A unified view of a family of soliton equations related to spin Calogero-Moser systems

Abstract: We study the interconnections between the spin Benjamin-Ono (sBO) and half-wave maps (HWM) equations, a pair of nonlinear partial integro-differential equations that have recently been found to permit multi-soliton solutions, where the time evolution of the constituent solitons can be described in terms of the well-known, completely integrable, spin Calogero-Moser (sCM) system. By considering a symmetry transformation of the sCM dynamics we are led to introduce a scale parameter into the sBO equation, yielding what we call the rescaled sBO (rsBO) equation, which has both the sBO and HWM equations as special cases. Together with the addition of a new constant background term in the multi-soliton ansatz for the sBO equation, this allows us to formulate a theorem for the rsBO equation that unifies and generalizes previously known soliton theorems for the sBO and HWM equations. The theorem offers a new perspective on these equations; we use it to show the emergence of HWM dynamics in a certain background-dominated limit of the sBO equation, and to suggest a generalization of the HWM equation. Along the way we discuss basic properties of the new multi-soliton solutions, and how to construct them. We spend some time proving that indeed all previously known multi-soliton solutions of the HWM equation are given by the new theorem, and not just a subset. We discuss, and state a conjecture about, possible physical interpretations of the sBO equation. Finally, we apply the same ideas to the spin non-chiral intermediate long-wave (sncILW) and non-chiral intermediate Heisenberg ferromagnet (ncIHF) equations, find that they are related in the same way as the sBO and HWM equations, and formulate a unified theorem for their multi-soliton solutions. For ease of exposition we keep the discussion to hermitian solutions of the sBO and sncILW equations and $\bb R^3$-valued solutions of the HWM and ncIHF equations, though readers familiar with the subject will have no problem generalizing to the non-hermitian and $\bb C^3$-valued cases.