nLab sheafification



Topos Theory

topos theory



Internal Logic

Topos morphisms

Extra stuff, structure, properties

Cohomology and homotopy

In higher category theory




Given a site CC, the sheafification functor universally turns presheaves on CC into sheaves.

It is characterized as being the left adjoint functor L:PSh(C)Sh(C)L : PSh(C) \to Sh(C) to the inclusion Sh(C)PSh(C)Sh(C) \hookrightarrow PSh(C) of sheaves into all presheaves, exhibiting this as a reflective subcategory.

Therefore sheafification is a special case of a very general phenonemon of localizations of categories. See category of sheaves for more.


Let (C,J)(C,J) be a site in the sense of: small category equipped with a coverage. Write PSh(C)PSh(C) for the category of presheaves on CC and

j:CPSh(C) j : C \to PSh(C)

for the Yoneda embedding. Write

i:Sh J(C)PSh(C) i : Sh_J(C) \hookrightarrow PSh(C)

for the category of sheaves: the full subcategory on those presheaves that are JJ-sheaves.



The inclusion of sheaves into presheaves admits a left adjoint functor, which hence exhibits Sh(C)Sh(C) as a reflective subcategory or reflective localization of PSh(C)PSh(C):

(Li):Sh J(C)LPSh(C). (L \dashv i) : Sh_J(C) \stackrel{\overset{L}{\leftarrow}}{\hookrightarrow} PSh(C) \,.

This follows by general properties discussed at reflective subcategory. We spell out the argument using the theory of localization at a set of morphisms satisfying a calculus of fractions.

Recall from the discussion at sheaf that Sh J(C)Sh_J(C) is by definition the full subcategory of PSh(C)PSh(C) on the local objects with respect to the morphisms

W={S({U i}):=lim ( i,jj(U i)× j(U)j(U j) ij(U i))j(U)|{U iU} iIJ} W = \left\{ S(\{U_i\}) := \lim_\to\left( \coprod_{i,j} j(U_i) \times_{j(U)} j(U_j) \stackrel{\to}{\to} \coprod_i j(U_i) \right) \to j(U) | \;\; \{U_i \to U\}_{i \in I} \in J \right\}

of sieve inclusions for all covering families of the coverage JJ. (Here the colimit lim \lim_\to is the coequalizer of the two injection maps, as indicated. This is spelled out in more detail at sheaf and at sieve.)

Now we invoke the following results:

  1. The localization proposition says that every full subcategory of a locally presentable category on the WW-local objects for a small set WW of morphisms is a reflective subcategory, given by the localization at these morphisms;

  2. By Gabriel-Zisman's theorem every reflective subcategory is the localization at the collection of morphisms inverted by the left adjoint (which by the localization proposition is the saturation of the original set of morphisms).

  3. If WW satisfies the axioms of a calculus of fractions then, by the discussion there, this localization is equivalently given by the category PSh(C)[W 1]PSh(C)[W^{-1}] whose objects are those of PSh(C)PSh(C) and whose morphisms are given by PSh(C)[W 1](X,A)lim X^wWXPSh C(X^,A)PSh(C)[W^{-1}](X,A) \simeq {\lim_{\to}}_{\hat X \stackrel{w \in W}{\to} X} PSh_C(\hat X,A).

Notice that an object is a local object with respect to the above set of morphisms WW precisely if it is local with respect to the set of all small colimits (in the arrow category Arr(PSh(C))Arr(PSh(C)) ) of such morphims (since the hom-functor PSh C(,A)PSh_C(-,A) sends colimits in the first argument to limits, and a limit of isomorphisms is an isomorphism).

Let hence W¯\bar W be the completion of AA under forming small colimits in Arr(PSh(C))Arr(PSh(C)).

We claim that the morphisms in W¯\bar W form a calculus of fractions. The first condition to check is that for all morphisms of presheaves Xj(U)X \to j(U) and every covering family {U iU}\{U_i \to U\} there is a morphism YXY \to X in W¯\bar W and a commuting diagram

Y S({U i}) s X j(U) \array{ Y &\to& S(\{U_i\}) \\ \downarrow && \downarrow^{\mathrlap{s}} \\ X &\to& j(U) }

in PSh(C)PSh(C). (It is sufficient to demand this for sWW¯s \in W \subset \bar W to deduce the stability conditions for all morphisms in W¯\bar W, since by universal colimits in the presheaf topos PSh(C)PSh(C) the pullback of a colimit is a colimit of pullbacks.)

Similarly, to see that we can find YXY \to X we use the co-Yoneda lemma to decompose XX as a colimit of representables Xlim jK jX \simeq {\lim_{\to}_j} K_j and then use universal colimits to deduce that we are looking at a diagram of the form

lim rs *K r S({U i}) s lim rj(K r) j(U). \array{ {\lim_{\to}}_r s^* K_r &\to& S(\{U_i\}) \\ \downarrow && \downarrow^{\mathrlap{s}} \\ {\lim_\to}_r j(K_r) &\to& j(U) } \,.

Since W¯\bar W is closed under colimits, it is hence sufficient that we show the stability condition for XX any representable. So we need to fill diagrams of the form

Y S({U i}) s j(K) j(U). \array{ Y &\to& S(\{U_i\}) \\ \downarrow && \downarrow^{\mathrlap{s}} \\ j(K) &\to& j(U) } \,.

For this now use the single condition on a coverage: that for {U iU}\{U_i \to U\} a covering family in the site (C,J)(C,J) we can find a covering family {K jK}\{K_j \to K\} such that every K jKK_j \to K factors through one of the U iUU_i \to U. But this means that also the sieves factor, and we have a commuting diagram

S({K j}) S({U i}) j(K) j(U). \array{ S(\{K_j\}) &\to& S(\{U_i\}) \\ \downarrow && \downarrow \\ j(K) &\to& j(U) } \,.

This shows that W¯\bar W satisfies the first condition at factorization system.

The second condition at calculus of fractions demands that if two composites of the form

XS({U i})j(U) X \stackrel{\to}{\to} S(\{U_i\}) \to j(U)

are equal in PSh(C)PSh(C), then there is a morphism YXY \to X in PSh(C)PSh(C) such that the two composites

YXS({U i}) Y \to X \stackrel{\to}{\to} S(\{U_i\})

are equal. But the sieve inclusions are monomorphisms, hence this condition is trivially satisyfied (choose YXY \to X to be the identity on XX). Again by decomposing into colimits by the co-Yoneda lemma and using universal colimits and the pasting law for pullbacks, the same follows for general morphisms in W¯\bar W

lim kX k X lim kS(U k,i) Y lim kj(U k) j(U). \array{ {\lim_\to}_k X_k &\stackrel{\simeq}{\to}& X \\ \downarrow \downarrow && \downarrow \downarrow \\ {\lim_\to}_k S(U_{k,i}) &\to& Y \\ \downarrow && \downarrow \\ {\lim_\to}_k j(U_k) &\stackrel{\simeq}{\to}& j(U) } \,.

by applying the above on each component kk of the colimit.

This gives us the localization L:PSh(C)Sh J(C)L : PSh(C) \to Sh_J(C) as described at calculus of fractions. By the discussion there we have that Sh J(C)Sh_J(C) is equivalently given by the category PSh(C)[W 1]PSh(C)[W^{-1}] with the same objects as PSh(C)PSh(C) and hom-sets given by

PSh(C)[W 1](X,A)lim {X^wW^X}PSh C(X^,A). PSh(C)[W^{-1}](X,A) \simeq {\lim_\to}_{\{\hat X \stackrel{w \in \hat W}{\to} X\}} PSh_C(\hat X,A) \,.

So we have that for XPSh(C)X \in PSh(C) a presheaf and ASh J(C)A \in Sh_J(C) the hom-set Sh C(L(X),A)Sh_C(L(X),A) is given by

Sh C(L(X),A)lim X^wW¯XPSh C(X^,A). Sh_C(L(X), A) \simeq {\lim_{\to}}_{\hat X \stackrel{w \in \bar W}{\to} X} PSh_C(\hat X,A) \,.

But if AA is a sheaf, it is a W¯\bar W-local object and hence PSh C(X^wX,A)PSh_C(\hat X \stackrel{w}{\to} X, A) is an isomorphism for all wW¯w \in \bar W. Hence the above colimit is over a diagram constant on its value at w=Id:XXw = Id : X \to X therefore we have a natural isomorphism

PSh C(X,A). \cdots \simeq PSh_C(X,A) \,.

This demonstrates the adjunction (Li)(L \dashv i).



For XPSh(C)X \in PSh(C) a presheaf on the site (C,J)(C,J), its sheafification L(X)L(X) is the presheaf given on any UCU \in C by

L(X):Ulim {U^wj(U)}PSh C(U^,X), L(X) : U \mapsto {\lim_\to}_{\{\hat U \stackrel{w}{\to} j(U)\}} PSh_C(\hat U, X) \,,

where the colimit on the right is over all wW¯w \in \bar W.


By the Yoneda lemma we have

L(X)(U)PSh C(j(U),L(X)). L(X)(U) \simeq PSh_C(j(U), L(X)) \,.

By the above proposition this is

Sh C(L(j(U)),L(X)). \cdots \simeq Sh_C(L(j(U)), L(X)) \,.

By the proof of the above proposition, using the formulas discussed at calculus of fractions, this hom-set is given by

lim {U^wW¯j(U)}PSh C(U^,X). \cdots \simeq {\lim_\to}_{\{ \hat U \stackrel{w \in \bar W}{\to} j(U)\}} PSh_C(\hat U, X) \,.


By the definition of W¯\bar W, the morphisms U^j(U)\hat U \to j(U) in W¯\bar W are colimits of diagrams of covering sieves

U^lim {KU}S({K i})lim {KU}j(K). \hat U \simeq {\lim_\to}_{\{K \to U\}} S(\{K_i\}) \to {\lim_\to}_{\{K \to U\}} j(K) \,.

This means (…) that the above colimit may be computed as two consecutive colimits of the form

lim {S({U i})j(U)}PSh C(S({U i}),X). {\lim_{\to}}_{\{S(\{U_i\}) \to j(U)\}} PSh_C(S(\{U_i\}), X) \,.

One such application is called the plus construction.


A morphism S({U i})XS(\{U_i\}) \to X out of a sieve into any presheaf is in components precisely a matching family of the presheaf XX on the covering {U iU}\{U_i \to U\}.


Use that the sieve is the coequalizer

S({U i}) ij(U i) i,jj(U i)× j(U)j(U j) S(\{U_i\}) \to \coprod_i j(U_i) \stackrel{\to}{\to} \coprod_{i,j} j(U_i) \times_{j(U)} j(U_j)

and that the hom-functor PSh C(,X)PSh_C(-,X) sends colimits to limits. More details on this computation are at sheaf.


The unit of the (Li)(L \dashv i)-adjunction has as components natural morphisms

XLX X \to L X

in PSh(C)PSh(C), from any presheaf into its sheafification. By general properties of reflective subcategories these morphisms are mapped to isomorphisms by L:PSh(C)Sh(C)L : PSh(C) \to Sh(C). Therefore these are local isomorphisms.

So every presheaf is related by a local isomorphism to its sheafification.

For sheaves with values in categories other than SetSet

For presheaves with values in categories other than Set, sheafification may be a difficult problem, unless one has some extra assumptions.

With values in models for finite-limit theories

Consider a type of structure TT defined in terms of an essentially algebraic theory finite limits (such as groups, algebras, modules, etc.), then internal TT-models are preserved by both direct images and inverse images of geometric morphisms. Therefore, the adjunction (Li):Sh J(C)PSh(C)(L \dashv i) : Sh_J(C) \to PSh(C) directly induces an adjunction between TT-models in sheaves and presheaves. And since finite limits of sheaves and presheaves are computed pointwise, TT-models in the category of (pre)sheaves are the same as (pre)sheaves of TT-models-in-SetSet.

Using the IPC-property

If a category AA satisfies the following assumptions, sheafification of presheaves in [S op,A][S^{op}, A] exists and is constructed analogously as for Set-valued sheaves.

This is true for instance for

  • the category Set of sets;

  • the category Grp of groups;

  • the category kAlgk Alg of kk-algebras;

  • the category Mod(R)Mod(R) of modules,

(but all of these are also TT-models for finite-limit theories TT).

One should say more: there are so many applications and fairly difficult theorems there; for example van
Osdol’s work.

Mike: Another way to think about this is: if you have a type of structure TT defined in terms of finite limits (such as groups, algebras, modules, etc.), then internal TT-models are preserved by both direct and inverse images of geometric morphisms. Therefore, the adjunction between sheaves and presheaves of sets directly induces an adjunction between TT-models in sheaves and presheaves. And since finite limits of sheaves and presheaves are computed pointwise, TT-models in the category of (pre)sheaves are the same as (pre)sheaves of TT-models-in-SetSet.

If TT is not defined in terms of finite limits, then internal TT-models in sheaves need not be the same as sheaves of TT-models-in-SetSet. My intuition would be that the former, rather than the latter, is the more interesting and important notion. For instance, a local ring in a topos of sheaves is a sheaf of rings whose stalks are local, rather than a sheaf taking values in the category of local rings, and this is usually what people care about. But since people have studied the other version, there must be important examples of it as well?

Urs: okay, I have added this

to the above now – so is the IPC-property business really unnecessary for the examples above?

Mike: I’m pretty sure it is not. Does anyone have any examples where the IPC-property business is important?


The description of sheafification in terms of local isomorphisms is in section 16.3 (for Set-valued presheaves) and section 17.4 (for more general presheaves) of

The description in terms of dense monomorphisms using Lawvere-Tierney topology is in section V.3 of

Extension of sheafification of presheaves with values in other categories has been advanced in

  • Alex Heller, K. A. Rowe, On the category of sheaves, Amer. J. Math. 84, 1962, 205-216

  • Barr, Grillet, and Van Osdol, Exact categories and categories of sheaves, Lecture Notes in Math., Vol. 236, Springer, Berlin, 1971

  • Friedrich Ulmer, On the existence and exactness of the associated sheaf functor, J. Pure Appl. Algebra 3, 1971, 295-306

  • Alexander Rosenberg, Almost quotient categories, sheaves and localizations, 181 p. Seminar on supermanifolds 25, University of Stockholm, D. Leites editor, 1988 (in Russian; partial remake in English exists)

Discussion in homotopy type theory is in

  • Kevin Quirin, Nicolas Tabareau, Lawvere-Tierney sheafification in Homotopy Type Theory, Journal of Formalized Reasoning, Vol 9, No 2, (2016) (web)
category: sheaf theory

Last revised on February 4, 2023 at 01:53:15. See the history of this page for a list of all contributions to it.