The ring of modular forms for $\Gamma_1=\text{SL}(2,\mathbf{Z})$ $$M(\Gamma_1)\ =\ \bigoplus_{k\ge 0} M(\Gamma_1)_k\ =\ k[E_4,E_6]$$ is a free ring, with a generator in weight $4$ and weight $6$. (The reason is that $\mathfrak{H}/\Gamma_1=\mathbf{P}(4,6)$ is a weighted projective space, and the above is $\bigoplus_{k\ge 0}H^0(\mathbf{P}(4,6),\mathcal{O}(k))$). The generators are Eisenstein series.

Question: what is the ring of modular forms of level $2$ $$M(\Gamma_2)\ =\ \bigoplus_{k\ge 0} M(\Gamma_2)_k\ ?$$ Since $\mathfrak{H}/\Gamma_2=\mathbf{P}(2,2)$ I believe it should be free on two generators of weight $2$, but I don't know what they are explicitly as functions on $\mathfrak{H}$.


Yes, the ring of holomorphic modular forms on $\Gamma(2) = \Gamma_2 \triangleleft \mathrm{SL}_2(\mathbb{Z})$ is equal to $\mathbb{C}[X,Y]$ where $X = \Theta_2^4$ and $Y= \Theta_3^4$ and where

$$ \Theta_2(z) = \sum_{n \in \mathbb{Z}}{e^{\pi i (n+1/2)^2 z}}, \quad \Theta_3(z) = \sum_{n \in \mathbb{Z}}{e^{\pi i n^2 z}}, \qquad \Theta_4(z) = \sum_{n \in \mathbb{Z}}{(-1)^ne^{\pi i n^2 z}} $$

are the classical theta constants (sometimes also denoted $\Theta_2 =\theta_{10}, \Theta_3 = \theta_{00}, \Theta_4 =\theta_{01}$)

Thus, for example $$ M_2(\Gamma_2)= \mathbb{C} \Theta_2^4 \oplus \mathbb{C} \Theta_3^4, \quad M_4(\Gamma_2) = \mathbb{C} \Theta_2^8 \oplus \mathbb{C} \Theta_2^4\Theta_3^4 \oplus \mathbb{C}\Theta_3^8. $$

You can prove this description using the following valency formula (which is in some sense simpler than the one for level 1 forms and also be proved via contour integration, as is done (in level 1) in Serre's course of arithmetic, for example)

$$ \nu_{[1]}(f) + \nu_{[0]}(f) + \nu_{[\infty]}(f) + \sum_{[p] \in \Gamma_2 \backslash \mathbb{H}}{\nu_{[p]}(f)}= \frac{k}{2} $$

valid for any nonzero meromorphic modular form $f$ of even integral weight $k$. It implies first that none of the theta constants $\Theta_2, \Theta_3, \Theta_4$ has a zero on the upper half plane. Another input is that multiplication by $\Theta_3^4$ defines an isomorphism from $M_k(\Gamma_2)$ onto $S_{k+2}^{[1]}(\Gamma_2) \subset M_{k+2}(\Gamma_2)$ where $S_{k+2}^{[1]}(\Gamma_2)$ denotes the subspace of forms vanishing at the cusp 1. (This can be seen as an analogue of multiplication by $\Delta \in S_{12}(\Gamma_1)$, shifting the weight by 12)

  • $\begingroup$ Thank you! There's a slight typo- $Y=\Theta_3^4$, and theta. Also, is $\Theta_4^4$ not a level 2 modular form, or is it just expressible in terms of $\Theta_2^4$ and $\Theta_3^4$ (in which case, what is the expression)? $\endgroup$
    – Meow
    Dec 16 '20 at 17:14
  • 1
    $\begingroup$ Thanks, I fixed some of the typos. $\Theta_4^4 = \Theta_3^4- \Theta_2^4$ is a modular form of level 2 and weight 2. The latter identity is also attributed to Jacobi. $\endgroup$
    – m.s
    Dec 16 '20 at 17:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.