Residues and Gorenstein Contractions of Genus One Curves

Adrian Neff · 2025

Read Residues and Gorenstein Contractions of Genus One Curves by Adrian Neff (2025) 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-2507-05326.txt.

Introduction

Adrian Neff and Jonathan Wise use residue data to shrink curves in algebraic geometry in a 2025 mathematics study. Their setting is a curve of genus one with nodes, sitting over a local artinian base, and inside it a smaller genus one piece they wish to collapse. Collapsing it cleanly requires the result to keep a mild singularity type called Gorenstein.

The authors first define residues for curves over such bases, then generalize them using tropical data from the curve, with values in line bundles. Those generalized residues guide a contraction that folds the chosen subcurve into a single Gorenstein point. Such contractions serve moduli problems for curves.

Catalog record

CREATORAdrian Neff
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE78 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-05326

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Residues and Gorenstein Contractions of Genus One Curves Adrian Neff Abstract . Let C C be a genus one nodal curve over a local artinian base and let E E be a proper subcurve of genus one. We define residues for curves over local artinian rings, then define generalized residues with values in line bundles over the local artinian ring that arise from tropical data on the curve. We then use these residues to construct a contraction of C C that collapses E E to a Gorenstein genus one singularity. 1. Introduction Given a genus one nodal curve C and a proper subcurve of genus one, E , we wish to construct a contraction of C collapsing E to a Gorenstein genus one singularity. Our construction begins with a family of centrally aligned genus one log curves \pi:C\to S=\mathrm{Spec}(A) , where A is a local artinian ring. One may think of the central alignment as a semistable modification \tilde{C}\to C that “orders the components” inside the subcurve E . Using this data, we produce a residue for singular curves valued in A , then use this to construct the contraction C\to\widebar{C} . Specifically, let C\to\mathrm{Spec}(A) be a family of curves over a local artinian ring A and let p\in C_{0} be a closed point in the fiber. We first construct the residue of a meromorphic differential \varphi along a branch X at p , denoted by \mathrm{Res}_{p,X}(\varphi) , that satisfies the following two properties: (i) if p is a node with branches X and Y , then \mathrm{Res}_{p,X}(\varphi)=-\mathrm{Res}_{p,Y}(\varphi); (ii) if C is smooth, then \sum_{p\in C}\mathrm{Res}_{p,C}(\varphi)=0. One may think of this as an algebraic analogue of integration over a vanishing cycle in the complex analytic setting. After defining our notion of residue, we construct the contraction over \mathrm{Spec}(A) via generalized residues with values in line bundles over \mathrm{Spec}(A) coming from tropical data, denoted \mathrm{Res}^{m} , as follows. There are n rational curves R_{1},...,R_{n} connecting the subcurve E to the rest of C . Let p_{i} be the node connecting R_{i} to E . The condition for a function f in a neighborhood of E to descend to the contraction is \sum_{p_{i}}\mathrm{Res}^{m}_{p_{i},R_{i}}(f\varphi)=0, where \varphi is a specially chosen differential. We then replace S with the spectrum of a local noetherian ring R with maximal ideal \mathfrak{m} . Setting S_{n}=\mathrm{Spec}(R/\mathfrak{m}^{n+1}) and C_{n}=C\times_{S}S_{n} , we have a diagram of cartesian squares {\lx@inpgf@ignorespaces C_{0}} {\lx@inpgf@ignorespaces C_{1}} {\lx@inpgf@ignorespaces C_{2}} {\lx@inpgf@ignorespaces\cdots} {\lx@inpgf@ignorespaces S_{0}} {\lx@inpgf@ignorespaces S_{1}} {\lx@inpgf@ignorespaces S_{2}} {\lx@inpgf@ignorespaces\cdots} and a contraction for each i , \tau_{i}:C_{i}\to\bar{C}_{i} . We then have the following characterization of the functions in \mathcal{O}_{\widebar{C}} . Theorem 1.1 . Let U\subset\bar{C}_{n} be an open subset, V=\tau_{n}^{-1}(U) , and f_{n}\in\mathcal{O}_{C_{n}}(V) . The following are equivalent: (i) f_{n} is in the image of \mathcal{O}_{C_{m}}(V)\to\mathcal{O}_{C_{n}}(V) for all m\geq n ; (ii) there exist compatible lifts f_{m}\in\mathcal{O}_{C_{m}}(V) of f_{n} for all m\geq n ; (iii) f_{n} is in \mathcal{O}_{\bar{C}_{n}}(U) ; (iv) f_{n} satisfies the residue condition: \sum_{p_{i}}\mathrm{Res}^{m}_{p_{i},R_{i}}(f_{n}\varphi_{n}|_{V})=0\in\mathcal{O}_{S}(-\delta) . Here the \varphi_{n} are specially chosen differentials on the C_{n} . 1.1. Relation to Other Works and Future Directions Contractions of genus one curves have been studied previously, for example, by Smyth in [ 17 ] , Ranganathan, Santos–Parker, and Wise in [ 16 ] , and Bozlee in [ 6 ] . Comparing approaches, the constructions in [ 17 ] and [ 16 ] both rely on putting the curve in a smoothing family and performing the contraction in that family, and therefore struggle to identify the contraction for a fixed curve. The construction in [ 6 ] remedies this by constructing the ring for the contraction by hand, but the viewpoint therein does not make clear the connections between the contraction and other related problems, such as smoothing of differentials. Centrally aligned curves were introduced in [ 16 ] in order to give a modular desingularization of \widebar{\mathcal{M}}_{1,n}(\mathbb{P}^{r},d) . By giving a more explicit construction of the contraction used in that paper, it makes clear why factorization through the contraction is related to smoothing stable maps, in terms of smoothing functions pulled back from projective space. We should also be able to use our viewpoint on the contraction to give a novel proof of the splitting of the virtual class of \widebar{\mathcal{M}}_{1,n}(\mathbb{P}^{r},d) over the main and boundary components. The technique for the contraction constructed within should also generalize to higher genera by considering residues of functions paired with multiple independent differentials on a curve. Such contractions are vital to constructing modular desingularizations of the moduli spaces of stable maps to projective space, see, e.g., the aforementioned construction in [ 16 ] and the similar construction of Battistella and Carocci in [ 4 ] . Extending this approach even to just genus two presents interesting challenges and questions, such as the presence of non-reduced structures on the contraction seen in [ 4 ] , coming from degenerating curves with a double cover of \mathbb{P}^{1} . Lastly, by making the connection between the contraction and residues of differentials explicit, we hope to be able to give a new proof of the differential smoothing conjecture (valid in any characteristic) posed by Ranganathan and Wise. This conjecture was stated by Battistella and Bozlee in [ 3 ] and proven in characteristic 0 by Chen and Chen in [ 8 ] . In the process, we should also be able to give an algebraic proof of the global residue condition for smoothing of differentials introduced in [ 5 ] . 1.2. Acknowledgements First, I would like to thank my advisor, Jonathan Wise, for pushing me in the direction of this paper and offering endless insights and guidance. I would also like to thank Luca Battistella, Francesca Carocci, Dawei Chen, and Dhruv Ranganathan for helpful conversations, encouragement, and general interest in the content of this paper. A special thanks to Sebastian Bozlee for providing helpful comments on an earlier draft of this paper and for suggesting a fix for the proof of Lemma 3.7 . 2. Background 2.1. Genus One Singularities We work over an algebraically closed field k throughout. Let C be a reduced curve over k and let \nu:\tilde{C}\to C be the normalization at a closed point p\in C . Let \{p_{1},...,p_{n}\} be the preimages of p under the map \nu . We define two quantities: (i) the number of branches of the singularity at p is m(p)=n ; (ii) the \delta -invariant is \delta(p)=\text{dim}_{k}(\nu_{*}\mathcal{O}_{\tilde{C}}/\mathcal{O}_{C}) . We then define the genus of the singularity at p to be g(p)=\delta(p)-m(p)+1. We will be concerned with Gorenstein genus one singularities, which have been characterized by Smyth ( [ 17 ] Proposition A.3) as follows. Proposition 2.1 . For each n\geq 1 , there is a unique Gorenstein genus one singularity with n branches. Specifically, these singularities are the following: (i) n=1 : the cusp V(y^{2}-x^{3}) ; (ii) n=2 : the tacnode V(y^{2}-yx^{2}) ; (iii) n\geq 3 : the union of n general lines through the origin in \mathbb{A}^{n-1} . 2.2. Tropical Curves We refer the reader to [ 9 ] for the details of the following approach to tropical curves. Let M be a finitely generated, integral, saturated, sharp monoid. We define a tropical curve with edge lengths in M . Definition 2.2

Read the complete text