Cohomology of a Stack¶

Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks.

Version: v1 · 2026-09-28 · History
Domain-specific #: 8532
Domain group: Formal Sciences
Origin domain: Mathematics
Subdomains: Algebraic Geometry, Homological Algebra → Mathematics

Core Idea¶

Cohomology of a stack extends sheaf cohomology from schemes and algebraic spaces to algebraic stacks, whose points may possess automorphisms and whose geometry is presented through a groupoid rather than a single ordinary space. Given a stack, an appropriate site such as its lisse–étale or étale site, and a sheaf or complex of coefficients, cohomology is obtained as the derived functors of global sections or by an equivalent derived-category construction. The resulting graded groups record obstructions to gluing local data while retaining the stack's stabilizer information.

The construction must be independent of the chosen presentation. If a stack is represented by a groupoid or simplicial object, descent produces a spectral sequence or total complex that combines the cohomology of its levels with the groupoid relations. For a quotient stack [X/G], this recovers an algebraic analogue of equivariant cohomology; the classifying stack BG is the limiting case with X a point, so its cohomology captures characteristic information about G. Coefficient choices—quasi-coherent, étale, ℓ-adic, de Rham, or others—and the topology control what theorem is being asserted.

Stack cohomology is not merely the cohomology of a coarse moduli space. Passing to that space can erase isotropy and descent data, so two stacks with the same coarse points may have different cohomology. It is also not the Chow ring, although cycle classes and comparison maps can relate intersection-theoretic and cohomological invariants. The abstraction is a presentation-invariant derived global invariant for sheaves on a stack, designed to make local-to-global reasoning compatible with automorphisms and descent.

How would you explain it like I'm…

No faithful explanation at this level. A five-year-old picture can only show an ordinary collection of points, erasing the automorphisms carried by stack points and collapsing the idea into cohomology of the coarse space, which is exactly what stack cohomology is not.

Symmetry-Aware Gluing Test

In geometry, mathematicians use cohomology to measure how small, local pieces of information fit together into a whole, and where they fail to fit. A stack is a special kind of shape whose points can carry their own symmetries, like a point that can be spun around without changing. Cohomology of a stack does this measuring while keeping track of those symmetries. That's why two stacks that look like they have the same points can still give different answers.

Sheaf Cohomology on Stacks

Cohomology of a stack extends sheaf cohomology, a tool that measures the obstructions to gluing local data into global data, from ordinary spaces to algebraic stacks. A stack is a geometric object whose points may have automorphisms (symmetries), and it is often described using a groupoid, a set of points together with the ways points are identified. To compute its cohomology, you choose a suitable site (a notion of 'open covers') and coefficients, then take derived global sections. The answer must not depend on how the stack is presented. For a quotient stack [X/G] it gives an algebraic version of equivariant cohomology, and for the classifying stack BG it captures information about the group G. Importantly, it is not just the cohomology of the stack's coarse space of points, which can forget the symmetries.

Cohomology of a stack extends sheaf cohomology from schemes and algebraic spaces to algebraic stacks, whose points may carry automorphisms and whose geometry is presented by a groupoid rather than a single space. Given a stack, a site such as the lisse-etale or etale site, and a sheaf or complex of coefficients, cohomology is defined as the derived functors of global sections or an equivalent derived-category construction. The graded groups record obstructions to gluing local data while retaining stabilizer information. The construction must be independent of presentation: for a groupoid or simplicial presentation, descent yields a spectral sequence or total complex combining the cohomology of the levels with the groupoid relations. For a quotient stack [X/G] this recovers an algebraic analogue of equivariant cohomology, and the classifying stack BG, the case where X is a point, captures characteristic information about G. The coefficient type (quasi-coherent, etale, l-adic, de Rham, or others) and the topology determine which theorem is being asserted. It is not the cohomology of the coarse moduli space, which can erase isotropy and descent data, and it is not the Chow ring, though cycle class maps can relate them.

Structural Signature¶

Sig role-phrases:

the algebraic stack — geometric object with points, automorphisms, and descent data presented through a groupoid
the chosen site — lisse–étale, étale, or another topology defining covers and sheaves
the coefficient object — sheaf or complex, such as quasi-coherent, étale, l-adic, or de Rham data
the global-sections functor — operation collecting compatible local sections over the stack
the derived construction — higher derived functors or equivalent derived-category machinery producing graded groups
the presentation complex — simplicial object, groupoid nerve, spectral sequence, or totalization used for computation
the descent compatibility — relations ensuring local information and automorphisms glue correctly
the presentation-invariance requirement — equivalent stack presentations yield the same cohomological invariant
the stabilizer-sensitive output — higher obstruction data retaining isotropy erased by a coarse moduli space
the theory boundary — distinction from coarse-space cohomology and from Chow rings despite comparison maps

What It Is Not¶

Not merely cohomology of the coarse moduli space. Coarse passage can erase stabilizers, automorphisms, and descent data retained by the stack.
Not dependent on one chosen presentation. Groupoid or simplicial calculations must produce a presentation-invariant result.
Not one coefficient theory. Quasi-coherent, étale, l-adic, de Rham, and other coefficients live on different sites and support different theorems.
Not ordinary point-set topology on a set of isomorphism classes. Stack geometry includes automorphisms at points and local descent relations.
Not the Chow ring. Cycle-theoretic and cohomological invariants can be related by comparison maps without becoming identical.
Not simple levelwise cohomology without descent. Spectral sequences or total complexes incorporate the maps and coherences among presentation levels.
Not merely equivariant cohomology as an analogy. Quotient stacks actually recover an algebraic form of equivariant reasoning, while general stacks extend beyond quotient presentations.

Scope of Application¶

Cohomology of a stack applies to local-to-global sheaf or complex invariants on stacks where stabilizers, automorphisms, and descent data carry geometric information lost by coarse spaces.

Quotient stacks. Presentations [X/G] retain group action and isotropy in the cohomological object.
Classifying stacks. BG connects stack cohomology with group, equivariant, and characteristic-class phenomena.
Descent and sheaves. Local data on a site are glued with automorphism-compatible conditions.
Obstruction theory. Cohomological classes control existence, lifting, and deformation problems in moduli settings.
Spectral sequences. Groupoid, simplicial, or atlas presentations provide computable pages when differentials and convergence are tracked.
Comparison theorems. Maps among stack, equivariant, coarse, and other theories require exact hypotheses.
Multiple coefficient theories. Etale, lisse-etale, quasi-coherent, de Rham, and l-adic variants have different objects and results.
Applicability boundary. Coarse-moduli cohomology and Chow rings are not substitutes; naive invariant-taking and levelwise presentation groups omit derived or stacky contributions, and presentation independence must be proved.

Clarity¶

Cohomology of a stack extends sheaf-cohomological gluing and obstruction data to objects presented with groupoid symmetry and stabilizers. Clarity requires the stack, chosen site or topology, coefficient sheaf or complex, and derived construction; saying ‘cohomology’ without these can hide materially different invariants. Presentation independence is load-bearing: a groupoid atlas is computational data, not the invariant itself. The sharper question is how descent combines local cohomology with automorphism information and whether a proposed computation is invariant under replacing the presentation by an equivalent one.

Manages Complexity¶

Cohomology of a stack compresses local geometric data, gluing obstructions, and stabilizer symmetry into graded groups defined on an appropriate site or through a presentation-independent derived construction. The analyst tracks the stack, topology, coefficients, and descent data rather than recomputing each atlas overlap by hand. Quotient, algebraic, and differentiable-stack branches use related machinery with different sites. Spectral sequences organize contributions by simplicial level or group action. This compression preserves automorphism information that coarse moduli spaces discard while ensuring equivalent groupoid presentations yield the same invariant rather than presentation-dependent bookkeeping.

Abstract Reasoning¶

Site-selection move. Choose the stack's lisse–étale, étale, or other appropriate site and coefficient object before inferring cohomology. Descent move. From a groupoid or simplicial presentation, build a total complex or spectral sequence and infer global classes from compatible local data. Invariance move. Replace the atlas by an equivalent presentation and require the resulting groups to agree. Stabilizer move. For quotient stacks, reason from group action and equivariant data rather than collapsing immediately to the coarse quotient. Boundary move. Cohomology without site, coefficients, and grading is incomplete, and presentation-level calculations are not themselves the invariant.

Knowledge Transfer¶

Within the home domain. Cohomology of a stack transfers across algebraic geometry, moduli problems, orbifolds, and descent settings where objects have automorphisms and local data glue only up to coherent isomorphism. Site, sheaves, derived functors, stabilizers, descent, and spectral sequences retain formal roles. Beyond the home domain (C — invariant machinery). It applies literally to stacks in any compatible geometric category, not metaphorically to ordinary software stacks. Its boundary is theory-dependent: topology, coefficients, and cohomology notion must be specified, and coarse moduli spaces can lose stabilizer information. Similar groups do not imply equivalent stacks or moduli interpretations.

Examples¶

Canonical¶

For a quotient stack [X/G], one may choose a presentation by the action groupoid and form its simplicial nerve X, G×X, G×G×X, and so on. A sheaf on the chosen site supplies compatible data on these levels. Applying global sections and totalizing the resulting complex yields higher cohomology, often through a spectral sequence. The answer retains information from stabilizer groups that cohomology of the coarse orbit space can lose. Replacing the presentation by an equivalent groupoid must not change the resulting invariant.

Mapped back: [X/G] is the algebraic stack, topology the chosen site, and sheaf the coefficient object. Sections use the global-sections functor; totalization the derived construction over the presentation complex. Compatibility is the descent compatibility, equivalence the presentation-invariance requirement, and isotropy the stabilizer-sensitive output.

Applied / In Practice¶

An algebraic geometer computes line-bundle obstruction classes on a moduli stack using an atlas and its iterated fiber products. Local cocycles are checked on overlaps and against automorphisms, then descended to stack cohomology. The same computation is repeated with a refined atlas as a consistency check. A class visible only through stabilizer action is not discarded merely because the coarse moduli space has trivial corresponding cohomology, and it is not automatically called a Chow class.

Mapped back: Atlas overlaps instantiate the presentation complex and cocycle checks the descent compatibility. Refinement tests the presentation-invariance requirement. Retained automorphism data is the stabilizer-sensitive output, while separating coarse cohomology and Chow theory preserves the theory boundary.

Structural Tensions¶

T1 — Identity versus admissible variation. Cohomology of a Stack must remain recognizable across legitimate variants. Admissible variation is bounded by this condition: Presentations [X/G] retain group action and isotropy in the cohomological object. The stable element is expressed by this invariant: Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks. Treating every surface change as a new abstraction fragments the identity, while allowing a change to the constitutive relation produces a false positive.

Diagnostic: After the proposed variation, can an analyst still establish this invariant: Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks?

T2 — Recognition versus proxy. The domain needs observable or inferential evidence for Cohomology of a Stack, but the evidence is not automatically the identity. The working recognition rule is: the theory boundary — distinction from coarse-space cohomology and from Chow rings despite comparison maps. A familiar indicator can occur without the defining relation, and the relation can persist when a customary detector is unavailable.

Diagnostic: Does the evidence establish the defining claim—Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks—or only a correlated sign?

T3 — Definition versus operational judgment. A compact definition aids reuse, whereas actual classification in algebraic geometry can require expert decisions about boundary conditions, measurements, conventions, or exceptions. The construction must be independent of the chosen presentation. The definition must constrain those judgments without pretending that every admissible case can be recognized from a label alone.

Diagnostic: Which observation would make a competent practitioner reject the classification under the stated definition?

T4 — Scope versus overextension. Cohomology of a Stack has a genuine habitat in which presentations [X/G] retain group action and isotropy in the cohomological object. Yet Coarse-moduli cohomology and Chow rings are not substitutes; naive invariant-taking and levelwise presentation groups omit derived or stacky contributions, and presentation independence must be proved. A useful application map therefore has to be broad enough to cover recurring practice and narrow enough to exclude merely topical or metaphorical occurrences.

Diagnostic: Can the claimed application fill the same carrier and relation roles, or has only the name traveled?

T5 — Transfer versus domain accent. Knowledge about Cohomology of a Stack can travel within its home domain, and some structural lessons may travel farther. Cohomology of a stack transfers across algebraic geometry, moduli problems, orbifolds, and descent settings where objects have automorphisms and local data glue only up to coherent isomorphism. What transfers must be separated from the specialist vocabulary, warrant, and closure conditions that remain anchored in algebraic geometry.

Diagnostic: Is the receiving case a literal instance of Cohomology of a Stack, a co-instance of Abstraction, or only an analogy?

T6 — Autonomous identity versus forced placement. Cohomology of a Stack has a stable source-domain identity—Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks.—but no current live node supplies a necessary genus or structural prerequisite without distortion. Leaving the node unattached preserves the accepted identity and exposes a real gap in the present DAG rather than hiding it under a merely topical parent.

Diagnostic: Would the proposed parent be true of every Cohomology of a Stack instance for a reason stronger than shared vocabulary or subject matter?

Structural–Framed Character¶

Cohomology of a Stack is mixed: structurally specifiable but materially dependent on its disciplinary frame. Its structural side consists of the carrier the algebraic stack — geometric object with points, automorphisms, and descent data presented through a groupoid and the constitutive relation Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks. Its framed side comes from algebraic geometry, which fixes what the terms denote, what counts as evidence, and when a qualification or exception defeats the classification.

Across the principal tests, the entry is not merely a free-floating pattern. Evaluative weight: the identity can be stated descriptively even when its use has practical or normative consequences. Practice dependence: the theory boundary — distinction from coarse-space cohomology and from Chow rings despite comparison maps. Institutional stabilization: disciplinary conventions may stabilize the name and test without necessarily creating every underlying event or relation. Vocabulary portability: the invariant is Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks. Import versus recognition: an outside case qualifies literally only if the same typed roles and collapse condition are available; otherwise the comparison is analogical.

No current parent captures the reusable remainder without losing or distorting the defining relation. Cohomology of a Stack is therefore admitted as an approved unparented root. This is an explicit graph disposition, not a claim that the abstraction has no relations or that a later densification pass cannot discover one.

Structural Core vs. Domain Accent¶

What is skeletal. The portable skeleton is a typed carrier organized by a constitutive relation, an invariant, a recognition test, and a collapse condition. Here the carrier is the algebraic stack — geometric object with points, automorphisms, and descent data presented through a groupoid. The decisive relation is Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks, which also states the controlling invariant at this level. Stripped of specialist nouns, this organization is represented by Abstraction.

What is domain-bound. algebraic geometry supplies the actual objects or agents, admissible transformations, units or conventions, standards of warrant, and named exceptions. In this case, recognition requires evidence for the theory boundary — distinction from coarse-space cohomology and from Chow rings despite comparison maps. Admissible variation is bounded by the condition that presentations [X/G] retain group action and isotropy in the cohomological object, and the classification collapses when coarse passage can erase stabilizers, automorphisms, and descent data retained by the stack. These are constitutive differentia, not illustrative decoration.

Why it remains a domain-specific node. The identity is stable within algebraic geometry, but no current live parent passes the necessary-relation test. The node is therefore an approved unparented root; future placement must preserve the theory boundary — distinction from coarse-space cohomology and from Chow rings despite comparison maps rather than attach the name by topical similarity.

Reviewed placement — approved unparented root. No current live node supplies a defensible necessary genus or structural prerequisite for Cohomology of a Stack. The reviewed identity is: Cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks. Attaching it to the accelerated suggestion would confuse topical similarity with hierarchy; the node is therefore admitted without a parent pending later graph densification.
Nearest catalog surface declined — Equivariant cohomology. Its rematch score was 0.284201. Retrieval proximity did not establish synonymy or parentage; the carrier, invariant, and collapse condition remain different.
Related reasoning operations. Evidence, comparison, boundary testing, and representation can support a case without becoming additional DAG parents.

Neighborhood in Abstraction Space¶

Cohomology of a Stack sits in a moderately populated region (54^th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Unclustered & Miscellaneous (2551 abstractions)

Nearest neighbors

Algebraic stack — 0.90
Burnside category — 0.85
Moduli Space — 0.85
Ringed Space — 0.85
Descent (Mathematics) — 0.85

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

Not to Be Confused With¶

A forced generic parent. No current live node passed the necessary-relation test. Tell: do not infer hierarchy from shared subject matter, method words, or retrieval proximity; preserve Cohomology of a Stack as an approved root until a genuine broader identity is available.
P Adic Hodge Theory. This is the closest catalog retrieval surface, not an accepted synonym or parent. Tell: Ask which entry's carrier, invariant, and collapse test the case actually satisfies; shared vocabulary or a score of 0.799412 is insufficient.
Not merely cohomology of the coarse moduli space. Coarse passage can erase stabilizers, automorphisms, and descent data retained by the stack. Tell: Require the positive recognition condition that the theory boundary — distinction from coarse-space cohomology and from chow rings despite comparison maps.
Not dependent on one chosen presentation. Groupoid or simplicial calculations must produce a presentation-invariant result. Tell: Replace the familiar surface feature and test whether cohomology of a stack extends sheaf-cohomological constructions to algebraic stacks by choosing an appropriate site and deriving global sections, including equivariant interpretations for quotient stacks.
A detector, representation, or consequence. A method may reveal Cohomology of a Stack, a notation may describe it, and an outcome may follow from it without any of those being identical to the abstraction. Tell: Would the defining relation remain if the present detector, notation, or downstream effect changed?
A metaphorical transfer. A case outside the home domain may resemble the structure while lacking its native role types and standards of warrant. Tell: If only the general organization survives, route the comparison to Abstraction rather than treating it as another Cohomology of a Stack instance.

References¶

Frozen Wikipedia revision: https://en.wikipedia.org/wiki/Cohomology_of_a_stack (revision 1339170696).
Supporting reference preserved in the packet: https://www.math.harvard.edu/~lurie/papers/tamagawa.pdf

The frozen Wikipedia revision is discovery provenance. The cited source set was reviewed for identity, formal or operational relation, and scope. The encyclopedia's structural synthesis is bounded to those claims; URL transport failure alone was not treated as substantive contradiction.