The extremal-point conjecture for the support of the invariant measure

Let F0F_0 and F1F_1 be the two Lotka–Volterra vector fields, and let GG be the polynomial defining the non-axis collinearity curve

C~={(x,y)R+2:G(x,y)=0}.\tilde C=\left\{(x,y)\in\mathbb{R}_+^2:G(x,y)=0\right\}.

Define

T={(x,y)R+2:G(x,y)=0 and (F0G)(x,y)=0}.T=\left\{(x,y)\in\mathbb{R}_+^2:G(x,y)=0\text{ and }(F_0\cdot\nabla G)(x,y)=0\right\}.

For zTz\in T, let C(z)C(z) be the bounded region enclosed by the trajectories of F0F_0 and F1F_1 starting at zz, together with the segment [1/a1,1/a0]×{0}[1/a_1,1/a_0]\times\{0\}.

Extremal-point conjecture. The set TT is a singleton {z0}\{z_0\}, and the support of the invariant measure that is not supported by one of the two axes is

C(z0)×{0,1}.C(z_0)\times\{0,1\}.

Numerical experiments suggest that there is a unique extremal point on the relevant part of C~\tilde C, whose two flow trajectories form the boundary of the support. The conjecture identifies this point and the support of the non-axis-supported invariant measure, while the preceding discussion only establishes inclusion of the relevant stable-equilibrium trajectories in the invariant support.

Sources & referencesView supporting material

Primary source

Florent Malrieu and Pierre-André Zitt, “On the persistence regime for Lotka-Volterra in randomly fluctuating environments”, arXiv:1601.08151 (2017).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.