Representability of codimension three cycles

Kalyan Banerjee · 2026

Read Representability of codimension three cycles by Kalyan Banerjee (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-2604-20223.txt.

Introduction

Banerjee builds the notion of representability for three-codimensional cycles on a fourfold, via zero cycles modulo rational equivalence on surfaces. A cycle of codimension three on a four-dimensional shape is thus called representable when it can be matched, up to rational equivalence, by zero cycles living on surfaces. The paper sets out this notion plainly, tying high-codimension cycles to the simplest cycles on surfaces. That link gives a clean way to test and sort such cycles through surface geometry.

Catalog record

CREATORKalyan Banerjee
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27721
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.20223

Browse the full library · About the Tower of Babel

Full text

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

Representability of codimension three cycles Kalyan Banerjee Abstract . In this paper, we develop the notion of representability of co-dimension three cycles on a fourfold in terms of zero cycles modulo rational equivalence on surfaces. 1. Introduction The representability question in the theory of Chow groups is an important question. Precisely, it means the following: let X be a smooth projective variety over the field of complex numbers and let {\rm{CH}}^{p}(X) denote the Chow group of co-dimension p algebraic cycles on X modulo rational equivalence. Let {\rm{CH}}^{p}(X)_{{\rm{alg}}} denote the subgroup of {\rm{CH}}^{p}(X) consisting of algebraically trivial cycles. We say that the group {\rm{CH}}^{p}(X)_{{\rm{alg}}} is representable if there exists a smooth projective curve C and a correspondence \Gamma on C\times X such that \Gamma_{*} from J(C)\cong{\rm{CH}}^{1}(C)_{{\rm{alg}}} to {\rm{CH}}^{p}(X)_{{\rm{alg}}} is surjective. The most interesting and intriguing case is the zero cycles on X . The first breakthrough result in this direction is by Mumford, [ M ] : the group {\rm{CH}}^{2}(S)_{{\rm{alg}}} of a smooth projective complex algebraic surface with geometric genus greater than zero is not representable. Roitman [ R1 ] further generalized for higher-dimensional varieties proving that if the variety X has a global holomorphic i -form on it, then the group {\rm{CH}}^{n}(X)_{{\rm{alg}}} is not representable. Here n is the dimension of X and we have 0<i\leq n . Then there is the famous converse question due to Spencer Bloch saying that: for a smooth projective complex algebraic surface S with geometric genus equal to zero, the group {\rm{CH}}^{2}(S)_{{\rm{alg}}} is representable. This question has been answered for the surfaces not of general type with geometric genus equal to zero by Bloch-Kas-Liebarman in [ BKL ] . The conjecture is still open in general for surfaces of general type, but it has been solved in many examples in [ B ] , [ IM ] , [ GP ] , [ PW ] , [ V ] , [ VC ] . In the case of highest codimensional cycles on a smooth projective variety X , the notion of representability can also be defined in another way. Definition 1.1 . We consider the natural map from the symmetric power {\rm{Sym}}^{d}X to {\rm{CH}}^{n}(X)_{{\rm{alg}}} , for d positive and n being the dimension of X . Suppose that this map is surjective for some d then we say that the group {\rm{CH}}^{n}(X)_{{\rm{alg}}} is representable. It can be proved as in [ [ Vo ] , Chapter 10, Section 1], that this notion of representability is equivalent to the first notion of representability introduced before. So following the approach of Voisin as in [ Vo ] , it is natural to ask whether there is a second notion of representability for lower co-dimensional cycles, specifically for co-dimension three cycles on a smooth projective fourfold. To do this, first, we assume that: (I) There exist a smooth, projective variety Y and a correspondence \Gamma of codimension p on Y\times X , such that \Gamma_{*} is surjective from {{\rm{CH}}_{0}(Y)}_{0}\cong{{\rm{CH}}_{0}(Y)}_{{\rm{alg}}} to {\rm{CH}}^{p}(X)_{0} , (here {{\rm{CH}}_{0}(Y)}_{0},{\rm{CH}}^{p}(X)_{0} denote the group of degree zero cycles on Y,X , respectively modulo rational equivalence). Under this assumption, the group {\rm{CH}}^{p}(X)_{0} is contained in {\rm{CH}}^{p}(X)_{{\rm{alg}}} . On the other hand, {\rm{CH}}^{p}(X)_{{\rm{alg}}} is always contained in {\rm{CH}}^{p}(X)_{0} (this fact is due to the existence of Chow varieties; for reference, see [ [ Fu ] , Example 10.3.3]). Now, the second notion of representability is as follows: Consider the two-fold product of the Chow variety C^{p}_{d}(X)\times C^{p}_{d}(X) where C^{p}_{d}(X) is the projective variety that parameterizes the codimension p and degree d cycles on X . Note that the degree of a cycle is determined by fixing a embedding of X into a projective space. Consider the natural map from this product to the group {\rm{CH}}^{p}(X)_{0} and hence (under assumption) to {\rm{CH}}^{p}(X)_{{\rm{alg}}} and ask the question: Does the surjectivity of a proper countable union (which is not a finite union) of these maps implies the representability in the first sense of {\rm{CH}}^{p}(X)_{{\rm{alg}}} up to dimension two, i.e: Let X be a smooth projective variety and let A^{p}(X) denote the algebraically trivial, codimension p algebraic cycles on X , modulo rational equivalence, which was previously denoted by CH^{p}(X)_{{\rm{alg}}} (for notational convenience). Then we say that A^{p}(X) is weakly representable up to dimension two if there exist finitely many curves C_{1},\cdots,C_{m} with correspondences \Gamma_{1},\cdots,\Gamma_{m} on C_{1}\times X,\cdots,C_{m}\times X and finitely many surfaces S_{1},\cdots,S_{n} with correspondences \Gamma_{j}^{\prime} on S_{j}\times X , such that \sum_{i}\Gamma_{i}+\sum_{j}\Gamma_{j}^{\prime} is surjective from \oplus_{i}A^{1}(C_{i})\oplus_{j}A^{2}(S_{j}) to A^{p}(X) . First, we prove the following in this direction: Suppose that: A) The Chow group of codimension p cycles is generated by linear subspaces, that is, the natural map (Abel-Jacobi) from {\rm{CH}}_{0}(F(X)) to {\rm{CH}}^{p}(X) is surjective. Here F(X) is the Fano variety of linear subspaces of codimension p . B) Suppose that A^{p}(X) is representable in the sense that the map from a proper countable union of the twofold product of the Chow variety to A^{p}(X) is surjective (which is not a finite union). Theorem 1.2 . With the above assumptions, there exist a smooth projective surface S and a correspondence \Gamma to S\times X , such that \Gamma_{*}:A^{2}(S)\to A^{p}(X) is surjective. In fact S is obtained as a hyperplane section of the Fano variety of co-dimension p -linear subspaces of X . As an application, we show that A^{3}(X) is REPRESENTABLE up to dimension two where X is a smooth, projective Fano fourfold such that A^{3}(X) is generated by differences of rational curves. This is due to the result of [ TZ ] over complex numbers. Our argument in this direction is a careful analysis of the argument present in the approach of Roitman in [ R ] , Voisin in [ [ Vo ] , Chapter 10, Section 1], where the authors deal with the case of zero cycles. First, we recall various notions of representability in the second sense, denoted as ”finite dimensionality” of Chow groups of codimension p cycles and show their equivalence. The key point is to use the Roitman’s result on the map from the twofold product of the Chow variety to A^{p}(X) saying that the fibers of this map are a countable union of Zariski closed subsets in the product of Chow varieties. Then after having these equivalent notions of ”finite dimensionality” in hand, we proceed to the main theorem. Throughout the text, we work in the field of complex numbers. 2. Notion of representability of codimension three cycles Let X be a smooth projective variety and let A^{i}(X) denote the algebraically trivial, codimension i algebraic cycles on X , modulo rational equivalence. Then we say that A^{i}(X) is weakly representable up to dimension two if there exist finitely many curves C_{1},\cdots,C_{m} with correspondences \Gamma_{1},\cdots,\Gamma_{m} on C_{1}\times X,\cdots,C_{m}\times X and finitely many surfaces S_{1},\cdots,S_{n} with correspondences \Gamma_{j}^{\prime} on S_{j}\times X , such that \sum_{i}\Gamma_{i}+\sum_{j}\Gamma_{j}^{\prime} is surjective from \oplus_{i}A^{1}(C_{i})\oplus_{j}A^{2}(S_{j}) to A^{i}(X) . If we assume that X is a fourfold, then the representability of A^{2}(X) is a birational invariant. This is because if we blow up X to \widetilde{X} , then A^{2}(\widetilde{X}) is isomorphic to A^{2}(X)\oplus A^{1}(Z) , where Z is the center of the blow-up. Since A^{1}(Z) is dominated by J(\Gamma) , for some smooth projective curve \Gamma , this will imply that

Read the complete text