Vector Fields

A vector field $F$ is a function that maps from $\mathbb{R^n}$ to $\mathbb{R}^n$, where $n\geq 2$. For $n=2$, we denote $F$ as

\[P(x,y)i+Q(x,y)j=\langle P(x,y),Q(x,y)\rangle\]

Here, $P$ and $Q$ are functions from $\mathbb{R}^2$ to $\mathbb{R}$ and they are the component functions for $F$. For $n=3$, $F$ is denoted as

\[P(x,y,z)i+Q(x,y,z)j+R(x,y,z)k=\langle P(x,y,z),Q(x,y,z),R(x,y,z)\rangle\]

We can visualize $F$ by drawing corresponding vectors in its space.

Example. $F(x,y)=-yi+xj=\langle -y,x\rangle$ are vectors perpendicular to $\langle x,y\rangle$. Then, the following graph visualizes $F$. (The following is a link to Desmos that draws the vector field $F$)

embed

Example. Gravitation can be considered a force, and force can be represented by vectors. Therefore, if we place a mass $M$ at the origin, each point in space will exert a force towards the origin. This mass induces a vector field in space, called gravitation field. So, what is this function $F$?

A force is defined by its scalar and direction(a unit vector). According to Newton’s Law of Gravity, the scalar of this force at a point $(x,y,z)$ is given by

\[\frac{mMG}{r^2}\]

In this formula, $m$ represents your mass, $G$ is the gravitational constant, and $r=\sqrt{x^2+y^2+z^2}$ is the distance from the point $(x,y,z)$ to the origin. The direction of gravity is the unit vector parallel to $-\langle x,y,z\rangle$, which is

\[\frac{\langle -x,-y,-z\rangle}{\sqrt{x^2+y^2+z^2}}.\]

Putting both information together, we have the gravitation field

\[F(x,y,z)=\frac{-xmMG}{(\sqrt{x^2+y^2+z^2})^3}i+\frac{-ymMG}{(\sqrt{x^2+y^2+z^2})^3}j+\frac{-zmMG}{(\sqrt{x^2+y^2+z^2})^3}k.\]

Example. The gradient $\nabla f$ of a differentiable function $f:\mathbb{R^2}\to \mathbb{R}$ is a vector field. This vector field is known as the gradient field. If $F$ is a gradient field and $\nabla f=F$, then $f$ is referred to as a potential function for $F$.

Line Integrals

Line Integrals with Respect to an Arc Length

A line integral signifies the integration of a function along a smooth curve $C$. We define the line integral using the Riemann sum, taking it as the limit. That is

\[\int_Cf(x,y)ds = \lim_{n\to\infty}\sum_{i=1}^nf(x_i^\ast,y_i^\ast)\Delta s_i.\]

The $s$ represents the length of the curve $C$. If $C$ can be parametrized by $(x(t),y(t))$ from $a$ to $b$, then we also have

\[s(t)=\int_{C_x} ds=\int_{a}^t\sqrt{x'(t)^2+y'(t)^2}dt\]

where $C_x$ represents a segment of the curve $C$ extending from $(x(a),y(a))$ to $(x(t),y(t))$. Clearly, it implies

\[ds = \sqrt{x'(t)^2+y'(t)^2}dt.\]

Therefore, it follows

\[\int_Cf(x,y)ds = \int_a^b f(x(t),y(t))\sqrt{x'(t)^2+y'(t)^2}dt.\]

Remark. If $C$ is a piecewise-smooth curve, i.e. $C$ is a disjoint union of finitely many smooth curves $C_1, C_2,\ldots, C_n$, then

\[\int_C fds = \int_{C_1}fds+\cdots\int_{C_n}fds.\]

Line Integral with respsect to $x$ and $y$

The definitions of line integrals along a smooth curve $C$ with respect to $x$ and $y$ are given by the limits

\[\begin{align*} \int_Cf(x,y)dy = \lim_{n\to\infty}\sum_{i=1}^nf(x_i^\ast,y_i^\ast)\Delta x_i\\\\ \int_Cf(x,y)dx = \lim_{n\to\infty}\sum_{i=1}^nf(x_i^\ast,y_i^\ast)\Delta y_i \end{align*}\]

where $\Delta s_i^2=\Delta x_i^2+\Delta y_i^2.$ If $C$ can be parametrized by $(x(t),t(t))$ from $a$ to $b$, then

\[\int_Cf(x(t),y(t))x'(t)dt\quad\text{and}\quad\int_Cf(x(t),y(t))y'(t)dt.\]

Let $F(x,y)=P(x,y)i+Q(x,y)j$. In this case, line integrals with respect to $x$ and $y$ often appear together. We then will write

\[\int_CP(x,y)dx+\int_CQ(x,y)dy=\int_CP(x,y)dx+Q(x,y)dy.\]

Remark. Line integrals along a smooth curve usually depend on the curve $C$, even if the curve has the same starting and ending points. It’s intriguing to know what kind of function $F$ has an integral that is independent of the curve and only depends on its starting and ending points.