We consider a newform of weight 5 and level 7, whose local representation at 7 is principal series.
> S := CuspidalSubspace(ModularSymbols(Gamma1(7), 5, 1));
> newforms := NewformDecomposition(S);
> Eigenform(newforms[1], 15);
q + q^2 - 15*q^4 + 49*q^7 - 31*q^8 + 81*q^9 - 206*q^11 + 49*q^14 + O(q^15)
> pi := LocalComponent(newforms[1], 7);
> pi;
Ramified Principal Series Representation of GL(2,Q_7)
> chi := CentralCharacter(pi);
> Conductor(chi);
7
> parameters := PrincipalSeriesParameters(pi);
These are Dirichlet characters on Z/7Z (the trivial character
and the character of order 2):
> Conductor(parameters[1]), Order(parameters[1]);
1 1
> Conductor(parameters[2]), Order(parameters[2]);
7 2
The principal series representation π is the induction up to GL
2(Q
7) of a character of the
Borel subgroup inflated from a character of the diagonal group Q
7 x x Q
7 x .
The restriction of this character to Z
7 x x Z
7 x gives the pair of Dirichlet characters above.
We now compute the Galois representation.
> rho := WeilRepresentation(pi); rho;
2-dim Galois representation (2,0) with G=C2, I=C2, conductor 7^1 over Q7[40]
> IsAbelian(Group(rho));
true
The Weil representation is simply the sum of the two characters above
(up to unramified twists), considered as characters of the Galois group
of Q
7 via local class field theory.
> Decomposition(rho);
[
1-dim trivial Galois representation 1 over Q7[40],
1-dim Galois representation (1,-1) with G=C2, I=C2, conductor 7^1 over Q7[40]
]
We consider a supercuspidal representation of conductor 121, associated to a newform of weight 2 and level 121.
> S := CuspidalSubspace(ModularSymbols(Gamma0(121), 2, 1));
> newforms := NewformDecomposition(S);
> newforms;
[
Modular symbols space for Gamma_0(121) of weight 2 and dimension 1
over Rational Field,
Modular symbols space for Gamma_0(121) of weight 2 and dimension 1
over Rational Field,
Modular symbols space for Gamma_0(121) of weight 2 and dimension 1
over Rational Field,
Modular symbols space for Gamma_0(121) of weight 2 and dimension 1
over Rational Field,
Modular symbols space for Gamma_0(121) of weight 2 and dimension 2
over Rational Field
]
> Eigenform(newforms[2], 11);
q + q^2 + 2*q^3 - q^4 + q^5 + 2*q^6 - 2*q^7 - 3*q^8 + q^9 + q^10 + O(q^11)
> pi := LocalComponent(newforms[2], 11);
> pi;
Supercuspidal Representation of GL(2,Q_11)
This means the representation of the Weil group associated to pi is irreducible.
> Conductor(pi);
121
> W := CuspidalInducingDatum(pi);
> W;
GModule W of dimension 10 over Rational Field
W is a module over a group which is a quotient of GL
2(Z
11), namely GL
2(Z/(11Z)).
The representation π is induced from some extension of W to the open subgroup Q
11 x GL
2(Z
11).
> Group(W);
MatrixGroup(2, IntegerRing(11)) of order 2^4 * 3 * 5^2 * 11
Generators:
[2 0]
[0 1]
[1 1]
[0 1]
[ 0 1]
[10 0]
> Group(W) eq GL(2, Integers(11));
true
> rho:=WeilRepresentation(pi);
This gives the Weil representation attached to π up to
multiplication by an unramified twist.
> rho;
2-dim Galois representation (2,0,-1) with G=S3, I=C3, conductor 11^2
over Q11[10]
We consider a supercuspidal representation of conductor 3
3, associated to a newform of weight 4 and level 27.
> S := CuspidalSubspace(ModularSymbols(Gamma0(27), 4, 1));
> newforms := NewformDecomposition(S);
> Eigenform(newforms[1], 13);
q + 3*q^2 + q^4 + 15*q^5 - 25*q^7 - 21*q^8 + 45*q^10 - 15*q^11 + O(q^13)
> pi:=LocalComponent(newforms[1], 3);
> pi;
Supercuspidal Representation of GL(2,Q_3)
> W:=CuspidalInducingDatum(pi);
> W;
GModule W of dimension 2 over Rational Field
> Group(W);
MatrixGroup(2, IntegerRing(9)) of order 2^2 * 3^5
Generators:
[1 1]
[0 1]
[2 0]
[0 1]
[1 0]
[0 2]
[1 0]
[3 1]
These matrices generate (topologically) the Iwahori subgroup of GL
2(Z
3) consisting of matrices which are upper-triangular modulo 3. W is an irreducible two-dimensional G-module. The representation π is induced from some extension of W to the normalizer of the Iwahori in GL
2(Q
3).
> E, chi:=AdmissiblePair(pi);
> E;
Totally ramified extension defined by the polynomial x^2 - 3
over 3-adic ring mod 3^10
> E.1^2;
3
> chi(1+E.1);
-zeta_3 - 1
Note that
chi can only be evaluated on units of E, so that
chi(E.1) would result in an error.
> WeilRepresentation(pi);
2-dim Galois representation (2,0,-1) with G=S3, I=S3, conductor 3^3 over Q3[10]
V2.28, 13 July 2023