24.04.2014 Views

Inter-universal Teichmuller Theory I: Construction of Hodge Theaters

Inter-universal Teichmuller Theory I: Construction of Hodge Theaters

Inter-universal Teichmuller Theory I: Construction of Hodge Theaters

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

INTER-UNIVERSAL TEICHMÜLLER THEORY I 61<br />

[cf. [FrdI], Proposition 1.11, (iv)]. In particular, one may pull-back sections <strong>of</strong> the<br />

monoid O ⊲ (−) on D Θ Cv<br />

Θ v <strong>of</strong> Example 3.2, (v), to B. Such pull-backs are useful, for<br />

instance, when one considers the roots <strong>of</strong> Θ v<br />

, as in the theory <strong>of</strong> [EtTh], §5.<br />

Remark 3.2.4. Before proceeding, we pause to discuss certain minor oversights<br />

on the part <strong>of</strong> the author in the discussion <strong>of</strong> the theory <strong>of</strong> tempered Frobenioids<br />

in [EtTh], §3, §4. Let Z log<br />

∞ be as in the discussion at the beginning <strong>of</strong> [EtTh], §3.<br />

Here, we recall that Z log<br />

∞ is obtained as the “<strong>universal</strong> combinatorial covering” <strong>of</strong><br />

the formal log scheme associated to a stable log curve with split special fiber over<br />

the ring <strong>of</strong> integers <strong>of</strong> a finite extension <strong>of</strong> an MLF <strong>of</strong> residue characteristic p [cf.<br />

loc. cit. for more details]; we write Z log for the generic fiber <strong>of</strong> the stable log curve<br />

under consideration.<br />

(i) First, let us consider the following conditions on a nonzero meromorphic<br />

function f on Z log<br />

∞ :<br />

(a) For every N ∈ N ≥1 , it holds that f admits an N-th root over some<br />

tempered covering <strong>of</strong> Z log .<br />

(b) For every N ∈ N ≥1 which is prime to p, it holds that f admits an N-th<br />

root over some tempered covering <strong>of</strong> Z log .<br />

(c) The divisor <strong>of</strong> zeroes and poles <strong>of</strong> f is a log-divisor.<br />

It is immediate that (a) implies (b). Moreover, one verifies immediately, by considering<br />

the ramification divisors <strong>of</strong> the tempered coverings that arise from extracting<br />

roots <strong>of</strong> f, that (b) implies (c). When N is prime to p, iff satisfies (c), then<br />

it follows immediately from the theory <strong>of</strong> admissible coverings [cf., e.g., [PrfGC],<br />

§2, §8] that there exists a finite log étale covering Y log → Z log whose pull-back<br />

Y∞<br />

log → Z∞<br />

log to the generic fiber Z∞<br />

log <strong>of</strong> Z log<br />

∞ is sufficient<br />

(R1) to annihilate all ramification over the cusps or special fiber <strong>of</strong> Z log<br />

∞<br />

might arise from extracting an N-th root <strong>of</strong> f, aswellas<br />

that<br />

(R2) to split all extensions <strong>of</strong> the function fields <strong>of</strong> irreducible components <strong>of</strong><br />

the special fiber <strong>of</strong> Z log<br />

∞ that might arise from extracting an N-th root <strong>of</strong><br />

f.<br />

That is to say, in this situation, it follows that f admits an N-th root over the<br />

tempered covering <strong>of</strong> Z log given by the “<strong>universal</strong> combinatorial covering” <strong>of</strong> Y log .<br />

In particular, it follows that (c) implies (b). Thus, in summary, we have:<br />

(a) =⇒ (b) ⇐⇒ (c).<br />

On the other hand, unfortunately, it is not clear to the author at the time <strong>of</strong> writing<br />

whether or not (c) [or (b)] implies (a).<br />

(ii) Observe that it follows from the theory <strong>of</strong> [EtTh], §1 [cf., especially, [EtTh],<br />

Proposition 1.3] that the theta function that forms the main topic <strong>of</strong> interest <strong>of</strong><br />

[EtTh] satisfies condition (a) <strong>of</strong> (i).

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!