Bifurcation diagram of the 2023 Cambridge quadratic-product system Created 2026-09-24 Updated 2026-09-24
The equilibria of , lie on
for every , and on for . They meet in a subcritical pitchfork bifurcation at ; the positive square-root branch meets the branch and exchanges stability in a transcritical bifurcation at .
For
the equilibrium branches are , , and wherever the square roots are real. If , pitchforks occur at and , and the two nonzero branch pairs meet in two transcritical bifurcations at . If , all five branches meet in one degenerate fifth-order bifurcation. If , separate pitchforks occur at and . A constant perturbation produces imperfect pitchforks and saddle nodes, so none of these unperturbed bifurcations is structurally stable without the reflection symmetry.
Bifurcations of the 2022 Cambridge quadratic-cubic system Created 2026-09-24 Updated 2026-09-24
For
the invariant center manifold near is , with reduced equation and hence a transcritical bifurcation. Near , put and . The extended center manifold begins
and its reduced equation is , a saddle-node bifurcation.
Chemostat washout 2026-10-14
Chemostat washout means the culture's biomass concentration tends to zero because growth cannot sustain replacement of the cells leaving the vessel. In the ideal Monod equation chemostat with , its equilibrium is . Its Jacobian matrix has eigenvalues and : a small inoculum grows when , whereas chemostat washout is locally asymptotically stable when . For nonnegative initial concentrations, chemostat washout occurs for ; the boundary case also loses biomass asymptotically.
At the positive chemostat steady state meets the chemostat washout equilibrium, and one eigenvalue is zero, so linearization alone does not decide the boundary case. On the invariant nutrient-equivalent balance , small positive obeys : nutrient consumption lowers the specific growth rate below , leading to slow decay rather than a positive biomass concentration. The continued mathematical branches exchange stability in a transcritical bifurcation; the positive-biomass branch becomes physically inadmissible beyond .
At near , use , , . The stable direction is the axis and the centre direction is the axis. The extended centre manifold has and reduced flow , a supercritical pitchfork bifurcation in parameter . At near , use , , . The centre line is , the stable line is , and the reduced flow is , a transcritical bifurcation. The branch , joins the two bifurcations and is stable for .
The equilibria are and . Their linear trace is and determinant . For nonzero the symmetry-protected pitchfork bifurcation splits into a transcritical bifurcation at and a saddle-node bifurcation at . The energy obeys . For positive , the two outer saddle equilibria have unequal barrier heights, so the symmetric heteroclinic cycle unfolds into separated one-way connections and a homoclinic orbit termination of the central stable limit cycle.
A transcritical bifurcation need not already have its persistent branch at coordinate zero. The smooth parameter-dependent shift turns the displayed crossing of into the standard form with parameter . The crossing branches exchange dynamical stability. A mere nonsmooth relabeling of stable and unstable branches as would conceal this smooth exchange.
At , the two branches meet at . The eigenvalues there are and . For , the nonzero quadratic term and transverse stability identify a saddle-node bifurcation. If , it creates a sink equilibrium and a saddle equilibrium as increases; if , it creates a source equilibrium and a saddle equilibrium. At , the zero eigenvalue is double and the ordinary saddle-node description needs the two-parameter unfolding.
At , the origin and lower branch cross. Since , the branches cross transversely in the diagram. The origin changes from a sink equilibrium to a saddle equilibrium as increases through zero; the lower branch changes from a positive- saddle equilibrium to a negative- sink equilibrium. This is a transcritical bifurcation. A local centre-manifold calculation also gives , because and the transverse eigenvalue is .
The figure uses a representative ; the upper branch's additional Hopf change is indicated. For the upper branch is unstable everywhere on its existence interval. The exact trace criterion in part (a) specifies stability for every fixed positive .
Figure 1.
Equilibrium branches and the two-parameter diagram of the quadratic-product flow
. Details.
Put , so and . The second equilibrium equation factors as . The resulting equilibria are
The mixed pair exists strictly when ; equality describes its merger with an axial equilibrium. At the eigenvalues of the Jacobian matrix are , and at they are . At a nonzero mixed equilibrium,
Thus both mixed equilibria are saddle equilibria. The axial equilibria are nodes when their two real eigenvalues have the same sign and saddles when the signs differ. In particular,
There are exactly three local bifurcation lines away from their common codimension-two intersection. On , and meet in a transcritical bifurcation, visible in on the invariant axis . On , the mixed pair meets in a pitchfork bifurcation. Eliminating near , for fixed nonzero , gives : this pitchfork is supercritical for and subcritical for . On , the mixed pair meets . Writing and , elimination gives . This pitchfork is subcritical for and supercritical for . A supercritical pitchfork here need not produce a stable mixed state: its other eigenvalue can already be positive. The negative mixed Jacobian determinant settles the full stability question. There is no Hopf bifurcation in this quadratic truncation.
Along , the axial curves are and . The mixed pair has the same plotted amplitude for its two signs of :
Consequently the mixed curves exist outside the interval whose endpoints are and . The stable portions of the axial curves are
For , stability passes directly from to at the transcritical bifurcation ; the two pitchfork bifurcations lie at and . For , loses stability at and becomes stable only at . Between these points none of the equilibria is stable. This local statement does not prescribe a bounded attractor of the quadratic truncation. All mixed portions are unstable.
Figure 1.
Even-odd mode bifurcations and amplitude branches; solid branches are stable, dashed branches unstable
. Details.
Write . The entire unfolded diagram is obtained from , with the same dynamical stability sign as before. A saddle-node bifurcation at requires , while a saddle-node bifurcation at requires .
For , the crossing separates into a positive stable branch and a negative unstable branch, with no saddle-node bifurcation near . Three equilibria exist for , and the two negative branches annihilate at the outer saddle-node bifurcations . Outside that interval only the positive stable equilibrium remains. If , the diagram has just that one branch everywhere; equality is the limiting saddle-node bifurcation contact at .
For , let . Two inner saddle-node bifurcations occur at , and two outer saddle-node bifurcations at . For there is a single stable equilibrium below ; for there are three equilibria with stable–unstable–stable ordering; outside the outer saddle-node bifurcations only the positive stable equilibrium remains. The preceding figure shows representative small unfoldings. In particular this unrestricted perturbation destroys the exact transcritical bifurcation crossing.
Set . Equilibria satisfy , and their scalar linear eigenvalue is
It vanishes only at or . Substitution in the equilibrium relation gives the three nonhyperbolic parameter-state points
The branches with and are stable, while is unstable. For there are three equilibria; beyond the two outer saddle-node bifurcations there is just the positive stable one. At there is also the distinct stable equilibrium .
Near the crossing, use the smooth coordinate , whose derivative is positive near zero. The equation becomes with . A positive time rescaling removes , and the parameter-dependent state shift gives
Thus the crossing at is a transcritical bifurcation, with parameter . The two smooth branches exchange dynamical stability; labeling the stable branch by instead would obscure that exchange. The parameter-dependent coordinate shift in a transcritical bifurcation explains why the original coordinate need not display a branch identically equal to zero.
Figure 1.
Scalar equilibrium diagrams before and after positive and negative unfolding, with stable branches solid and unstable branches dashed
. Details.
Put . The planar vector field and its equilibrium eigenvalue data are
For a nonzero equilibrium, use to rewrite the determinant as . These formulas determine the local bifurcations of the imperfect soft Duffing-van der Pol oscillator directly.
When , the origin is a saddle equilibrium for and has determinant for ; the two equilibria are saddle equilibria. They meet the origin in a symmetry-protected pitchfork bifurcation at for . At the origin undergoes a supercritical Hopf bifurcation. To determine criticality, take : averaging gives , so the stable small limit cycle has .
The potential energy of the conservative part and the exact energy balance are
At zero imperfection the stable limit cycle grows to a heteroclinic cycle connecting the two outer saddle equilibria. Near the double-zero point its leading connection curve can be calculated, rather than inferred just from symmetry. In the Hamiltonian system limit the heteroclinic orbit is , and
The weak-damping persistence condition is therefore , not an exact formula at order-one parameters.
Now take ; negative imperfection reflects the phase portrait. The pitchfork bifurcation unfolds into a transcritical bifurcation crossing at and a saddle-node bifurcation at . The latter occurs at . Between those values the origin and are saddle equilibria, while lies in and is a sink or source according to the trace. At , exchanges its saddle equilibrium/non-saddle character with the origin. The double-zero points are now at and .
The original Hopf bifurcation line remains, still supercritical. A second Hopf bifurcation curve occurs on the intermediate nonzero equilibrium:
It is important not to assign the same criticality to its entire length. Write , so . Around that center the restoring quadratic coefficient is , the damping linear coefficient is , and . The Hopf criticality for an asymmetric Lienard center calculation gives cubic amplitude coefficient
Hence it is supercritical for and subcritical for , with a generalized Hopf bifurcation point at . Higher-order terms are needed exactly there. For small imperfection the leading quintic coefficient is positive: writing and using the coordinate that normalizes the potential energy, , with and , the coefficient of in at vanishing cubic coefficient is . Weak-damping averaging therefore supplies the nearby saddle-node bifurcation of stable and unstable limit cycles in the generalized Hopf bifurcation unfolding. None of these small features intersects either requested parameter line.
Globally, the two saddle equilibrium barriers are no longer equal. Their leading energy difference is . The two one-way connections consequently separate; matching their energy differences with the connection integral gives
These weak-damping heteroclinic splitting of a tilted quartic oscillator formulas apply for small with . Between the separated connection curves, the central stable limit cycle terminates in a homoclinic orbit to the lower-barrier saddle equilibrium, for positive imperfection. The simultaneous symmetric heteroclinic-cycle destruction is therefore replaced by two basin-changing one-way connections and a distinct homoclinic orbit loss of the limit cycle. Its period diverges at that loss.
Along , decrease from above one. The sequence is supercritical Hopf bifurcation at ; first heteroclinic orbit; homoclinic orbit loss of the stable limit cycle; second heteroclinic orbit; transcritical bifurcation crossing at zero; saddle-node bifurcation at . The connection locations are global, not obtained by treating the Melnikov approximation as exact. For illustration, direct stable/unstable manifold matching at places them at approximately , respectively. The diagram below gives the qualitative ordering without prescribing those example values for arbitrary imperfection.
Along , the sequence is simply transcritical bifurcation crossing at zero, followed by saddle-node bifurcation at . Before the latter, in the relevant region, so decreases strictly on nonstationary trajectories and there is no periodic orbit or saddle equilibrium loop. After it only a saddle equilibrium remains, which cannot be enclosed by a periodic orbit by the Poincare-index obstruction to a periodic orbit. The origin does not undergo a Hopf bifurcation at , since its determinant there is negative.
Figure 1.
Bifurcation sequences along the two diagonal parameter lines after a small positive reflection-symmetry imperfection; arrows indicate decreasing lambda
. Details.
Write . The admissible equilibrium points are , when , and when . The Jacobian matrix is
At the eigenvalues are and ; at they are and . At a positive coexistence equilibrium point,
It is therefore asymptotically stable. It is a node for , a focus for the reverse inequality, and has repeated real eigenvalues at equality. The node–focus change does not alter the attracting equilibrium point's topology.
There are three open physical parameter regions:
  • : extinction. Only is admissible and it is a sink.
  • , : predator extinction. is a saddle, and is a sink.
  • : coexistence. and are saddles on the boundary, while is a sink.
The axes are invariant. On , decays exponentially; on , prey obey the logistic differential equation. Positivity follows because each right-hand side has its population as a factor. Also is bounded by logistic comparison, while satisfies , so all forward trajectories are bounded. The Dulac function has divergence in the interior, excluding periodic orbits by the Bendixson-Dulac criterion.
In the extinction region, , and then is eventually negative. In the prey-only region, the upper bound implies predator decay and convergence of interior prey to . In the coexistence region the Lyapunov function
has . Its positive-quadrant sublevel sets are compact, and the only invariant subset with also has . The LaSalle invariance principle proves convergence of every interior trajectory to . These facts justify the complete qualitative phase portraits, not just their local arrows.
At , and meet in a transcritical bifurcation; positive prey decay algebraically. At with , and exchange stability. On its centre manifold, gives , so the physically positive side is still attracting. There is no interior Hopf bifurcation since the coexistence trace never vanishes.
Figure 1.
Three continuous predator–prey phase portraits and their parameter regions
. Details.
The quadratic discrete predator-prey map equals the identity plus the vector field of part (a), so its admissible fixed points are the same . Their Jacobian matrices are the identity plus the corresponding continuous-time Jacobian matrices, but map stability requires eigenvalues inside the unit circle rather than the left half-plane.
At the multipliers are : it is a sink for and a saddle for . At they are and . Therefore
For , the prey multiplier is below : is a saddle when and a repeller when . For , makes it a saddle. The crossings at and are the two transcritical bifurcations already encountered. The crossing at is a prey period-doubling bifurcation.
At positive coexistence,
For clarity, derive the relevant Schur stability criterion directly: setting in gives
The open unit disk maps to . A real quadratic has both roots there precisely when all three displayed coefficients have the same positive sign. Since , the coexistence fixed point is a sink precisely when
This region exists only for . In the coexistence triangle , put . If , one root lies below and the other in , so is a saddle. If but , it is a repeller; if , it is a sink. These signs give the complete local fixed-point partition, including real and complex multipliers.
The coexistence period-doubling bifurcation lies at ; its segment adjoining the sink region has . The other multiplier there is and lies strictly inside the unit circle on this segment. This coexistence flip is subcritical on that segment. To calculate its direction, normalize the eigenvector by and the left eigenvector by . The symmetric quadratic derivative is given explicitly in part (d). Eliminating the quadratic term from the one-dimensional centre manifold map gives the effective cubic flip coefficient
For a scalar reduced map , its second iterate satisfies . Thus on the stable-fixed-point side , and the two-cycle multiplier is . An unstable coexistence two-cycle meets the stable fixed point as increases to the flip curve. The other multiplier remains inside the unit circle. The complex unit-modulus crossing lies at , with . The two curves meet at with a double multiplier . The prey flip and predator invasion curves meet at with multipliers .
The sketches mark the sink regions and all these local thresholds. They are a local atlas, not a claim that the map has only these global attractors. Indeed the full nonnegative quadrant is not invariant: becomes negative when . On the prey axis the logistic map already supplies new dynamics. Its nontrivial two-cycle, obtained by factoring after removing the fixed-point factors, is
Its longitudinal multiplier is , so it is stable on the axis for . The transverse multiplier is , making the cycle stable in the plane when additionally . Further flips and more complicated logistic dynamics show why the differential equation's three global regions cannot be transferred unchanged to its unit-step map.
Figure 1.
Local stability regions and fixed-point bifurcation curves of the quadratic predator–prey map
. Details.
At a fixed point , and the first equation factors. The branches are
The three branches meet at , and the positive square-root branch meets the quadratic branch at . The Jacobian matrix at a general fixed point is
On , its trace is and determinant : the branch is attracting for and a saddle for or . On , the eigenvalues are and . The negative branch is always a saddle for ; the positive branch is a saddle for and attracting for . At the bifurcation values the zero eigenvalue requires the nonlinear analysis below.
For , , , the equations are
The extended centre manifold has with no linear terms. Since has degree at least three, its invariance equation to degree two is . Thus
The reduced quadratic factors as . Its two roots have reduced eigenvalues and , respectively, and the transverse eigenvalue remains negative. They exchange stability at , a transcritical bifurcation. The exact branches have and , consistent with this second-order evolution equation.
Near , the slow equation has leading form , so this is a subcritical pitchfork bifurcation with the parameter orientation displayed. For there is one nearby saddle, whose unstable direction is close to and stable direction close to vertical. For the central branch is attracting and the outer two equilibria near are saddles; their separatrices distinguish attraction toward the central equilibrium from departure outwards. The actual central location is , not exactly the origin away from .
Adding positive breaks the pitchfork symmetry. The leading equilibrium equation near zero is : the diagram unfolds into a single branch plus a saddle-node pair, with the latter appearing for positive of order . More precisely the fold is at , at leading order. Near one the reduced equation becomes . Its discriminant is ; the crossing is replaced by two saddle-node endpoints with a no-equilibrium gap for to leading order. Outside that gap, the lower reduced root is stable and the upper one unstable. These local approximations describe the sketch; the exact perturbed curves have higher-order corrections.
Figure 1.
Fixed-point branches, local phase portraits and the positive-forcing unfoldings of the pitchfork and transcritical bifurcations
. Details.
Figure 2.
Nearby planar flows on either side of the subcritical pitchfork, with exact equilibrium locations
. Details.
Representative normal forms are for a saddle-node bifurcation, for a transcritical bifurcation, and for a supercritical pitchfork bifurcation. Replacing the last cubic term by gives the subcritical pitchfork bifurcation.
Treat as a variable with . The extended center manifold is tangent to , so write . Its invariance equation is . At quadratic order this reads
Matching coefficients gives , , . Substitution into the first equation therefore gives
Here denotes total degree at least three in . With , and , this becomes . Thus , , , and the bifurcation is a saddle-node. For negative parameter, the negative- branch is stable and the positive- branch unstable along the center manifold; the transverse eigenvalue is near .
The fixed-point equation is . Thus
when the discriminant is nonnegative. At zero the multiplier is , so it is linearly stable for and unstable for . At a nonzero fixed point, eliminate to obtain the branch and its multiplier:
The branch is stable precisely when . These formulas specify stability throughout the diagram, including any extra flip points outside the region requested.
If and , the two nonzero branches exist for and have multiplier . Thus is a subcritical pitchfork bifurcation. If , the branch through zero has and ; its stability exchanges with zero at , a transcritical bifurcation. The two nonzero roots meet at
where . The graph of has nonzero second derivative , proving a saddle-node bifurcation. At zero and , the multiplier is , indicating a period-doubling bifurcation; generically its nondegeneracy coefficient is . If , necessarily here, composition at instead gives . For small, the leading return equation is , producing a two-cycle with leading amplitudes . Its return multiplier is . Thus this special case still has a flip bifurcation, but it is degenerate and subcritical rather than the generic cubic normal form.
For , the fold lies above . Between it and the transcritical point, the branch is stable provided its multiplier remains above , while the other branch is unstable. For , the fold lies below : immediately above it the branch is stable and the branch between and zero is unstable; the branch crossing zero becomes stable for . All these statements are subject to the displayed exact multiplier criterion if the diagram is extended. The sketch uses representative parameter values and stops before additional flips, as stipulated.
Figure 1.
Fixed-point branches and stability for three representative cubic maps, with folds, transcritical or pitchfork crossings, and the origin flip
. Details.
For a flow, a hyperbolic fixed point has no eigenvalue of its linearization with zero real part; a nonhyperbolic fixed point has at least one such eigenvalue. At the eigenvalues are , while at they are . Thus nonhyperbolicity occurs at the two stated parameters. The following centre manifold reductions show the associated actual bifurcations.
Near put , , . The transformed equations are
At , the non-extended stable subspace is the axis and the centre subspace is the axis. In the extended centre manifold, write , with . The symmetry allows an even graph. At quadratic order its invariance equation has left side of degree at least three and right side , so . Substitution gives
This is a supercritical pitchfork bifurcation in parameter . The central branch is stable for and unstable for near one; the two nonzero branches for are stable. At the bifurcation itself the cubic term attracts on the centre manifold.
Near put , , . Then
At , the stable subspace is and the centre line is . Thus the leading extended centre graph is , and the reduced flow is
This is a transcritical bifurcation: is stable for , while the branch is stable for ; the opposite branches are unstable. At the critical point the quadratic term is only one-sided attracting.
Figure 1.
Local pitchfork and transcritical bifurcations, with stable branches solid and unstable branches dashed
. Details.
The exact additional branch in is
It meets at and at , displaying the connecting pitchfork and transcritical branches in a quadratic-product flow. Its Jacobian has determinant and trace . Hence it is stable for and a saddle for . The negative- companion is needed for the pitchfork sketch near ; no separate analysis of is required.
Now . The zero point exists for all parameters and has eigenvalue , so it is stable for and unstable for . At every nonzero fixed point,
Both outer branches are stable and both inner branches unstable.
For , the nonzero locations are as in part (i). At the local equation is : these are subcritical pitchfork bifurcations, with the two unstable inner branches present on the side where zero is stable. At the bifurcation itself , so zero is unstable. At the inner and outer branches cross and exchange stability in transcritical bifurcations. The positive crossing is semistable, attracting from the right; the negative crossing is semistable, attracting from the left, because the additional factor reverses the local flow there.
For , the outer stable pair exists only for . The points are supercritical pitchfork bifurcations, since the cubic coefficient is negative. Zero loses stability as increases across , and a stable symmetric pair appears. At the threshold zero remains nonlinearly stable, with . There are no further bifurcations. The second row of the preceding figure gives the diagrams. The plots use ; the general locations are obtained from the formulas above.