Smooth integral proper covers of every positive power node
ProvedMvPolynomial.CrossingQuotient.exists_smooth_integral_proper_cover_of_parameter_poweralgebraic-geometrynodal-curvesphilippon-multiplicitysmooth-morphisms
Let be a commutative integral domain and let . Consider the affine power node
There exist an integral scheme and a proper surjective morphism
such that the composite of with the original structure map to is smooth. Moreover, the actual restriction of over the principal open is an isomorphism:
No field, characteristic, perfectness, noetherianity or regularity assumption on is required. The case is included.
This gives the repeated-one-parameter local cover needed in nodal boundary constructions, with the full unchanged principal open retained for further gluing. Global relative projectivity and arbitrary products of several boundary parameters remain separate requirements of the general alteration construction.
Preamble
import Mathlib import Definitions.Def_MvPolynomial_CrossingResolution set_option autoImplicit false set_option backward.isDefEq.respectTransparency.types false open AlgebraicGeometry CategoryTheory CategoryTheory.Limits open MvPolynomial MvPolynomial.CrossingQuotient universe u
Formal statement
theorem MvPolynomial.CrossingQuotient.exists_smooth_integral_proper_cover_of_parameter_power
(K : Type u) [CommRing K] [IsDomain K] (n : ℕ) :
let R := Polynomial K
let c : R := Polynomial.X ^ (n + 1)
∃ (Y : Scheme.{u}) (q : Y ⟶ Spec (.of (CrossingQuotient R c))),
IsIntegral Y ∧ IsProper q ∧ Surjective q ∧
Smooth (q ≫ Spec.map (CommRingCat.ofHom
((algebraMap R (CrossingQuotient R c)).comp (Polynomial.C : K →+* R)))) ∧
IsIso (q ∣_ PrimeSpectrum.basicOpen (U c)) := by sorrySource
Derived polynomial-base local case of A. J. de Jong, Smoothness, semi-stability and alterations, Publications Mathématiques de l’IHÉS 83 (1996), Sections 3.3–3.6, pp. 63–65, especially the nodal product equations of 3.3 and the two-chart factorization in the proof of Proposition 3.6; https://www.numdam.org/article/PMIHES_1996__83__51_0.pdf . This is an auxiliary proper-cover statement for uv=t^(n+1), not a verbatim numbered assertion. The proof works over any integral domain by polynomial-coordinate calculations and actual gluing of proper modifications, without invoking regularity of the coefficient ring. It retains an actual isomorphism over D(u) and makes no projectivity or blowup-identification assertion for the iterated cover. The binary projective construction is the same one recorded in https://prove2.me/theorems/531b8729-e3ef-409f-a060-dc1b9f2c6dcd ; its complete helpers are included rather than imported as unproved assumptions.