Vector Fields

A vector field assigns a vector to every point of a region in space. In symbols, $F : U \subseteq R^{n} \to R^{n}$ . Vector fields model flows, forces, and gradients — velocity of a fluid, the electric field, the gradient of temperature. They are the central objects of vector calculus, manipulated via the operators gradient, divergence, and curl, and integrated via line and surface integrals. Foundational work: Gauss, Green, Stokes, Maxwell, and Gibbs/Heaviside (19th century).

A vector field on an open set $U \subseteq R^{n}$ is a function $F$ that attaches to each point $x \in U$ a vector $F (x) \in R^{n}$ , written in components as $F = (F_{1}, \dots, F_{n})$ . Vector fields matter because nearly every continuous physical influence that has both magnitude and direction at each location — fluid velocity, gravitational pull, the electromagnetic field — is naturally a vector field, and the entire machinery of Divergence, Curl, and integral theorems exists to extract physical meaning from them. A surprising fact: in three dimensions, any smooth vector field that decays at infinity can be uniquely split into a "curl-free" part (a gradient) plus a "divergence-free" part (a curl) — this is the Helmholtz decomposition, which silently underlies electromagnetism and fluid dynamics.

The Core Insight

A scalar field gives you one number at each point (temperature, pressure); a vector field gives you a whole arrow — magnitude and direction — at each point. The geometry of how those arrows spread, swirl, and align is captured entirely by three derivatives: the gradient ( $\nabla f$ , building a field from a scalar), the divergence ( $\nabla \cdot F$ , measuring outflow), and the curl ( $\nabla \times F$ , measuring rotation). Mastering vector fields is mastering the dictionary between these local derivatives and global integrals.

Formal Statement

Vector Field

Let $U \subseteq R^{n}$ be an open set. A vector field on $U$ is a map

F : U \to R^{n}, F (x) = (F_{1} (x), F_{2} (x), \dots, F_{n} (x)),

⟳this is a definition;

F_{i} : U \to R

are the component (scalar) functions, and

x = (x_{1}, \dots, x_{n})

We say $F$ is continuous (resp. $C^{1}$ , $C^{\infty}$ ) if every component $F_{i}$ is continuous (resp. continuously differentiable, smooth). In three dimensions the standard notation, using basis vectors $i, j, k$ , is

F (x, y, z) = P (x, y, z) i + Q (x, y, z) j + R (x, y, z) k .

⟳rewrite

(F_{1}, F_{2}, F_{3})

P, Q, R

— the traditional letters for the three components in vector calculus.

The Del Operator and Its Three Children

Define the symbolic vector operator $\nabla$ ("del" or "nabla") in $R^{3}$ by

\nabla = (\frac{\partial}{\partial x}, \frac{\partial}{\partial y}, \frac{\partial}{\partial z}) .

⟳

\nabla

is not a number but a recipe — it acts on whatever stands to its right.

The gradient of a scalar field $f$ , the divergence and curl of a vector field $F = (P, Q, R)$ are

\nabla f = (\frac{\partial f}{\partial x}, \frac{\partial f}{\partial y}, \frac{\partial f}{\partial z}),

⟳apply each component of

\nabla

to the scalar

f

— produces a vector field.

\nabla \cdot F = \frac{\partial P}{\partial x} + \frac{\partial Q}{\partial y} + \frac{\partial R}{\partial z},

⟳formal dot product of

\nabla

with

F

— produces a scalar field.

\nabla \times F = (\frac{\partial R}{\partial y} - \frac{\partial Q}{\partial z}, \frac{\partial P}{\partial z} - \frac{\partial R}{\partial x}, \frac{\partial Q}{\partial x} - \frac{\partial P}{\partial y}) .

⟳formal cross product

\nabla \times F

via the determinant of the matrix with rows

(i, j, k)

(\partial_{x}, \partial_{y}, \partial_{z})

(P, Q, R)

— produces a vector field.

Conservative (Gradient) Field

A vector field $F$ on $U$ is conservative if there exists a scalar field $f$ (a potential) with

F = \nabla f on U .

⟳definition; the sign convention

F = - \nabla ϕ

is common in physics (force = minus gradient of potential energy).

Fundamental Theorem for Line Integrals

If $F = \nabla f$ is continuous on $U$ and $C$ is a piecewise-smooth curve from $a$ to $b$ inside $U$ , then the line integral depends only on the endpoints:

\int_{C} F \cdot d r = \int_{C} \nabla f \cdot d r = f (b) - f (a) .

⟳parametrize

C

r (t)

, apply the chain rule to

\frac{d}{d t} f (r (t)) = \nabla f \cdot r^{'} (t)

, then integrate using the single-variable FTC.

Consequently, line integrals of conservative fields around any closed loop vanish.

Curl Test for Conservativeness

Let $F$ be $C^{1}$ on a simply connected open set $U \subseteq R^{3}$ . Then

F = \nabla f for some f ⟺ \nabla \times F = 0 on U .

⟳the forward direction is the identity

\nabla \times (\nabla f) = 0

(mixed partials commute); the converse uses simple-connectedness via Stokes' Theorem to construct

f

by path integration.

Why It Holds / Derivation

The Two Fundamental Vector Identities

We prove $\nabla \times (\nabla f) = 0$ and $\nabla \cdot (\nabla \times F) = 0$ , the backbone of the curl/divergence tests.

Step 1 — Curl of a gradient. Let

f

C^{2}

so mixed partials commute (Clairaut's theorem). Compute the first component of

\nabla \times (\nabla f)

using [6] with

(P, Q, R) = (f_{x}, f_{y}, f_{z})

(\nabla \times \nabla f)_{1} = \frac{\partial f _{z}}{\partial y} - \frac{\partial f _{y}}{\partial z} = f_{z y} - f_{y z} = 0.

⟳substitute

R = f_{z}, Q = f_{y}

into the first slot of [6], then invoke

f_{z y} = f_{y z}

Step 2. By symmetry the other two components also vanish, so

\nabla \times (\nabla f) = 0

. Step 3 — Divergence of a curl. With

F = (P, Q, R)

apply [5] to the vector [6]:

\nabla \cdot (\nabla \times F) = \frac{\partial}{\partial x} (R_{y} - Q_{z}) + \frac{\partial}{\partial y} (P_{z} - R_{x}) + \frac{\partial}{\partial z} (Q_{x} - P_{y}) .

⟳differentiate each component of [6] and add.

Step 4. Expand: the six terms are

R_{y x} - Q_{z x} + P_{z y} - R_{x y} + Q_{x z} - P_{y z}

. By equality of mixed partials,

R_{y x} = R_{x y}

Q_{z x} = Q_{x z}

P_{z y} = P_{y z}

, so all six cancel in pairs, giving

\nabla \cdot (\nabla \times F) = 0

■

Fundamental Theorem for Line Integrals

Step 1. Parametrize

C

r (t)

t \in [a, b]

, with

r (a) = a

r (b) = b

. By definition,

\int_{C} \nabla f \cdot d r = \int_{a}^{b} \nabla f (r (t)) \cdot r^{'} (t) d t .

⟳the line-integral definition replaces

d r

with

r^{'} (t) d t

Step 2. The multivariable chain rule gives

\frac{d}{d t} f (r (t)) = \nabla f (r (t)) \cdot r^{'} (t)

, so the integrand is an exact derivative. Step 3. Apply the single-variable Fundamental Theorem of Calculus:

\int_{a}^{b} \frac{d}{d t} f (r (t)) d t = f (r (b)) - f (r (a)) = f (b) - f (a) . ■

⟳the result is path-independent because only endpoint values survive.

James Clerk Maxwell | 1831–1879

Maxwell turned the abstract vector field into the governing language of physics. Building on Faraday's intuitive "lines of force," he formulated the electromagnetic field as coupled vector fields $E$ and $B$ , and his 1865 equations — later compactly written with $\nabla \cdot$ and $\nabla \times$ by Heaviside and Gibbs — predicted that electromagnetic disturbances propagate as waves at the speed of light. His Treatise on Electricity and Magnetism (1873) is the historical birthplace of the field concept as we teach it.

Building Intuition

Picture a windy weather map: at every city an arrow shows wind speed and direction. That is a vector field. Now read its three derivatives physically:

Gradient turns a "hill" (scalar height) into arrows pointing uphill, steepest first — it always points perpendicular to level curves.
Divergence at a point asks: is this a source or a sink? Drop a tiny balloon; if air rushes out faster than it comes in, divergence is positive (a source); if it collapses inward, negative.
Curl asks: would a tiny paddlewheel spin here? Place a microscopic wheel in the flow; if neighboring arrows are stronger on one side it rotates — curl points along the spin axis (right-hand rule) with magnitude twice the local angular velocity.

A field can have zero divergence yet swirl violently (pure rotation), or huge divergence yet never spin (radial outflow). These two behaviors are independent — that independence is exactly what the Helmholtz decomposition formalizes.

A planar vector field: rotational part (closed flow lines) versus the radial source. The gold curve traces one integral curve (flow line) of the swirling component.

Worked Examples

Divergence and Curl of an Explicit Field

Given:

F (x, y, z) = (x y, y^{2} z, x z^{2})

at the point

(1, 2, 3)

. Find:

\nabla \cdot F

and

\nabla \times F

there. Step 1 — Divergence (use [5]):

P = x y, Q = y^{2} z, R = x z^{2}, P_{x} = y, Q_{y} = 2 y z, R_{z} = 2 x z .

\nabla \cdot F = y + 2 y z + 2 x z .

(1, 2, 3)

2 + 2 (2) (3) + 2 (1) (3) = 2 + 12 + 6 = 20.

Step 2 — Curl (use [6]):

1st: $R_{y} - Q_{z} = 0 - y^{2} = - y^{2}$
2nd: $P_{z} - R_{x} = 0 - z^{2} = - z^{2}$
3rd: $Q_{x} - P_{y} = 0 - x = - x$

So $\nabla \times F = (- y^{2}, - z^{2}, - x)$ . At $(1, 2, 3)$ : $(- 4, - 9, - 1)$ .

\nabla \cdot F = 20 (s^{- 1} if F is a velocity in s^{- 1} -scaled units), \nabla \times F = (- 4, - 9, - 1) .

Finding a Potential and a Line Integral

Given:

F (x, y) = (2 x y + 3, x^{2} - 1)

. Compute

\int_{C} F \cdot d r

along any path from

(0, 0)

(2, 1)

. Step 1 — Test conservativeness:

P = 2 x y + 3

Q = x^{2} - 1

. Then

P_{y} = 2 x

and

Q_{x} = 2 x

, equal everywhere;

R^{2}

is simply connected, so

F = \nabla f

exists (planar curl test, special case of [9]). Step 2 — Build $f$ : Integrate

P

x

f = x^{2} y + 3 x + g (y)

. Differentiate in

y

f_{y} = x^{2} + g^{'} (y)

must equal

Q = x^{2} - 1

, so

g^{'} (y) = - 1

g (y) = - y

. Thus

f (x, y) = x^{2} y + 3 x - y .

Step 3 — Apply the Fundamental Theorem [8]:

\int_{C} F \cdot d r = f (2, 1) - f (0, 0) = (4 \cdot 1 + 6 - 1) - 0 = 9.

\int_{C} F \cdot d r = 9 (units of work, e.g. joules if F is a force in N).

Common Misconceptions

Zero curl does NOT automatically mean conservative. Equation [9] requires the domain to be simply connected (no holes). On a domain with a hole, a field can satisfy

\nabla \times F = 0

everywhere yet fail to be a gradient. Likewise, divergence and curl are not "the derivative" of a vector field — they are two specific scalar/vector summaries of the full

n \times n

Jacobian matrix

\partial F_{i} / \partial x_{j}

. The Jacobian carries strictly more information.

Irrotational but Not Conservative

Consider on $U = R^{2} ∖ {(0, 0)}$ the field

F (x, y) = (\frac{- y}{x ^{2} + y ^{2}}, \frac{x}{x ^{2} + y ^{2}}) .

⟳this is

\nabla θ

in regions where the polar angle

θ

is single-valued, but

θ

jumps by

2 π

around the origin.

A direct computation gives $Q_{x} - P_{y} = 0$ everywhere on $U$ (irrotational). Yet integrating around the unit circle counterclockwise gives

\oint_{C} F \cdot d r = \int_{0}^{2 π} 1 d θ = 2 π \neq = 0.

A nonzero loop integral is impossible for a true gradient field (by [8]), so

F

is not conservative on

U

. The culprit is the hole at the origin —

U

is not simply connected.

Connections

Vector fields are the meeting point of nearly all of vector calculus. The three integral theorems — Green's Theorem (plane), Stokes' Theorem (surfaces), and the Divergence Theorem — each relate an integral of a derivative of $F$ over a region to an integral of $F$ over its boundary; together with Differential Forms they unify into the generalized Stokes theorem $\int_{M} d ω = \int_{\partial M} ω$ . The flow generated by integrating $\dot{x} = F (x)$ links vector fields to Ordinary Differential Equations and Dynamical Systems. On curved spaces, vector fields become sections of the Tangent Bundle studied in Differential Geometry, and the Helmholtz decomposition generalizes to the Hodge Decomposition.

Real-World Applications

Electromagnetism: $E$ and $B$ in Maxwell's equations; Gauss's law $\nabla \cdot E = ρ / ε_{0}$ ties divergence to charge density.
Fluid dynamics: the velocity field $u (x, t)$ in the Navier–Stokes equations; incompressible flow means $\nabla \cdot u = 0$ , and vorticity is $ω = \nabla \times u$ .
Meteorology & oceanography: wind and current fields, with divergence diagnosing convergence zones and storm formation.
Gravitation: Newtonian gravity is a conservative field $g = - \nabla ϕ$ ; orbit computation uses the potential $ϕ$ .
Computer graphics & games: flow fields steer crowds, particle systems, and fluid simulations; curl noise generates divergence-free turbulent smoke.
Medical imaging (MRI / DTI): diffusion-tensor and gradient fields map water flow along neural fibers.
Robotics & path planning: artificial potential fields guide robots — attractive gradient toward goals, repulsive gradient from obstacles.

Historical Development

1760s — Euler & Lagrange: velocity fields appear implicitly in early fluid mechanics; the Euler equations describe an ideal fluid's velocity field.
1813 — Gauss: states the divergence theorem in the context of gravitational potential.
1828 — George Green: Essay on the Application of Mathematical Analysis to Electricity and Magnetism introduces potential theory and Green's theorem.
1831–1855 — Michael Faraday: experimental "lines of force," the physical seed of the field concept (though Faraday used no equations).
1850s — William Thomson (Lord Kelvin) & Stokes: formulate and popularize Stokes' theorem (it appears as an 1854 Smith's Prize exam question).
1865/1873 — James Clerk Maxwell: unifies electricity, magnetism, and light into field equations; coins much field terminology including "curl" and "convergence."
1880s — Josiah Willard Gibbs & Oliver Heaviside: independently develop modern vector analysis, the $\nabla$ notation, and the compact form of Maxwell's equations — defeating Hamilton's quaternions for the textbook standard.
1858 — Hermann von Helmholtz: proves the decomposition of a vector field into curl-free and divergence-free parts (vorticity theorems).
1900s — Élie Cartan: the calculus of differential forms recasts all vector-field theorems in coordinate-free language.

Subtleties & Limits

The deepest open question tied directly to vector fields is the Navier–Stokes existence and smoothness problem — a Clay Millennium Prize problem: do smooth, finite-energy initial velocity fields in $R^{3}$ always produce solutions that remain smooth for all time, or can the velocity field develop a singularity (blow up to infinite speed) in finite time? No one knows. More broadly, the topology of the domain controls which irrotational fields are gradients (de Rham cohomology measures the failure), and on curved manifolds the naive operators must be replaced by the exterior derivative and Hodge star, where the index theorems (e.g. Poincaré–Hopf, relating zeros of a vector field to the Euler characteristic) impose global constraints — you cannot, for instance, comb a hairy sphere flat.

Vector Fields

The Core Insight

Formal Statement

Vector Field

The Del Operator and Its Three Children

Conservative (Gradient) Field

Fundamental Theorem for Line Integrals

Curl Test for Conservativeness

Why It Holds / Derivation

The Two Fundamental Vector Identities

Fundamental Theorem for Line Integrals

James Clerk Maxwell | 1831–1879

Building Intuition

A planar vector field: rotational part (closed flow lines) versus the radial source. The gold curve traces one integral curve (flow line) of the swirling component.

Worked Examples

Divergence and Curl of an Explicit Field

Finding a Potential and a Line Integral

Common Misconceptions

Irrotational but Not Conservative

Connections

Real-World Applications

Historical Development

Subtleties & Limits

See Also

Before this

Where this leads