Table of Contents
Fetching ...

Counting newforms with prescribed ramified supercuspidal components

Andrew Knightly, Kimball Martin

TL;DR

The paper derives explicit dimension formulas for spaces of newforms S_k^{new}(N) with prescribed ramified supercuspidal local components at a set T of primes. By blending a simple trace formula with local Galois-orbit analysis, the authors show that dim S_k^{new}(N;π_T) equals a main term plus an Atkin–Lehner–trace correction that depends only on the ramified data E_p and local root numbers ε_p, and is explicit in cases with M=1 or T prime. The key insight is that the global dimension is governed by the Galois orbits of ramified supercuspidals and the trace of W_T on the full newspace, revealing root-number biases and invariance under local Galois conjugacy. The results connect local representation theory (ramified dihedral supercuspidals) with global automorphic dimensions, and extend prior work by treating ramified odd-conductor cases and outlining extensions to depth-zero components. This provides exact counts and structural understanding of newforms with prescribed ramified components, with consequences for root-number bias and Galois decomposition in spaces of cusp forms.

Abstract

We give a formula for the number of newforms in $S_k^{\mathrm{new}}(N)$ that have prescribed ramified supercuspidal components $π_p$ at a set $T$ of primes dividing $N$. This dimension is given in terms of the trace of the Atkin--Lehner operator at $T$ on $S_k^{\mathrm{new}}(N)$. It depends only upon the weight, the level, the ramified quadratic extensions $E_p/{\mathbb Q}_p$ attached to the $π_p$, and the root number of each $π_p$. The formula is completely explicit when $T$ consists of either a single prime or all prime factors of $N$.

Counting newforms with prescribed ramified supercuspidal components

TL;DR

The paper derives explicit dimension formulas for spaces of newforms S_k^{new}(N) with prescribed ramified supercuspidal local components at a set T of primes. By blending a simple trace formula with local Galois-orbit analysis, the authors show that dim S_k^{new}(N;π_T) equals a main term plus an Atkin–Lehner–trace correction that depends only on the ramified data E_p and local root numbers ε_p, and is explicit in cases with M=1 or T prime. The key insight is that the global dimension is governed by the Galois orbits of ramified supercuspidals and the trace of W_T on the full newspace, revealing root-number biases and invariance under local Galois conjugacy. The results connect local representation theory (ramified dihedral supercuspidals) with global automorphic dimensions, and extend prior work by treating ramified odd-conductor cases and outlining extensions to depth-zero components. This provides exact counts and structural understanding of newforms with prescribed ramified components, with consequences for root-number bias and Galois decomposition in spaces of cusp forms.

Abstract

We give a formula for the number of newforms in that have prescribed ramified supercuspidal components at a set of primes dividing . This dimension is given in terms of the trace of the Atkin--Lehner operator at on . It depends only upon the weight, the level, the ramified quadratic extensions attached to the , and the root number of each . The formula is completely explicit when consists of either a single prime or all prime factors of .

Paper Structure

This paper contains 14 sections, 15 theorems, 103 equations.

Key Result

Theorem 1.1

Suppose $T\ge 5$ is odd, and fix a tuple $\pi_T=(\pi_p)_{p|T}$ as above. Let $\varepsilon_{\pi_T}=\prod_{p|T}\varepsilon_p$ be the product of the root numbers of the $\pi_p$. Define Let $k\ge 4$ be even and let $N$ be as in N (with $v_3(N)=3$ if $3\, | \, T$). Then where $W_T=\prod_{p|T}W_p$ is the Atkin-Lehner operator at $T$ of level $N$, and $\psi^{\mathrm{new}}$ is the multiplicative functio

Theorems & Definitions (40)

  • Theorem 1.1
  • Remark 1.2
  • Corollary 1.3
  • Remark 1.4
  • Example 1.5
  • Example 1.6
  • Proposition 2.1
  • proof
  • Proposition 2.2
  • Remark 2.3
  • ...and 30 more