Irreducibility, genus, and global geometry of the spectral curves of the quantum KdV hierarchy

The quantum Korteweg-de Vries (KdV) hierarchy constructed by Buryak and Rossi — the quantization of the KdV hierarchy with respect to its first (Gardner–Zakharov–Faddeev) Poisson structure, built from double ramification cycles in the moduli space of stable curves — admits a complete set of common eigenvectors, the deformed Schur polynomials, constructed however only as formal power series at the origin of the deformation parameter $\sigma=-\epsilon^2/\sqrt{\hbar}$. The joint spectrum on each finite-dimensional graded piece defines plane algebraic curves $\Sigma_{k,m}=\{(\sigma,\rho):\det(K_m(\sigma)-\rho)=0\}$, the spectral curves of the hierarchy. Ruzza and Yang conjectured that for each weight $k$ these curves are irreducible with geometric genus $g(\Sigma_{k,m})=(k-1)p(k)+1-\sum_{\lambda\vdash k}\ell(\lambda)$, independent of the Hamiltonian index $m$. This problem asks for a proof of that conjecture, together with the global geometric characterization of the curves — branch points, singularities, and monodromy of the degree-$p(k)$ covering of the $\sigma$-line — from which the analyticity in $\sigma$ of the quantum KdV spectral data, beyond its formal construction, would follow. As of 2026 the conjecture and the associated global questions are open: the follow-up literature computes the spectral data order by order in the dispersion parameter but does not address the geometry of the curves.

Background

The KdV hierarchy is the archetypal integrable system: an infinite family of commuting nonlinear evolution equations, whose first member is $u_t=6uu_x-u_{xxx}$. It carries two compatible Hamiltonian structures; in terms of the Fourier modes $u(x)=\sum_{k\in\mathbb Z}U_k e^{ikx}$, the first (Gardner–Zakharov–Faddeev) Poisson bracket is simply $\{U_k,U_l\}_1=ik\,\delta_{k+l,0}$. Quantizing a Hamiltonian hierarchy means promoting the commuting Hamiltonian densities to mutually commuting operators after canonical quantization of the modes (creation operators $\widehat U_k=q_k$ for $k>0$, derivatives $\widehat U_{-k}=-\hbar k\,\partial/\partial q_k$ for $k>0$, so the Poisson bracket becomes a commutator). The naive quantizations of the classical KdV densities fail to commute, and the quantization problem, posed by Dubrovin, is to find $\hbar$-deformed densities that do. For the second Poisson structure this was achieved by Bazhanov, Lukyanov and Zamolodchikov within two-dimensional conformal field theory; for the first Poisson structure, Buryak and Rossi constructed a solution, for arbitrary dispersion parameter $\epsilon$, out of the intersection theory of double ramification cycles — cycle classes in the moduli space $\overline{\mathcal M}_{g,n}$ of stable complex curves obtained as fundamental classes of loci of curves carrying a meromorphic function with prescribed residues $a_1,\dots,a_n$ summing to zero. For the simplest input this construction produces the quantum KdV Hamiltonians $H_m$ ($m\ge -1$): a commuting family of operators on a bosonic Fock space of polynomials in the variables $q_1,q_2,\dots$.

The spectral problem asks for the joint eigenvalues and eigenvectors of the $H_m$. In the dispersionless limit $\epsilon=0$ (also called the quantum Hopf hierarchy) Dubrovin solved it: the eigenvectors are the Schur polynomials $s_\lambda(T)$, the standard basis of the ring $\widetilde\Lambda=\mathbb C[T_1,T_2,\dots]$ ($T=q/\sqrt{\hbar}$) indexed by integer partitions $\lambda$, and the eigenvalues are shifted symmetric functions — polynomials in the Bloch–Okounkov generators $Q_j(\lambda)$, where $Q_j$ is a symmetrized power sum of the shifted parts $\lambda_i-i+\tfrac12$ of the partition. For nonvanishing dispersion, Ruzza and Yang proved a symmetry property of the quantum Hamiltonians (with respect to the inner product making the Schur basis orthonormal) and a non-degeneracy property of the dispersionless spectrum, and deduced the existence of a unique basis of common eigenvectors $r_\lambda(T;\sigma)$ — the deformed Schur polynomials — with eigenvalues $E_m(\lambda;\sigma)$, both obtained only as formal power series in the single deformation parameter $\sigma=-\epsilon^2/\sqrt{\hbar}$ (after restricting the zero mode $U_0=0$ and rescaling).

Since the Hamiltonian $H_0$ is a grading operator, the rescaled operators $K_m(\sigma)$ preserve each homogeneous component $\widetilde\Lambda_k$ of degree $k$, which has dimension $p(k)$, the number of partitions of $k$. The weight-$k$ spectral curves of the hierarchy are the plane algebraic curves

$$\Sigma_{k,m}=\left\{(\sigma,\rho)\in\mathbb C^2:\ \det_{\widetilde\Lambda_k}\bigl(K_m(\sigma)-\rho\bigr)=0\right\},\qquad m\ge 1,$$

ramified coverings of the $\sigma$-sphere of degree $p(k)$: the eigenvalues are the branches of the coordinate $\rho$ over the $\sigma$-line, and the eigenvector coefficients are meromorphic functions on the curve (here and below, the geometric genus of a singular curve means the genus of its desingularization). Dubrovin (in a 2016 private communication) computed the geometric genus of $\Sigma_{k,1}$ for $0\le k\le 7$, and Ruzza and Yang extended the computation to $0\le k\le 10$, finding the values $0,0,0,1,4,9,21,37,69,113,187$; they observed that these agree with sequence A238641 of the On-Line Encyclopedia of Integer Sequences and conjectured the closed formula below. The local structure of the curves is understood: at $\sigma=0$ the eigenvectors are the Schur polynomials; as $\sigma\to\infty$ they approach the monomial basis $T_\lambda$ with branches $\rho\sim\sigma^m\sum_i\lambda_i^{2m+1}$; symmetry of the Hamiltonians forces branch points into complex-conjugate pairs, while collisions of eigenvalues at real $\sigma$ produce singular points. The global picture — irreducibility, the genus, the complete branch locus, the singularities, and the monodromy as $\sigma$ varies — is unknown, and Ruzza and Yang explicitly defer it to future investigations. They also observe that the analyticity in $\sigma$ of the spectral data, which their construction leaves formal, should be answered through these curves, since the eigenvalue series are the Taylor expansions at $\sigma=0$ of the branches of $\rho$ on $\Sigma_{k,m}$.

Problem Statement

Prove Conjecture 1.1 of Ruzza–Yang (2021) and settle the global geometry of the spectral curves of the Buryak–Rossi quantum KdV hierarchy. For the curves

$$\Sigma_{k,m}=\left\{(\sigma,\rho)\in\mathbb C^2:\ \det_{\widetilde\Lambda_k}\bigl(K_m(\sigma)-\rho\bigr)=0\right\},\qquad k\ge 1,\ m\ge 1,$$

where $K_m(\sigma)$ are the rescaled quantum KdV Hamiltonians (zero mode $U_0=0$, rescaling $T=q/\sqrt{\hbar}$) acting on the degree-$k$ homogeneous piece $\widetilde\Lambda_k$ of $\mathbb C[T_1,T_2,\dots]$:

1. prove that for every fixed $k\ge 1$ and every $m\ge 1$ the curve $\Sigma_{k,m}$ is irreducible;
2. prove that its geometric genus is independent of $m$ and equal to

$$g\bigl(\Sigma_{k,m}\bigr)=(k-1)\,p(k)+1-\sum_{\lambda\vdash k}\ell(\lambda),$$

where $p(k)$ is the number of partitions of $k$ and $\ell(\lambda)$ is the length (number of nonzero parts) of the partition $\lambda$; 3. determine the global geometry of the curves as the Hamiltonian index $m$ and the weight $k$ vary: locate the branch points of the degree-$p(k)$ covering $\Sigma_{k,m}\to\mathbb P^1_{\sigma}$ — the values of $\sigma$ at which two sheets merge, which by the symmetry of the Hamiltonians lie off the real axis — and the singular points of $\Sigma_{k,m}$, in particular the real values of $\sigma$ at which distinct eigenvalues coalesce; report these branch and singular loci as explicit polynomials in $\sigma$ (with their factorization) and describe the monodromy of the covering as a permutation representation of the fundamental group of the punctured $\sigma$-line on the $p(k)$ sheets; the deliverable is the general description valid for all $k\ge 1$ and $m\ge 1$, results for individual pairs $(k,m)$ counting only as partial progress; 4. establish the analyticity in $\sigma$ of the spectral data — the eigenvalues $E_m(\lambda;\sigma)$ and the deformed Schur eigenvectors $r_\lambda(T;\sigma)$ — beyond their construction as formal power series at $\sigma=0$, for example by realizing them as single-valued (meromorphic) functions on the curves $\Sigma_{k,m}$; this item is entailed by item 3, since the eigenvalue series are the Taylor expansions at $\sigma=0$ of the branches of $\rho$ on $\Sigma_{k,m}$, so it need not be established independently of the global geometric description.

These deliverables jointly serve one objective: to upgrade the local, formal solution of the quantum KdV spectral problem around $\sigma=0$ to a global analytic and geometric description.

References

1. Giulio Ruzza and Di Yang, “On the spectral problem of the quantum KdV hierarchy”, Journal of Physics A: Mathematical and Theoretical 54 (2021), no. 37, 374001. DOI: 10.1088/1751-8121/ac190a; https://doi.org/10.1088/1751-8121/ac190a (preprint arXiv:2104.01480, https://arxiv.org/abs/2104.01480)
2. Boris Dubrovin, “Symplectic Field Theory of a Disk, Quantum Integrable Systems, and Schur Polynomials”, Annales Henri Poincaré 17 (2016), no. 7, 1595–1613. DOI: 10.1007/s00023-015-0449-2; https://doi.org/10.1007/s00023-015-0449-2
3. Alexandr Buryak, “Double ramification cycles and integrable hierarchies”, Communications in Mathematical Physics 336 (2015), no. 3, 1085–1107. DOI: 10.1007/s00220-014-2235-2; https://doi.org/10.1007/s00220-014-2235-2
4. Alexandr Buryak and Paolo Rossi, “Double ramification cycles and quantum integrable systems”, Letters in Mathematical Physics 106 (2016), no. 3, 289–317. DOI: 10.1007/s11005-015-0814-6; https://doi.org/10.1007/s11005-015-0814-6
5. Spencer Bloch and Andrei Okounkov, “The Character of the Infinite Wedge Representation”, Advances in Mathematics 149 (2000), no. 1, 1–60. DOI: 10.1006/aima.1999.1845; https://doi.org/10.1006/aima.1999.1845
6. Vladimir V. Bazhanov, Sergei L. Lukyanov and Alexander B. Zamolodchikov, “Integrable structure of conformal field theory, quantum KdV theory and Thermodynamic Bethe Ansatz”, Communications in Mathematical Physics 177 (1996), no. 2, 381–398. DOI: 10.1007/BF02101898; https://doi.org/10.1007/BF02101898
7. Jan-Willem M. van Ittersum and Giulio Ruzza, “Quantum KdV hierarchy and quasimodular forms”, Communications in Number Theory and Physics 18 (2024), no. 2, 405–439. DOI: 10.4310/CNTP.2024.v18.n2.a4; https://doi.org/10.4310/CNTP.2024.v18.n2.a4 (preprint arXiv:2202.03213, https://arxiv.org/abs/2202.03213)
8. Jan-Willem M. van Ittersum and Giulio Ruzza, “Quantum KdV Hierarchy and Shifted Symmetric Functions”, International Mathematics Research Notices 2025 (2025), no. 9, rnaf102. DOI: 10.1093/imrn/rnaf102; https://doi.org/10.1093/imrn/rnaf102 (preprint arXiv:2407.02266, https://arxiv.org/abs/2407.02266)
9. Zhe Wang, “Quantum Dispersionless KdV Hierarchy Revisited”, Communications in Mathematical Physics 407 (2025), no. 1, article 4. DOI: 10.1007/s00220-025-05484-2; https://doi.org/10.1007/s00220-025-05484-2 (preprint arXiv:2410.16747, https://arxiv.org/abs/2410.16747)
10. Davide Masoero and Giulio Ruzza, “ODE/IM Correspondence at the Free-Fermion Point. Laguerre Wronskians, Shifted Symmetric Functions, and Quantum KdV”, preprint arXiv:2605.24563 (2026); https://arxiv.org/abs/2605.24563
11. Jean-Yves Thibon, “Stable Symmetric Series, Differential Operators, and Jack Deformations”, preprint arXiv:2608.25651 (2026); https://arxiv.org/abs/2608.25651