Abstract:We obtain explicit upper and lower bounds on the size of the coefficients of the Drinfeld modular polynomials $\Phi_N$ for any monic $N\in\mathbb{F}_q[t]$. These polynomials vanish at pairs of $j$-invariants of Drinfeld $\mathbb{F}_q[t]$-modules of rank 2 linked by cyclic isogenies of degree $N$. The main term in both bounds is asymptotically optimal as $\mathrm{deg}(N)$ tends to infinity. We also obtain precise estimates on the Weil height and Taguchi height of Hecke images of Drinfeld modules of rank 2.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to estimate the magnitude of the coefficients of the Drinfeld modular polynomial \( \Phi_N \), that is, the height of \( \Phi_N \). Specifically, the author aims to obtain explicit upper and lower bounds for the height of \( \Phi_N \) and prove that these bounds are asymptotically optimal as \( \deg N \) tends to infinity. In addition, the author also studies the Weil height of Hecke images and the Taguchi stable height to compare the heights of Drinfeld modules and their Hecke images.
### Problem Background
The Drinfeld modular polynomial \( \Phi_N(X, Y) \) is a binary polynomial defined over \( F_q[t] \), and it vanishes on pairs of \( j \)-invariants under certain specific conditions. The coefficients of these polynomials grow rapidly as \( |N| \) increases, similar to the case of classical modular polynomials.
### Main Contributions
1. **Explicit Upper and Lower Bounds**: The author obtains explicit upper and lower bounds for the height \( h(\Phi_N) \) of \( \Phi_N \) and proves that these bounds are asymptotically optimal as \( \deg N \) tends to infinity. The specific results are as follows:
\[
h(\Phi_N) = \frac{q^2 - 1}{2} \psi(N) (\deg N - 2\lambda_N + b_q(N)),
\]
where,
\[
0 \leq b_q(N) \leq \frac{4 + 2q^3 + q^2 - 2q + 1}{q(q - 1)^2}.
\]
2. **Height Comparison of Hecke Images**: The author also compares the Weil height of the \( j \)-invariant of the Drinfeld module with the height of its Hecke image. The specific result is:
\[
\frac{q^2 - 1}{2} \left( 2\lambda_N - \deg N - b_q(N) \right) \leq h(j(\phi)) - \frac{1}{\psi(N)} \sum_{C} h(j(\phi/C)) \leq \frac{q + q^2 - 1}{2} \min \left[ 0, 2\lambda_N - \deg N + \log \left( \frac{1}{q h(j(\phi))} + 1 \right) \right].
\]
3. **Taguchi Stable Height**: The author proves that the Taguchi stable height provides a more concise expression when estimating the height of Hecke images. The specific result is:
\[
h_{\text{Tag}}(\phi) - \frac{1}{\psi(N)} \sum_{C} h_{\text{Tag}}(\phi/C) = \lambda_N - \frac{1}{2} \deg N.
\]
### Methods and Techniques
To obtain the above results, the author uses the following techniques and methods:
- **Mahler Measure Estimation**: Derive upper and lower bounds for the height by estimating the Mahler measure of \( \Phi_N(X, j(z)) \).
- **Geometric Mean and Discriminant Function**: Simplify the proof by using the property that the geometric mean of the discriminant function of Hecke images is equal to itself.
- **Farey Sequence**: Use the Farey sequence over the function field for estimation.
- **Bruhat - Tits Tree**: Handle the problem of "large" representatives by controlling the basic representatives on the Bruhat - Tits tree.
These techniques together ensure the accuracy and asymptotic optimality of the obtained results.