Geometry of the Kahan discretizations of planar quadratic Hamiltonian systems. II: Systems with a linear Poisson tensor

From MaRDI portal
Publication:2297882



Abstract: Kahan discretization is applicable to any quadratic vector field and produces a birational map which approximates the shift along the phase flow. For a planar quadratic Hamiltonian vector field with a linear Poisson tensor and with a quadratic Hamilton function, this map is known to be integrable and to preserve a pencil of conics. In the paper `Three classes of quadratic vector fields for which the Kahan discretization is the root of a generalised Manin transformation' by P. van der Kamp et al., it was shown that the Kahan discretization can be represented as a composition of two involutions on the pencil of conics. In the present note, which can be considered as a comment to that paper, we show that this result can be reversed. For a linear form ell(x,y), let B1,B2 be any two distinct points on the line ell(x,y)=−c, and let B3,B4 be any two distinct points on the line ell(x,y)=c. Set B0=frac12(B1+B3) and B5=frac12(B2+B4); these points lie on the line ell(x,y)=0. Finally, let Binfty be the point at infinity on this line. Let mathfrakE be the pencil of conics with the base points B1,B2,B3,B4. Then the composition of the Binfty-switch and of the B0-switch on the pencil mathfrakE is the Kahan discretization of a Hamiltonian vector field with a quadratic Hamilton function H(x,y). This birational map Phif:mathbbCP2dashrightarrowmathbbCP2 has three singular points B0,B2,B4, while the inverse map Phif−1 has three singular points B1,B3,B5.


In unpublished lecture notes, W. Kahan introduced a method of discretization applicable to any system of ordinary differential equations in \(\mathbb{R}^n\) that has a quadratic vector field. Such equations have the form \[\dot{x} = f(x) = Q(x) + Bx + c,\] where each component of \(Q : \mathbb{R}^n \rightarrow \mathbb{R}^n\) is a quadratic form, \(B\) is an \(n \times n\) real matrix and \(c \in \mathbb{R}^n\). The discretization that Kahan uses has the following form: \[\frac{\tilde{x} - x}{2 \epsilon} = Q(x, \tilde{x}) + \frac{1}{2} B(x + \tilde{x}) + c,\] where \[Q(x, \tilde{x}) = \frac{1}{2} ((Q(x + \tilde{x}) - Q(x) - Q(\tilde{x}))\] is the symmetric bilinear form that corresponds to the quadratic form \(Q\). The discretization is linear with respect to \(\tilde{x}\) and so defines a rational map \(\tilde{x} = \Phi_f(x, \epsilon)\). This can be written explicitly as \(\tilde{x} = \Phi_f (x, \epsilon) = x + 2 \epsilon (I - \epsilon f'(x))^{-1} f(x) \), where \(f'(x)\) represents the Jacobian of \(f\). This mapping approximates a time \(\epsilon\) shift along solutions of the original differential equation. Since the discretization is invariant with respect to interchange of \(x\) and \(\tilde{x}\), a reversibility property holds: \(\Phi^{-1}_f (x, \epsilon) = \Phi_f (x, -\epsilon)\). In their prior work the authors studied properties of Kahan's method when applied to integrable systems. They determined that in a number of cases Kahan's approach preserves integrability in the sense that the mapping \(\Phi_f(x, \epsilon)\) has as many independent integrals of motion as the original system. In [J. Phys. A, Math. Theor. 52, No. 4, Article ID 045204, 10 p. (2019; Zbl 1422.70012)] \textit{P. H. van der Kamp} et al. show that the Kahan map \(\Phi_f\) can be represented as a composition of two involutions on a pencil of conics in \(\mathbb{C}^2\). The current paper shows that the reverse is also true: a Kahan discretization can be reconstructed from a linear form and a pencil of conics. For Part I, see [the authors et al., Proc. R. Soc. Lond., A, Math. Phys. Eng. Sci. 475, No. 2223, Article ID 20180761, 13 p. (2019; Zbl 1501.37068)].











This page was built for publication: Geometry of the Kahan discretizations of planar quadratic Hamiltonian systems. II: Systems with a linear Poisson tensor

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2297882)