This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

The Number of Rational Points of Two Parameter Calabi-Yau manifolds as Toric Hypersurfaces

Yuan-Chun Jing Xuan Li and Fu-Zhong Yang corresponding author           E-mail:[email protected]
Abstract

The number of rational points in toric data are given for two-parameter Calabi-Yau nn-folds as toric hypersurfaces over finite fields 𝔽p\mathbb{F}_{p} . We find that the fundamental period is equal to the number of rational points of the Calabi-Yau nn-folds in zeroth order pp-adic expansion. By analyzing the solution set of the GKZ-system given by the enhanced polyhedron, we deduce that under type II/F-theory duality the 3D and 4D Calabi-Yau manifolds have the same number of rational points in zeroth order. Taking the quintic and its duality as an example, the number of rational points in some specific complex moduli are given by numerical calculation to support our results.

1 Introduction

The period integrals of holomorphic 3-form on Calabi-Yau manifolds play an important role in topological string theory. The mirror symmetry between two Calabi-Yau manifolds (M3,W3)(M_{3},W_{3}) makes it possible to use the pure geometric data of W3W_{3} to calculate the Yukawa coupling of the low-energy effective theory after the superstring compactification on M3M_{3} [1]. The Yukawa coupling is of great interests in physics because it contains non-perturbative quantum corrections from holomorphic instantons, mathematically corresponding to Gromov-Witten invariants.

A classic result in arithmetic geometry is that the number of rational points of elliptic curves over finite field 𝔽p\mathbb{F}_{p} can be given by periods as hypergeometric functions, as shown in [2]. Recent work [3] has extended this result to higher genus curves over 𝔽pn\mathbb{F}_{p^{n}}. As a generalization of this result from elliptic curves to Calabi-Yau manifolds, [4] gives the number of rational points for quintic in terms of periods in pp-adic expansion. These results connect complex geometry with arithmetical geometry, and give a number theoretic meaning to the geometric quantities that superstring theory cares about.

Then, ζ\zeta functions, as the generating functions for the number of rational points, of Calabi-Yau manifolds over finite fields are studied [5], [6], [7]. In particular, some propositions and conjecture about arithmetical mirror symmetry are given. Recently, the local ζ\zeta functions are investigated using the Picard-Fuchs equations satisfied for several single-parameter manifolds, and gives rather comprehensive numerical results [8].

In superstring theory , Calabi-Yau manifolds with more than one moduli are interests for physics and mathematics , such as in type II/F-theory duality [9], the dimension of moduli space of the Calabi-Yau 4-fold W4W_{4} is equal to the sum of the dimension of moduli space of the Calabi-Yau 3-fold W3W_{3} and its divisor DD. We wish to generalize the relation between periods and the number of rational points to multi-parameter Calabi-Yau manifolds. In this article, the number of rational points ν0(ϕ,ψ)\nu_{0}(\phi,\psi) in toric data is given for two-parameter Calabi-Yau nn-folds as toric hypersurfaces. We find that ν0(ϕ,ψ)\nu_{0}(\phi,\psi) is equal to the fundamental period ϖ0(ϕ,ψ)\varpi_{0}(\phi,\psi) of Calabi-Yau nn-folds over finite fields in zeroth order pp-adic expansion.

When considering type II/F-theory duality, our results can give some insights on the number of rational points of the dual manifolds W4W_{4} and (W3,D)(W_{3},D) . Under the zeroth order pp-adic expansion, the number of rational points ν0(W4),ν0(W3)\nu_{0}(W_{4}),\nu_{0}(W_{3}) of W4W_{4} and W3W_{3} are equal, and the complex structure of divisor DD has no contribution to ν0(W4)\nu_{0}(W_{4}). This has been checked by numerical method for a D-brane system (W3,D)(W_{3},D) on quntic and its dual W4W_{4} .

This article is organized as follows. In Section 2, we briefly introduce the basic about pp-adic numbers and calculations of the number of rational points over finite fields. In Section 3, we introduce the results in [4] at first. Then by toric geometry, we derive a general expression for the relation between the number of rational points and periods of two-parameters Calabi-Yau nn-folds. In section 4, some corollaries of our results are described by analyzing the structure of the enhanced GKZ-system, and numerical results are given for specific models. In Section 5, we summarize the main results of this article and some possible future studies.

2 Finite fields and pp-adic numbers

Finite fields

In order to study the arithmetic properties of Calabi-Yau manifolds, we need to define them over finite fields rather than complex number fields as usual. Discussion of finite fields can be found in standard textbook, sufficient introductions also are provided in [4], but for completeness, this article contains the most essential knowledge.

The structure of a finite field 𝔽pn\mathbb{F}_{p^{n}} for prime number pp and n+n\in\mathbb{Z}^{+} is completely determined by the number of its elements pnp^{n} . The simplest finite field is the integer modp\mathrm{mod}\ p

𝔽p=modp={0,1,,p1}\mathbb{F}_{p}=\mathbb{Z}\ \mathrm{mod}\ p=\{0,1,...,p-1\}

In this article, we just discuss finite fields containing prime elements.

In elementary number theory, Fermat’s little theorem is well known:

ap=amodpa^{p}=a\ \mathrm{mod}\ p (1)

In 𝔽p\mathbb{F}_{p}, the calculation of modp\mathrm{mod}\ p is automatically included, so this theorem can be expressed as:

ap1={1,ifa00,ifa=0a^{p-1}=\left\{\begin{array}[]{l}1,\ \mathrm{if}\ \ a\neq 0\\ 0,\ \mathrm{if}\ \ a=0\end{array}\right. (2)

This conclusion is important for us to calculate the number of rational points.

pp-adic numbers

In the study of the relation between the number of rational points and periods, one can find that the relation between the complete numbers of rational points and periods is relatively complicated, but if the number of rational points reduced by modpn\mathrm{mod}\ p^{n} is considered:

νn1=νmodpn\nu_{n-1}=\nu\ \mathrm{mod}\ p^{n} (3)

the relation become clear. Obviously, after modp\mathrm{mod}\ p reduction, most of the information about rational points is lost, and ν0\nu_{0} can only be regarded as a “zeroth order approximation” of the number of rational points. In order to recover the complete rational points, we want the following expansion to exist:

ν=ν0+ν1p+ν2p2+\nu=\nu_{0}+\nu_{1}p+\nu_{2}p^{2}+... (4)

Calculating the expansion of order nn is equivalent to calculating rational points under modpn+1\mathrm{mod}\ p^{n+1}. However, prime pp is not a infinitesimal in the general sense. In order to abtain the expansion, we introduce the concept of pp-adic number and redefine the norm. Notice that any rational number rr can always be written as:

r=mn=m0n0pαr=\frac{m}{n}=\frac{m_{0}}{n_{0}}p^{\alpha}

where pp is prime. We define the pp-adic norm of the rational number rr as:

rp=pα,0p=0\|r\|_{p}=p^{-\alpha}\ ,\|0\|_{p}=0

It is easy to verify that it satisfies the definition of norm. The pp-adic norm can be used to complete the rational number field: p\mathbb{Q}\to\mathbb{Q}_{p}. p\mathbb{Q}_{p} is called a pp-adic number field, its elements are called pp-adic number and can be expressed as series:

A=n=n0anpn;anA=\sum_{n=n_{0}}^{\infty}a_{n}p^{n}\ ;\ a_{n}\in\mathbb{Z} (5)

This is exactly the power series we need, the rational points expansioned by pp. After defining the pp-adic norm, the concept of pp-adic analysis naturally follows, but we will not discuss it in detail. Now, the prime pp can be treated as infinitesimal, which allows us to write modp\mathrm{mod}\ p reduced as:

ν0=νmodpν=ν0+o(p)\nu_{0}=\nu\ \mathrm{mod}\ p\Leftrightarrow\nu=\nu_{0}+o(p) (6)

The number of rational points

For algebraic surfaces over 𝔽p\mathbb{F}_{p} defined by zeros of polynomials P(x;ψ)P(x;\psi) with moduli ψ𝔽p\psi\in\mathbb{F}_{p} :

M(ψ)={x𝔽p5|P(x;ψ)=0}M(\psi)=\{x\in\mathbb{F}_{p}^{5}|P(x;\psi)=0\}

the number of rational points are defined as:

ν(ψ)=#{x𝔽p5|P(x;ψ)=0}\nu(\psi)=\#\{x\in\mathbb{F}_{p}^{5}|P(x;\psi)=0\} (7)

Of course, the numerical value of rational points can be easily obtained by computer program, but in order to study the relation between the number of rational points and periods, it is necessary to put the problem on pp-adic number field and calculate the approximation of rational points in each order. The specific method is to use Fermat’s little theorem, x𝔽p5\forall x\in\mathbb{F}_{p}^{5}, we have:

P(x;ψ)p1={1+o(p),ifxXψ0+o(p),ifxXψ\displaystyle P(x;\psi)^{p-1}=\left\{\begin{array}[]{l}1+o(p),\ \mathrm{if}\ x\notin X_{\psi}\\ 0+o(p),\ \mathrm{if}\ x\in X_{\psi}\end{array}\right. (10)

The count of rational points can then be constructed:

ν(ψ)=x𝔽p5(1P(x;ψ)p1)+o(p)\nu(\psi)=\sum_{x\in\mathbb{F}_{p}^{5}}(1-P(x;\psi)^{p-1})+o(p) (11)

This expression is equivalent to the zeroth order approximation in the sense of pp-adic analysis. To obtain an exact solution for the number of rational points, we can consider higher-order corrections:

P(x;ψ)p1\displaystyle P(x;\psi)^{p-1} =\displaystyle= 1+o(p)\displaystyle 1+o(p)
P(x;ψ)p(p1)\displaystyle P(x;\psi)^{p(p-1)} =\displaystyle= 1+o(p2)\displaystyle 1+o(p^{2})
\displaystyle...

This is the main method to be used later in this article.

3 The number of rational points for Calabi-Yau

3.1 The number of rational points for quntic

For the elliptic curve X(τ)X(\tau) defined over finite field 𝔽p\mathbb{F}_{p} with complex moduli τ\tau, the number of rational points of the curve X(τ)X(\tau) can be expressed as [2] :

ν(τ)=(1)(1)(p1)/2n=0(1/2r)2τn\nu(\tau)=(-1)(-1)^{(p-1)/2}\sum_{n=0}^{\infty}\binom{-1/2}{r}^{2}\tau^{n}

This is the same as the expression of the period π(τ)\pi(\tau) of X(τ)X(\tau). For the quintic Calabi-Yau manifold M3(ψ)M_{3}(\psi) in [4], the defining equation is:

P(x;ψ)=i=15xi55ψi=15xiP(x;\psi)=\sum_{i=1}^{5}x_{i}^{5}-5\psi\prod_{i=1}^{5}x_{i}

Where x𝔽p5x\in\mathbb{F}^{5}_{p} , and ψ𝔽p\psi\in\mathbb{F}_{p} is the complex moduli. According to (11) in Section 2, one can write the number of rational points of M3(ψ)M_{3}(\psi) as :

ν(ψ)=p5x𝔽p5(i=15xi55ψi=15xi)p1+o(p)\displaystyle\nu(\psi)=p^{5}-\sum_{x\in\mathbb{F}_{p}^{5}}\left(\sum_{i=1}^{5}x_{i}^{5}-5\psi\prod_{i=1}^{5}x_{i}\right)^{p-1}+o(p)

the number of rational points can be obtained by simplification:

ν(ψ)=n=0[p5](5n)!(n!)5zn+o(p)\displaystyle\nu(\psi)=\sum_{n=0}^{[\frac{p}{5}]}\frac{(5n)!}{(n!)^{5}}z^{n}+o(p) (12)

where z=(5ψ)5z=(5\psi)^{-5}. This is formally corresponds to the fundamental period of quintic:

f0(ψ)=n=0(5n)!(n!)5znf_{0}(\psi)=\sum_{n=0}^{\infty}\frac{(5n)!}{(n!)^{5}}z^{n} (13)

the truncation of the summationthe is the only difference. It can be proved by the formula (ap+b)!=a!b!(p!)a+o(p2)(ap+b)!=a!b!(p!)^{a}+o(p^{2}) in pp-adic analysis that the truncation is generated by transferring the series n=0anzn\sum_{n=0}^{\infty}a_{n}z^{n} to the finite field 𝔽p\mathbb{F}_{p}. Then one can find the equation between the number of rational points and fundamental period of quntic over finite field under zeroth order approximation:

ν(ψ)=[p/5]f0(ψ)+o(p)\nu(\psi)=\ ^{[p/5]}f_{0}(\psi)+o(p) (14)

where [p/5][p/5] indicates the above truncation.

There is a similar discussion about first-order corrections of the number of points, and the exact expression of the number of rational points can be guess as:

ν(ψ)=k=041k!(p1p)kfk(k)(p1)(zp4)modp5\nu(\psi)=\sum_{k=0}^{4}\frac{1}{k!}\left(\frac{p}{1-p}\right)^{k}\ {}^{(p-1)}f^{(k)}_{k}(z^{p^{4}})\ \mathrm{mod}\ p^{5} (15)

Where fk(k)f^{(k)}_{k} represents the kk logarithmic derivative (zddz)kfk(z\frac{d}{dz})^{k}f_{k} of the kk-th period fkf_{k}, and this formula is verified by numerical calculation.

This result relates the important geometric quantity, periods, for the calculation of effective superpotential in physical superstring theory to the number of rational points in arithmetic geometry for the first time, and gives a fairly accurate equation. It is still one of the few studies of Calabi-Yau manifolds, the inner spaces of superstrings, over finite fields.

3.2 The number of rational points for nn dimensional two-parameter hypersurface

In this article, Calabi-Yau manifolds will be described by the language of toric geometry [10]. Compact Calabi-Yau can be defined as hypersurfaces in toric variety [11]. Consider a Laurent polynomials:

f(x)=cmxmf(x)=\sum c_{m}x^{m} (16)

its monomial corresponds to a convex polyhedron Δ\Delta in lattice space MM_{\mathbb{Q}} :

Δ(f)M\Delta(f)\subset M_{\mathbb{Q}}

which is the convex hull of monomial whose coefficients are not equal to zero. In fact, this polyhedron Δ\Delta is the same one that defines the toric variety PΔ=ProjSΔP_{\Delta}=\mathrm{Proj}\ S_{\Delta}. The zeros set of each Laurent polynomial corresponds to an affine hypersurface Zf,ΔZ_{f,\Delta} in PΔP_{\Delta} :

Zf,Δ={xPΔ|f(x)=0}Z_{f,\Delta}=\{x\in P_{\Delta}\ |\ f(x)=0\}

a compact Calabi-Yau manifold can then be defined. If the polyhedron Δ\Delta is a simplex, or if its vertices give only one linear relation, then the corresponding Calabi-Yau manifold has one-parameter. Otherwise, a multi-parameter object can be obtained.

In [6], some examples of the number of rational points of Calabi-Yau manifolds are examined in detail by Gauss sum. For our purposes, the relation between rational points and periods of multi-parameter hypersurfaces is expected to be given. Taking the two-parameter case as an example, for the nn-dimensional toric variety embedded in the N+nN+n-dimensional ambient space, there is a general two-parameter hypersurface equation:

P(x;ϕ,ψ)=𝕞Δα𝕞x𝕞=𝕞Δα𝕞x𝕞ϕQψU\displaystyle P(x;\phi,\psi)=\sum_{\mathbb{m}\in\Delta}\alpha_{\mathbb{m}}x^{\mathbb{m}}=\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}}-\phi Q-\psi U (17)

We use Fermat’s little theorem (10) in Section 2 , instead of the Gauss sum, to compute the number of rational points in zeroth order:

ν0(ϕ,ψ)\displaystyle\nu_{0}(\phi,\psi) =\displaystyle= x𝔽pN+n(1(P(x;ϕ,ψ))p1)\displaystyle\sum_{x\in\mathbb{F}_{p}^{N+n}}(1-(P(x;\phi,\psi))^{p-1})
=\displaystyle= pN+nx𝔽pN+n(𝕞Δα𝕞x𝕞ϕQψU)p1\displaystyle p^{N+n}-\sum_{x\in\mathbb{F}_{p}^{N+n}}(\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}}-\phi Q-\psi U)^{p-1}

bt polynomial theorem , above equation can be expand as:

ν0(ϕ,ψ)\displaystyle\nu_{0}(\phi,\psi) =\displaystyle= x𝔽pN+nn1,n2(p1)!(p1n1n2)!n1!n2!(𝕞Δα𝕞x𝕞)n1+n2(ϕQ)n1(ψU)n2\displaystyle\sum_{x\in\mathbb{F}_{p}^{*N+n}}\sum_{n_{1},n_{2}}\frac{(p-1)!}{(p-1-n_{1}-n_{2})!n_{1}!n_{2}!}(\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{1}+n_{2}}(-\phi Q)^{-n_{1}}(-\psi U)^{-n_{2}}

considering that pp is infinitesimal in pp-adic analysis, we has the following formula:

(p1)!(p1n1n2)!\displaystyle\frac{(p-1)!}{(p-1-n_{1}-n_{2})!} =\displaystyle= (p1)(p2)(p1n1n2)\displaystyle(p-1)(p-2)...(p-1-n_{1}-n_{2}) (18)
=\displaystyle= (1)n1+n1(n1+n2)!+o(p)\displaystyle(-1)^{n_{1}+n_{1}}(n_{1}+n_{2})!+o(p)

in terms of (18), the number of rational points can be reduced to:

ν0(ϕ,ψ)\displaystyle\nu_{0}(\phi,\psi) =\displaystyle= x𝔽pN+nn1,n2(1)n1+n1(n1+n2)!n1!n2!(𝕞Δα𝕞x𝕞)n1+n2(ϕQ)n1(ψU)n2\displaystyle\sum_{x\in\mathbb{F}_{p}^{*N+n}}\sum_{n_{1},n_{2}}\frac{(-1)^{n_{1}+n_{1}}(n_{1}+n_{2})!}{n_{1}!n_{2}!}(\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{1}+n_{2}}(-\phi Q)^{-n_{1}}(-\psi U)^{-n_{2}}
=\displaystyle= x𝔽pN+nn1,n2(n1+n2)!n1!n2!n𝕞=n1+n2(n1+n2)!𝕞n𝕞!𝕞Δ(α𝕞x𝕞)n𝕞1ϕn1ψn2Qn1Un2\displaystyle\sum_{x\in\mathbb{F}_{p}^{*N+n}}\sum_{n_{1},n_{2}}\frac{(n_{1}+n_{2})!}{n_{1}!n_{2}!}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\frac{(n_{1}+n_{2})!}{\prod_{\mathbb{m}}n_{\mathbb{m}}!}\prod_{\mathbb{m}\in\Delta^{\prime}}(\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{\mathbb{m}}}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}Q^{n_{1}}U^{n_{2}}}
=\displaystyle= n1,n2[(n1+n2)!]2n1!n2!1ϕn1ψn2n𝕞=n1+n2𝕞Δα𝕞n𝕞n𝕞!x𝔽pN+nx𝕞Δn𝕞𝕞n1𝕢n2𝕦\displaystyle\sum_{n_{1},n_{2}}\frac{[(n_{1}+n_{2})!]^{2}}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\prod_{\mathbb{m}\in\Delta^{\prime}}\frac{\alpha^{n_{\mathbb{m}}}_{\mathbb{m}}}{n_{\mathbb{m}}!}\sum_{x\in\mathbb{F}_{p}^{*N+n}}x^{\sum_{\mathbb{m}\in\Delta^{\prime}}{n_{\mathbb{m}}}\mathbb{m}-n_{1}\mathbb{q}-n_{2}\mathbb{u}}

further expansion by polynomial theorem, finally we have:

ν0(ϕ,ψ)\displaystyle\nu_{0}(\phi,\psi) =\displaystyle= x𝔽pN+nn1,n2(n1+n2)!n1!n2!n𝕞=n1+n2(n1+n2)!𝕞n𝕞!𝕞Δ(α𝕞x𝕞)n𝕞1ϕn1ψn2Qn1Un2\displaystyle\sum_{x\in\mathbb{F}_{p}^{*N+n}}\sum_{n_{1},n_{2}}\frac{(n_{1}+n_{2})!}{n_{1}!n_{2}!}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\frac{(n_{1}+n_{2})!}{\prod_{\mathbb{m}}n_{\mathbb{m}}!}\prod_{\mathbb{m}\in\Delta^{\prime}}(\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{\mathbb{m}}}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}Q^{n_{1}}U^{n_{2}}}
=\displaystyle= n1,n2[(n1+n2)!]2n1!n2!1ϕn1ψn2n𝕞=n1+n2𝕞Δα𝕞n𝕞n𝕞!x𝔽pN+nx𝕞Δn𝕞𝕞n1𝕢n2𝕦\displaystyle\sum_{n_{1},n_{2}}\frac{[(n_{1}+n_{2})!]^{2}}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\prod_{\mathbb{m}\in\Delta^{\prime}}\frac{\alpha^{n_{\mathbb{m}}}_{\mathbb{m}}}{n_{\mathbb{m}}!}\sum_{x\in\mathbb{F}_{p}^{*N+n}}x^{\sum_{\mathbb{m}\in\Delta^{\prime}}{n_{\mathbb{m}}}\mathbb{m}-n_{1}\mathbb{q}-n_{2}\mathbb{u}}

This is the zeroth order expression of the number of rational points of n1n-1 dimensional two-parameter hypersurfaces.

3.3 The fundamental period for nn dimensional two-parameter hypersurface

For the hypersurface Zf,ΔZ_{f,\Delta} in toric variety, such as the mirror manifold W3W_{3} in usual, which defined by the dual polyhedron Δ\Delta^{*} in mirror symmetry, the period integral is defined as:

Πi=γi1f(x)j=1ndxjxj\Pi_{i}=\int_{\gamma_{i}}\frac{1}{f(x)}\prod_{j=1}^{n}\frac{dx_{j}}{x_{j}} (20)

These periods, as solutions of the Picard-Fuchs equation, can be obtained from the GKZ-generalized hypergeometric system related to Δ\Delta^{*} [12], [13]. But in order to compare the number of rational points in zeroth order ν0(ϕ,ψ)\nu_{0}(\phi,\psi) , we use the definition (20) to calculate the fundamental period ϖ0(ϕ,ψ)\varpi_{0}(\phi,\psi) of two-parameter hypersurface directly :

ϖ0(ϕ,ψ)\displaystyle\varpi_{0}(\phi,\psi) =\displaystyle= C(2πi)N+ndN+nxP(ϕ,ψ)\displaystyle\frac{C}{(2\pi i)^{N+n}}\int\frac{d^{N+n}x}{-P(\phi,\psi)} (21)

Plug the polynomial equation (17) into (21) :

ϖ0(ϕ,ψ)\displaystyle\varpi_{0}(\phi,\psi) =\displaystyle= C(2πi)N+ndN+nxϕQ+ψU𝕞Δα𝕞x𝕞\displaystyle\frac{C}{(2\pi i)^{N+n}}\int\frac{d^{N+n}x}{\phi Q+\psi U-\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}}}
=\displaystyle= C(2πi)N+ndN+nxϕQ+ψU(1𝕞Δα𝕞x𝕞ϕQ+ψU)1\displaystyle\frac{C}{(2\pi i)^{N+n}}\int\frac{d^{N+n}x}{\phi Q+\psi U}\left(1-\frac{\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}}}{\phi Q+\psi U}\right)^{-1}

consider the power series expansion of the integrand:

(1𝕞Δα𝕞x𝕞ϕQ+ψU)1\displaystyle\left(1-\frac{\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}}}{\phi Q+\psi U}\right)^{-1}
=n1+n21(n1+n2)!n1,n2(n1+n2)!n1!n2!1(Qϕ)n1(Uψ)n21/Qϕn11/Uψn2(1𝕞Δα𝕞x𝕞ϕQ+ψU)1\displaystyle=\sum_{n_{1}+n_{2}}\frac{1}{(n_{1}+n_{2})!}\sum_{n_{1},n_{2}}\frac{(n_{1}+n_{2})!}{n_{1}!n_{2}!}\frac{1}{{(Q\phi)}^{n_{1}}{(U\psi)}^{n_{2}}}\partial^{n_{1}}_{1/{Q\phi}}\partial^{n_{2}}_{1/{U\psi}}\left(1-\frac{\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}}}{\phi Q+\psi U}\right)^{-1}
=n1,n21n1!n2!1ϕn1ψn2(n1+n2)!(𝕞Δα𝕞x𝕞)n1+n2Qn1Pn2\displaystyle=\sum_{n_{1},n_{2}}\frac{1}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}(n_{1}+n_{2})!(\sum_{\mathbb{m}\in\Delta^{\prime}}\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{1}+n_{2}}Q^{-n_{1}}P^{-n_{2}}

expand by polynomial theorem:

n1,n21n1!n2!1ϕn1ψn2(n1+n2)!n𝕞=n1+n2(n1+n2)!𝕞n𝕞𝕞Δ(α𝕞x𝕞)n𝕞n𝕞Qn1Pn2\displaystyle\sum_{n_{1},n_{2}}\frac{1}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}(n_{1}+n_{2})!\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\frac{(n_{1}+n_{2})!}{\prod_{\mathbb{m}}n_{\mathbb{m}}}\prod_{\mathbb{m}\in\Delta^{\prime}}\frac{(\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{\mathbb{m}}}}{n_{\mathbb{m}}}Q^{-n_{1}}P^{-n_{2}}
=n1,n2[(n1+n2)!]2n1!n2!1ϕn1ψn2n𝕞=n1+n2𝕞Δ(α𝕞x𝕞)n𝕞n𝕞Qn1Pn2\displaystyle=\sum_{n_{1},n_{2}}\frac{[(n_{1}+n_{2})!]^{2}}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\prod_{\mathbb{m}\in\Delta^{\prime}}\frac{(\alpha_{\mathbb{m}}x^{\mathbb{m}})^{n_{\mathbb{m}}}}{n_{\mathbb{m}}}Q^{-n_{1}}P^{-n_{2}}

Finally, we find that the explicit expression of the fundamental period and the number of rational points in zeroth order approximation, namely:

ν0(ϕ,ψ)\displaystyle\nu_{0}(\phi,\psi) =\displaystyle= n1,n2[(n1+n2)!]2n1!n2!1ϕn1ψn2n𝕞=n1+n2𝕞Δα𝕞n𝕞n𝕞!\displaystyle\sum_{n_{1},n_{2}}\frac{[(n_{1}+n_{2})!]^{2}}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\prod_{\mathbb{m}\in\Delta^{\prime}}\frac{\alpha^{n_{\mathbb{m}}}_{\mathbb{m}}}{n_{\mathbb{m}}!}
×x𝔽pN+nx𝕞Δn𝕞𝕞n1𝕢n2𝕦\displaystyle\times\sum_{x\in\mathbb{F}_{p}^{*N+n}}x^{\sum_{\mathbb{m}\in\Delta^{\prime}}{n_{\mathbb{m}}}\mathbb{m}-n_{1}\mathbb{q}-n_{2}\mathbb{u}}
ϖ0(ϕ,ψ)\displaystyle\varpi_{0}(\phi,\psi) =\displaystyle= n1,n2[(n1+n2)!]2n1!n2!1ϕn1ψn2n𝕞=n1+n2𝕞Δα𝕞n𝕞n𝕞!\displaystyle\sum_{n_{1},n_{2}}\frac{[(n_{1}+n_{2})!]^{2}}{n_{1}!n_{2}!}\frac{1}{\phi^{n_{1}}\psi^{n_{2}}}\sum_{\sum n_{\mathbb{m}}=n_{1}+n_{2}}\prod_{\mathbb{m}\in\Delta^{\prime}}\frac{\alpha_{\mathbb{m}}^{n_{\mathbb{m}}}}{n_{\mathbb{m}}!}
×C(2πi)N+ndN+nxϕQ+ψUx𝕞Δn𝕞𝕞n1𝕢n2𝕦\displaystyle\times\frac{C}{(2\pi i)^{N+n}}\int\frac{d^{N+n}x}{\phi Q+\psi U}x^{\sum_{\mathbb{m}\in\Delta^{\prime}}{n_{\mathbb{m}}}\mathbb{m}-n_{1}\mathbb{q}-n_{2}\mathbb{u}}

Since the same restricts are given by xx-sum and integral in the formula above [4] :

𝕞Δn𝕞𝕞=n1𝕢+n2𝕦\displaystyle\sum_{\mathbb{m}\in\Delta^{\prime}}{n_{\mathbb{m}}}\mathbb{m}=n_{1}\mathbb{q}+n_{2}\mathbb{u} (24)

It means that we prove that the equation of rational points and fundamental period still holds for n1n-1 dimensional two-parameter hypersurface under zeroth order approximation:

ν(ϕ,ψ)=(p1)ϖ0(ϕ,ψ)+o(p)\displaystyle\nu(\phi,\psi)=\ ^{(p-1)}\varpi_{0}(\phi,\psi)+o(p) (25)

4 The number of rational points for 4-fold in F-theory

4.1 The type II/F-theory duality

In toric geometry, one way to construct a multi-parameter Calabi-Yau manifold is to embed the polyhedron Δ\Delta in a 4D lattic space corresponding to the one-parameter Calabi-Yau 3-fold W3W_{3} into the 5D lattic space. Then add extra vertices {ρi}\{\rho_{i}\} to make it an enhanced polyhedron Δ~\tilde{\Delta} [14]. This means that additional linear relations among vertices are introduced and therefore additional complex moduli parameters are provided.

The above configuration is called the type II/F-theory duality [9] , Δ~\tilde{\Delta} relates two completely different geometry. One is the Calabi-Yau 4-fold W4W_{4} which is related to F-theory compactification. W4W_{4} is the multi-parameter manifold obtained by elliptic fibration of W3W_{3} [15]. The other is the manifold pair (W3,E)(W_{3},E), open strings is compact on W3W_{3} , and end on the D-brane wrapping the submanifold EE. The periods satisfies the same GKZ-system given by the same enhanced polyhedron Δ~\tilde{\Delta}, therefore, although the relative periods Π^(W3,D)\hat{\Pi}(W_{3},D) and the absolute periods Π(W4)\Pi(W_{4}) has different geometry meaning, the same low-energy effective theory can be given. Here DD is a divisor on W3W_{3} and parameterizes the deformation of D-brane EE .

For the charge vector li(Δ)l_{i}(\Delta) given by the linear relation Δli(Δ)vi=0\sum_{\Delta}l_{i}(\Delta)v_{i}=0 of the vertex viΔv_{i}\in\Delta, since the linear relation is maintained in the embedding viui=(vi,0)v_{i}\mapsto u_{i}=(v_{i},0) from 4D to 5D , the same charge vector l^i(Δ~)=(li(Δ),0k)\hat{l}_{i}(\tilde{\Delta})=(l_{i}(\Delta),0^{k}) holds for extended polyhedra Δ~\tilde{\Delta} , where k=#{ρi}k=\#\{\rho_{i}\} . In GKZ-system, it means the same Picard-Fuchs operators:

{lja>0(i=0lja1(θji))i=1|l0a|(i|l0a|θ0)lja<0,j0(i=0|lja1(θj+|lja|i))za}Π(z)=0\displaystyle\{\prod_{l_{j}^{a}>0}(\prod_{i=0}^{l_{j}^{a}-1}(\theta_{j}-i))-\prod_{i=1}^{|l_{0}^{a}|}(i-|l_{0}^{a}|-\theta_{0})\prod_{l_{j}^{a}<0,j\neq 0}(\prod_{i=0}^{|l_{j}^{a}\mid-1}(\theta_{j}+|l_{j}^{a}|-i))z_{a}\}\Pi(z)=0 (26)

where θj=zjddzj\theta_{j}=z_{j}\frac{d}{dz_{j}} . Note that all periods of W3W_{3} consists of a subset of the periods of W4W_{4}. In particular, the two manifolds have the same fundamental period, ϖ0(W3)=ϖ0(W4)\varpi_{0}(W_{3})=\varpi_{0}(W_{4}). By the conclusion of Section 3 , the relation between complex structures of two manifolds will be expressed on the number of rational points over finite fields.

4.2 The corollary

By the formula (25) at the end of Section 3, we can immediately obtain some corollaries about the relation between rational points of W3(ψ)W_{3}(\psi) and W4(ϕ,ψ)W_{4}(\phi,\psi) over finite fields 𝔽p\mathbb{F}_{p}. The one-parameter polynomials and two-parameter polynomials

P3(x;ψ)=𝕞Δα𝕞x𝕞P_{3}(x;\psi)=\sum_{\mathbb{m}\in\Delta^{*}}\alpha_{\mathbb{m}}x^{\mathbb{m}}
P4(x;ϕ,ψ)=𝕞Δ~α𝕞x𝕞P_{4}(x;\phi,\psi)=\sum_{\mathbb{m}\in\tilde{\Delta}^{*}}\alpha_{\mathbb{m}}x^{\mathbb{m}}

given by the polyhedron Δ\Delta^{*} and its enhanced Δ~\tilde{\Delta}^{*} define 3D and 4D Calabi-Yau hypersurfaces are dual space in the sense of type II/F-theory duality and have the same fundamental period ϖ0=[p/n]f0\varpi_{0}=\ ^{[p/n]}f_{0} . Therefore, their the number of rational points are:

ν(P3)\displaystyle\nu(P_{3}) =\displaystyle= #{x𝔽pn|P3(x;ψ)=0modp}\displaystyle\#\{x\in\mathbb{F}_{p}^{n}|P_{3}(x;\psi)=0\ \mathrm{mod}\ p\}
ν(P4)\displaystyle\nu(P_{4}) =\displaystyle= #{x𝔽pn~|P4(x;ϕ,ψ)=0modp}\displaystyle\#\{x\in\mathbb{F}_{p}^{\tilde{n}}|P_{4}(x;\phi,\psi)=0\ \mathrm{mod}\ p\}

Where nn is the number of vertices of Δ\Delta , n~\tilde{n} is the number of vertices of Δ~\tilde{\Delta} dual of Δ~\tilde{\Delta}^{*}. Since they are equal to the same fundamental period in the zeroth order approximation, we can write ν0(P3)=ν0(P4)\nu_{0}(P_{3})=\nu_{0}(P_{4}), or equivalent:

ν(P3)=ν(P4)modp\displaystyle\nu(P_{3})=\nu(P_{4})\ \mathrm{mod}\ p (27)

This result relate the arithmetic properties of two toric varieties with different dimensions, and is not obvious in mathematics. But a simple proof is given by type II/F-theory duality in physics. In fact, the whole reasoning does not depends on the dimension of toric hypersurfaces, so (27) can be generalized to nn-dimension.

One can also observe the effect of elliptic fibration process from 3- to 4-dimensional manifold on the number of rational points over finite field. In toric geometry, elliptic fibration is determined by the extra vertices {ρi}\{\rho_{i}\}, and thus is related to the divisor DD on the Calabi-Yau 3-fold W3W_{3}. However, over finite fields, the different geometry of DD , or the different elliptic fibration, has no influence on the zeroth order pp-adic expansion of the number of rational points.

4.3 Example: Quintic and the dual 4-fold

For specific models under the type II/F-theory duality, equation (27) can be tested numerically. For example, one can consider the vertices of the polyhedron Δ\Delta and its dual Δ\Delta^{*} are:

Δ\Delta v1=(1,1,1,1)v_{1}=(-1,-1,-1,-1) Δ\Delta^{*} v1=(1,1,1,1)v_{1}^{*}=(-1,-1,-1,-1)
v2=(1,1,1,4)v_{2}=(-1,-1,-1,4) v2=(0,0,0,1)v_{2}^{*}=(0,0,0,1)
v3=(1,1,4,1)v_{3}=(-1,-1,4,-1) v3=(0,0,1,0)v_{3}^{*}=(0,0,1,0)
v4=(1,4,1,1)v_{4}=(-1,4,-1,-1) v4=(0,1,0,0)v_{4}^{*}=(0,1,0,0)
v5=(4,1,1,1)v_{5}=(4,-1,-1,-1) v5=(1,0,0,0)v_{5}^{*}=(1,0,0,0)
Table 1: the vertices of Δ\Delta and Δ\Delta^{*}

Calabi-Yau hypersurface corresponds to Δ\Delta^{*} is given by the formula:

P(Δ)=vjΔaiviΔxjv,vi+1\displaystyle P(\Delta^{*})=\sum_{v_{j}\in\Delta^{*}}a_{i}\prod_{v_{i}\in\Delta}x_{j}^{\langle v,v_{i}^{*}\rangle+1} (28)

In this case it is known as the quintic :

P3(Δ)=x15+x25+x35+x45+x55+a0x1x2x3x4x5P_{3}(\Delta^{*})=x_{1}^{5}+x_{2}^{5}+x_{3}^{5}+x_{4}^{5}+x_{5}^{5}+a_{0}x_{1}x_{2}x_{3}x_{4}x_{5}

Embedding Δ\Delta^{*} into the 5D lattic space and specify the extra vertices corresponding to the divisor DD as:

ρ1=(0,0,0,0,1),ρ2=(1,1,1,1,1)\rho_{1}=(0,0,0,0,1)\ ,\ \rho_{2}=(-1,-1,-1,-1,1)

Then there are extended polyhedra Δ~\tilde{\Delta}^{*} and their duals Δ~\tilde{\Delta} :

Δ~\tilde{\Delta} v1=(1,1,1,1,1)v_{1}=(-1,-1,-1,-1,-1) Δ~\tilde{\Delta}^{*} v1=(1,1,1,1,0)v_{1}^{*}=(-1,-1,-1,-1,0)
v2=(1,1,1,3,1)v_{2}=(-1,-1,-1,3,-1) v2=(0,0,0,1,0)v_{2}^{*}=(0,0,0,1,0)
v3=(1,1,1,4,0)v_{3}=(-1,-1,-1,4,0) v3=(0,0,1,0,0)v_{3}^{*}=(0,0,1,0,0)
v4=(1,1,3,1,1)v_{4}=(-1,-1,3,-1,-1) v4=(0,1,0,0,0)v_{4}^{*}=(0,1,0,0,0)
v5=(1,1,4,1,0)v_{5}=(-1,-1,4,-1,0) v5=(1,0,0,0,0)v_{5}^{*}=(1,0,0,0,0)
v6=(1,3,1,1,1)v_{6}=(-1,3,-1,-1,-1) ρ1=(0,0,0,0,1)\rho_{1}=(0,0,0,0,1)
v7=(1,4,1,1,0)v_{7}=(-1,4,-1,-1,0) ρ2=(1,1,1,1,1)\rho_{2}=(-1,-1,-1,-1,1)
v8=(3,1,1,1,1)v_{8}=(3,-1,-1,-1,-1)
v9=(4,1,1,1,0)v_{9}=(4,-1,-1,-1,0)
Table 2: Δ~\tilde{\Delta} and Δ~\tilde{\Delta}^{*} ’s vertices

The defining equation of to the Calabi-Yau 4-folds of F-theory are also given by (28):

P4(Δ~)=x14+x24x35+x44x55+x64x75+x84x95+x15x2x4x6x8+b0x3x5x7x9+a0x1x2x3x4x5x6x7x8x9P_{4}(\tilde{\Delta}^{*})=x_{1}^{4}+x_{2}^{4}x_{3}^{5}+x_{4}^{4}x_{5}^{5}+x_{6}^{4}x_{7}^{5}+x_{8}^{4}x_{9}^{5}+x_{1}^{5}x_{2}x_{4}x_{6}x_{8}+b_{0}x_{3}x_{5}x_{7}x_{9}+a_{0}x_{1}x_{2}x_{3}x_{4}x_{5}x_{6}x_{7}x_{8}x_{9}

By selecting different complex moduli parameters a0,b0a_{0},b_{0}, we calculate the number of rational points for the first five prime numbers, see Appendix A. For all a0,b0a_{0},b_{0} and prime pp, the calculation results show that equation (27) is valid.

5 Summary and outlook

The main motivation of this article is to generalize the results of [4] to more general cases. In Section 3, we use toric geometry to give a relatively general proof for the proposition that over pp-adic field, the fundamental period of two-parameter Calabi-Yau nn-folds is equal to the number of rational points in the zeroth order approximation.

The 4D Calabi-Yau manifolds corresponding to F-theory is an example of this case, so our result can be used to study open string and type II/F-theory duality over finite fields. In particular, by the formula (25) , we deduce that the 3D and 4D Calabi-Yau manifolds of type II/F-theory duality have the same number of rational points in zeroth order over a finite field. Taking quintic hypersurface as an example, we verify this conclusion under different complex moduli parameters.

For manifold pair (W3,E)(W_{3},E) or (W3,D)(W_{3},D) that is interesting for open string compactification, we expect that similar conclusion holds for Calabi-Yau manifold MM and divisor DD respectively. To verify this, a method based on type II/F-theory duality will be presented in the follow-up work. This is not only the application of results in this article, but also the evidence of type II/F-theory duality over finite fields.

As the formula (25) in Section 3 is only accurate to zeroth order, its improvement is also what we need to do in the next step. In the case of Calabi-Yau manifolds with more than one moduli, the periods given by different charge vectors form several subsets. For the higher-order correction of rational points, how to combine the periods belonging to different subsets is the primary problem we need to solve. Processing of higher-order corrections will help us to see the effect of extra periods, or ”opening periods” in open string case, on the number of rational points. Clarification of this structure is necessary to analyze the properties of rational points in elliptic fibration and type II/F-theory duality.

In addition, it might make sense to consider the mirror symmetry of ζ\zeta function in the open string case.

Appendix A numerical results

a0=0a_{0}=0 b0=0b_{0}=0 b0=1b_{0}=1 b0=2b_{0}=2 b0=3b_{0}=3 b0=4b_{0}=4
p=2p=2 ν(P3)\nu(P_{3}) 16 16
ν(P4)\nu(P_{4}) 256 254
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (0,0) (0,0)
p=3p=3 ν(P3)\nu(P_{3}) 81 81 81
ν(P4)\nu(P_{4}) 6561 6609 6465
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (0,0) (0,0) (0,0)
p=5p=5 ν(P3)\nu(P_{3}) 625 625 625 625 625
ν(P4)\nu(P_{4}) 390625 386785 389345 394465 390625
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (0,0)(0,0) (0,0) (0,0) (0,0) (0,0)
p=7p=7 ν(P3)\nu(P_{3}) 2401 2401 2401 2401 2401
ν(P4)\nu(P_{4}) 5764801 5810161 5846449 5755729 5665009
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (0,0)(0,0) (0,0) (0,0) (0,0) (0,0)
p=11p=11 ν(P3)\nu(P_{3}) 19251 19251 19251 19251 19251
ν(P4)\nu(P_{4}) 215248881 214038881 214478881 214038881 214038881
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1)(1,1) (1,1) (1,1) (1,1) (1,1)
Table 3: The number of rational points of P3P_{3} and P4P_{4} when a0=0a_{0}=0
a0=1a_{0}=1 b0=0b_{0}=0 b0=1b_{0}=1 b0=2b_{0}=2 b0=3b_{0}=3 b0=4b_{0}=4
p=2p=2 ν(P3)\nu(P_{3}) 17 17
ν(P4)\nu(P_{4}) 255 255
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1)(1,1)
p=3p=3 ν(P3)\nu(P_{3}) 73 73 73
ν(P4)\nu(P_{4}) 6625 6481 6529
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1) (1,1)
p=5p=5 ν(P3)\nu(P_{3}) 681 681 681 681 681
ν(P4)\nu(P_{4}) 387041 401121 385761 390881 387041
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1)(1,1) (1,1) (1,1) (1,1)
p=7p=7 ν(P3)\nu(P_{3}) 2191 2191 2191 2191 2191
ν(P4)\nu(P_{4}) 5810161 5746657 5710369 5782945 5710369
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (0,0) (0,0) (0,0) (0,0) (0,0)
p=11p=11 ν(P3)\nu(P_{3}) 25501 25501 25501 25501 25501
ν(P4)\nu(P_{4}) 214498881 213288881 214828881 213728881 215488881
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (3,3) (3,3)(3,3) (3,3) (3,3) (3,3)
Table 4: The number of rational points of P3P_{3} and P4P_{4} when a0=1a_{0}=1
a0=2a_{0}=2 b0=0b_{0}=0 b0=1b_{0}=1 b0=2b_{0}=2 b0=3b_{0}=3 b0=4b_{0}=4
p=3p=3 ν(P3)\nu(P_{3}) 91 91 91
ν(P4)\nu(P_{4}) 6481 6529 6625
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1) (1,1)
p=5p=5 ν(P3)\nu(P_{3}) 641 641 641 641 641
ν(P4)\nu(P_{4}) 389601 385761 393441 393441 389601
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1) (1,1) (1,1) (1,1)
p=7p=7 ν(P3)\nu(P_{3}) 2407 2407 2407 2407 2407
ν(P4)\nu(P_{4}) 5763505 5808865 5700001 5763505 5808865
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (6,6) (6,6) (6,6) (6,6) (6,6)
p=11p=11 ν(P3)\nu(P_{3}) 33001 33001 33001 33001 33001
ν(P4)\nu(P_{4}) 214038881 212828881 215248881 215028881 214588881
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1) (1,1) (1,1) (1,1)
Table 5: The number of rational points of P3P_{3} and P4P_{4} when a0=2a_{0}=2
a0=3a_{0}=3 b0=0b_{0}=0 b0=1b_{0}=1 b0=2b_{0}=2 b0=3b_{0}=3 b0=4b_{0}=4
p=5p=5 ν(P3)\nu(P_{3}) 561 561 561 561 561
ν(P4)\nu(P_{4}) 394721 390881 393441 378081 394721
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1) (1,1) (1,1) (1,1)
p=7p=7 ν(P3)\nu(P_{3}) 2371 2371 2371 2371 2371
ν(P4)\nu(P_{4}) 5771281 5671489 5671489 5834785 5852929
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (5,5) (5,5) (5,5) (5,5) (5,5)
p=11p=11 ν(P3)\nu(P_{3}) 25501 25501 25501 25501 25501
ν(P4)\nu(P_{4}) 214498881 215488881 214828881 213288881 213728881
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (3,3) (3,3) (3,3) (3,3) (3,3)
Table 6: The number of rational points of P3P_{3} and P4P_{4} when a0=3a_{0}=3
a0=4a_{0}=4 b0=0b_{0}=0 b0=1b_{0}=1 b0=2b_{0}=2 b0=3b_{0}=3 b0=4b_{0}=4
p=5p=5 ν(P3)\nu(P_{3}) 621 621 621 621 621
ν(P4)\nu(P_{4}) 390881 387041 389601 394721 389601
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (1,1) (1,1) (1,1) (1,1) (1,1)
p=7p=7 ν(P3)\nu(P_{3}) 2461 2461 2461 2461 2461
ν(P4)\nu(P_{4}) 5751841 5833489 5833489 5751841 5688337
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (4,4) (4,4) (4,4) (4,4) (4,4)
p=11p=11 ν(P3)\nu(P_{3}) 25501 25501 25501 25501 25501
ν(P4)\nu(P_{4}) 214498881 213728881 213728881 215488881 213288881
ν0(P3),ν0(P4)\nu_{0}(P_{3}),\nu_{0}(P_{4}) (3,3) (3,3) (3,3) (3,3) (3,3)
Table 7: The number of rational points of P3P_{3} and P4P_{4} when a0=4a_{0}=4

References

  • [1] Candelas, P., Xenia, C., Green, P. S., & Parkes, L. (1991). A pair of Calabi-Yau manifolds as an exactly soluble superconformal theory. Nuclear Physics B, 359(1), 21-74.
  • [2] Clemens, C. H. . (1981). A scrapbook of complex curve theory. American Mathematical Society.
  • [3] Sung, Y. . (2017). Rational points over finite fields on a family of higher genus curves and hypergeometric functions. Taiwanese Journal of Mathematics, 21(1).
  • [4] Candelas, P., De La Ossa, X., & Rodriguez-Villegas, F. (2000). Calabi-Yau manifolds over finite fields, I. arXiv preprint hep-th/0012233.
  • [5] Candelas, P., de la Ossa, X., & Rodriguez-Villegas, F. (2003). Calabi-Yau manifolds over finite fields, II. Calabi-Yau varieties and mirror symmetry (Toronto, ON, 2001), 38, 121-157.
  • [6] Kadir, S. N. (2004). The Arithmetic of Calabi–Yau Manifolds and Mirror Symmetry. arXiv preprint hep-th/0409202.
  • [7] Fu, L., & Wan, D. (2006). Mirror congruence for rational points on Calabi-Yau varieties. Asian Journal of Mathematics, 10(1), 1-10.
  • [8] Candelas, P., de la Ossa, X., & van Straten, D. (2021). Local Zeta Functions From Calabi-Yau Differential Equations. arXiv preprint arXiv:2104.07816.
  • [9] Alim, M., Hecht, M., Jockers, H., Mayr, P., Mertens, A., & Soroush, M. (2010). Hints for off-shell mirror symmetry in type II/F-theory compactifications. Nuclear Physics B, 841(3), 303-338.
  • [10] Cox, D. A., Little, J. B., & Schenck, H. K. (2011). Toric varieties (Vol. 124). American Mathematical Soc.
  • [11] Batyrev, V. V. (1993). Dual polyhedra and mirror symmetry for Calabi-Yau hypersurfaces in toric varieties. arXiv preprint alg-geom/9310003.
  • [12] Hosono, S., Klemm, A., Thiesen, S., & Yau, S. T. (1995). Mirror symmetry, mirror map and applications to Calabi-Yau hypersurfaces. Communications in Mathematical Physics, 167(2), 301-350.
  • [13] Hosono, S., Lian, B. H., & Yau, S. T. (1996). GKZ-generalized hypergeometric systems in mirror symmetry of Calabi-Yau hypersurfaces. Communications in Mathematical Physics, 182(3), 535-577.
  • [14] Alim, M. , Hecht, M. , Mayr, P. , & Mertens, A. . (2009). Mirror symmetry for toric branes on compact hypersurfaces. Journal of High Energy Physics, 909(9), 1295-1301.
  • [15] Vafa, C. (1996). Evidence for F-theory. Nuclear Physics B, 469(3), 403-415.
  • [16] Morrison, D. R. (1991). Picard-Fuchs equations and mirror maps for hypersurfaces. arXiv preprint hep-th/9111025.