and $h_1(y)\le x\le h_2(y)$. The limits of integration for the inner integral become these bounding functions.

Consider the region bounded by $y=x^2$ and $y=2x$. To set up the integral for a function $f(x,y)$, you must first sketch and choose integration order. As a Type I region, $y$ varies from the bottom curve $y=x^2$ to the top curve $y=2x$. The $x$-limits are the intersection points, found by solving $x^2=2x$, giving $x=0$ and $x=2$. The iterated integral is:

$${\int }_{x=0}^2{\int }_{y=x^2}^{2x}f(x,y)dydx.$$

As a Type II region, you solve for $x$ in terms of $y$: $x=y/2$ (right) and $x=\sqrt{y}$ (left), with $y$ from 0 to 4. The setup is:

$${\int }_{y=0}^4{\int }_{x=y/2}^{\sqrt{y}}f(x,y)dxdy.$$

The choice of order is critical; one order often leads to a much simpler integral than the other, or may be necessary if the antiderivative is only expressible with a certain variable order.

Core Applications: Area, Volume, and Average Value

The double integral provides elegant solutions to fundamental geometric and analytic problems.

• Area: The area of a planar region $R$ is found by integrating the constant function 1 over $R$: $\text{Area}(R)={\iint }_{R}1dA$.
• Volume: As introduced, if $f(x,y)\ge 0$, the volume under the surface $z=f(x,y)$ and above $R$ is $V={\iint }_{R}f(x,y)dA$. This directly models material deposition, fluid storage, or structural fill.
• Average Value: The average value of a multivariable function $f$ over a region $R$ is a direct analog to the single-variable case. It is given by:

$$f_{\text{avg}}=\frac{1}{\text{Area}(R)}{\iint }_{R}f(x,y)dA.$$ This is essential for finding the mean temperature over a plate, the average stress over a cross-section, or the average density of an object.

Common Pitfalls

1. Misapplying Fubini's Theorem: Fubini's Theorem guarantees equality of iterated integrals for continuous functions over rectangles. Over general regions, the limits change with the order, and the theorem must be applied with careful attention to those bounds. Assuming you can freely swap limits without redefining the region is a major error.
2. Incorrectly Identifying Region Bounds: The most common computational mistake is misidentifying the curves that form the upper and lower (or left and right) bounds for the inner integral. Always sketch the region. For Type I, ask: "For a fixed $x$, where does $y$ enter and leave the region?" The answer must be in the form $y=\text{[function of x]}$.
3. Choosing the Harder Order of Integration: Some integrals are impossible (or very difficult) to evaluate in one order. If you encounter an inner integral you cannot antidifferentiate (e.g., $\int e^{x^2}dx$), stop and consider switching the order of integration. Sketch the region and re-describe it in the opposite type.
4. Confusing Area with Volume: When asked for the area of a region, you integrate $1dA$. When asked for the volume under a surface, you integrate $f(x,y)dA$. Using $f(x,y)$ to find area will give an incorrect, function-dependent result.

Summary

• The double integral ${\iint }_{R}f(x,y)dA$ is defined as the limit of a two-dimensional Riemann sum and fundamentally represents a summed quantity (like mass) or a volume.
• Fubini's Theorem allows computation via iterated integrals. On rectangular regions, the order of integration ($dydx$ or $dxdy$) can be chosen freely.
• Over general regions (Type I: vertically simple, Type II: horizontally simple), the limits of the inner integral are functions that describe the region's boundaries. Choosing the optimal integration order is a key problem-solving skill.
• Primary applications include calculating the area of a region (${\iint }_{R}1dA$), the volume under a surface (${\iint }_{R}f(x,y)dA$ where $f\ge 0$), and the average value of a function over a region.
• Success depends on accurately sketching the region of integration, correctly establishing bounds for iterated integrals, and strategically selecting the order of integration to simplify the antiderivative process.