Bifurcation Theory — Mathematics Knowledge System

Overview

Bifurcation theory studies how the qualitative structure of a dynamical system's equilibria — how many there are, and whether each is stable or unstable — changes as a parameter is varied. A bifurcation is a value of the parameter at which this qualitative picture changes abruptly: equilibria can appear, disappear, merge, split, or swap stability. Because such transitions correspond to a linearization eigenvalue crossing zero (or the imaginary axis, for oscillatory bifurcations), bifurcation analysis builds directly on the eigenvalue-based stability criteria from equilibrium and stability theory, but tracks how those eigenvalues move as a parameter r changes rather than fixing r once and for all.

Intuition

Think of an equilibrium's stability as being read off the slope f′(x*) at that equilibrium — negative slope means the flow pushes back toward x* (stable), positive slope means it pushes away (unstable). A bifurcation happens exactly when the graph of f(x,r) as a function of x, for varying r, changes how many times it crosses the x-axis (creating or destroying equilibria) or when a crossing's slope passes through zero (flipping stability). The saddle-node bifurcation is the simplest way equilibria can appear: as r decreases through 0, the parabola r+x² dips down to touch the x-axis at exactly one point (r=0, a single degenerate equilibrium at x=0) and then crosses it twice for r<0, creating a stable/unstable pair out of nothing. The transcritical bifurcation instead keeps the same two equilibria (x=0 and x=r) for every r, but the two lines' relative position swaps as r crosses 0, so which one is 'the stable branch' switches.

Formal Definition

Definition

For a one-parameter family of systems x′ = f(x, r), a bifurcation occurs at a parameter value r₀ where the number or stability of equilibria (solutions of f(x,r)=0) changes. Two of the simplest and most common types are the saddle-node and transcritical bifurcations:

x' = r + x^2 \quad \text{(saddle-node normal form)}

Two equilibria merge and annihilate as r crosses 0

x^* = \pm\sqrt{-r} \ \ (\text{real only for } r \le 0), \qquad f'(x^*) = 2x^*

Equilibria and their stability (sign of f') for the saddle-node case

x' = rx - x^2 \quad \text{(transcritical normal form)}

Two equilibria persist for all r but exchange stability at r=0

x^* \in \{0,\ r\}, \qquad f'(0)=r,\ \ f'(r)=-r

Equilibria and their stability for the transcritical case

Derivation

Analyze the saddle-node normal form x′ = r + x² by finding its equilibria and classifying stability via f′(x*) = 2x*:

f(x,r) = r + x^2 = 0 \implies x^* = \pm\sqrt{-r}

Equilibria exist only when r ≤ 0 (no real square root of a positive number)

f'(x) = 2x \implies f'(-\sqrt{-r}) = -2\sqrt{-r} < 0 \ (\text{for } r<0)

The negative branch x=-√(-r) has negative slope: stable equilibrium

f'(+\sqrt{-r}) = 2\sqrt{-r} > 0 \ (\text{for } r<0)

The positive branch x=+√(-r) has positive slope: unstable equilibrium

r > 0: \text{no equilibria at all} \quad (\text{since } r+x^2 > 0 \text{ for all real } x)

For r>0, the two equilibria have vanished entirely — a saddle-node bifurcation at r=0

Applications

The buckling of a loaded beam under increasing compressive force is a pitchfork bifurcation: below a critical load the straight configuration is the unique stable equilibrium, and above it the beam buckles into one of two new stable bent equilibria while the straight configuration becomes unstable.

Worked Examples

For r>0, r+x² is always positive (sum of a positive number and a square), so there are no equilibria.
$r > 0 \implies f(x,r) = r + x^2 > 0 \ \forall x \implies \text{no equilibria}$
At r=0, the only equilibrium is the degenerate double root x*=0 (f=x², touching the axis).
$r=0 \implies x^{*}=0 \ \text{(semi-stable: } f'(0)=0\text{)}$
For r<0, two equilibria exist at x*=±√(-r), with the negative one stable and the positive one unstable (computed above).
$r<0 \implies x^*_{-}=-\sqrt{-r} \text{ (stable)}, \quad x^*_{+}=\sqrt{-r} \text{ (unstable)}$

Answer: As r decreases through 0, two equilibria (one stable, one unstable) are born together out of nothing — a saddle-node bifurcation, the simplest mechanism by which equilibria can appear or disappear.

Practice Problems

Difficulty 7/10

For x′ = r + x² at r = -4, find both equilibria and classify their stability.

Difficulty 7/10

For x′ = rx - x² at r = 3, find both equilibria and classify their stability.

Difficulty 8/10

A population model x′ = rx - x² (logistic-type growth with per-capita growth rate r) has x=0 representing extinction. Using the transcritical analysis above, explain what happens to the extinction state as r crosses from negative to positive, and interpret it biologically.

Common Mistakes

Common Mistake

Assuming every bifurcation destroys or creates equilibria (like the saddle-node case).

The transcritical bifurcation is a counterexample: the same two equilibria (x*=0 and x*=r) exist on both sides of the bifurcation value; what changes is which one is stable, not how many there are.

Common Mistake

Classifying stability using f'(x*) evaluated at the wrong equilibrium, or forgetting f' must be evaluated AT each specific equilibrium (not as a general formula).

Stability is always local: compute f'(x) as a function of x, then substitute the specific equilibrium value x* being classified — different equilibria of the same system can have different stability even though f'(x) is a single formula.

Quiz

Summary

Bifurcation theory tracks how the number and stability of equilibria of x′=f(x,r) change as a parameter r varies.
A saddle-node bifurcation (normal form x′=r+x²) creates or destroys a stable/unstable pair of equilibria as r crosses a critical value.
A transcritical bifurcation (normal form x′=rx-x²) keeps the same equilibria for all r but exchanges their stability at the critical value.
Both types are detected by tracking f'(x*) at each equilibrium as r varies — a bifurcation occurs where f'(x*)=0, mirroring the linearization eigenvalue criteria of general equilibrium and stability analysis.

References

WebsiteWikipedia — Bifurcation theory
BookStrogatz, S. Nonlinear Dynamics and Chaos, 2nd ed., Ch. 3.