• No results found

Manin conjectures for K3 surfaces

N/A
N/A
Protected

Academic year: 2021

Share "Manin conjectures for K3 surfaces"

Copied!
27
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Manin Conjectures for K3 surfaces

Ronald van Luijk CRM, Montreal MSRI, Berkeley PIMS, UBC/SFU

May 17, 2006 Banff

(2)

A K3 surface

with 2 singular curves of genus 0 and

(3)

Question 1. Is there a K3 surface, defined over a number field, such that its set of rational points is neither empty, nor dense?

(4)

Question 1. Is there a K3 surface, defined over a number field, such that its set of rational points is neither empty, nor dense?

Examples of K3 surfaces are smooth quartic surfaces in P3. Question 2 (Swinnerton-Dyer).

Does X ⊂ P3 given by

x4 + 2y4 = z4 + 4w4 have more than 2 rational points?

(5)

Question 1. Is there a K3 surface, defined over a number field, such that its set of rational points is neither empty, nor dense?

Examples of K3 surfaces are smooth quartic surfaces in P3. Question 2 (Swinnerton-Dyer).

Does X ⊂ P3 given by

x4 + 2y4 = z4 + 4w4 have more than 2 rational points?

Answer (Elsenhans, Jahnel, 2004):

(6)

We will look at the growth of the number of rational points of bounded height. Consider a surface X/K, choose a height H, and set

NU(B) = #{x ∈ U(K) : H(x) ≤ B}.

Conjecture 1 (Batyrev, Manin). Let X be a smooth, geometrically integral, projective variety over a number field K, and let D be a hyper-plane section. Assume that the canonical sheaf KX satisfies −KX = aD for some a > 0. Then there exists a finite field extension L, a constant C, and an open subset U ⊂ X, such that with b = rk Pic(XL) we have

(7)

We will look at the growth of the number of rational points of bounded height. Consider a surface X/K, choose a height H, and set

NU(B) = #{x ∈ U(K) : H(x) ≤ B}.

Conjecture 1 (Batyrev, Manin). Let X be a smooth, geometrically integral, projective variety over a number field K, and let D be a hyper-plane section. Assume that the canonical sheaf KX satisfies −KX = aD for some a > 0. Then there exists a finite field extension L, a constant C, and an open subset U ⊂ X, such that with b = rk Pic(XL) we have

NUL(B) ≈ CBa(log B)b−1.

(8)

Define the height-zeta function Z(U, s) = X x∈U(K) H(x)−s. From 1 2πi Z c+i∞ c−i∞ x s ds s =      1 if x > 1 1 2 if x = 1 0 if x < 1 (c > 0) we get N (U, x) = 1 2πi Z c+i∞ c−i∞ Z(U, s) x s ds s (c >> 0)

(9)

Assuming Z(U, s) is analytic on <(s) > a − ², except for a pole of order b at a, we can write

N (U, x) = 1 2πi Z c+i∞ c−i∞ Z(U, s) x s ds s (c >> 0) = ress=a hZ(U, s)s−1 exp(s log x)i + 1

2πi

Z a−²+i∞

a−²−i∞ Z(U, s) x

s ds

(10)

Assuming Z(U, s) is analytic on <(s) > a − ², except for a pole of order b at a, we can write

N (U, x) = 1 2πi Z c+i∞ c−i∞ Z(U, s) x s ds s (c >> 0) = ress=a hZ(U, s)s−1 exp(s log x)i + 1

2πi

Z a−²+i∞

a−²−i∞ Z(U, s) x

s ds

s .

The main term is

xap(log x)

for some polynomial p of degree (

b − 1 if a 6= 0, b if a = 0.

(11)

Assuming Z(U, s) is analytic on <(s) > a − ², except for a pole of order b at a, we can write

N (U, x) = 1 2πi Z c+i∞ c−i∞ Z(U, s) x s ds s (c >> 0) = ress=a hZ(U, s)s−1 exp(s log x)i + 1

2πi

Z a−²+i∞

a−²−i∞ Z(U, s) x

s ds

s .

The main term is

xap(log x)

for some polynomial p of degree (

b − 1 if a 6= 0, b if a = 0.

(12)

For X a K3 surface: N (U, B) ∼ C(log B)rk Pic X ?

(13)

For X a K3 surface: N (U, B) ∼ C(log B)rk Pic X ?

Not if X admits an elliptic fibration (in particular, if rk Pic X ≥ 5). More problems:

• infinitely many curves of genus 0 or 1,

• infinitely many automorphisms (cf. A. Baragar), • Swinnerton-Dyer’s surface has very slow growth, • which height to use?

(14)

For X a K3 surface: N (U, B) ∼ C(log B)rk Pic X ?

Not if X admits an elliptic fibration (in particular, if rk Pic X ≥ 5). More problems:

• infinitely many curves of genus 0 or 1,

• infinitely many automorphisms (cf. A. Baragar), • Swinnerton-Dyer’s surface has very slow growth, • which height to use?

We will run experiments on quartic surfaces, comparing exponents, and the main coefficient C to a relatively naive heuristic coefficient ˜C (part of Peyre’s constant for del Pezzo).

(15)

10 20 30 40 # pt’s ˜ C ∼ 6.24 line : C(log B) − 9˜ ρ(XQ) = 1 # Aut XQ < ∞ no elliptic fibration f = x3w − 3x2y2 − x2yw − x2zw + x2w2

−xy2z − xy2w − 4xyz2 − xyzw + 2xyw2 − 2xz3 +xz2w + 2xzw2 + y3w + y2zw − y2w2 − 5yz3 +yz2w + yzw2 − yw3 − z4 + z2w2 + zw3 + 2w4

(16)

2 4 6 8 10 # pt’s ˜ C ∼ 0.57 line : C(log B)˜ ρ(X) = 1 ρ(XQ) = 20 # Aut XQ = ∞

no elliptic fibration over Q

(17)

10 20 30 40 # pt’s ˜ C ∼ 7.8 line : 34C(log B) − 10˜ ρ(XQ) = 1 # Aut XQ < ∞ no elliptic fibration f = x3w − 3x2y2 + 4x2yz − x2z2 −x2z2 + x2zw − xy2z − xyz2 +xw3 + y3w + y2z2 + z3w 17

(18)

Consider X given by

f + 6m(w4 + 2x4 − (y + w)4 − 4z4) for m ∈ {±1, ±2, ±3} with f as on the previous page.

Then X has geometric Picard number 1.

m C˜ C · log(12000) #{x : H(x) ≤ 12000}˜ 0 7.8 73 46 1 0.30 2.8 5 −1 0.58 5.4 9 2 0.39 3.7 2 −2 0.35 3.3 6 3 0.18 1.7 4

(19)

5 10 15 20 25 30 # pt’s ˜ C ∼ 6.66 line : 12C(log B)˜ ρ(X) = 1 ρ(XQ) = 2 (over Q(√3)) Aut XQ = {1} no elliptic fibration f = x3w − x2y2 − 2x2yz −x2z2 + x2zw + xy2z + xyz2 xw3 + y3w − y2z2 + z3w 19

(20)

20 40 60 80 100 # pt’s ˜ C ∼ 0.96 parabola : 1.1(log B)2 ρ(X) = 2 ρ(XQ) = 2 Aut XQ = {1} no elliptic fibration f = 3x3w − x2y2 − 2x2yz

−x2z2 − 4xy2z + 3xy2w − 4xyz2 +4xyw2 + 3xzw2 + 4xw3 + y3w −3y2z2 + 4y2w2 + 2z3w + w4 20

(21)

1 2 3 4 5 log(# pt’s) ˜ C ∼ 0.96

line : 2(log log B) + log 1.1 ρ(X) = 2

ρ(XQ) = 2 Aut XQ = {1}

no elliptic fibration

f = 3x3w − x2y2 − 2x2yz

−x2z2 − 4xy2z + 3xy2w − 4xyz2 +4xyw2 + 3xzw2 + 4xw3 + y3w −3y2z2 + 4y2w2 + 2z3w + w4 21

(22)

10 20 30 40 # pt’s ˜ C ∼ 0.53 parabola : 32C((log B) − 2)˜ 2 + 2 ρ(X) = 2 ρ(XQ) = 2 Aut XQ = {1} no elliptic fibration f = −6x3y − 6x3z − 3x3w − x2y2 + 16x2yz +35x2z2 + 2xy2z + 3xy2w + 8xyz2 − 2xyw2 +6xz3 + 3xzw2 − 2xw3 + y3w + 3y2z2

(23)

1 2 3 4 y log(# pt’s) ˜ C ∼ 0.53

line : 2(log log B) − 0.75 ρ(X) = 2

ρ(XQ) = 2 Aut XQ = {1}

no elliptic fibration

(24)

50 100 150 200 250 # pt’s ˜ C ∼ 0.012 cubic : 0.9((log B) − 3)3 + 10 ρ(X) = 3 ρ(XQ) = 3 # Aut XQ = ∞ (A.Baragar) no elliptic fibration f = −7x3z + 11x3w + 7x2y2 − 11x2yz − 11x2yw − x2z2 +16x2zw − 9x2w2 + 10xy2z + 5xy2w − 44xyz2 + 10xyzw

−14xyw2 − 10xz3 + 18xz2w − 15xzw2 + 5xw3 + 2y4 + 23y3z − 7y3w + 16y2z2 −3y2zw + 12y2w2 − yz3 − 15yz2w − 21yzw2 − 3yw3 + 18z4 + 16z3w + 3z2w2

(25)

1 2 3 4 5 y log(# pt’s) ˜ C ∼ 0.012

line : 3(log log B) − 1.2 ρ(X) = 3

ρ(XQ) = 3

# Aut XQ = ∞

no elliptic fibration

(26)

10 20 30 40 50 60 # pt’s ˜ C ∼ 0.00061 parabola : 0.074(log B)3 ρ(X) = 3 ρ(XQ) = 3 # Aut XQ = ∞ (A.Baragar) no elliptic fibration f = · · · 26

(27)

1 2 3 4 log(# pt’s) ˜ C ∼ 0.00061

line : 3(log log B) − 2.6 ρ(X) = 3

ρ(XQ) = 3

# Aut XQ = ∞

no elliptic fibration

Referenties

GERELATEERDE DOCUMENTEN

Shift in hub nodes from posterior to more anterior brain regions in the alpha but mainly in the beta band, has also been described in AD, and related to disease severity (Engels

EDF-FR-i- and M T D R (Multitraffic Dynamic Reservation) have been investigated for the transmission of multiserv ice traffic over TDM.A./TDD channels in broadband

When the analysis was conducted for each month (results not shown), obvious dense region of positive tests were observed only in the months of June and July of 2005, thus

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly

06/2017 Pharmacy Day 2017, Groningen, The Netherlands ( oral presentation ) 11/2013 4th National Conference on Medicinal Chemistry in China, Jinan, China.. A bout

In conclusion, we showed that cardiovascular risk factors including total cholesterol, triglycerides and HbA1c are associated with 18 F-NaF uptake and calci- fication in the

The main aim of the study is to compare the extent to which the need for autonomy, relatedness and competence of baby boomers, generation X, Y and Z

The oocyte pool hypothesis, which states that a lower number of available oocytes leads to an increased risk of aneuploidy, does not apply for preg- nancies following ovarian