The modular-integral integrand is integrable on the vertical ray
ProvedMTT.modularIntegral_integrableFor a cusp form f of weight k and level N, an arbitrary polynomial P, and a rational base point r, the function t maps to f(r + it) P(r + it) is integrable over the open half line t > 0. This is what makes the modular integral of Mazur-Tate-Teitelbaum a genuine Lebesgue integral rather than a formal symbol, and it is the hypothesis needed to know that the modular integral is additive in f.
Two separate decay statements are involved. As t tends to infinity the cusp form is exponentially small because its q-expansion at the cusp at infinity has no constant term, and this dominates the polynomial growth of P. As t tends to zero the vertical ray approaches the rational point r, which is a cusp of Gamma_1(N); the cusp form vanishes there as well, again exponentially in 1/t after passing to the local parameter. The second statement is the substantive one: the Petersson bound alone gives only |f(r+it)| = O(t^{-k/2}), which is not integrable near zero for k at least 2, so a genuine expansion at the cusp r is required.
import Definitions.Def_MTT_Cohomology import Mathlib.RingTheory.Flat.Basic set_option autoImplicit false noncomputable section open scoped BigOperators TensorProduct open MTT.Cohomology
theorem MTT.modularIntegral_integrable
{N k : ℕ} (hN : 0 < N) (hk : 2 ≤ k) (f : CuspForm (MTT.GammaOne N) (k : ℤ))
(P : Polynomial ℂ) (r : ℚ) :
MeasureTheory.IntegrableOn
(fun t : ℝ => f (UpperHalfPlane.ofComplex ((r : ℂ) + Complex.I * t)) *
P.eval ((r : ℂ) + Complex.I * t))
(Set.Ioi (0 : ℝ)) MeasureTheory.volume := by sorry