Lemma 63.9.4. Consider a cartesian square
of quasi-compact and quasi-separated schemes with f separated and of finite type. Then there is a canonical isomorphism
Moreover, these isomorphisms are compatible with the isomorphisms of Lemma 63.9.2.
Lemma 63.9.4. Consider a cartesian square
of quasi-compact and quasi-separated schemes with f separated and of finite type. Then there is a canonical isomorphism
Moreover, these isomorphisms are compatible with the isomorphisms of Lemma 63.9.2.
Proof. Choose a compactification j : X \to \overline{X} over Y and denote \overline{f} : \overline{X} \to Y the structure morphism. Let j' : X' \to \overline{X}' and \overline{f}' : \overline{X}' \to Y' denote the base changes of j and \overline{f}. Since Rf_! = R\overline{f}_* \circ j_! and Rf'_! = R\overline{f}'_* \circ j'_! the isomorphism can be constructed via
where the first arrow is the isomorphism given to us by the proper base change theorem (Étale Cohomology, Lemma 59.91.12 in the bounded below torsion case and Étale Cohomology, Lemma 59.92.3 in the case that \Lambda is torsion) and the second arrow is the isomorphism of Lemma 63.3.12.
To finish the proof we have to show two things: first we have to show that the isomorphism of functors so obtained does not depend on the choice of the compactification and second we have to show that if we vertically stack two base change diagrams as in the lemma, then these base change isomorphisms are compatible with the isomorphisms of Lemma 63.9.2. A straightforward argument which we omit shows that both follow if we can show that the isomorphisms
Rg_* \circ Rf_* = R(g \circ f)_* for f : X \to Y and g : Y \to Z proper,
g_! \circ f_! = (g \circ f)_! for f : X \to Y and g : Y \to Z separated and quasi-finite, and
g_! \circ Rf'_* = Rf_* \circ g'_! for f : X \to Y and f' : X' \to Y' proper and g : Y' \to Y and g' : X' \to X separated and quasi-finite with f \circ g' = g \circ f'
are compatible with base change. This holds for (1) by Cohomology on Sites, Remark 21.19.4, for (2) by Remark 63.3.14, and (3) by Lemma 63.8.5. \square
Comments (0)
There are also: