Intersection of curves in ℙ 4 \mathbb{P}^{4}
Luca Chiantini, Łucja Farnik, Giuseppe Favacchio, Brian Harbourne, Juan Migliore, Tomasz Szemberg and Justyna Szpond · 2025
Read Intersection of curves in ℙ 4 \mathbb{P}^{4} by Luca Chiantini, Łucja Farnik, Giuseppe Favacchio, Brian Harbourne, Juan Migliore, Tomasz Szemberg and Justyna Szpond (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-15801.txt.
Introduction
Luca Chiantini, Łucja Farnik, and five co-authors ask how many points two distinct curves in four-dimensional projective space can share. Both curves have given degrees, neither pair lies in a flat hyperplane, and the question asks for the largest possible intersection count. The plane case is trivial and the three-dimensional case was solved in the nineteen eighties, with extremal curves lying on a quadric surface.
The authors define a bound from the two degrees and prove it when both curves lie on a surface of minimal degree, here a cubic surface. They conjecture the bound always holds and prove it in many cases, including when at least one curve is arithmetically Cohen-Macaulay, a strong algebraic regularity condition.
Catalog record
| CREATOR | Luca Chiantini, Łucja Farnik, Giuseppe Favacchio, Brian Harbourne, Juan Migliore, Tomasz Szemberg and Justyna Szpond |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 125 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-15801
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Intersection of curves in ℙ 4 \mathbb{P}^{4} Luca Chiantini, Łucja Farnik, Giuseppe Favacchio, Brian Harbourne, Juan Migliore, Tomasz Szemberg and Justyna Szpond Abstract . Given positive integers n n , d 1 d_{1} and d 2 d_{2} , it is natural to seek a sharp upper bound for the number of intersection points of reduced, irreducible curves C 1 C_{1} and C 2 C_{2} of those degrees in ℙ n \mathbb{P}^{n} . When n = 2 n=2 this is trivial. When n = 3 n=3 this problem was solved independently by Diaz and by Giuffrida. They showed that two curves achieving the maximum number of intersection points have to be rational curves on a smooth surface of minimal degree, i.e., a quadric surface. As such, these curves are far from being arithmetically Cohen-Macaulay (ACM). It is noteworthy that Hartshorne and Miró-Roig also addressed this problem for curves in ℙ 3 \mathbb{P}^{3} under the assumption that the curves are ACM; this latter paper produced very different results from those of Diaz and Giuffrida, and introduced very deep techniques. Diaz and Giuffrida also gave initial results when n > 3 n>3 . In this note we continue this study when n = 4 n=4 . We introduce an invariant, B ( d 1 , d 2 ) B(d_{1},d_{2}) . We prove that when both curves lie on a surface of minimal degree (now a cubic surface) then | C 1 ∩ C 2 | ≤ B ( d 1 , d 2 ) |C_{1}\cap C_{2}|\leq B(d_{1},d_{2}) , and for any d 1 , d 2 d_{1},d_{2} sharp examples exist (although only one of our two curves is rational). We conjecture that B ( d 1 , d 2 ) B(d_{1},d_{2}) is always an upper bound, among arbitrarily chosen irreducible curves of the given degrees. We prove this conjecture in many cases; one of these is when at least one of the two curves is ACM, giving a link to the work of Hartshorne and Miró-Roig. Our approach focuses on the genera of C 1 C_{1} , C 2 C_{2} and especially the genus, g g , of C 1 ∪ C 2 C_{1}\cup C_{2} . We introduce a second invariant, B g ( d 1 , d 2 ) B_{g}(d_{1},d_{2}) , which is always an upper bound; however, it may happen that B g ( d 1 , d 2 ) > B ( d 1 , d 2 ) B_{g}(d_{1},d_{2})>B(d_{1},d_{2}) so in these cases this approach does not lead immediately to a proof of the conjecture for the degrees d 1 , d 2 d_{1},d_{2} in question. We give some situations where we can get around this problem, one being the ACM situation just mentioned. We also show, when d 1 , d 2 d_{1},d_{2} are both odd, that smooth, irreducible, rational curves achieving | C 1 ∩ C 2 | = B ( d 1 , d 2 ) |C_{1}\cap C_{2}|=B(d_{1},d_{2}) exist not lying on a cubic surface, giving a stark contrast to the situation in ℙ 3 \mathbb{P}^{3} . Our main approach is to study the h h -vector and the geometry of the general hyperplane section Γ \Gamma of C 1 ∪ C 2 C_{1}\cup C_{2} . Finally, using a lifting result of Huneke and Ulrich, we make conclusions about what surfaces may contain C 1 ∪ C 2 C_{1}\cup C_{2} . \markleft L. Chiantini, Ł. Farnik, G. Favacchio, B. Harbourne, J. Migliore, T. Szemberg and J. Szpond 1. Introduction In this work, a curve refers to a reduced, equidimensional variety of dimension 1 . Throughout the paper, we assume that C_{1} and C_{2} are reduced, irreducible, and nondegenerate curves of degrees d_{1} and d_{2} , respectively. A curve is arithmetically Cohen-Macaulay (ACM) if its coordinate ring is Cohen-Macaulay, and several cohomological criteria exist to determine this. We are interested in finding an upper bound for the number, |C_{1}\cap C_{2}| (counted without multiplicity), of intersection points of curves C_{1},C_{2} of degrees d_{1},d_{2} in \mathbb{P}^{n} over the complex numbers. The following natural questions have been addressed by various authors (as we will see). Questions 1.1 . 1. What kinds of upper bounds can be given for |C_{1}\cap C_{2}| and how do they depend on the dimension, n , of the ambient projective space? 2. Given n , is there a sharp bound that depends only on d_{1} and d_{2} ? 3. If C_{1} and C_{2} attain the maximal number of intersection points, what additional geometric or algebraic constraints must they satisfy? 4. What roles do the arithmetic genera of the curves play? The following two additional questions were motivated by analogous questions posed by Hartshorne and Miró-Roig [ 16 ] for ACM curves in \mathbb{P}^{3} . 5. If two curves have the maximum number of intersection points, does this force the two curves to lie in a common surface of minimal degree? 6. If the maximum is attained, is the union of the two curves necessarily arithmetically Cohen-Macaulay? Quoting Hartshorne and Miró-Roig, “There seems to be scant attention to these questions in the literature.” Our goal is to address all these questions in the case of curves in \mathbb{P}^{4} . We begin here with a quick review of the known results. If the ambient space is \mathbb{P}^{2} , then of course the set C_{1}\cap C_{2} may contain up to d_{1}d_{2} points, with the upper bound being achieved when the curves intersect transversally. The arithmetic genus of either curve does not play a relevant role here. For curves C_{1},C_{2} in \mathbb{P}^{3} , Diaz [ 6 ] and, independently, Giuffrida [ 9 ] established the upper bound: (1.1) |C_{1}\cap C_{2}|\leq(d_{1}-1)(d_{2}-1)+1. Furthermore, they showed that this bound is achieved if and only if both curves lie on the same smooth quadric, in particular a surface of minimal degree in \mathbb{P}^{3} and have arithmetic genus 0, with divisor types (d_{1}-1,1) and (1,d_{2}-1) . Moving beyond this, Giuffrida [ 8 ] showed that the maximum intersection for curves on a cubic surface in \mathbb{P}^{3} but not on a quadric is on the order of \frac{5}{9}d_{1}d_{2} . The paper [ 16 ] by Hartshorne and Miró-Roig addressed an interesting variant of this problem, looking for the maximum number of intersection points of two reduced, irreducible ACM curves in \mathbb{P}^{3} . The approach of their paper was through a careful study of the h -vectors of C_{1},C_{2} and C_{1}\cup C_{2} . This is a classical approach and is also at the heart of this paper. One of our main results is in the context of ACM curves in \mathbb{P}^{4} (see Theorem D ). The situation for curves in higher-dimensional projective spaces \mathbb{P}^{n} is much less clear. Giuffrida established the following bound in [ 10 , Theorem] : (1.2) |C_{1}\cap C_{2}|\leq(d_{1}-n+2)(d_{2}-n+2)+n-2, which, for n=3 , coincides exactly with ( 1.1 ). Moreover, Giuffrida demonstrated in [ 10 , Example] that the bound in ( 1.2 ) is sharp and is attained when C_{1} and C_{2} are rational normal curves (of degree n ) meeting in n+2 points. Diaz also addressed curves in \mathbb{P}^{n} , giving the bound (d_{1}-n+1)d_{2}+1 [ 6 , (13) Corollary] . In the situation of nondegenerate, irreducible