The Central Idea
A level set consists of all points where a function has a fixed value.
For a function of two variables,
\[ z=f(x,y), \]a level curve is obtained by setting
\[ \boxed{f(x,y)=c}, \]where \(c\) is a constant.
For a scalar-valued function of three variables,
\[ F(x,y,z), \]a level surface is obtained from
\[ \boxed{F(x,y,z)=c}. \]Why Level Sets Matter
A multivariable function assigns a numerical value to each point in its domain.
A level set reverses the viewpoint:
\[ \boxed{\text{Which points share the same output value?}} \]This converts numerical information into geometric structure.
Level Curves of a Function \(z=f(x,y)\)
Suppose
\[ z=f(x,y). \]If we choose a constant height
\[ z=c, \]then the corresponding level curve satisfies
\[ \boxed{f(x,y)=c}. \]Geometrically, this is the projection into the \(xy\)-plane of the intersection between the graph and the horizontal plane \(z=c\).
Level Curves as Horizontal Slices
Imagine slicing the surface
\[ z=f(x,y) \]with a horizontal plane
\[ z=c. \]The intersection is a space curve lying at height \(c\).
Projecting that intersection down to the \(xy\)-plane gives the level curve.
Contour Maps
A collection of level curves for several values of \(c\) forms a contour map.
For example, we might plot
\[ f(x,y)=0,\qquad f(x,y)=1,\qquad f(x,y)=2,\qquad f(x,y)=3. \]Each contour marks a constant output value.
Example: Circular Level Curves
Consider
\[ f(x,y)=x^2+y^2. \]The level equation is
\[ x^2+y^2=c. \]For
\[ c>0, \]this is a circle of radius
\[ \boxed{\sqrt c}. \]The Level Curve at \(c=0\)
For
\[ x^2+y^2=0, \]the only point is
\[ \boxed{(0,0)}. \]Thus a level curve can degenerate to a single point.
No Real Level Curve for Negative \(c\)
Since
\[ x^2+y^2\ge0, \]the equation
\[ x^2+y^2=c \]has no real points when
\[ c<0. \]Not every constant necessarily produces a nonempty level set.
What the Circular Contours Tell Us
The level curves
\[ x^2+y^2=c \]are concentric circles.
Therefore, the corresponding graph
\[ z=x^2+y^2 \]has rotational symmetry around the \(z\)-axis.
The contours already reveal the bowl-like geometry of an elliptic paraboloid.
Example: Elliptical Level Curves
Consider
\[ f(x,y)=\frac{x^2}{4}+\frac{y^2}{9}. \]The level equation is
\[ \frac{x^2}{4}+\frac{y^2}{9}=c. \]For \(c>0\), divide by \(c\):
\[ \frac{x^2}{4c}+\frac{y^2}{9c}=1. \]The level curves are ellipses.
Scaling of Elliptical Contours
As \(c\) increases, the semiaxes become
\[ 2\sqrt c \]and
\[ 3\sqrt c. \]Thus every level curve has the same shape ratio but a different scale.
Example: Hyperbolic Level Curves
Consider
\[ f(x,y)=x^2-y^2. \]The level equation is
\[ \boxed{x^2-y^2=c}. \]The geometry changes according to the sign of \(c\).
Positive Hyperbolic Levels
If
\[ c>0, \]then
\[ x^2-y^2=c \]is a hyperbola opening left and right.
Negative Hyperbolic Levels
If
\[ c<0, \]rewrite as
\[ y^2-x^2=-c. \]The level curve is now a hyperbola opening upward and downward.
The Zero Level of \(x^2-y^2\)
If
\[ c=0, \]then
\[ x^2-y^2=0. \]Factor:
\[ (x-y)(x+y)=0. \]Therefore, the zero level set is
\[ \boxed{y=x\quad\text{or}\quad y=-x}. \]Contours Reveal the Saddle
The changing hyperbola orientation for positive and negative levels is a signature of the surface
\[ z=x^2-y^2. \]The contour map reveals saddle geometry even without viewing the full three-dimensional graph.
Example: Parallel-Line Level Curves
Consider
\[ f(x,y)=2x-y. \]The level equation is
\[ 2x-y=c. \]Solving for \(y\):
\[ \boxed{y=2x-c}. \]The level curves are parallel lines.
What Parallel Contours Indicate
The graph
\[ z=2x-y \]is a plane.
Equal-value slices of a plane produce evenly oriented parallel contour lines.
Level Curves Need Not Be Closed
A level curve can be:
- closed, such as a circle or ellipse;
- open, such as a hyperbola;
- unbounded, such as a line;
- disconnected;
- a single point;
- empty.
Disconnected Level Curves
One level value may produce several separate components.
For example,
\[ x^2-y^2=1 \]has two disconnected branches.
A level set is defined by equality of function value, not by connectedness.
Level Surfaces
For a scalar function of three variables,
\[ F(x,y,z), \]a level surface is defined by
\[ \boxed{F(x,y,z)=c}. \]It consists of all points in space where the scalar field has the same value.
Example: Spherical Level Surfaces
Let
\[ F(x,y,z)=x^2+y^2+z^2. \]Then
\[ F(x,y,z)=c \]gives
\[ x^2+y^2+z^2=c. \]For \(c>0\), this is a sphere of radius
\[ \boxed{\sqrt c}. \]Nested Level Surfaces
For
\[ F(x,y,z)=x^2+y^2+z^2, \]the levels
\[ c=1,\quad c=4,\quad c=9 \]produce spheres of radii
\[ 1,\quad2,\quad3. \]The surfaces are nested around the origin.
Example: Ellipsoidal Level Surfaces
Consider
\[ F(x,y,z)= \frac{x^2}{4} + \frac{y^2}{9} + \frac{z^2}{16}. \]The level equation
\[ F(x,y,z)=c \]gives a family of ellipsoids for
\[ c>0. \]Example: Planar Level Surfaces
Consider
\[ F(x,y,z)=x+2y-3z. \]The level equation is
\[ x+2y-3z=c. \]Every level surface is a plane.
Different values of \(c\) produce parallel planes.
Scalar Fields
A scalar field assigns one numerical value to each point in a region of space.
Examples can represent:
- temperature;
- pressure;
- density;
- electric potential;
- gravitational potential;
- concentration;
- height or elevation.
Level sets organize points having equal scalar value.
Topographic Contours
On a topographic map, each contour line joins points having the same elevation.
If elevation is modeled by
\[ h(x,y), \]then a contour line satisfies
\[ \boxed{h(x,y)=c}. \]Thus topographic maps are practical level-curve diagrams.
Closely Spaced Contours
If neighboring contour values differ by a fixed amount, closely spaced contour lines usually indicate that the function changes rapidly over a short horizontal distance.
On a topographic map, this corresponds to a steep slope.
Widely Spaced Contours
If equal increments in function value are represented by contours that are far apart, the function changes more gradually with position.
On a topographic map, this corresponds to gentler terrain.
Contour Spacing Depends on the Chosen Levels
Geometric spacing should be interpreted together with the numerical interval between contour values.
If one map uses increments of \(1\) and another uses increments of \(100\), visual spacing alone does not support direct comparison.
Contours Do Not Usually Cross
Suppose two distinct level curves of a function intersect at a point.
If that point belongs to both
\[ f(x,y)=c_1 \]and
\[ f(x,y)=c_2, \]then
\[ c_1=c_2. \]Therefore, distinct level curves of a single-valued function cannot cross at the same domain point.
Different Components of the Same Level May Meet
A single level equation can contain several branches or components.
At singular values, these components can meet.
For example,
\[ x^2-y^2=0 \]gives two intersecting lines, but they belong to the same level value
\[ c=0. \]The Gradient
For a differentiable scalar function
\[ f(x,y), \]the gradient is
\[ \boxed{ \nabla f= \left\langle \frac{\partial f}{\partial x}, \frac{\partial f}{\partial y} \right\rangle }. \]In three variables,
\[ \boxed{ \nabla F= \left\langle F_x,F_y,F_z \right\rangle }. \]The Gradient Is Normal to a Regular Level Set
At a regular point where
\[ \nabla f\neq\mathbf0, \]the gradient is perpendicular to the level curve.
Likewise, in three dimensions,
\[ \nabla F \]is normal to the level surface when the gradient is nonzero.
Why the Gradient Is Perpendicular
Along a level curve,
\[ f(x,y)=c \]remains constant.
Therefore, moving tangent to the level curve produces no first-order change in \(f\).
The gradient points in the direction of greatest first-order increase, so it must be perpendicular to every tangent direction.
Example: Gradient of Circular Levels
For
\[ f(x,y)=x^2+y^2, \]the gradient is
\[ \boxed{\nabla f=\langle2x,2y\rangle}. \]This vector points radially outward.
The level curves are circles, whose normal direction is radial.
Gradient Magnitude and Rate of Change
The magnitude
\[ \|\nabla f\| \]measures the maximum first-order rate of increase per unit distance at a point.
Large gradient magnitude often corresponds to closely packed contours when the contour interval is fixed.
Level Curves and Directional Change
Moving along a level curve keeps the function value fixed.
Moving across level curves changes the function value.
The gradient points across the level curves in the direction of steepest increase.
Implicit Curves Are Level Curves
Any implicit curve
\[ F(x,y)=0 \]is the zero level curve of \(F\).
More generally,
\[ F(x,y)=c \]is a level curve.
Thus level-set language unifies implicit curve descriptions.
Implicit Surfaces Are Level Surfaces
Any implicit surface
\[ F(x,y,z)=0 \]is the zero level surface of \(F\).
For example,
\[ x^2+y^2+z^2-R^2=0 \]is the zero level surface of
\[ F(x,y,z)=x^2+y^2+z^2-R^2. \]Quadratic Surfaces as Level Surfaces
The previous lesson studied equations such as
\[ \frac{x^2}{a^2} + \frac{y^2}{b^2} + \frac{z^2}{c^2} =1. \]This is the level surface
\[ F(x,y,z)=1 \]for
\[ F(x,y,z)= \frac{x^2}{a^2} + \frac{y^2}{b^2} + \frac{z^2}{c^2}. \]Families of Level Sets
A single function generates an entire family of geometric objects as \(c\) varies.
For
\[ F(x,y,z)=x^2+y^2+z^2, \]the family
\[ F=c \]produces nested spheres.
For
\[ f(x,y)=x^2-y^2, \]the family produces hyperbolas and intersecting lines.
Level Sets Can Change Topology
As the level value \(c\) changes, the qualitative structure of a level set can change.
For example,
\[ x^2-y^2=c \]changes from one hyperbola orientation to two intersecting lines and then to the perpendicular hyperbola orientation as \(c\) passes through zero.
Critical Values and Structural Change
Values of \(c\) associated with critical points can mark changes in the geometry or topology of level sets.
In introductory work, this can appear as:
- a loop appearing or disappearing;
- separate components joining;
- a closed curve collapsing to a point;
- the orientation of open curves changing.
Level Curves of a Radial Function
Suppose
\[ f(x,y)=g\left(\sqrt{x^2+y^2}\right). \]The function depends only on distance from the origin.
Therefore, its level curves are typically circles centered at the origin whenever the relevant equation for radius has real solutions.
Level Surfaces of a Radial Function
If
\[ F(x,y,z)=g\left(\sqrt{x^2+y^2+z^2}\right), \]then the scalar field depends only on distance from the origin.
Its regular level surfaces are typically spheres centered at the origin.
Symmetry from Level Sets
Level-set families often reveal symmetry more clearly than the original function formula.
Concentric circles reveal radial symmetry.
Parallel lines reveal a constant directional structure.
Repeated ellipses reveal anisotropic scaling.
Level Sets and Coordinate Changes
A coordinate transformation can simplify the geometry of level sets.
For example, a rotated quadratic expression may produce tilted ellipses in the original coordinates.
Rotating to principal axes can convert those contours into standard ellipses.
Level Sets and Matrices
A quadratic function such as
\[ f(x,y)= ax^2+2bxy+cy^2 \]can be written
\[ \boxed{\mathbf x^T A\mathbf x}. \]Its level curves are determined by the geometry of the quadratic form.
Eigenvectors identify principal directions, while eigenvalues control scaling along those directions.
Level Curves and Optimization
Level curves are useful for visualizing optimization.
If an objective function has contours
\[ f(x,y)=c, \]then increasing \(c\) moves through higher objective values.
Constraints can be drawn as separate curves, and optimal points often occur where the highest attainable contour touches a constraint.
Level Sets in Physics
Many physical scalar fields are visualized using level sets.
Examples include:
- equipotential curves and surfaces;
- isothermal curves and surfaces;
- constant-pressure surfaces;
- constant-density surfaces.
Equipotential Surfaces
If
\[ V(x,y,z) \]represents potential, then an equipotential surface satisfies
\[ \boxed{V(x,y,z)=c}. \]Every point on that surface has the same potential value.
Isotherms
If
\[ T(x,y) \]represents temperature over a region, then
\[ \boxed{T(x,y)=c} \]defines an isotherm: a curve joining points of equal temperature.
Three-Dimensional Isothermal Surfaces
For a temperature field
\[ T(x,y,z), \]the equation
\[ T(x,y,z)=c \]gives an isothermal surface.
Such surfaces can be nested, disconnected, curved, or unbounded depending on the field.
Level Sets as Dimension Reduction
A scalar equation generally lowers dimension by one under regular conditions.
In two variables,
\[ f(x,y)=c \]typically gives a one-dimensional curve.
In three variables,
\[ F(x,y,z)=c \]typically gives a two-dimensional surface.
But Dimension Reduction Can Fail at Special Values
At singular or degenerate values, a level set may collapse or change dimension.
For example,
\[ x^2+y^2+z^2=0 \]is only one point rather than a regular surface.
Common Error: Confusing a Level Curve with the Graph of \(f\)
The graph
\[ z=f(x,y) \]is a surface in three dimensions.
A level curve
\[ f(x,y)=c \]is a curve in the \(xy\)-plane.
They are related but not the same object.
Common Error: Forgetting the Constant \(c\)
Level sets form a family.
Changing
\[ c \]changes the geometric set.
The equation
\[ f(x,y)=c \]should therefore always be interpreted together with the chosen level value.
Common Error: Assuming Every Level Value Exists
For
\[ f(x,y)=x^2+y^2, \]negative level values produce no real points.
Always check whether the equation
\[ f(x,y)=c \]is feasible.
Common Error: Assuming Different Contours Can Cross
A single-valued function cannot have two different output values at the same input point.
Therefore, level curves for distinct values
\[ c_1\neq c_2 \]cannot cross.
Common Error: Reading Steepness Without Checking the Contour Interval
Closely spaced contours indicate rapid change only when the numerical difference between successive contours is understood.
Visual spacing and numerical spacing must be interpreted together.
Common Error: Assuming Every Level Set Is Smooth
Level sets can contain cusps, corners, crossings, isolated points, or other singular behavior.
Smoothness is typically guaranteed only under additional conditions, such as a nonzero gradient.
Common Error: Assuming the Gradient Is Always Defined and Nonzero
The gradient may fail to exist or may equal
\[ \mathbf0. \]At such points, the usual statement that the gradient gives a normal direction requires additional care.
A Reliable Procedure for Level Curves
- Start with \(f(x,y)\).
- Choose a level value \(c\).
- Set \(f(x,y)=c\).
- Simplify the resulting equation.
- Identify the curve type if possible.
- Check whether real points exist.
- Repeat for several values of \(c\).
- Compare size, orientation, spacing, and connectedness.
- Use the family to infer the shape of the original graph.
A Reliable Procedure for Level Surfaces
- Start with \(F(x,y,z)\).
- Choose \(c\).
- Set \(F(x,y,z)=c\).
- Rewrite in recognizable standard form when possible.
- Check intercepts and symmetry.
- Use planar traces to understand the surface.
- Check feasibility and degeneracy.
- Compare several values of \(c\) to understand the family.
Example: Full Level-Curve Workflow
Consider
\[ f(x,y)=4x^2+y^2. \]Choose
\[ c=16. \]Then
\[ 4x^2+y^2=16. \]Divide by \(16\):
\[ \boxed{ \frac{x^2}{4} + \frac{y^2}{16} =1 }. \]The level curve is an ellipse with semiaxes \(2\) and \(4\).
Example: Full Level-Surface Workflow
Consider
\[ F(x,y,z)=x^2+y^2-z^2. \]At level
\[ c=1, \]we have
\[ x^2+y^2-z^2=1. \]This is a
\[ \boxed{\text{hyperboloid of one sheet}}. \]At level
\[ c=0, \]the surface becomes
\[ \boxed{x^2+y^2=z^2}, \]a double cone.
Practice 1: Circular Level Curve
For
\[ f(x,y)=x^2+y^2, \]describe the level curve
\[ f(x,y)=25. \]Practice 2: Elliptical Level Curve
For
\[ f(x,y)=\frac{x^2}{9}+\frac{y^2}{4}, \]describe the level curve
\[ f(x,y)=1. \]Practice 3: Hyperbolic Level Curve
Describe
\[ x^2-y^2=4. \]Practice 4: Zero Level
Find the level set
\[ x^2-y^2=0. \]Practice 5: Parallel-Line Contours
Find the level curve of
\[ f(x,y)=3x+y \]corresponding to
\[ c=6. \]Practice 6: Spherical Level Surface
For
\[ F(x,y,z)=x^2+y^2+z^2, \]describe the level surface
\[ F=49. \]Practice 7: Planar Level Surface
Describe
\[ x+2y-z=5. \]Practice 8: Gradient Direction
For
\[ f(x,y)=x^2+y^2, \]find the gradient at
\[ (3,4). \]Practice 9: Empty Level Set
Does the function
\[ f(x,y)=x^2+y^2 \]have a real level curve for
\[ c=-3? \]Practice 10: Identify from a Level Surface
Identify the surface
\[ x^2+y^2-z^2=1. \]Practice Answers
Practice 1
The equation is
\[ x^2+y^2=25. \]Therefore, the level curve is a
\[ \boxed{\text{circle of radius }5}. \]Practice 2
The equation is already in standard form:
\[ \frac{x^2}{9}+\frac{y^2}{4}=1. \]Therefore, the level curve is an
\[ \boxed{\text{ellipse}} \]with semiaxes \(3\) and \(2\).
Practice 3
Rewrite:
\[ \frac{x^2}{4}-\frac{y^2}{4}=1. \]Therefore, the level curve is a
\[ \boxed{\text{hyperbola opening left and right}}. \]Practice 4
Factor:
\[ x^2-y^2=(x-y)(x+y). \]Therefore,
\[ \boxed{y=x\quad\text{or}\quad y=-x}. \]Practice 5
Set
\[ 3x+y=6. \]Solving for \(y\):
\[ \boxed{y=6-3x}. \]Practice 6
The level equation is
\[ x^2+y^2+z^2=49. \]Therefore, the level surface is a
\[ \boxed{\text{sphere of radius }7}. \]Practice 7
The equation
\[ x+2y-z=5 \]is a
\[ \boxed{\text{plane}}. \]Practice 8
The gradient is
\[ \nabla f=\langle2x,2y\rangle. \]At \((3,4)\),
\[ \boxed{\nabla f(3,4)=\langle6,8\rangle}. \]Practice 9
Since
\[ x^2+y^2\ge0, \]it cannot equal \(-3\).
Therefore, the level set is
\[ \boxed{\text{empty}}. \]Practice 10
There are two positive squared terms and one negative squared term, and the equation equals \(1\).
Therefore, the surface is a
\[ \boxed{\text{hyperboloid of one sheet}}. \]What to Remember
Level sets turn function values into geometry.
- A level curve of \(f(x,y)\) satisfies \[ f(x,y)=c. \]
- A level surface of \(F(x,y,z)\) satisfies \[ F(x,y,z)=c. \]
- Contour maps display several level curves at once.
- Level sets can be closed, open, disconnected, degenerate, or empty.
- Distinct level values of a single-valued function cannot cross at the same domain point.
- At regular points, the gradient is normal to the level curve or level surface.
- Closely spaced equal-interval contours indicate rapid change.
The central habit is: fix the output value, then study the geometry of all points that share it.
Where We Go Next
Level sets help us understand complicated forms through slices and families of simpler geometric objects. The next lesson turns from analytic description to decomposition of spatial form.
In Polyhedra and Spatial Decomposition, we will study three-dimensional objects built from faces, edges, vertices, cells, and component pieces, and how complex forms can be analyzed by breaking them into simpler spatial units.