Theta-functions for singular curves
Indranil Biswas, Jacques Hurtubise · 2026
Read Theta-functions for singular curves by Indranil Biswas, Jacques Hurtubise (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-11152.txt.
Introduction
Indranil Biswas and Jacques Hurtubise build theta functions for singular curves. Start with X, an irreducible singular Riemann surface, and its desingularisation X-tilde. Its generalised Jacobian J(X) fibers over the Jacobian of X-tilde, and an Abel map of X-tilde to J(X) lifts the Abel map to the smooth Jacobian.
The authors construct a theta function over a compactified generalised Jacobian, and it supplies a section of a fitting positive line bundle. Translation on J(X) then yields all line bundles of that degree. Shifting the theta function and restricting it to the Abel image yields one section that works across all line bundles of that degree over X. This extends Riemann's classical result to the singular case. Algebraic geometers get the classical theta construction where curves are allowed to be singular.
Catalog record
| CREATOR | Indranil Biswas, Jacques Hurtubise |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65562 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.11152
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Theta-functions for singular curves Indranil Biswas, Jacques Hurtubise Abstract . Let X X be an irreducible singular Riemann surface, with desingularisation X ~ \widetilde{X} . The generalised Jacobian J ( X ) J(X) of X X fibers over the Jacobian J ( X ~ ) J(\widetilde{X}) of X ~ \widetilde{X} , and there is an Abel map A A of X ~ \widetilde{X} to J ( X ) J(X) , lifting the Abel map to J ( X ~ ) J(\widetilde{X}) . We build a theta function on a compactification of the generalised Jacobian J ( X ) J(X) (giving a section of a suitable positive line bundle). The translation action on J ( X ) J(X) then yields all line bundles of that degree, and the translates of the theta function, restricted to A ( X ~ ) A(\widetilde{X}) , give a “universal section” of the line bundles of that degree over X X . This extends to the singular case a classical result of Riemann. 2010 Mathematics Subject Classification 14H60, 14D21, 70G55, 70H06 1. Introduction Theta functions and their various generalizations have a wide variety of uses. In their most classical form, on the Jacobian of a compact Riemann surface, they give — in some sense — a universal construction of the function theory of the curve. The curve embeds into its Jacobian by the well-known Abel map: A\,:\,X\,\hookrightarrow\,J(X) . The theta function represents a section of a line bundle L on the curve, in a degree in which the translation action on J(X) gives all line bundles of that degree. Thus translating A(X) by T_{a} , and pulling back the theta-function, we get a section of the pull back A^{*}T_{a}^{*}L on the curve, represented as a quasi-periodic function. From this, taking products, quotients, etc of these various translates, we can recreate all the function theory of the curve. For beautiful expositions of all this, the reader is referred to several books of Mumford, notably the three books of Tata lectures [ Mu1 , Mu2 , Mu3 ] , and his monograph “Curves and their Jacobians” [ Mu4 ] . One particular use of these functions lies in the solution of various problems in integrable systems, most notably when one considers the flows in several parameters associated to equations of KP type. There are many ways of doing this (see, among many others, [ Be , Kr , Du ] ); one simple version, expounded in [ AHH ] , is to consider Lax equations \frac{dL(\lambda)}{dt}\ =\ [P(L(\lambda)),\,L(\lambda)] (1.1) where L(\lambda)\,=\,L(\lambda,\,t) is some flow (in t ) of matricial rational functions in \lambda . Here P is some function of L(\lambda) — typically a projection. The flow has as an invariant the spectral curve S in the (\lambda,\,z) -plane given by \det(L\lambda-z{\mathbb{I}})\,=\,0 , and the cokernel of (L(\lambda)-z{\mathbb{I}}) will typically be a line bundle E supported on S . The cases of interest give linear flows of line bundles on the Jacobian, and one can then write out the solutions with a bit of effort in terms of the \theta –functions — first writing out a basis of sections of E and from there an expression for L . The scheme has various generalisations, for example the generalised Hitchin systems [ BR , Bo , Ma ] , where the parameter \lambda lies on a Riemann surface, and L(\lambda) is a section of some twist of the endomorphisms of a bundle. This method has been very successful in producing quasi-periodic solutions to KP-type equations (combining several flows of the same type). The solutions which piqued interest in these equations many years ago were, however, solitons. They correspond here to a degeneration of the curve into a rational, typically nodal, curve, with the period lattice of the Jacobian expanding out to infinity in some directions. Indeed, various ingenious degenerations from the quasi-periodic to the solitonic flows have been presented over the years. A small, recent, sample could include [ Ab ] , [ Ga ] , [ Ko ] ; there remains the fact that the basic geometry of ( 1.1 ) lies in the limit, and it would be nice to solve in these terms. The question then arises of understanding ( 1.1 ) with a singular spectral curve. There already exists a suitable generalised Jacobian (see [ AK ] , [ Ro ] , [ Ca ] , [ Gi ] , [ OS ] ) and its concomitant Abel map, defined on the desingularisation \widetilde{X} of X , the question then arises of how to generalise the theta function. Indeed, there are already several examples of this in the literature, starting as far as we can tell with a paper of Clebsch [ Cl ] , dealing with a curve with a simple node, and other cases have followed. We propose here to deal with the general case; indeed, once one sees what is required, finding a generalisation is not too difficult. If we look at the usual theta function, we have that translating the image of the curve in its Jacobian by the Abel map, we obtain by pull back (1) a realisation of sections of regular line bundles in degree equal to the genus as quasi-periodic functions, with suitable automorphy factors, and (2) an actual section of that line bundle, given by the translate (modulo the Riemann constant) of the theta function, with a divisor given by the vanishing of the theta-function. Our aim is to obtain a similar construction here, and indeed our theta-functions will exhibit the same quasi-periodicity as in the smooth case. The Jacobian of the singular curves has a higher dimension compared to that of its desingularisation. Roughly, pulling back from the singular curve to its desingularisation, the various rank one locally free sheaves (line bundles) on the singular curve correspond on the desingularisation to different equivalence classes of trivialisations at the preimage of the singular points of line bundles. In our case these trivialisations will be simply the pull-backs of the shifts of our generalised theta-functions. We will in the course of our construction be compactifying the generalised Jacobian; there are several ways of doing this in a natural fashion, adding on torsion free sheaves, and so on. Our compactification will be much more elementary, and we are not sure if it has much geometric meaning. The aim is to have a compactification which works, i.e., with a theta-function that gives us what we want, that is some sort of universal section of line bundles over the curve. The theta-function will be a section of a line bundle which is non-trivial on the fibers of this compactification, and so when pulled back to the desingularised curve will be a section of degree greater than that of the line bundle for the usual theta function. This is quite normal, as we are looking for sections on the singular curve, and this has higher arithmetic genus. The construction we have is fairly explicit, at least in the sense that ordinary theta functions are explicit: there will be in the construction, periods of Abelian integrals, generalisation of the Riemann constant, and indeed the basic ingredient