Source

Adapted from my own handwritten and hand-typed notes from Year 1 to Year 3 of my EEE degree

The gradient of a scalar field is a vector field that points uphill. Dotting it with a direction gives the directional derivative, the slope of the field in that direction.

Scalar and Vector Fields

Two kinds of field:

  • a scalar field gives a number at each point, e.g. temperature in a room
  • a vector field gives a vector at each point, e.g. gravity , fluid velocity, electric field
  • examples in 2D: (horizontal flow), (a vortex)

Del Operator

It acts in three ways:

OperationActs onResult
gradient scalar fieldvector field
divergence vector fieldscalar field
curl vector fieldvector field

Do not confuse with , which usually means a small change.

Gradient

E.g. gives .

Properties at a point where :

  • points in the direction of greatest increase of
  • is that maximum rate of change
  • is perpendicular to the level surface through the point, so it is a surface normal

Following uphill to find a maximum is the gradient ascent (hill climbing) method.

Directional Derivative

The rate of change of in the direction of a unit vector is

If the given direction is not a unit vector, use . This is the same idea as the component of one vector along another.

Unit Normal to a Surface

Method

Steps:

  1. Compute the partial derivatives to get
  2. Substitute the point
  3. Normalise the direction vector
  4. Dot product for the directional derivative, or divide by its length for the unit normal

Worked Example

at , direction .

Unit direction: . Directional derivative:

Unit normal to the surface at that point:

The del operator applied to vector fields gives the divergence.