ConceptComplete

Ramification and Decomposition - Key Properties

The decomposition and inertia groups provide a refined understanding of how primes behave in Galois extensions.

DefinitionInertia Group

For a Galois extension L/KL/K and prime P\mathfrak{P} of LL above p\mathfrak{p} of KK, the inertia group is: IP={ΟƒβˆˆDP:Οƒ(Ξ±)≑α(modP)Β forΒ all α∈OL}I_\mathfrak{P} = \{\sigma \in D_\mathfrak{P} : \sigma(\alpha) \equiv \alpha \pmod{\mathfrak{P}} \text{ for all } \alpha \in \mathcal{O}_L\}

This is a normal subgroup of DPD_\mathfrak{P}. The extension is unramified at p\mathfrak{p} if and only if IP={1}I_\mathfrak{P} = \{1\}.

The quotient DP/IP≅Gal((OL/P)/(OK/p))D_\mathfrak{P}/I_\mathfrak{P} \cong \text{Gal}((\mathcal{O}_L/\mathfrak{P})/(\mathcal{O}_K/\mathfrak{p})) is cyclic, generated by the Frobenius element.

TheoremRamification Formula

In a Galois extension L/KL/K, all primes Pi\mathfrak{P}_i above p\mathfrak{p} have the same ramification index ee and inertia degree ff. Moreover: efg=n=[L:K]efg = n = [L : K]

where gg is the number of primes above p\mathfrak{p}. The orders satisfy: ∣DP∣=ef,∣IP∣=e|D_\mathfrak{P}| = ef, \quad |I_\mathfrak{P}| = e

ExampleCyclotomic Extensions

For L=Q(ΞΆp)L = \mathbb{Q}(\zeta_p) with pp prime, the only ramified prime is (p)(p), which factors as: (p)=(1βˆ’ΞΆp)pβˆ’1(p) = (1 - \zeta_p)^{p-1}

with ramification index e=pβˆ’1e = p - 1, inertia degree f=1f = 1, and g=1g = 1. All other rational primes are unramified.

For a prime qβ‰ pq \neq p, the decomposition depends on the order of qq modulo pp (smallest ff with qf≑1(modp)q^f \equiv 1 \pmod{p}).

DefinitionDifferent and Discriminant

The different DL/K\mathfrak{D}_{L/K} is the inverse of the ideal: {α∈L:TrL/K(Ξ±OL)βŠ†OK}\{\alpha \in L : \text{Tr}_{L/K}(\alpha \mathcal{O}_L) \subseteq \mathcal{O}_K\}

The discriminant is Ξ”L/K=NL/K(DL/K)\Delta_{L/K} = N_{L/K}(\mathfrak{D}_{L/K}). These measure ramification: DL/K=∏PPePβˆ’1+Ξ΄P\mathfrak{D}_{L/K} = \prod_{\mathfrak{P}} \mathfrak{P}^{e_\mathfrak{P} - 1 + \delta_\mathfrak{P}}

where Ξ΄Pβ‰₯0\delta_\mathfrak{P} \geq 0 measures wild ramification (Ξ΄P>0\delta_\mathfrak{P} > 0 when ePe_\mathfrak{P} is divisible by the residue characteristic).

Remark

Tame vs. wild ramification: If ePe_\mathfrak{P} is coprime to the residue characteristic, ramification is tame and Ξ΄P=0\delta_\mathfrak{P} = 0. Otherwise, it's wild with Ξ΄P>0\delta_\mathfrak{P} > 0.

Wild ramification is more complicated, involving higher ramification groups. Tame ramification is well-understood via Kummer theory.

TheoremHilbert's Ramification Theory

In an abelian extension L/KL/K, the prime p\mathfrak{p} splits completely if and only if p\mathfrak{p} is a norm from LL: there exists β∈L\beta \in L with NL/K(β)∈pN_{L/K}(\beta) \in \mathfrak{p}.

This is a special case of the Artin reciprocity law, connecting splitting behavior to ideal-theoretic conditions.

ExampleRamification in Number Field Towers

Consider QβŠ‚KβŠ‚L\mathbb{Q} \subset K \subset L. If p\mathfrak{p} ramifies in L/QL/\mathbb{Q}, it may ramify in K/QK/\mathbb{Q}, in L/KL/K, or both.

The ramification indices satisfy: eL/Q=eK/Qβ‹…eL/Ke_{L/\mathbb{Q}} = e_{K/\mathbb{Q}} \cdot e_{L/K}.

For QβŠ‚Q(i)βŠ‚Q(i,2+i)\mathbb{Q} \subset \mathbb{Q}(i) \subset \mathbb{Q}(i, \sqrt{2+i}): the prime (2)(2) ramifies at each level, with ramification indices multiplying appropriately.