Pith. sign in

REVIEW 4 major objections 5 minor 25 references

Composition of Sarkisov links between del Pezzo surfaces

T0 review · 4 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Any two birationally equivalent del Pezzo surfaces of Picard rank one over a perfect field are joined by at most two Sarkisov links.

desk verdict A credible proof of the optimal N=2 bound for Sarkisov links over perfect fields; the real work is field-tracking, but the conclusion leans on an imported classification whose characteristic scope is not stated. read the letter →

arxiv 2607.13270 v1 pith:2ELTVYUS submitted 2026-07-14 math.AG

Anastasia V. Vikulova This is my paper · ORCID
classification math.AG MSC 14E3014J2614J45
keywords SarkisovlinksdelPezzosurfacesPicardrankoneperfectfieldsbirationalmapsSeveri-BrauerdatatypeIIMorifibrespaces
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proves a bounded factorization theorem in the Sarkisov program over any perfect field: any birational map between two minimal del Pezzo surfaces of Picard rank one can be written as a composition of at most two Sarkisov links. The bound is optimal: over certain fields, a degree-6 del Pezzo surface without a rational point and the projective plane require exactly two links. This settles the dimension-2 analogue of a conjecture on Fano threefolds and gives a uniform statement that does not depend on the base field being algebraically closed. The proof works by classifying the possible degrees and field invariants and showing that any two surfaces can be connected through a degree-8 or degree-5 intermediate.

What carries the argument

The central object is a Sarkisov link of type II between minimal del Pezzo surfaces, drawn as a two-ray spin in which a del Pezzo surface T of Picard rank 2 is obtained by blowing up a point (of controlled degree) on each side; the point must be in Sarkisov general position, meaning that the blow-up is again a del Pezzo surface. The load-bearing technique is to prove that such general-position points exist for the right field extensions on degree-5 and degree-6 surfaces, and to use the Severi–Brauer data of a degree-6 surface to show that each link changes exactly one field invariant. This lets the author move the pair (K,L) of a degree-6 surface to any other pair in at most two steps, and s

What would settle it

Find a perfect field, for instance of positive characteristic, on which there exists a Sarkisov link from a minimal del Pezzo surface of degree 6 to P^2; then the lower-bound example collapses and the N=2 bound would need re-examination. Alternatively, produce two birationally equivalent minimal del Pezzo surfaces of Picard rank one that provably require three links under the same classification.

Watch

Extended reading notes

Core claim

The paper establishes Theorem 1.2: for any two smooth del Pezzo surfaces of Picard rank one over a perfect field that are birational to each other, there exists a birational map that decomposes as a composition of N=2 Sarkisov links, and N=2 is the best possible constant. The author combines the classification of Sarkisov links between minimal rational surfaces with explicit constructions of type-II links: for degree-5 surfaces, points in Sarkisov general position of degree 2 and degree 5 exist; for degree-6 surfaces, the Severi–Brauer data (a pair of fields encoding the field structure of the six (−1)-curves) can be adjusted one field at a time by links that blow up points of degree 2 or 3.

Load-bearing premise

The proof assumes that the full classification of Sarkisov links between minimal rational surfaces over perfect fields—especially the absence of a link from a degree-6 surface to P^2—is complete and valid in every characteristic.

Editorial extensions

If this is right

  • The Sarkisov graph of minimal del Pezzo surfaces of Picard rank one over any perfect field has diameter at most two.
  • The constant N=2 is optimal and independent of the base field, solving the dimension-2 case of the bounded-factorization conjecture.
  • Birational maps between such surfaces can be factored constructively through a degree-8 or degree-5 intermediate surface, giving an explicit normal form.
  • The result shows that the gap between degree-6 del Pezzo surfaces and P^2 is exactly two links, consistent with the absence of direct links in the classification of Sarkisov links.
  • For non-rational degree-6 surfaces, the minimal number of links is at most one, so the only obstruction to a single link is rationality over the given field.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The field-invariant viewpoint suggests a direct arithmetic interpretation: the minimal number of links measures how many field degrees of freedom (the two fields K and L in the Severi–Brauer data) must be changed, so surfaces differing in both fields are exactly those that require two links.
  • The lower-bound example could be turned into a testable prediction: any algorithm that computes a Sarkisov factorization of a birational map from a pointless degree-6 del Pezzo surface to P^2 must pass through a degree-8 surface.
  • A natural extension would be to check whether the same N=2 bound persists for del Pezzo surfaces of Picard rank one over imperfect fields, where the imported classification input may no longer hold.
  • The paper's own example of Hirzebruch surfaces shows that the two-link bound is special to Picard-rank-one del Pezzo surfaces and does not extend to conic bundles.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proves Theorem 1.2: for any two birationally equivalent smooth del Pezzo surfaces of Picard rank one over a perfect field, there is a birational map that decomposes into at most two Sarkisov links, and this bound is optimal. The proof splits into a rational and a non-rational case. For rational surfaces of degree 5, 6, 8, it uses the classification of Sarkisov links (Theorem 2.9, quoted from Iskovskikh and from Lamy–Schneider), together with new field-tracking lemmas (Lemmas 6.1–6.6) and results on points in Sarkisov general position (Lemmas 5.1–5.7). For non-rational surfaces it uses Proposition 3.7, quoted from Kurz–Yasinsky, and several imported classification theorems. The optimality example is Corollary 2.12: a minimal degree-6 del Pezzo surface and P^2 over Q, between which no single Sarkisov link exists and a composition of two type-II links is constructed in Example 2.10.

Significance. If Theorem 1.2 is correct, it gives a striking dimension-2 analogue of Corti's conjecture with an explicit and optimal bound N=2. The paper's own contributions—Lemmas 5.4–5.7 on general position and Lemmas 6.1–6.6 on how fields of definition of (−1)-curves behave under type-II links—are useful and appear internally coherent, assuming the imported results they cite. The paper is well organized, and the field-tracking chain through the degree-6 Severi–Brauer data is a sensible strategy. However, the main theorem and the optimality statement are almost entirely conditional on heavy imported classification results whose characteristic scope is never stated. This makes the manuscript, as it stands, not fully verifiable for the full generality claimed.

major comments (4)
  1. [§2, Theorem 2.9; §7, Lemma 7.1] Theorem 1.2 is stated for every perfect field, but the proof relies on the imported classification of Sarkisov links in Theorem 2.9 (quoted from [14, Theorem 2.4] and [9, Theorem 2.6]) without saying in which characteristics those results are proved. The load-bearing assertions obtained purely by referring to this theorem include: there is no link from a degree-6 minimal del Pezzo surface to P^2 (Lemma 7.1(vi), Corollary 2.12), there is no link between degrees 5 and 6 (Lemma 7.1(vii)), the existence of the type-II edges D5–P^2, D5–D8, D6–D8, D6–D6, D8–P^2 (Lemmas 6.1–6.6), and the first-paragraph statement in Lemma 7.1 that any composition containing a non-type-II link has length at least 3. If any of these edges is missing, or an extra edge D6–P^2 exists, both the upper bound N=2 and the optimality example fail. The manuscript must either quote the precise characteristic hypotheses of [
  2. [§8, proof of Theorem 1.2; §3, Proposition 3.7] The same lack of characteristic hypotheses affects other imported results used to close the proof: [9, Theorems 1.6, 4.4, 4.5] for d≤3, [18, Theorems 1.1, 1.2] for d=4, [23, Theorem 1.6] for degree 8, and Lemma 2.13 from [19]. In particular, [18] is a recent arXiv preprint, and its hypotheses are not quoted, so a reader cannot check whether it covers, for example, positive characteristic. Proposition 3.7 is stated as a proved proposition, but its proof is essentially a citation to [13, Proposition 7.4]; it should be presented explicitly as an imported theorem with its precise field and characteristic hypotheses. The same applies to [13, Corollaries 6.7 and 6.11] used in Lemmas 6.4 and 6.5.
  3. [§3, Lemma 3.5] Lemma 3.5 is used in Lemma 6.3 to identify the quadratic field K attached to a degree-6 surface with the field K' attached to a degree-8 surface. Its proof is too compressed to verify: from the assumption K'≠K it says 'This means that there are two contractions on minimal del Pezzo surfaces of the same degree d'. By Theorem 2.9 we obtain d'=9.' I cannot follow why two contractions force d'=9, nor how Remark 3.4 then implies K=K'. Since Lemma 6.3 is a key ingredient for Lemma 7.1(vi)–(viii), this step needs a complete proof or a precise citation.
  4. [§5, Lemma 5.7] The proof of Lemma 5.7 contains a notational gap: η(P) is a Galois orbit of three points, not a point on V_K, but the proof then says 'η(P) lies on a line L' and later writes 'P∈L∩...' as though P were a single point. The argument can likely be repaired by working with the orbit and invariant lines, but as written the case analysis is not rigorously checkable. Since this lemma is needed for Lemma 6.5 and hence for Lemma 7.1(ix), it should be rewritten carefully.
minor comments (5)
  1. [§2, Figure 2] The diagram in Theorem 2.9 is hard to read and the legend is terse. Since the paper later relies on specific nonexistence statements (e.g., no D6–P^2 edge), it would help to spell out the needed consequences in a numbered statement rather than leaving the reader to decode the graph.
  2. [§7, Lemma 7.1(v)] In the proof of the (8,8) case, the reference to Lemma 6.1 appears to be a typo: that lemma handles degree 5. The desired link from a degree-8 surface to P^2 is the inverse of the link constructed in Lemma 6.6.
  3. [§5, Lemma 5.7] In addition to the gap noted above, the proof should distinguish notationally between the degree-3 scheme P and its geometric points P1,P2,P3, and between η(P) and the set {η(P1),η(P2),η(P3)}.
  4. [§2, Example 2.10] The diagram uses X5, X7, X8 without explanation. A few words identifying which morphisms are blow-ups and which are contractions would make the example easier to follow.
  5. [§1, Remark 1.3] The assertion that over Q every pair of Hirzebruch surfaces can be connected by a single Sarkisov link is stated without proof or reference; since it is not used in the main theorem, it could simply be deleted or given a citation.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: Theorem 1.2 is proved from the external Iskovskikh–Lamy–Schneider classification and independent lemmas. The only circular step is a quoted proposition whose proof invokes itself, a minor proof-of-record defect.

  1. other [Section 3, Proposition 3.7 (proof), p. 9]
    "By Lemma 3.6 we have Ind(X)̸= 1. By [9, Theorem 2.6] we get that any Sarkisov link starting from X ends up with minimal del Pezzo surface of degree 6. Therefore, by [13, Proposition 7.4], which says that any birational non-rational non-isomorphic minimal del Pezzo surfaces of degree 6 can be connected with a Sarkisov link, we get the desired."

    Proposition 3.7 is stated verbatim as [13, Proposition 7.4]. Its proof concludes by invoking [13, Proposition 7.4] as the final justification, so the proof assumes the very statement it is establishing. This is a circular proof-of-record. However, the proposition is an imported external theorem rather than a result derived inside the paper, and the main theorem's dependency is on the cited statement, not on this proof, so this does not make the N=2 bound circular.

full rationale

The central claim—that any birational pair of minimal del Pezzo surfaces over a perfect field is connected by at most two Sarkisov links—is not assumed. Its proof in Section 8 splits into rational and non-rational cases. The rational case is handled by Lemma 7.1, whose estimates are built from Lemmas 6.1–6.6, and these lemmas construct explicit links (degree-5 to P^2, degree-5 to degree-8, degree-6 to degree-8, degree-6 to degree-6 with degree-2 or degree-3 blowups). The lower bound and the optimality example (Corollary 2.12) use the absence of a degree-6-to-P^2 edge, which is taken from the classification Theorem 2.9. All of these are external inputs or independently argued lemmas; no fitted parameter is relabelled as a prediction, and no equation is rigged by definition. The paper's dependence on Theorem 2.9 (quoted from [14, Theorem 2.4], going back to Iskovskikh [9, Theorem 2.6]) and on the very recent [13, Prop 7.4; Cor 6.7, 6.11] is a soundness/completeness risk over arbitrary perfect fields, but it is not circular reasoning. The only concrete circular step is Proposition 3.7's proof, which cites the proposition itself; because the proposition is a quoted theorem, this is a presentation flaw rather than a load-bearing derivation. Accordingly the circularity score is low.

Assumptions & free parameters 0 free parameters · 9 assumptions · 0 invented entities

The paper contributes field-of-definition lemmas and a global argument, but the proof is built on a chain of imported classification results: Iskovskikh's Sarkisov graph, Kurz–Yasinsky's Severi–Brauer data, Zaitsev's degree-5 forms, Trepalin's degree-8 classification, and Shramov–Trepalin's degree-4 classification. No free parameters are fitted; no new entities are postulated. The key risk is whether all cited results hold over every perfect field, including positive characteristic.

assumptions (9)
  • domain assumption Theorem 2.9: complete classification of Sarkisov links between minimal rational surfaces over a perfect field.
    Imported from [14, Theorem 2.4] and [9, Theorem 2.6]. Used throughout §7–8 for both the upper bound and the lower-bound optimality example.
  • domain assumption Theorem 2.3: a del Pezzo surface of degree ≥5 with a K-point is K-birational to P^2.
    Cited from [24, Theorem 2.1]; used in §8 to divide into rational and non-rational cases.
  • domain assumption Theorem 2.2: a del Pezzo surface of degree 5 always has a K-point.
    Cited from [16]/[22]; used for the rational degree-5 case (Lemma 6.1) and in §8.
  • domain assumption Proposition 3.7 ([13, Prop 7.4]): non-isomorphic birational minimal degree-6 del Pezzo surfaces without K-points are connected by one Sarkisov link.
    Restated as Proposition 3.7; used for the non-rational d=6 case in §8. This is an imported result, not proved in this paper.
  • domain assumption [13, Cor 6.7 and 6.11]: field equalities K and L are preserved/transformed as stated by degree-2 and degree-3 Sarkisov links on degree-6 del Pezzo surfaces.
    Directly quoted in the proofs of Lemmas 6.4 and 6.5; load-bearing for Lemma 7.1(viii)-(ix).
  • domain assumption [18, Theorems 1.1 and 1.2]: birational del Pezzo surfaces of degree 4 are isomorphic.
    Used in §8 to dispose of the case d=4. Very recent (arXiv 2512.19660).
  • domain assumption [23, Theorem 1.6]: pointless del Pezzo surfaces of degree 8 that are birational are isomorphic.
    Used in §8 for the non-rational d=8 case.
  • domain assumption [19, Corollary 2.10]: a birational map from a Severi–Brauer surface of degree 9 to a minimal del Pezzo surface is an isomorphism or a type-II Sarkisov link.
    Used as Lemma 2.13 for the non-rational d=9 case.
  • domain assumption [25, Lemmas 2.3, 8.4, Cor 8.3, 8.5]: classification of degree-5 del Pezzo forms by Galois data and minimality criteria.
    Used in Lemma 4.1 and Lemma 5.3; ensures degree-5 surfaces are determined by the splitting field M and remain minimal after quadratic extension.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Composition of Sarkisov links between del Pezzo surfaces." pith.science (2026). https://pith.science/paper/2ELTVYUS

@misc{pith2026260713270,
  author       = {Pith},
  title        = {Pith review of: Composition of Sarkisov links between del Pezzo surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2ELTVYUS}},
  note         = {Machine review of arXiv:2607.13270}
}
read the original abstract

We prove that for any two birationally equivalent del Pezzo surfaces of Picard rank one over a perfect field there is a birational map between them such that it is decomposed in a composition of at most two Sarkisov links.

Figures

Figures reproduced from arXiv: 2607.13270 by the authors.

Figure 1
Figure 1. The four types of Sarkisov links for dimension 2. where “div” and “fib” mean divisorial contraction and fibration, respectively. Fig￾ure 1 is taken from [2, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Sarkisov links between minimal rational surfaces over a perfect field where an edge between two vertexes means that there is a Sarkisov link between some minimal rational surfaces which classes are written in the vertexes. Roman number in the grey box on the edge means the type of the Sarkisov link and Arabic numbers mean the degree of points which are blown up. As we see from Theorem 2.9 there are no Sarkisov links… view at source ↗
Figure 3
Figure 3. Anticanonical morphism ϕ|−KP1 | Since 5 points ϕ(Q1), ϕ(Q2), P1, P2, P3 are in Sarkisov general position on P 2 K we can blow them up over K and get a del Pezzo surface X4 of degree 4 of Picard rank 3. Let h1 : X4 → P 2 be this blowup. Contracting (h1) −1 ∗ (l) and (h1) −1 ∗ (Q) we get a minimal del Pezzo surface X6 of degree 6. Let h2 : X4 → X6 be this contraction. Let g1 be a blowup of ϕ(Q) and g2 be a contraction… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 4 linked inside Pith

  1. [13]

    E. Kurz, E. Yasinsky,Birational Geometry of sextic del Pezzo surfaces, arXiv: 2507.21737 [math:AG] (2025)

    arXiv 2025

  2. [19]

    C. A. Shramov,Birational automorphisms of Severi–Brauer surfaces, Sb. Math.,211, no.3 (2020), 466–480

    2020

  3. [18]

    Shramov, A

    C. Shramov, A. Trepalin,Birational geometry of del Pezzo surfaces of degree 4, arXiv: 2512.19660 [math:AG] (2025)

    arXiv 2025

  4. [1]

    Bernasconi, H

    F. Bernasconi, H. Tanaka,Geometry and arithmetic of geometrically integral regular del Pezzo surfaces, arXiv: 2408.11378 [math:AG] (2024)

    arXiv 2024

  5. [2]

    Blanc, J

    J. Blanc, J. Schneider, E. Yasinsky,Birational maps of Severi–Brauer surfaces, with appli- cations to Cremona groups of higher rank, arXiv: 2211.17123 [math:AG] (2022)

    arXiv 2022

  6. [3]

    Corti,Factoring birational maps of threefolds after Sarkisov, Journal Alg

    A. Corti,Factoring birational maps of threefolds after Sarkisov, Journal Alg. Geom.4, (1995), 223–254

    1995

  7. [4]

    Corti,Three equivalent conjectures on the birational geometry of Fano 3-folds, Proc

    A. Corti,Three equivalent conjectures on the birational geometry of Fano 3-folds, Proc. Steklov Inst. Math.264(2009), no. 1, 45–47

    2009

  8. [5]

    C. D. Hacon, J. McKernan,The Sarkisov program, J. Algebraic Geom.22(2013), no. 2, 389–405

    2013

Show all 25 references
  1. [6]

    Hassett,Special cubic fourfolds, Compos

    B. Hassett,Special cubic fourfolds, Compos. Math.120(2000), no. 1, 1–23

    2000

  2. [7]

    Hassett,Cubic fourfolds, K3 surfaces, and rationality questions, Rationality problems in algebraic geometry, Lecture Notes in Math.,2172, Fond

    B. Hassett,Cubic fourfolds, K3 surfaces, and rationality questions, Rationality problems in algebraic geometry, Lecture Notes in Math.,2172, Fond. CIME/CIME Found. Subser., Springer, Cham (2016), 29–66

    2016

  3. [8]

    V. A. Iskovskikh,Minimal models of rational surfaces over arbitrary field, Math. USSR Izv.,43(1979), 19–43 (in Russian); translation in Math. USSR Izv.,14, no. 1 (1980), 17–39. COMPOSITION OF SARKISOV LINKS BETWEEN DEL PEZZO SURF ACES 19

    1979

  4. [9]

    V. A. Iskovskikh,Factorization of birational maps of rational surfaces from the viewpoint of Mori theory, Russian Math. Surveys,51, no. 4 (1996), 585–652

    1996

  5. [10]

    Katzarkov, M

    L. Katzarkov, M. Kontsevich, T. Pantev, T. Y. Yue,Birational invariants from Hodge struc- tures and quantum multiplication, arXiv: 2508.05105 [math:AG] (2025)

    2025

  6. [11]

    Koll´ ar,Severi–Brauer varieties; a geometric treatment, arXiv:1606.04368 [math.AG] (2016)

    J. Koll´ ar,Severi–Brauer varieties; a geometric treatment, arXiv:1606.04368 [math.AG] (2016)

    2016 arXiv

  7. [12]

    Kuznetsov,Derived categories of cubic fourfolds, Cohomological and geometric approaches to rationality problems, Progr

    A. Kuznetsov,Derived categories of cubic fourfolds, Cohomological and geometric approaches to rationality problems, Progr. Math.,282, Birkh¨ auser Boston, Boston, MA, (2010), 219–243

    2010

  8. [14]

    S. Lamy, J. Schneider,Generating the plane Cremona groups by involutions, Algebr. Geom. 11(2024), no. 1, 111–162

    2024

  9. [15]

    Russo, G

    F. Russo, G. Staglian` o,Explicit rationality of some special Fano fourfolds, Rationality of varieties, Progr. Math.,342, Birkh¨ auser/Springer, Cham (2021), 323–343

    2021

  10. [16]

    N. I. Shepherd-Barron,The rationality of quintic Del Pezzo surfaces — a short proof, Bull. London Math. Soc.24(1992), no. 3, 249–250

    1992

  11. [17]

    Schneider, S

    J. Schneider, S. Zimmermann,Algebraic subgroups of the plane Cremona group over a perfect field, ´Epijournal G´ eom. Alg´ ebrique5(2021), Art. 14, 48 pp

    2021

  12. [20]

    Shramov, V

    C. Shramov, V. Vologodsky,Automorphisms of pointless surfaces, arXiv: 1807.06477 [math:AG] (2018)

    2018 arXiv

  13. [21]

    Skorobogatov,Torsors and rational points, Cambridge Tracts in Mathematics,144, Cam- bridge University Press, Cambridge (2001)

    A. Skorobogatov,Torsors and rational points, Cambridge Tracts in Mathematics,144, Cam- bridge University Press, Cambridge (2001)

    2001

  14. [22]

    H. P. F. Swinnerton-Dyer,Rational points on del Pezzo surfaces of degree5, Algebraic geome- try, Oslo 1970 (Proc. Fifth Nordic Summer School in Math.), pp. 287–290, Wolters-Noordhoff Publishing, Groningen (1972)

    1970

  15. [23]

    Trepalin,Birational classification of pointless del Pezzo surfaces of degree8, Eur

    A. Trepalin,Birational classification of pointless del Pezzo surfaces of degree8, Eur. J. Math.9(2023), no. 1, 21 pp

    2023

  16. [24]

    V´ arilly-Alvarado,Arithmetic of del Pezzo surfaces, Birational geometry, rational curves, and arithmetic, 293–319, Simons Symp., Springer, Cham (2013)

    A. V´ arilly-Alvarado,Arithmetic of del Pezzo surfaces, Birational geometry, rational curves, and arithmetic, 293–319, Simons Symp., Springer, Cham (2013)

    2013

  17. [25]

    A. V. Zaitsev,Forms of del Pezzo surfaces of degree5and6, Sb. Math.214(2023), no. 6, 816–831. Department of Mathematics and Computer Science, University of Basel, Spiegelgasse 1, 4051 Basel, Switzerland Email address:vikulovaav@gmail.com

    2023

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.