Néron–Severi Group¶

The Picard-group quotient that identifies algebraically equivalent line bundles on a smooth projective complex variety.

Version: v1 · 2026-10-07 · History
Domain-specific #: 13958
Domain group: Formal Sciences
Origin domain: Mathematics
Subdomain: Algebraic Geometry → Mathematics
Aliases: Néron–Severi group, Neron-Severi group

Core Idea¶

For a smooth projective complex variety \(X\), the Néron–Severi group is \(\operatorname{NS}(X)=\operatorname{Pic}(X)/\operatorname{Pic}^{0}(X)\). The Picard group comprises isomorphism classes of algebraic line bundles under tensor product. Its connected identity component comprises algebraically trivial classes. The quotient identifies line bundles that differ by an algebraically trivial class, so an NS element is a class modulo algebraic equivalence, not a particular bundle. The resulting abelian group is finitely generated; its rank is the Picard number, an invariant of the quotient rather than its definition.[^ref-c97667977c8b]

Scope of Application¶

The construction applies across unlike smooth projective complex varieties. For a positive-dimensional complex abelian variety, \(\operatorname{Pic}^{0}(A)\) is its dual abelian variety and has positive dimension, so quotienting discards a genuine family of Picard classes. For a complex projective K3 surface, \(\operatorname{Pic}^{0}(X)=0\), so \(\operatorname{NS}(X)=\operatorname{Pic}(X)\). The K3 lattice has rank \(1\leq\rho(X)\leq20\) and intersection signature \((1,\rho(X)-1)\) under the projectivity condition. A nonprojective complex K3 may have rank zero.[^{ref-c97667977c8b][}ref-5f5da888f95f]

The sources define NS in settings wider than this entry's shared smooth projective complex scope. That scope keeps the two cases and their Picard-component interpretations comparable without silently extending a rank or signature statement across fields.[^{ref-c97667977c8b][}ref-5f5da888f95f]

Clarity¶

There are two classification steps. Picard identifies isomorphic line bundles. NS then identifies Picard classes whose difference lies in \(\operatorname{Pic}^{0}(X)\), the algebraic-equivalence kernel. On a K3 this kernel happens to vanish; on a positive-dimensional abelian variety it does not. The same quotient rule covers both, so neither the K3 equality nor the abelian symmetric-map realization should replace the definition.[^{ref-c97667977c8b][}ref-5f5da888f95f]

The Picard number is only the rank of the finitely generated group. Numerical equivalence or a cohomological realization may relate to NS under additional hypotheses, but each is a different test until those hypotheses are supplied.[^{ref-c97667977c8b][}ref-5f5da888f95f]

Manages Complexity¶

Quotienting separates algebraically trivial variation of line-bundle classes from the finitely generated class group. This can turn a continuously varying Picard family into discrete coset data useful for rank or intersection calculations. The cost is precise: an NS coset does not recover which member of \(\operatorname{Pic}^{0}(X)\) represented the original Picard class when the kernel is nontrivial. One may keep the full Picard group when that information is needed.[^ref-c97667977c8b]

Abstract Reasoning¶

To test a proposed NS calculation, fix \(X\), form its Picard group, identify \(\operatorname{Pic}^{0}(X)\), and ask whether two line-bundle classes have algebraically trivial difference. Form cosets before computing a rank or using a variety-specific description. Removing the kernel step leaves the Picard source rather than NS; replacing it with a different equivalence relation defines a different quotient. A zero kernel still gives a valid NS quotient, as the K3 case shows.[^{ref-c97667977c8b][}ref-5f5da888f95f]

As a group, the quotient has a closed associative coset operation, identity, inverses, and its ordinary translation action on itself. It therefore strictly instantiates the live Group Prime. It also strictly presupposes the domain-specific Picard Group as its source; the quotient is not generally a Picard group itself, nor does its group action imply an action on the underlying variety.[^ref-c97667977c8b]

Knowledge Transfer¶

The positive-dimensional abelian variety and the projective K3 carry the same roles: base variety, Picard source, algebraically trivial subgroup, and quotient. What changes is the subgroup and the useful realization. On an abelian variety, line bundles yield symmetric maps to the dual; on a K3, the quotient coincides with a lattice of integral \((1,1)\)-classes. These case accents cannot be transferred wholesale, but the quotient test does transfer within the evidenced algebraic-geometric scope. The more portable group and quotient logic is already broader than this named geometry-specific entry.[^{ref-c97667977c8b][}ref-5f5da888f95f]

Example¶

Complex abelian variety. For a positive-dimensional \(A\), the carrier is \(A\), the source group is \(\operatorname{Pic}(A)\), the nontrivial kernel is \(\operatorname{Pic}^{0}(A)=\widehat A\), and the output is \(\operatorname{NS}(A)\). The map from a line bundle \(L\) to a symmetric homomorphism \(\phi_L:A\to\widehat A\) gives a specialized realization of its quotient class over the algebraically closed field. A polarization needs further positivity; symmetric homomorphism alone is not that claim.[^ref-c97667977c8b]

Complex projective K3 surface. The carrier is \(X\), the source is \(\operatorname{Pic}(X)\), the kernel is zero, and the output is \(\operatorname{NS}(X)=\operatorname{Pic}(X)\). The integral \((1,1)\)-class and intersection-lattice description is specific to this geometry; projectivity supports the positive rank and signature above. The example shows that the quotient definition does not require a nontrivial kernel.[^ref-5f5da888f95f]

Relationships to Other Abstractions¶

Current abstraction Néron–Severi Group Domain-specific

Parents (2) — more general patterns this builds on

Néron–Severi Group is a kind of Group Prime

Every Néron–Severi quotient is an abelian group, whereas groups need not arise from line bundles.
Néron–Severi Group presupposes Picard Group Domain-specific

Forming NS(X) necessarily starts from the live Picard Group Pic(X).

Neighborhood in Abstraction Space¶

Néron–Severi Group sits in a sparse region of the domain-specific corpus (88^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2551 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With¶

Picard Group is the numerator, Picard variety describes the connected geometric component, and Picard number is only the rank. A single line bundle is an object represented before the quotient. Do not silently replace algebraic equivalence with numerical equivalence or identify NS with all integral cohomology. Exceptional coincidences in a K3 case do not make these definitions synonymous for every variety.[^{ref-c97667977c8b][}ref-5f5da888f95f]

References¶

[^ref-c97667977c8b]: Brian Conrad, lecture notes by Tony Feng, Abelian Varieties, Stanford University, Spring 2015, author-edited course notes. §5.3 Definition 5.3.1 and Theorem 5.3.2, printed pp.49–50, define NS and prove finite generation; §5.1 and §5.2 cover the dual and the \(\phi_L\) kernel; §7.4 Proposition 7.4.3 and Remark 7.4.5, printed p.68, give the symmetric-map realization over an algebraically closed field. [^ref-5f5da888f95f]: Daniel Huybrechts, Lectures on K3 Surfaces, author-hosted book PDF, Chapter 17 §1.1 and Remark 1.1, printed pp.358–359. These establish \(\operatorname{Pic}(X)\cong\operatorname{NS}(X)\) for complex K3, the integral \((1,1)\) description, and the projective rank/signature qualifications.