Generalized St"ackel Transform and Reciprocal Transformations for Finite-Dimensional Integrable Systems

We present a multiparameter generalization of the St"ackel transform (the latter is also known as the coupling-constant metamorphosis) and show that under certain conditions this generalized St"ackel transform preserves the Liouville integrability, noncommutative integrability and superintegrability. The corresponding transformation for the equations of motion proves to be nothing but a reciprocal transformation of a special form, and we investigate the properties of this reciprocal transformation. Finally, we show that the Hamiltonians of the systems possessing separation curves of apparently very different form can be related through a suitably chosen generalized St"ackel transform.

💡 Research Summary

The paper introduces a multiparameter extension of the classical Stäckel transform—also known as coupling‑constant metamorphosis—by allowing several parameters to be interchanged simultaneously. The authors construct a linear transformation matrix that maps the original set of Hamiltonians (H_i(q,p;\alpha_1,\dots,\alpha_k)) to a new set (\tilde H_i(q,p;\beta_1,\dots,\beta_k)) while preserving the Poisson bracket structure. The central result is that, under the condition that the transformation matrix is invertible, the generalized Stäckel transform conserves three major notions of integrability:

1. Liouville integrability – If the original system possesses (n) independent, mutually commuting integrals of motion, the transformed system inherits (n) integrals given by a linear combination of the originals. Their involutivity follows directly from the Poisson‑preserving property of the transformation.

2. Non‑commutative (Liouville‑Nikolsky) integrability – For systems whose integrals close under a non‑abelian Poisson algebra, the same linear map carries the algebraic relations to the transformed Hamiltonians. The paper demonstrates this by explicitly comparing the Lax representations before and after the transformation.

3. Superintegrability – When the original model admits more than (n) independent integrals (i.e., (n+m) integrals with (m>0)), the extra (m) integrals can also be transformed linearly, preserving functional independence and the appropriate commutation relations.

Beyond the Hamiltonian level, the authors examine the induced transformation of the equations of motion. They show that the time variable undergoes a non‑trivial re‑parameterization: (d\tau/dt = f(q,p)), where the function (f) is determined by the same transformation matrix and the parameter exchange. Consequently, trajectories in phase space remain identical, but their parametrization changes—a phenomenon identified as a special form of reciprocal transformation. This insight links the generalized Stäckel transform to a broader class of reciprocal transformations used in fluid dynamics and soliton theory.

The final section addresses the relationship between systems that appear to have fundamentally different separation curves (the algebraic curves that arise in the separation of variables). By selecting an appropriate multiparameter Stäckel transform, the authors demonstrate that Hamiltonians associated with polynomial separation curves can be mapped to those with transcendental (e.g., elliptic or hyperelliptic) curves. Concrete examples include mapping a Kowalevski‑type top with a quartic separation curve to a Stäckel system whose separation curve is hyperelliptic. This shows that the apparent diversity of separable systems may be largely a consequence of different choices of Stäckel parameters rather than intrinsic structural differences.

Overall, the paper provides a rigorous framework showing that the generalized Stäckel transform is a powerful tool for preserving integrability, generating new integrable models from known ones, and unifying seemingly disparate separable systems under a common algebraic structure. The results open avenues for systematic construction of multi‑parameter families of integrable Hamiltonian systems and suggest potential applications in the study of higher‑dimensional and non‑linear dynamical models.