Distribution of integer points on determinant surfaces and a $\text{mod-}p$ analogue

Basic Information

• Paper ID: 2508.14793
• Title: Distribution of integer points on determinant surfaces and a $\text{mod-}p$ analogue
• Authors: Satadal Ganguly, Rachita Guria
• Classification: math.NT (Number Theory)
• Publication Date: August 2025 (arXiv v2: August 24, 2025)
• Paper Link: https://arxiv.org/abs/2508.14793

Abstract

This paper establishes asymptotic formulas with smooth weights for counting integer solutions to the determinant equation $xy-zw=r$ (where $r$ is a nonzero integer), providing explicit main terms and strong error bounds in terms of the variables $x, y, z, w$ and the size of $r$. Similarly, for the congruence $xy-zw \equiv 1 \pmod{p}$ (where $p$ is a large prime), analogous asymptotic formulas and strong error bounds are established.

Research Background and Motivation

Core Problem

This paper investigates the classical problem of representing integers by quadratic forms, specifically focusing on counting integer points on the determinant surface $Q(x_1,x_2,x_3,x_4) = x_1x_2 - x_3x_4 = r$.

Significance

1. Theoretical Value: The problem of representing integers by quadratic forms has been central to number theory since Fermat's time, intimately connected with deep theoretical tools such as modular form theory and the circle method.
2. Practical Applications: Counting solutions to determinant equations naturally arises in multiple areas of analytic number theory:
   • Moment estimates for L-functions (BC18, DFI97)
   • Shifted convolution sum estimates for divisor functions
   • Matrix representation theory problems
3. Technical Challenges: For indefinite quadratic forms with $n \geq 4$, the Hasse principle holds (Siegel's theorem). However, obtaining explicit strong error bounds depending on $r$ is a technical difficulty.

Limitations of Existing Methods

1. Heath-Brown HB96's Circle Method:
   • Only handled special cases $r=0$ and $r=X^2$
   • For general quaternary quadratic forms, error term is $O(X^{3/2+\varepsilon})$
   • Cannot effectively handle $r$ as an independent variable
2. Getz Get18's Improvement:
   • Improved error term via "double Kloosterman refinement"
   • But only applicable to $r=0$ (since $S_q(c)$ is multiplicative in $q$ only when $c=0$)
3. Authors' Previous Work GG23:
   • Handled sharp-cut case but required $|r| \leq X^{1/3}$
   • Needed more refined spectral-theoretic tools

Innovations in This Paper

This paper transforms the problem into Kloosterman sum estimates by directly applying Poisson summation rather than the circle method, then utilizes the Kuznetsov trace formula to obtain:

1. For any $r = o(X^2)$, error term $O(r^\theta X^{1+\varepsilon})$ where $\theta \leq 7/64$ is the Ramanujan-Petersson exponent
2. Handles the general case with $r$ as an independent variable
3. For mod-$p$ congruences, proves that only $X \gg p^{1/2+\varepsilon}$ is needed for asymptotic formulas

Core Contributions

1. Theorem 1.1 (Main Theorem): For the equation $ad-bc=r$, establishes the asymptotic formula $S_V(X,r) = M_V(X,r) + O_\varepsilon(r^\theta X^{1+\varepsilon})$ where the main term is $M_V(X,r) = \sum_{l|r}\sum_{k>0}\frac{\mu(k)}{k}\int\!\!\!\int\!\!\!\int \frac{1}{z}V\left(\frac{x}{X}\right)V\left(\frac{lky}{X}\right)V\left(\frac{lkz}{X}\right)V\left(\frac{r+lkxy}{zlkX}\right)dxdydz$
2. Corollary 1.2: When $r = o(X^2)$, $S_V(X,r) = K(V,r)X^2 + O(\sigma(|r|)) + O(r^\theta X^{1+\varepsilon})$ where $K(V,r) = \frac{1}{\zeta(2)}\frac{\sigma(|r|)}{|r|}\int\!\!\!\int\!\!\!\int \frac{1}{t}V(u)V(v)V(t)V\left(\frac{uv}{t}\right)dudvdt$
3. Theorem 1.3 (mod-$p$ Analogue): For the congruence $ad-bc \equiv 1 \pmod{p}$, when $p^{1/100} < X < p/2$, $\sum_{ad-bc\equiv 1\pmod{p}}V\left(\frac{a}{X}\right)V\left(\frac{b}{X}\right)V\left(\frac{c}{X}\right)V\left(\frac{d}{X}\right) = \frac{X^4}{p}\left(\int V(t)dt\right)^4 + O(X^2g(p))$ where $g(p) \to \infty$ arbitrarily slowly.
4. Methodological Contributions:
   • Avoids the circle method, directly applying Fourier analysis
   • Refined analysis of Bessel transforms in the Kuznetsov formula
   • Methods generalize to $\alpha x_1x_2 - \beta x_3x_4 = r$ and ternary forms $\alpha x_1x_2 - \beta x_3^2$

Detailed Methodology

Problem Formulation

Problem 1 (Determinant Equation): Given a nonzero integer $r$ and growth parameter $X \to \infty$, count integer solutions to $ad - bc = r, \quad a,b,c,d \in \mathbb{Z}$ with $a,b,c,d \sim X$ (with smooth weights $V$).

Problem 2 (mod-$p$ Congruence): Given a large prime $p$ and parameter $X$, count integer solutions to $ad - bc \equiv 1 \pmod{p}$ with $|a|,|b|,|c|,|d| \leq X$ (with smooth weights).

Core Methodological Architecture

Step One: Poisson Summation and Dimension Reduction

Substituting $d = (r+bc)/a$, the condition becomes $bc \equiv -r \pmod{a}$: $S_V(X,r) = \sum_a \sum_c V\left(\frac{a}{X}\right)V\left(\frac{c}{X}\right)\sum_{\substack{b \in \mathbb{Z}\\b \equiv -r\bar{c} \pmod{a}}}V\left(\frac{b}{X}\right)V\left(\frac{r+bc}{aX}\right)$

Applying the Poisson summation formula (Lemma 2.1) to the $b$-sum: $\sum_{b \equiv -r\bar{c} \pmod{a}}V_{a,c}(b) = \frac{1}{a}\sum_{n \in \mathbb{Z}}e\left(\frac{-nr\bar{c}}{a}\right)\hat{V}_{a,c}\left(\frac{n}{a}\right)$

Separating the $n=0$ term (main term) from $n \neq 0$ terms (error term): $S_V(X,r) = M_V(X,r) + E_V(X,r)$

Step Two: Main Term Analysis (Section 4)

For the main term, after removing coprimality conditions and applying Möbius inversion: $M_V(X,r) = \sum_{l|r}\sum_{k>0}\frac{\mu(k)}{k}\int V\left(\frac{x}{X}\right)\sum_{a_1}\frac{1}{a_1}V\left(\frac{lka_1}{X}\right)\sum_{c_2}W(c_2,a_1)dx$

Applying Poisson summation again to the $c_2$-sum and $a_1$-sum, separating zero-frequency terms yields the explicit main term in Proposition 4.1.

Step Three: Error Term Analysis (Sections 5-7)

Key Transformation: Applying Poisson summation to the $c_1$-sum (Lemma 2.2) yields Kloosterman sums: $E_V(X,r) = \sum_{l|r}\sum_{n \neq 0}\int\!\!\int \sum_{a_1}\frac{1}{a_1^2}V_{a_1}(x,y)e\left(\frac{-nx}{a_1}\right)S(nr_1,0,a_1)dxdy$$\quad\quad\quad\quad + \sum_{l|r}\sum_{\substack{m,n \neq 0}}\int\!\!\int \sum_{a_1}\frac{1}{a_1^2}V_{a_1}(x,y)e\left(\frac{-nx-my}{a_1}\right)S(nr_1,-m,a_1)dxdy$

The first term is estimated using standard bounds on Ramanujan sums as $O(\tau(r)X^{1+\varepsilon})$.

Core Estimate (Proposition 6.1): Need to prove $\sum_{|n| \ll X^\varepsilon/l}\sum_{|m| \ll X^\varepsilon}\sum_{c>0}\frac{1}{c}S(nr_1,-m,c)f\left(\frac{4\pi\sqrt{|mnr_1|}}{c}\right) \ll r^\theta X^\varepsilon$

Technical Innovations

1. Application of Kuznetsov Formula (Section 6)

Applying the Kuznetsov trace formula (Lemma 2.6) decomposes Kloosterman sums into three parts: $\sum_c \frac{1}{c}S(nr_1,-m,c)f\left(\frac{4\pi\sqrt{|mnr_1|}}{c}\right) = \Sigma_{\text{Maass}} + \Sigma_{\text{cont.}} + \Sigma_{\text{hol.}}$

• Maass Form Contribution: $\Sigma_{\text{Maass}} = \sum_{j=1}^\infty \rho_j(nr_1)\rho_j(m)\check{f}(\kappa_j)$
• Continuous Spectrum Contribution: $\Sigma_{\text{cont.}} = \frac{1}{\pi}\int_{-\infty}^\infty \frac{(nmr_1)^{-i\eta}\sigma_{2i\eta}(nr_1)\sigma_{2i\eta}(m)\cosh(\pi\eta)\check{f}(\eta)}{|\zeta(1+2i\eta)|^2}d\eta$
• Holomorphic Form Contribution: $\Sigma_{\text{hol.}} = \sum_{k \text{ even}}\tilde{f}(k)\sum_{j}\psi_{j,k}(m)\psi_{j,k}(nr_1)$

2. Refined Estimates of Bessel Transforms (Sections 6.2-6.3)

Key Technique: Via Mellin inversion and Stirling's formula, prove:

Proposition 6.2: $\check{f}(\eta) \ll X^{2\varepsilon}\frac{e^{-\pi|\eta|}}{|\eta|^{2+\varepsilon}}$

Proof Strategy:

1. Represent $K_{2i\eta}(t)$ via Mellin inversion (Formula 6.7)
2. Shift integration line to $\sigma = -1-\delta$, collect residues at $s = \pm 2i\eta$
3. Integrate by parts twice (using $f''(t) \ll t^{-2}$)
4. Apply Stirling asymptotics (Lemma 2.3)

Proposition 6.3: $\ddot{f}(\eta) \ll X^\varepsilon \frac{e^{-\pi|\eta|}}{|\eta|^{5/2}}$

3. Maass Form Contribution Estimates (Section 6.4)

Lemma 6.4 (Key Technical Lemma): $\sum_{1 \leq n \leq N}|\rho_j(nq)|^2/\cosh(\pi\kappa_j) \ll q^{\theta+\varepsilon}N(N\kappa_j)^\varepsilon$

Proof: Uses Hecke relations (2.7) and Kim-Sarnak bounds ($\theta \leq 7/64$).

Proposition 6.5: Applying Cauchy-Schwarz and Weyl's law (Lemma 2.7): $M(X) \ll X^{2\varepsilon}\sum_{K \text{ dyadic}}\frac{1}{K^{2+\varepsilon}}\sum_{\kappa_j \sim K}\kappa_j^\varepsilon \ll r^\theta X^\varepsilon$

4. Continuous Spectrum and Holomorphic Form Treatment

• Continuous Spectrum (Proposition 6.6): Using asymptotics of $\zeta(1+2i\eta)$ and exponential decay, obtain $\Sigma_{\text{cont.}} \ll X^\varepsilon$
• Holomorphic Forms (Proposition 6.8): Apply Petersson trace formula, Deligne bounds, and J-Bessel function sum formulas (Lemma 6.7), obtain $\Sigma_{\text{hol.}} \ll X^\varepsilon$

Methodology for mod-$p$ Problem (Section 8)

Strategy Difference: For mod-$p$ problems, apply orthogonality: $S(X,1,V;p) = \frac{1}{p}\sum_{a,b,c,d}V_X(a,b,c,d) + \frac{1}{p}\sum_{h \neq 0}^*e\left(\frac{-h}{p}\right)\sum V_X(a,b,c,d)e\left(\frac{h(ad-bc)}{p}\right)$

Key Techniques:

1. Apply Poisson summation to only two variables ($c,d$) to avoid duplication
2. Eliminate $h$, transform to $ac+bd \equiv 0 \pmod{p}$
3. Apply Poisson summation again and Lemma 8.1, using smallness of dual variables ($\ll pg(p)/X$)

Experimental Setup

This is a pure mathematics theoretical paper with no numerical experiments. All results are obtained through rigorous mathematical proofs.

Theoretical Verification

1. Consistency with Known Results:
   • When $r=0$ or $r=X^2$, results agree with Heath-Brown HB96
   • Error term $O(r^\theta X^{1+\varepsilon})$ improves upon $O(X^{3/2+\varepsilon})$ for general quaternary forms
2. Limiting Cases:
   • When $r = o(X^2)$, main term $\sim K(V,r)X^2$ as expected
   • When $X < \sqrt{p}$, mod-$p$ problem reduces to determinant equation

Experimental Results

Main Results

Meaning of Theorem 1.1:

• Error Term Strength: $O(r^\theta X^{1+\varepsilon})$ is much smaller than main term $\sim X^2$ when $r = o(X^{2-\theta-\varepsilon})$
• Currently Optimal: $\theta = 7/64$ gives asymptotic formula when $r = o(X^{121/64-\varepsilon})$
• Conjectured Optimal: If Ramanujan-Petersson conjecture holds ($\theta=0$), asymptotic formula holds for all $r = o(X^2)$

Precise Main Term in Corollary 1.2: $K(V,r) = \frac{1}{\zeta(2)}\frac{\sigma(|r|)}{|r|}\int\!\!\!\int\!\!\!\int \frac{1}{t}V(u)V(v)V(t)V\left(\frac{uv}{t}\right)dudvdt$

• Factor $\frac{\sigma(|r|)}{|r|}$ reflects divisor structure of $r$ on solution count
• $\frac{1}{\zeta(2)} = \frac{6}{\pi^2}$ comes from probability of coprimality

Significance of Theorem 1.3:

• Improves upon Ahmadi-Shparlinski AS07 (who required fixed $\varepsilon$ in $X \gg p^{1/2+\varepsilon}$)
• Paper proves that for any $g(p) \to \infty$ (arbitrarily slowly), asymptotic formula holds when $X \gg p^{1/2}g(p)$
• Application: Abundance of matrices of height $O(p^{1/2+o(1)})$ belonging to $\text{SL}_2(\mathbb{F}_p)$ mod $p$

Comparison with Related Work

Advantages:

1. Wider range for $r$ (nearly reaching $X^2$)
2. Better dependence of error term on both $r$ and $X$
3. More direct method, avoiding circle method's technical complexity

Related Work

Quadratic Form Representation Theory

1. Classical Theory:
   • Jacobi-Siegel Theory: Study via theta series and modular forms (Kit86)
   • Siegel's Theorem: Indefinite quadratic forms satisfy Hasse principle for $n \geq 4$ (Sie51)
2. Circle Method:
   • Invented by Hardy-Ramanujan-Littlewood, developed by Kloosterman et al.
   • Heath-Brown HB96's "new form circle method"
3. Spectral Methods:
   • Pioneering work by Deshouillers-Iwaniec DI82, DI83
   • Applications of Kuznetsov formula

Applications of Determinant Equations

1. L-Function Moments: Duke-Friedlander-Iwaniec DFI97
2. Trilinear Forms in Kloosterman Fractions: Bettin-Chandee BC18
3. Shifted Convolution Sums: Related to divisor functions

Matrix Representations mod-$p$

1. Ahmadi-Shparlinski AS07:
   • Proved asymptotic formula when $X \gg p^{1/2+\varepsilon}$ (1.6)
   • Applications to matrix distribution over finite fields
2. This Paper's Improvement: Reduces $\varepsilon$ to arbitrarily slow $g(p)$

Conclusions and Discussion

Main Conclusions

1. Theoretical Contributions:
   • Establishes nearly optimal asymptotic formula for $xy-zw=r$ (under Ramanujan-Petersson conjecture)
   • Error term $O(r^\theta X^{1+\varepsilon})$ achieves strong bounds in both $r$ and $X$ dimensions
2. Methodological Contributions:
   • Demonstrates power of Poisson summation + Kuznetsov formula
   • Avoids technical complexity of circle method
   • Refined Bessel transform analysis is key technical innovation
3. Applied Value:
   • Provides better tools for L-function moment estimates
   • Gives nearly optimal results for mod-$p$ matrix distribution

Limitations

1. Range Restrictions:
   • Theorem 1.1 requires $r = o(X^2)$ (though already quite broad)
   • Theorem 1.3 requires $X > p^{1/100}$ (technical condition, can be relaxed)
2. Smooth Weights:
   • Uses smooth functions $V$ rather than sharp cutoffs
   • Sharp cutoff case is more difficult (see GG23)
3. Dependence on Unresolved Conjecture:
   • Error term depends on progress on Ramanujan-Petersson conjecture
   • Currently $\theta = 7/64$, conjectured value $\theta = 0$
4. Ternary Forms Unhandled:
   • Remark 1.4 mentions method extends to $\alpha x_1x_2 - \beta x_3^2$
   • Detailed proof not provided in paper

Future Directions

1. Generalization to Other Forms:
   • Ternary forms $\alpha x_1x_2 - \beta x_3^2$ (using Salié sums + Proskurin formula)
   • More general indefinite quadratic forms
2. Improving Error Terms:
   • Utilize progress on Ramanujan-Petersson conjecture
   • Explore alternative spectral methods
3. Sharp Cutoffs:
   • Extend range in GG23 (currently $|r| \leq X^{1/3}$)
   • Requires more refined spectral-theoretic tools
4. Applications:
   • Improved estimates for L-function moments
   • New results on shifted convolution sums

In-Depth Evaluation

Strengths

1. Methodological Innovation:
   • Direct application of Fourier analysis rather than circle method, with clear logic
   • Refined Bessel transform estimates (Propositions 6.2-6.3) demonstrate sophisticated analytic techniques
   • Transformation to Kloosterman sums followed by spectral theory application is clever
2. Result Strength:
   • Error term $O(r^\theta X^{1+\varepsilon})$ is strong bound in both parameters
   • Range $r = o(X^2)$ is nearly optimal
   • mod-$p$ results improve upon best existing results
3. Technical Depth:
   • All three parts of Kuznetsov formula (Maass, continuous spectrum, holomorphic) receive refined treatment
   • Lemma 6.4 (on $\rho_j(nq)$ estimates) is key technical contribution
   • Combination of Cauchy-Schwarz inequality with Weyl's law is effective
4. Writing Clarity:
   • Clear structure, logical flow from motivation to proof
   • Thorough comparison with related work (Remarks 1.1-1.6)
   • Complete technical details, highly reproducible

Weaknesses

1. Smooth Weight Restriction:
   • Sharp cutoff case has more applied value but not handled here
   • Though authors address part of this in GG23, range is narrower
2. Dependence on Deep Conjecture:
   • Specific error term value depends on Ramanujan-Petersson conjecture
   • Current $\theta = 7/64$ gives range $r = o(X^{121/64})$, room for improvement
3. Incomplete Generalizations:
   • Remark 1.4 mentions extension to ternary forms but proof not provided
   • Generalization to arbitrary indefinite quadratic forms unclear
4. Limited Application Demonstration:
   • Though L-function moment connection mentioned, no specific applications given
   • Could benefit from concrete examples or application scenarios

Impact

1. Theoretical Impact:
   • Provides new powerful tool for indefinite quadratic form representation problems
   • Bessel transform estimation techniques may inspire solutions to other problems
   • Demonstrates advantages of spectral methods in this context
2. Methodological Impact:
   • Poisson summation + Kuznetsov formula paradigm likely to be widely adopted
   • Avoids circle method complexity, making method more generalizable
3. Practical Value:
   • Provides stronger theoretical foundation for L-function moment estimates and related applications
   • mod-$p$ results have direct applications to matrix representation theory over finite fields
4. Reproducibility:
   • Complete proofs, clear technical details
   • Relies on standard tools (Kuznetsov formula, Weyl's law, etc.)
   • Highly reproducible

Applicable Scenarios

1. Theoretical Research:
   • Indefinite quadratic form representation problems
   • Kloosterman sum estimates
   • Spectral theory applications of automorphic forms
2. Analytic Number Theory:
   • L-function moment estimates
   • Shifted convolution sums
   • Exponential sum estimates
3. Algebraic Number Theory:
   • Quantitative versions of Hasse principle
   • Integral Brauer-Manin obstructions
4. Finite Fields:
   • Matrix distribution over finite fields
   • Congruence problems modulo primes

Key References

1. HB96 Heath-Brown, D. R. "A new form of the circle method, and its application to quadratic forms." J. Reine Angew. Math. 481 (1996): 149-206.
   • Circle method application to quadratic forms, main comparison point for this paper
2. DI83 Deshouillers, J.-M., and Iwaniec, H. "Kloosterman sums and Fourier coefficients of cusp forms." Invent. Math. 70 (1982/83): 219-288.
   • Classical reference for Kuznetsov formula
3. Kim03 Kim, H. H. "Functoriality for the exterior square of GL₄ and the symmetric fourth of GL₂." J. Amer. Math. Soc. 16 (2003): 139-183.
   • Best known result on Ramanujan-Petersson conjecture: $\theta \leq 7/64$
4. IK04 Iwaniec, H., and Kowalski, E. Analytic Number Theory. AMS, 2004.
   • Standard reference for analytic number theory
5. GG23 Ganguly, S., and Guria, R. "Counting lattice points on determinant surfaces and spectral methods of automorphic forms." Submitted, 2023.
   • Authors' prior work handling sharp cutoff case
6. AS07 Ahmadi, O., and Shparlinski, I. E. "Distribution of matrices with restricted entries over finite fields." Indag. Math. 18 (2007): 327-337.
   • Prior work on mod-$p$ problem

Overall Assessment: This is a high-quality analytic number theory paper that cleverly combines Fourier analysis with spectral theory of automorphic forms to obtain nearly optimal results for the classical determinant equation solution-counting problem. The method is novel, techniques are deep, results are strong, and the contribution to the field is significant. Particularly noteworthy are the refined Bessel transform estimates and the handling of $r$-dependence, demonstrating the authors' sophisticated analytic skills. The paper is clearly written with rigorous logic and represents an excellent example of spectral method applications.