1: \documentclass{amsart}

2: \usepackage[alphabetic]{amsrefs}

3: \usepackage{amssymb,amscd}

4: \usepackage[all]{xy}

5: 

6: %

7: %  we should redo rems environment to include the enumerated list, 

8: % so it's \begin{rems}\item ...\item ... \end{rems}

9: %

10: 

11: %\include{decs}

12: \numberwithin{equation}{subsection}

13: 

14: \swapnumbers

15: \theoremstyle{plain}

16: \newtheorem{thm}[subsection]{Theorem}

17: \newtheorem{prop}[subsection]{Proposition}

18: \newtheorem{cor}[subsection]{Corollary}

19: %%%%%% from R2

20: \newtheorem*{thm*}{Theorem}

21: \newtheorem{lemma}[subsection]{Lemma}

22: %%%%%%

23: 

24: \theoremstyle{definition}

25: \newtheorem{defn}[subsection]{Definition}

26: 

27: \theoremstyle{remark}

28: \newtheorem{rem}[subsection]{Remark}

29: \newtheorem{rems}[subsection]{Remarks}

30: 

31: 

32: \newcommand{\Curve}{\mathcal{C}}

33: \renewcommand{\O}{\mathcal{O}}

34: \newcommand{\FF}{\mathcal{F}}

35: \newcommand{\EE}{\mathcal{E}}

36: \newcommand{\LL}{\mathcal{L}}

37: \newcommand{\MM}{\mathcal{M}}

38: \newcommand{\NN}{\mathcal{N}}

39: \newcommand{\F}{\mathbb{F}}

40: \newcommand{\Fp}{{\mathbb{F}_p}}

41: \newcommand{\Fptimes}{{\mathbb{F}_p^\times}}

42: \newcommand{\Fq}{{\mathbb{F}_q}}

43: \newcommand{\Fqtimes}{{\mathbb{F}_q^\times}}

44: \newcommand{\Z}{\mathbb{Z}}

45: \newcommand{\Q}{\mathbb{Q}}

46: \newcommand{\R}{\mathbb{R}}

47: \newcommand{\C}{\mathbb{C}}

48: \newcommand{\A}{\mathbb{A}}

49: \renewcommand{\P}{\mathbb{P}}

50: \newcommand{\f}{\mathfrak{f}}

51: \renewcommand{\d}{\mathfrak{d}}

52: \newcommand{\m}{\mathfrak{m}}

53: \newcommand{\n}{\mathfrak{n}}

54: \newcommand{\p}{\mathfrak{p}}

55: \newcommand{\AF}{\mathcal{A}}

56: \newcommand{\K}{K} % compact open for level

57: \newcommand{\G}{\backslash G}

58: \newcommand{\T}{\mathcal{T}}

59: \newcommand{\Nbar}{\overline{N}}

60: \newcommand{\sha}{{\hbox to 10pt{\rlap{\hskip2.8pt\vrule

61: height6pt\hskip1.6pt\vrule height6pt\hskip1.6pt

62: \vrule height6pt}\hskip1pt\vrule height0.8pt width 8pt\hskip1pt}}}

63: 

64: \newcommand{\into}{\hookrightarrow}

65: \newcommand{\onto}{\twoheadrightarrow}

66: \newcommand{\isoto}{\tilde{\to}}

67: \newcommand{\tensor}{\otimes}

68: \newcommand{\compose}{\circ}

69: \newcommand{\nodiv}{\not|}

70: \def\nodiv{\mathrel{\mathchoice{\not|}{\not|}{\kern-.2em\not\kern.2em|}{\kern-.2em\not\kern.2em|}}}

71: \newcommand{\PGL}{\mathrm{PGL}}

72: \newcommand{\SL}{\mathrm{SL}}

73: \newcommand{\GL}{\mathrm{GL}}

74: \newcommand{\Sp}{\mathrm{Sp}}

75: \DeclareMathOperator{\id}{Id}

76: \DeclareMathOperator{\im}{Im}

77: \DeclareMathOperator{\tr}{Tr}

78: \DeclareMathOperator{\N}{N}

79: \DeclareMathOperator{\res}{Res}

80: \DeclareMathOperator{\fun}{Fun}

81: \DeclareMathOperator{\cond}{Cond}

82: \DeclareMathOperator{\ord}{ord}

83: \DeclareMathOperator{\rk}{Rank}

84: \DeclareMathOperator{\dvsr}{div}

85: \DeclareMathOperator{\RP}{Re}

86: \DeclareMathOperator{\ext}{Ext}

87: \DeclareMathOperator{\Hom}{Hom}

88: \DeclareMathOperator{\aut}{Aut}

89: \DeclareMathOperator{\pic}{Pic}

90: \DeclareMathOperator{\stab}{Stab}

91: \DeclareMathOperator{\lcm}{lcm}

92: \DeclareMathOperator{\gal}{Gal}

93: \DeclareMathOperator{\spec}{Spec}

94: \DeclareMathOperator{\en}{End}

95: \DeclareMathOperator{\mor}{Mor}

96: \DeclareMathOperator{\disc}{Disc}

97: 

98: 

99: 

100: \renewcommand{\vector}[1]{\mathbf{#1}}

101: \renewcommand{\v}{\vector{v}}

102: \newcommand{\w}{\vector{w}}

103: \newcommand{\<}{\langle}

104: \renewcommand{\>}{\rangle}

105: \newcommand{\abs}[1]{\lvert#1\rvert}

106: 

107: \newcommand{\pfrac}[2]{\genfrac{(}{)}{}{}{#1}{#2}}

108: \newcommand{\smat}[1]{\begin{smallmatrix}#1\end{smallmatrix}}

109: \newcommand{\psmat}[1]{\bigl(\begin{smallmatrix}#1\end{smallmatrix}\bigr)}

110: \newcommand{\pmat}[1]{\begin{pmatrix}#1\end{pmatrix}}

111: 

112: %%%%%%%%%%% from R2.tex and R3.tex %%%%%%%%%%%%%%%%%%%%%%

113: \newcommand{\Fbar}{{\overline{\mathbb{F}}}}

114: \newcommand{\Fpbar}{{\overline{\mathbb{F}}_p}}

115: \newcommand{\Fqbar}{{\overline{\mathbb{F}}_q}}

116: \newcommand{\Fpf}{{\mathbb{F}}_{p^f}}

117: \newcommand{\Fr}{{\mathbb{F}}_{r}}

118: \newcommand{\Qbar}{{\overline{\mathbb{Q}}}}

119: \newcommand{\EEbar}{{\overline{\EE}}}

120: \newcommand{\kbar}{{\overline{k}}}

121: \newcommand{\ratto}{{\dashrightarrow}}

122: \newcommand{\Ql}{{\mathbb{Q}_\ell}}

123: \newcommand{\Zl}{{\mathbb{Z}_\ell}}

124: \newcommand{\Qlbar}{{\overline{\mathbb{Q}}_\ell}}

125: %\DeclareMathOperator{\res}{Res}

126: \DeclareMathOperator{\ind}{Ind}

127: \DeclareMathOperator{\sgn}{Sign}

128: \DeclareMathOperator{\swan}{Swan}

129: \DeclareMathOperator{\sel}{Sel}

130: 

131: \newcommand{\XX}{\mathcal{X}}

132: \newcommand{\YY}{\mathcal{Y}}

133: \newcommand{\ZZ}{{\mathcal{Z}}}

134: \newcommand{\GG}{{\mathcal{G}}}

135: \newcommand{\HH}{{\mathcal{H}}}

136: %\newcommand{\FF}{\mathcal{F}}

137: 

138: \newcommand{\Zhatp}{{\hat\Z^{(p)}}}

139: 

140: %\newcommand{\p}{{\mathcal{p}}}

141: \renewcommand{\a}{{\mathbf{a}}}

142: 

143: % This needs fixed:

144: \newcommand{\sdp}{{\rtimes}}

145: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

146: 

147: 

148: \theoremstyle{plain}

149: \newtheorem{subsublemma}[subsubsection]{Lemma}

150: 

151: 

152: \begin{document}

153: \title[Small ranks over function fields]{Jacobi sums, Fermat Jacobians, 

154: \\and ranks of abelian varieties 

155: \\over towers of function fields} 

156: \author{Douglas Ulmer}

157: \address{Department of Mathematics \\ University of Arizona \\ Tucson,

158: AZ  85721}

159: \email{ulmer@math.arizona.edu}

160: \thanks{This paper is based upon work supported by the National

161: Science Foundation under Grant No. DMS 0400877}

162: \date{September 23, 2006}

163: \subjclass[2000]{Primary 14G05, 11G40; 

164: Secondary 11G05, 11G10, 11G30, 14G10, 14G25, 14K12, 14K15}

165: %\begin{abstract}

166: %\end{abstract}

167: \maketitle

168: 

169: \section{Introduction}

170: 

171: \subsection{}

172: Given an abelian variety $A$ over a function field $K=k(\Curve)$ with

173: $\Curve$ an absolutely irreducible, smooth, proper curve over a field

174: $k$, it is natural to ask about the behavior of the Mordell-Weil group

175: of $A$ in the layers of a tower of fields over $K$.  The simplest

176: case, which is already very interesting, is when $A$ is an elliptic

177: curve, $K=k(t)$ is a rational function field, and one considers the

178: towers $k(t^{1/d})$ or $\overline k(t^{1/d})$ as $d$ varies through

179: powers of a prime or through all integers not divisible by the

180: characteristic of $k$.

181: 

182: When $k=\Q$ or more generally a number field, several authors (e.g.,

183: \cite{Shioda}, \cite{Stiller}, \cite{Fastenberg}, \cite{Silverman1},

184: \cite{Silverman2}, and \cite{Ellenberg}) have considered this question

185: and given bounds on the rank of $A$ over $\Q(t^{1/d})$ or

186: $\Qbar(t^{1/d})$.  In some interesting cases it can be shown that $A$

187: has rank bounded independently of $d$ in the tower $\Qbar(t^{1/d})$.

188: Of course no example is yet known of an elliptic curve over $\Q(t)$

189: with unbounded ranks in the tower $\Q(t^{1/d})$, nor of an elliptic

190: curve over $\Qbar(t)$ with non-constant $j$-invariant and unbounded

191: ranks in the tower $\Qbar(t^{1/d})$.

192: 

193: When $k$ is a finite field, examples of Shioda and the author show

194: that there are non-isotrivial elliptic curves over $\Fp(t)$ with

195: unbounded ranks in the towers $\Fpbar(t^{1/d})$ \cite{Shioda}*{Remark

196:   10} and $\Fp(t^{1/d})$ \cite{UlmerR1}*{1.5}.  More recently, the

197: author has shown \cite{UlmerR2} that high ranks over function fields

198: over finite fields are in some sense ubiquitous.  For example, for

199: every prime $p$ and every integer $g>0$ there are absolutely simple

200: abelian varieties of dimension $g$ over $\Fp(t)$ with unbounded ranks

201: in the tower $\Fp(t^{1/d})$, and given any non-isotrivial elliptic

202: curve $E$ over $\Fq(t)$, there exists a finite extension $\Fr(u)$ such

203: that $E$ has unbounded (analytic) ranks in the tower $\Fr(u^{1/d})$.

204: 

205: One obvious difference between number fields and finite fields which

206: might be relevant here is the complexity of their absolute Galois

207: groups: that of a finite field is pro-cyclic while that of a number

208: field is highly non-abelian.  Ellenberg uses this non-abelianess in a

209: serious way in his work on bounding ranks and, in a private

210: communication, he asked whether it might be the case that, say, a

211: non-isotrivial elliptic curve over $\Fq(t)$ always has unbounded rank

212: in the tower $\Fp(t^{1/d})$.

213: 

214: Our goal in this note, which is a companion to \cite{UlmerR2}, is to

215: give a number of examples of abelian varieties over function fields

216: $\Fq(t)$ which have bounded ranks in the towers $\Fqbar(t^{1/d})$ as

217: $d$ ranges through powers of a suitable prime or through all integers

218: not divisible by $p$, the characteristic of $\Fq$.  We also get some

219: information about ranks in towers $k(t^{1/d})$ for arbitrary fields

220: $k$.  Along the way we prove some new results on Fermat curves which

221: may be of independent interest.  The main results are

222: Theorems~\ref{thm:GaussJacobiBounds}, \ref{thm:Jd0}, \ref{thm:Jdp},

223: \ref{thm:isotrivial}, \ref{thm:nonisol}, and \ref{thm:nonisod}

224:  

225:  

226: \subsection{}%thanks

227: It is a pleasure to thank Jordan Ellenberg for his stimulating

228: questions about ranks of elliptic curves as well as Brian Conrey, Bill

229: McCallum, and Dinesh Thakur for their help.  Special thanks are due to

230: Bjorn Poonen for several incisive remarks and for pointing out that

231: some arguments originally given for elliptic curves apply more

232: generally to higher-dimensional abelian varieties.

233: 

234: 

235: 

236:  

237: \section{Jacobi sums}

238:  

239: \subsection{} %characters

240: Throughout the paper $p$ will be a rational prime number, $\Fp=\Z/p\Z$

241: will be the prime field of characteristic $p$, and $q=p^f$ will be a

242: power of $p$.  Fix an algebraic closure $\Qbar$ of $\Q$.  All number

243: fields considered will tacitly be assumed to be subfields of $\Qbar$.

244: We denote by $\mu_d$ the group of $d$-th roots of unity in $\Qbar$.

245: 

246: Let $\p$ be a prime of $\O_\Qbar$, the ring of integers of $\Qbar$,

247: over $p$.  The field $\O_\Qbar/\p$ is an algebraic closure of $\Fp$

248: which we denote by $\Fpbar$ and we write $\Fq$ for its subfield of

249: cardinality $q$.

250: 

251: Reduction modulo $\p$ induces an isomorphism between the group of all

252: roots of unity of order prime to $p$ in $\O_\Qbar$ and the

253: multiplicative group $(\O_\Qbar/\p)^\times=\Fpbar^\times$.  We let

254: $t:\Fpbar^\times\to\Qbar^\times$ denote the inverse of this

255: isomorphism.  We will use the same letter $t$ for the restriction to

256: any of the finite fields $\Fqtimes$.  Every character of $\Fqtimes$ is

257: a power of $t$.

258: 

259: 

260: \subsection{}%Gauss sums

261: Fix a non-trivial additive character $\psi_p:\Fp\to\Qbar^\times$.  For

262: each $q$ we define an additive character $\psi_q$ as

263: $\psi_q=\psi_p\compose\tr_{\Fq/\Fp}$.

264:  

265: For each $q$ and each character $\chi$ of $\Fqtimes$, we define a

266: Gauss sum

267: $$G_q(\chi)=-\sum_{x\in\Fqtimes}\chi(x)\psi_q(x)\in\Q(\mu_{p(q-1)}).$$

268: It is well known that $G_q(\chi)=1$ if $\chi$ is the trivial character

269: and that $G_q(\chi)$ is an algebraic integer with absolute value

270: $q^{1/2}$ in every complex embedding if $\chi\neq1$.

271: 

272: For $d$ prime to $q$, $a\in\Z/d\Z$, and any $q\equiv1\pmod d$ we write

273: $G_q(a)$ for $G_q(t^{-a(q-1)/d})$ which lies in $\Q(\mu_{pd})$.  The

274: analysis leading to Stickelberger's theorem \cite{Washington}*{6.2}

275: shows that if $\wp$ is the prime of $\Q(\mu_{pd})$ under $\p$,

276: $q=p^f$, and $a\not\equiv0\pmod{d}$ then

277: $$\ord_{\wp}G_q(a)=(p-1)\sum_{j=0}^{f-1}

278: \left\langle\frac{p^ja}{d}\right\rangle$$

279: where $\langle x\rangle$ is the fractional part of $x$, i.e.,

280: $0\le\langle x\rangle<1$ and $x-\langle x\rangle\in\Z$.

281: 

282: 

283: \subsection{}\label{ss:JacobiSums}

284: Fix a positive integer $w$.  For each $q$ and each tuple of

285: non-trivial characters $\chi_0,\dots,\chi_{w+1}$ of $\Fqtimes$ such

286: that the product $\chi_0\cdots\chi_{w+1}$ is trivial, we define a

287: Jacobi sum

288: $$J_q(\chi_0,\dots,\chi_{w+1})=\frac1{q-1}

289: \sum_{\substack{x_0,\dots,x_{w+1}\in\Fpf^\times\\x_0+\cdots+x_{w+1}=0}}

290: \chi_0(x_0)\cdots\chi_{w+1}(x_{w+1})\in\Q(\mu_{q-1}).$$

291: 

292: It is well-known and elementary (see \cite{WeilNS}*{p.~501} for

293: example) that

294: $$J_q(\chi_0,\dots,\chi_{w+1})=\frac{(-1)^w}q\prod_{i=0}^{w+1}G_q(\chi_i).$$

295: In particular, the Jacobi sum is an algebraic integer with absolute

296: value $q^{w/2}$ in every complex embedding.

297: 

298: Let $A_{d,w}\subset(\Z/d\Z)^{w+2}$ be the set of tuples

299: $\a=(a_0,\dots,a_{w+1})$ such that $a_i\neq0$ for all $i$ and $\sum

300: a_i=0$.  If $\a\in A_{d,w}$ and $q\equiv1\pmod d$, we write $J_q(\a)$

301: for $J_q(t^{-a_0(q-1)/d},\dots,t^{-a_{w+1}(q-1)/d})$; clearly

302: $J_q(\a)\in\Q(\mu_d)$.  If $\wp$ is the prime of $\Q(\mu_{d})$ under

303: $\p$ and $q=p^f$, then

304: $$\ord_\wp J_q(\a)=\sum_{i=0}^{w+1}\sum_{j=0}^{f-1}

305: \left\langle\frac{p^ja_i}{d}\right\rangle-f.$$

306:  

307: We write $A'_{d,w}$ for those $\a\in A_{d,w}$ such that

308: $\gcd(d,a_0,\dots,a_{w+1})=1$.  Note that if $\a\in A_{d,w}$ and if

309: $e=\gcd(d,a_0,\dots,a_{w+1})$, $d'=d/e$ and

310: $\a'=(a_0/e,\dots,a_{w+1}/e)\in A'_{d',w}$ then for any $q\equiv1\pmod

311: d$ we have $J_q(\a)=J_q(\a')$.

312:  

313:  

314: Many of our results on ranks will be based on part (2) of the

315: following theorem about the distribution of Gauss and Jacobi sums.

316: Roughly speaking, it says that sums involving characters of large

317: order must either have large degree over $\Q$ or have valuation

318: bounded away from 0.

319: 

320: 

321: \begin{thm}\label{thm:GaussJacobiBounds}

322: \hfill\break

323: \begin{enumerate}

324: \item \vskip-12pt Fix a real number $\epsilon>0$ and a positive

325:   integer $n$.  There exists a constant $C_{\epsilon,n}$ depending

326:   only on $\epsilon$ and $n$ such that if $d>C_{\epsilon,n}$,

327:   $q=p^f\equiv1\pmod d$, $a\in(\Z/d\Z)^\times$, and the degree of

328:   $G_q(a)$ over $\Q(\mu_p)$ is $\le n$, then

329: $$\left|\frac{\ord_\wp G_q(a)}{(p-1)f}-\frac12\right|<\epsilon.$$ 

330: Here $\wp$ is the prime of $\Q(\mu_{pd})$ under $\p$.  Note that

331:   $\ord_\wp(q)=(p-1)f$.

332: \item Fix a positive integer $n$.  There exist constants $C_n$ and

333:   $\epsilon_n>0$ depending only on $n$ such that if $d>C_n$,

334:   $q=p^f\equiv1\pmod d$, $w\ge1$, $\a\in A'_{d,w}$, and the degree of

335:   $J_q(\a)$ over $\Q$ is $\le n$, then

336: $$\frac{\ord_\wp J_q(\a)}{f}>\epsilon_n.$$

337: Here $\wp$ is the prime of $\Q(\mu_d)$ under $\p$.  Note that

338: $\ord_\wp(q)=f$.

339: \end{enumerate}

340: \end{thm}

341:  

342: \begin{rems}

343: \hfill\break

344: \begin{enumerate}

345: \item \vskip-12pt The constants appearing in the theorem are {\it

346:     independent of\/} $p$ and effectively computable.

347: \item In part (2) of the theorem, we may replace ``the degree of

348:   $J_q(\a)$ over $\Q$ is $\le n$'' with ``the degree of the largest

349:   subfield of $\Q(J_q(\a))$ in which $p$ splits completely is $\le

350:   n$'' and similarly in part (1).  I do not know whether this has any

351:   applications to geometry.

352: \end{enumerate}

353: \end{rems}

354: 

355:  

356: The theorem is a consequence of Stickelberger's theorem and the

357: following simple estimate.

358:  

359: \begin{prop}\label{prop:BasicEstimate}

360:   Fix a real number $\epsilon>0$ and a positive integer $n$.  There

361:   exists a constant $C_{\epsilon,n}$ depending only on $\epsilon$ and

362:   $n$ such that if $d>C_{\epsilon,n}$ and $H\subset G=(\Z/d\Z)^\times$

363:   is a subgroup of index $\le n$, then for all $a\in G$,

364: $$\left|\frac{1}{|H|}\sum_{t\in H}\left\langle\frac{ta}{d}

365: \right\rangle-\frac12\right|< \epsilon.$$

366: \end{prop}

367:  

368: \begin{proof}

369: We have

370: \begin{align*}

371:  A:=\frac{1}{|H|}\sum_{t\in H}\left\langle\frac{ta}{d}\right\rangle

372:  &=\frac{1}{|H|}\sum_{\substack{s=1\\(s,d)=1}}^{d-1}\frac sd\frac1{[G:H]}

373:  \sum_{\chi\in\widehat{G/H}}\chi(sa^{-1})\\

374:  &=\frac12 +\frac1{d\phi(d)}\sum_{1\neq\chi\in\widehat{G/H}}\chi(a^{-1})

375:  \sum_{\substack{s=1\\(s,d)=1}}^{d-1}\chi(s)s

376: \end{align*}

377: where $|H|$ denotes the order of $H$, $\widehat{G/H}$ denotes the

378: group of characters of $G/H$ (which we view as characters of $G$

379: trivial on $H$), and $\phi(d)=|G|$ is Euler's function.  Partial

380: summation and the Polya-Vinogradov inequality \cite{Davenport}*{\S23}

381: show that there is an absolute constant $C$ such that the inner sum

382: above is $<Cd^{3/2}\log d$ and so the quantity $A$ to be estimated

383: satisfies

384:  $$\left|A-\frac12\right|\le\frac{Cnd^{1/2}\log d}{\phi(d)}.$$

385:  Well-known estimates for $\phi(d)$ \cite{HardyWright}*{Thm~327} say

386:  that for all $\delta>0$, $\phi(d)/d^{1-\delta}\to\infty$ as

387:  $d\to\infty$ so there is a constant $C_{\epsilon,n}$ depending only

388:  on $n$ and $\epsilon $ such that

389:  $$\frac{Cnd^{1/2}\log d}{\phi(d)}< \epsilon$$

390: whenever $d>C_{\epsilon,n}$.  This completes the proof of the proposition.

391: \end{proof}

392:  

393: \begin{cor}\label{cor:alld}

394:   Given $n$ there exists a constant $\delta_n>0$ depending only on $n$

395:   such that for any $d\ge2$, any $0\neq a\in\Z/d\Z$, and any subgroup

396:   $H\subset G=(\Z/d\Z)^\times$ of index $\le n$,

397: $$\frac{1}{|H|}\sum_{t\in H}\left\langle\frac{ta}{d}\right\rangle>\delta_n$$

398:  \end{cor}

399:  

400: \begin{proof}

401:   For $0\neq a\in(\Z/d\Z)$, set $e=\gcd(a,d)$, $d'=d/e$,

402:   $G'=(\Z/d'\Z)^\times$, $a'=a/e$, and $H'=\im(H\to G')$.  Then the

403:   index of $H'$ in $G'$ is $\le n$ and we have

404:  $$A:=\frac{1}{|H|}\sum_{t\in H}\left\langle\frac{ta}{d}\right\rangle=

405:  \frac{1}{|H'|}\sum_{t\in H'}\left\langle\frac{ta'}{d'}\right\rangle$$

406: and so we may assume that $\gcd(a,d)=1$, i.e., that $a\in G$.

407:  

408: Given $n$, let $C_{1/4,n}$ be the constant furnished by the

409: proposition for $n$ and $\epsilon =1/4$.  If $d>C_{1/4,n}$ then by the

410: proposition, $A>1/4$.  On the other hand, there are only finitely many

411: $d\le C_{1/4,n}$ and for each $d$, only finitely many subgroups

412: $H\subset(\Z/d\Z)$ of index $\le n$.  Since $A>0$ for each of these

413: finitely many possibilities, there is a $\delta_n>0$ such that

414: $A>\delta_n$ for all $d$ and $a$.

415:  \end{proof}

416:  

417:  

418: \subsection{Proof of Theorem~\ref{thm:GaussJacobiBounds} (1)}

419: Given $\epsilon$ and $n$, suppose that $d$, $q=p^f\equiv1\pmod d$, and

420: $a\in G=(\Z/d\Z)^\times\cong\gal(\Q(\mu_{pd})/\Q(\mu_p))$, are such

421: that $G_q(a)\in\Q(\mu_{pd})$ has degree $\le n$ over $\Q(\mu_p)$.  Let

422: $H\subset G$ be the subgroup of $G$ fixing $\Q(\mu_p,G_q(a))$, so that

423: $H$ has index $\le n$ in $G$.  If $\wp$ is the prime of $\Q(\mu_{pd})$

424: under $\p$, then for every $t\in H$, we have

425: $\ord_{\wp^t}(G_q(a))=\ord_{\wp}(G_q(a))$.  Therefore,

426:  \begin{align*}

427:  \frac{\ord_{\wp}G_q(a)}{(p-1)f}

428:  &=\frac1{|H|}\sum_{t\in H} \frac{\ord_{\wp^t}G_q(a)}{(p-1)f}\\

429:  &=\frac1{|H|f}\sum_{t\in H}\sum_{j=0}^{f-1}

430: \left\langle \frac{p^jta}{d}\right\rangle

431:  \end{align*}

432:  where the second equality comes from Stickelberger's theorem.  Let

433:  $P$ be the subgroup of $(\Z/d\Z)^\times$ generated by $p$ and $HP$

434:  the subgroup generated by $H$ and $P$.  The last displayed sum is

435:  then equal to

436:  $$\frac{1}{|HP|}\sum_{t\in

437:   HP}\left\langle\frac{ta}{d}\right\rangle.$$ 

438: Since $H$ has index $\le n$ in $G$, the same is true of $HP$ and so

439: Proposition~\ref{prop:BasicEstimate} shows that if $d>C_{\epsilon,n}$

440:  % and $G_q(a)$ has degree $\le n$ 

441: then

442: $$\left|\frac{\ord_{\wp}G_q(a)}{(p-1)f}-\frac12\right|<\epsilon$$

443: as was to be shown.

444: \qed

445:  

446:   

447: \subsection{Proof of Theorem~\ref{thm:GaussJacobiBounds} (2)}

448: Given $\a\in A'_{d,w}$, set $d_i=d/\gcd(d,a_i)$.  The following lemma

449: tells us that if $d$ is large then at least two of the $d_i$ are also

450: large.

451:  

452: \begin{subsublemma} 

453: With notation as above, there exists an absolute constant $C$ such

454: that at least two of the $d_i$ are $\ge C\log d$.

455: \end{subsublemma}

456:  

457: \begin{proof}

458: If $\ell$ divides $d$ then from the definitions, there are at least

459: two $i$'s such that $\ell$ does not divide $a_i$.  Therefore the

460: largest prime power dividing $d$ also divides at least two of the

461: $d_i$.

462:  

463: To finish we note that Chebyschev's theorem

464: \cite{HardyWright}*{Thm.~7} implies that the the largest prime power

465: dividing $d$ is $\ge C'\log d$ for some absolute constant $C'$.

466: Indeed, let $M$ be a positive number, let $p_1,\dots,p_{\pi(M)}$ be

467: the primes less than $M$, and let $p_i^{e_i}$ be the largest power of

468: $p_i$ less than $M$.  If $N=\prod_{i=1}^{\pi(M)}p_i^{e_i}$ then

469:  $$\log N=\sum_{i=1}^{\pi(M)}e_i\log p_i\le\pi(M)\log M\le C'M$$ 

470: by Chebyschev.  This shows that if $N$ is a product of prime powers

471: less than $M$, then $N\le e^{C'M}$.  Therefore the largest prime power

472: dividing $N$ is at least $C\log N$ where $C=1/C'$.

473: \end{proof}

474: 

475: Now fix $n$ and consider those $q\equiv1\pmod d$ and $\a$ such that

476: $J_q(\a)$ has degree $\le n$ over $\Q$.  Let $H\subset

477: G=(\Z/d\Z)^\times\cong\gal(\Q(\mu_d)/\Q)$ be the subgroup fixing

478: $\Q(J_q(\a))$ so that $H$ has index $\le n$ in $G$.  Then we have

479: \begin{align*}

480: A(q,\a):=\frac{\ord_{\wp}J_q(\a)}{f}

481: &=\frac{1}{|H|}\sum_{t\in H}\frac{\ord_{\wp^t}(J_q(\a))}{f}\\

482: &=\frac{1}{|H|}\sum_{t\in H}\frac 1f \left(\sum_{i=0}^{w+1}

483: \sum_{j=0}^{f-1}\left\langle\frac{tp^ja_i}{d}\right\rangle-f\right)\\

484: &=\left(\sum_{i=0}^{w+1}\frac{1}{|HP|}

485: \sum_{t\in HP}\left\langle\frac{ta_i}{d}\right\rangle\right)-1

486: \end{align*}

487: where as before $P$ is the subgroup of $(\Z/d\Z)^\times$ generated by

488: $p$ and $HP$ is the subgroup generated by $H$ and $P$.  Reindexing

489: $\a$, we may assume that $d_0$ and $d_1$ are $\ge C\log d$.  Since $H$

490: has index $\le n$, so does $HP$ and so we get bounds on the inner sums

491: in the last displayed equation.  More precisely, by

492: Corollary~\ref{cor:alld}, the inner sum is $>\delta_n$ for

493: $i=2,\dots,w+1$, and by Proposition~\ref{prop:BasicEstimate}, if $d_0$

494: and $d_1$ are sufficiently large (so that $C\log d>C_{\epsilon,n}$),

495: the $i=0$ and $i=1$ terms are $>1/2-\epsilon$.  Applying this with

496: $\epsilon=\delta_n/4\le w\delta_n/4$, we see that for sufficiently

497: large $d$, $A(q,\a)\ge(w-1/2)\delta_n\ge\delta_n/2$.  This completes

498: the proof of part (2) of the theorem.  \qed

499: 

500: 

501: \section{Fermat Jacobians}

502: 

503: \subsection{}%Fermat curves and Jacobians

504: Let $k$ be an arbitrary field with separable closure $\overline k$.

505: For each positive integer $d$ not divisible by the characteristic of

506: $k$ we consider the Fermat curve $F_d$ of degree $d$ over $k$ (the

507: zero locus of $\sum_{i=0}^2x_i^d$ in $\P^2$) and its Jacobian $J_d$.

508: If $A$ is an abelian variety over $k$, we say that ``$A$ appears in

509: $J_d$'' if there is a homomorphism of abelian varieties $A\to J_d$

510: with finite kernel.  We say ``$A$ appears in $J_d$ with multiplicity

511: $m$'' if $m$ is the largest integer such that $A^m$ appears in $J_d$.

512: The multiplicity with which $A$ appears in $J_d$ obviously depends

513: only on the $k$-isogeny class of $A$.

514: 

515: The following two theorems are the main results of this section.

516: 

517: \begin{thm}\label{thm:Jd0}

518:   Suppose that $k$ is a field of characteristic zero.  Then for every

519:   positive integer $g$, only finitely many $k$-isogeny classes of

520:   abelian varieties of dimension $\le g$ appear in $J_d$ as $d$ varies

521:   through all positive integers.  If $A$ is an abelian variety over

522:   $k$, then the multiplicity with which $A$ appears in $J_d$ is

523:   bounded by a constant depending only on the dimension of $A$.

524: \end{thm}

525: 

526: If $k$ has characteristic $p$ and $A$ is an abelian variety over $k$,

527: the $p$-rank of $A$ is by definition the dimension over $\Fp$ of the

528: group of $\overline k$-rational $p$-torsion points on $A$.  It is

529: known that the $p$-rank lies in the interval $[0,\dim A]$ and that it

530: is invariant under isogeny.

531: 

532: \begin{thm}\label{thm:Jdp}

533:   Suppose that $k$ is a field of characteristic $p>0$.  Then for every

534:   positive integer $g$, only finitely many $k$-isogeny classes of

535:   abelian varieties with positive $p$-rank and dimension $\le g$

536:   appear in $J_d$ as $d$ varies through all positive integers prime to

537:   $p$.  If $A$ is an abelian variety over $k$ with positive $p$-rank,

538:   then the multiplicity with which $A$ appears in $J_d$ is bounded by

539:   a constant depending only on the dimension of $A$.

540: \end{thm}

541: 

542: \begin{rems}

543: \hfill\break

544: \begin{enumerate}

545: \item \vskip-12pt We repeat that the constants in the theorems depend

546:   only on the dimension $g$.  In particular, they are independent of

547:   the characteristic of $k$.  As will be clear from the proof, they

548:   are also effectively computable.

549: \item Theorem~\ref{thm:Jd0} is already known in a more precise

550:   quantitative form by results of Aoki \cite{Aoki}, building on work

551:   of Koblitz, Rohrlich, and Shioda.  Theorem~\ref{thm:Jdp} may be

552:   known to experts but to my knowledge is not in the literature.  We

553:   will give a very simple proof of Theorem~\ref{thm:Jdp} for $k$

554:   finite and use this to deduce the general case and

555:   Theorem~\ref{thm:Jd0}.

556: \item It is proven in \cite{TS}, and by a different method in

557:   \cite{UlmerR2}, that over a field of characteristic $p$, the

558:   multiplicity with which a supersingular elliptic curve appears in

559:   $J_d$ is unbounded as $d$ varies.  Thus the last part of

560:   Theorem~\ref{thm:Jdp} would be false without the hypothesis of

561:   positive $p$-rank.  It is not clear what to expect for abelian

562:   varieties with $p$-rank zero which are not $\overline k$-isogenous

563:   to a product of supersingular elliptic curves.

564: \end{enumerate}

565: \end{rems}

566: 

567: The proofs of the theorems will be given in rest of this section.

568: 

569: 

570: \subsection{}\label{ss:oldnew}%old and new

571: If $d'<d$ is a divisor of $d$, then there is a canonical surjective

572: morphism $F_d\to F_{d'}$ ($x_i\mapsto x_i^{d/d'}$) which (because

573: $F_d\to F_{d'}$ is totally ramified at some place) induces an

574: injection of Jacobians $J_{d'}\into J_d$.  We define the {\it old

575:   part\/} $J_d^\text{old}$ to be the abelian subvariety of $J_d$

576: generated by the images of the morphisms $J_{d'}\into J_d$ as $d'$

577: varies through proper divisors of $d$ and we define the {\it new

578:   part\/} $J_d^\text{new}$ of $J_d$ to be the abelian variety over $k$

579: (well-defined only up to $k$-isogeny) such that $J_d$ is isogenous to

580: $J_d^\text{new}\times J_d^\text{old}$.  It is not hard to check, for

581: example by using the zeta function calculation mentioned in

582: \ref{ss:FermatZetas} below, that $J_d$ is isogenous to

583: $\prod_{d'|d}J_{d'}^\text{new}$.

584: 

585: Theorem~\ref{thm:Jd0} therefore follows from the statement that there

586: is a constant $C_g$ depending only on $g$ such that no abelian variety

587: $A$ of dimension $\le g$ appears in $J_d^\text{new}$ for any $d>C_g$.

588: Theorem~\ref{thm:Jdp} follows from the same statement with the

589: additional hypotheses that $A$ has positive $p$-rank and $d$ is not

590: divisible by $p$.

591: 

592: 

593: \subsection{}\label{ss:algfields}%reduction to algebraic fields

594: Given a field $k$, let $\F$ be its prime field and $k_0$ be the

595: algebraic closure of $\F$ in $k$.  Then $k_0$ is a perfect field and

596: so the extension $k/k_0$ is regular.  The Fermat Jacobian $J_d$ and

597: its new part $J_d^\text{new}$ are defined over $\F$ and so if $A$ is

598: an abelian variety over $k$ which appears in

599: $J_d^\text{new}\times_{\F} k$ then there is an abelian variety $A_0$

600: defined over $k_0$ which appears in $J_d^\text{new}\times_{\F} k_0$

601: and with $A_0\times_{k_0} k\cong A$.  (This is an old result of Chow

602: which has been given a detailed modern treatment by Conrad, see

603: \cite{Conrad}*{3.21}.)  Moreover, the abelian variety $A_0$ and the

604: morphism $A_0\to J_d^\text{new}$ are both defined over some finite

605: extension of $\F$.  Thus it will suffice to prove the existence of the

606: constants $C_g$ mentioned at the end of Subsection~\ref{ss:oldnew}

607: (depending only on $g$, not on $k$) for the cases when $k$ is a number

608: field or a finite field.

609: 

610: 

611: \subsection{}\label{ss:HondaTate}%Weil numbers and Honda-Tate

612: Let $k$ be $\Fq$, the subfield of $\Fpbar=\O_{\Qbar}/\p$ with $q$

613: elements.  A {\it Weil $q$-integer of weight 1\/} is an algebraic

614: integer $\alpha$ whose absolute value in every complex embedding is

615: $q^{1/2}$.  For the rest of this section we will call these simply

616: {\it Weil numbers\/}.

617: 

618: Honda-Tate theory \cite{TateHT} says that $\Fq$-isogeny classes of

619: $\Fq$-simple abelian varieties are in bijection with $\gal(\Qbar/\Q)$

620: orbits of Weil numbers.  If $A$ corresponds to $\alpha$, then

621: $E=\en_{\Fq}(A)\tensor\Q$ is a central simple algebra over

622: $\Q(\alpha)$ whose invariants in the Brauer group of $\Q(\alpha)$ can

623: be calculated in terms of the decomposition of $p$ in $\Q(\alpha)$.

624: The dimension of $A$ is $(1/2)[E:\Q(\alpha)]^{1/2}[\Q(\alpha):\Q]$ and

625: the eigenvalues of Frobenius on $H^1(A\times\Fqbar,\Ql)$ are the

626: conjugates of $\alpha$, each appearing with multiplicity

627: $[E:\Q(\alpha)]^{1/2}$.  The $p$-rank of $A$ is equal to the number of

628: eigenvalues of Frobenius which are units at $\p$ and so $A$ has

629: positive $p$-rank if and only if some conjugate of $\alpha$ is a unit

630: at $\p$.

631: 

632: If $C$ is a curve of genus $g$ over $\Fq$ and the $Z$-function of $C$

633: is

634: $$\frac{\prod_{i=1}^{2g}(1-\alpha_iT)}{(1-T)(1-qT)}$$ 

635: then the Weil numbers of the $\Fq$-simple factors of the Jacobian $J$

636: of $C$ are precisely the $\alpha_i$.  The multiplicity of $\alpha_i$

637: in the numerator is the multiplicity of the corresponding $A$ in $J$

638: up to $\Fq$-isogeny times $[E_A:\Q(\alpha)]^{1/2}$.

639: 

640: 

641: \subsection{}\label{ss:FermatZetas}%Fermat zetas

642: Given a positive integer $d$ and a prime power $q$ such that $q\equiv

643: 1\pmod{d}$ we consider the Fermat curve $F_d$ over $\Fq$.  By a

644: theorem of Weil \cite{WeilNS}, the $Z$-function of $F_d$ over $\Fq$ is

645: $$\frac{\prod_{\a\in A_{d,1}}(1-J_q(\a)T)}{(1-T)(1-qT)}$$

646: where $A_{d,1}$ was defined in Subsection~\ref{ss:JacobiSums}.

647: 

648: It is clear from Weil's computation of the $Z$-function that the Weil

649: numbers of $J_d^\text{new}$ are precisely the $J_q(\a)$ as $\a$ runs

650: through

651: $$A'_{d,1}=\{\a=(a_0,a_1,a_2)\in A_{d,1}|\gcd(d,a_0,a_1,a_2)=1\}.$$

652: 

653: \subsection{The case of finite fields}

654: We assume $k=\Fq$ and that $A$ is an abelian variety over $k$ which

655: has positive $p$-rank and dimension $\le g$ and appears in

656: $J_d^\text{new}$.  In this case, the Weil numbers of $A$ are among the

657: Weil numbers of $J_d^\text{new}$.  Extending $k$ if necessary, we may

658: assume that $d|(q-1)$ and so the Weil numbers of $J_d^\text{new}$ are

659: the Jacobi sums $J_q(\a)$ where $\a\in A'_{d,1}$.  By the results

660: recalled in Subsections~\ref{ss:HondaTate} and \ref{ss:FermatZetas} it

661: follows that some $J_q(\a)$ has degree $\le 2g$ over $\Q$ and is a

662: unit at the prime $\p$.  This implies that $d$ is $\le C_{2g}$ where

663: $C_{2g}$ is the constant appearing in

664: Theorem~\ref{thm:GaussJacobiBounds}(2) for $n=2g$.  Therefore no

665: abelian variety of positive $p$-rank and dimension $\le g$ appears in

666: $J_d^\text{new}$ for large $d$ and this establishes

667: Theorem~\ref{thm:Jdp} for finite fields.  The argument in

668: Subsection~\ref{ss:algfields} shows that the theorem also holds for

669: arbitrary fields of positive characteristic.

670: 

671: 

672: \subsection{The case of number fields}

673: Suppose that $A$ is an abelian variety of dimension $\le g$ defined

674: over a number field $k$.  Suppose that $d$ is larger than the constant

675: $C(g)=C_{2g}$ appearing in Theorem~\ref{thm:Jdp} and that $A$ appears

676: in $J_d^\text{new}$.  Then for every prime $\wp$ of $k$ where $A$ has

677: good reduction, by Theorem~\ref{thm:Jdp} the reduction $A\times

678: \F_\wp$ has $p$-rank 0.  This would violate the following result,

679: which appears in \cite{Ogus}*{2.7.1}:

680: 

681: \begin{subsublemma} \textup{(}Katz\textup{)} 

682: If $A$ is an abelian variety over a number field $k$, then for

683: infinitely many primes of $k$, the reduction of $A$ has positive

684: $p$-rank.

685: \end{subsublemma}

686: 

687: For the convenience of the reader, we sketch the proof of the lemma.

688: Choose a prime $\ell$ larger than $2g$.  Let $L$ be a finite extension

689: of $k$ such that $\gal(\Qbar/L)$ acts trivially on the $\ell$-torsion

690: of $A$.  If $\wp$ is a prime of $L$ over the rational prime $p$ where

691: the reduction of $A$ has $p$-rank zero, then the trace of the

692: Frobenius at $\wp$ on $H^1(A\times\overline k,\Ql)$ is an integer

693: $\equiv0\pmod{p}$ and $\le2g(\N\wp)^{1/2}$.  If $\wp$ has absolute

694: degree 1 over $\Q$ (i.e., $\N\wp=p$), and $\sqrt{p}>2g$ then we see

695: that the trace must be zero.  On the other hand, since $\gal(\Qbar/L)$

696: acts trivially on $\ell$-torsion, the trace must be

697: $\equiv2g\pmod\ell$.  Since $\ell>2g$ this is impossible.  The

698: conclusion is that the reduction of $A$ at a prime of absolute degree

699: one over a large $p$ must have positive $p$-rank.  Such primes have

700: density one in $L$ and the primes under them in $k$ are an infinite

701: set satisfying the conclusion of the lemma.

702: 

703: We note that a stronger version of this result for abelian varieties

704: over $\Q$ is proven in \cite{BayerGonzalez}*{Prop.~5.1}.

705: 

706: The lemma completes the proof of Theorem~\ref{thm:Jd0} for number

707: fields and, as explained in Subsection~\ref{ss:algfields}, therefore

708: also for arbitrary fields of characteristic zero.

709: 

710: 

711: 

712: \section{Isotrivial abelian varieties with bounded ranks in 

713: $\hat\Z$ or $\Zhatp$-towers}

714: 

715: \subsection{}

716: In the rest of the paper we will give examples of abelian varieties

717: with bounded ranks in towers of function fields over various fields

718: $k$.  Before doing so, let us dispense with a trivial situation: if

719: $A$ is an abelian variety over $k(t)$ with good reduction away from

720: $0$ and $\infty$ and at worst tame ramification at $0$ and $\infty$,

721: then for any $d$ prime to the characteristic of $k$, the degree of the

722: conductor of $A$ over $k(t^{1/d})$ is bounded independently of $d$.

723: Geometric rank bounds then show that the rank of $A$ over $k(t^{1/d})$

724: is also bounded independently of $d$.  Therefore it is only

725: interesting to consider situations where the degree of the conductor

726: grows in the tower under consideration.  All our examples below are of

727: this type.

728: 

729: 

730: 

731: \subsection{}

732: We review some well-known facts about constant and isotrivial abelian

733: varieties.  Let $k$ be any field, let $L$ be the function field of a

734: geometrically irreducible curve $\Curve$ smooth and proper over $\spec

735: k$, and let $J$ be the Jacobian of $\Curve$.  Let $A_0$ be an abelian

736: variety over $k$ and let $A=A_0\times_kL$.  Then it is clear that

737: $A(L)$, the group of $L$-rational points of $A$, is canonically

738: isomorphic to $\mor_k(\Curve,A_0)$, the group of $k$-scheme morphisms

739: from $\Curve$ to $A_0$.  Moreover, we have an exact sequence

740: \begin{equation*}

741: 0\to A_0(k)\to\mor_k(\Curve,A_0)\to\Hom_{k\text{-av}}(J,A_0)

742: \end{equation*}

743: where a $k$ point of $A_0$ is sent to the constant map with that value

744: and a morphism from $\Curve$ to $A_0$ is sent to the homomorphism of

745: abelian varieties induced by Albanese functoriality.  If $\Curve$ has

746: a $k$-rational divisor of degree 1 (for example if $k$ is finite) then

747: the last map above is surjective.  If $k$ is finitely generated over

748: its prime field, then by the Lang-N\'eron theorem, $A_0(k)$ is

749: finitely generated.  (See \cite{Conrad} for a modern treatment of

750: the Lang-N\'eron theorem.)  For any $k$, $\Hom_{k\text{-av}}(J,A_0)$ is

751: finitely generated and torsion free.  If $A_0$ is $k$-simple, then the

752: rank of $\Hom_{k\text{-av}}(J,A_0)$ is equal to the rank of the

753: endomorphism ring of $A_0$ times the multiplicity with which $A_0$

754: appears in $J$ up to $k$-isogeny.

755: 

756: 

757: \subsection{}

758: Continuing with the notation of the last subsection, suppose that

759: $\Curve$ is hyperelliptic, i.e., we are given a degree 2 morphism

760: $\Curve\to\P^1$.  Let $A'$ be the twist of $A=A_0\times_k

761: k(t)$ by the quadratic extension $L/k(t)$.  Since there are no

762: non-constant morphisms from $\P^1$ to an abelian variety, we have

763: $A(k(t))=A_0(k)$.  Since

764: \begin{equation*}

765: A(L)\tensor\Q\cong \left(A(k(t))\tensor\Q\right) \bigoplus 

766: \left(A'(k(t))\tensor\Q\right)

767: \end{equation*}

768: we conclude that $A'(k(t))$ has finite rank, bounded above by

769: \begin{equation}\label{eq:isotrank}

770: \dim_\Q A'(k(t))\tensor\Q=\dim_\Q \Hom_{k\text{-av}}(J,A_0)\tensor\Q

771: =\rk_\Z \Hom_{k\text{-av}}(J,A_0)

772: \end{equation}

773: with equality when $\Curve$ has a $k$-rational divisor of degree 1.

774: 

775: 

776: \subsection{}

777: We can now apply the rank formula above and our results about Fermat

778: Jacobians to give examples of bounded ranks in towers.  Let $K_1=k(t)$

779: and for every positive integer $d$ not divisible by the characteristic

780: of $k$, let $K_d=k(t^{1/d})$.  If the characteristic of $k$ is not 2,

781: let $L_1=k(u)$ with $u^2=t-1$; if the characteristic of $k$ is 2, let

782: $L_1=k(u)$ with $u^2+u=t$.  For all $d$ prime to the characteristic of

783: $k$, let $L_d=L_1K_d=k(t^{1/d},u)$.  Note that $L_d$ is the function

784: field of a hyperelliptic curve $C_d$ over $k$. Using ideas analogous

785: to \cite{UlmerR2}*{\S6}, one checks easily that there is a totally

786: ramified, surjective morphism from a Fermat curve $F_{n}\to\Curve_d$;

787: here $n=2d$ if the characteristic of $k$ is not 2 and $n=d$ if the

788: characteristic of $k$ is 2.  It follows that the Jacobian of $C_d$ is

789: an isogeny factor of $J_{n}$.  Applying the rank formula

790: \ref{eq:isotrank} and Theorems~\ref{thm:Jd0} and \ref{thm:Jdp} we have

791: the following.

792: 

793: 

794: \begin{thm}\label{thm:isotrivial}  

795: Let $k$ be a field and $A_0$ an abelian variety over $k$.  If the 

796: characteristic of $k$ is $p>0$, assume that $A_0$ is isogenous to a 

797: product of $k$-simple abelian varieties each with positive $p$-rank.  

798: Let $A=A\times_k k(t)$ and let $A'$ be the twist of $A$ by the quadratic 

799: extension $k(u)/k(t)$ where $u$ satisfies $u^2=t-1$ if the characteristic 

800: of $k$ is not 2 and $u^2+u=t$ if the characteristic of $k$ is 2.  Then the 

801: rank of the Mordell-Weil group $A'(k(t^{1/d}))$ is bounded as $d$ varies 

802: through all positive integers relatively prime to the characteristic of $k$.

803: \end{thm}

804: 

805: 

806: \section{Non-isotrivial elliptic curves with bounded ranks in $\Z_\ell$-towers}

807: 

808: \subsection{}

809: For examples of non-isotrivial elliptic curves with bounded ranks in

810: $\Z_\ell$ extensions, we consider the curve $E$ discussed in

811: \cite{UlmerR1} with affine equation

812: $$y^2+xy=x^3-t$$

813: over $\Fp(t)$.  

814: 

815: \bigskip

816: \begin{thm}~\label{thm:nonisol} 

817: Given $p$ let $S$ be the set of primes $\ell>3$ such that

818: $p\equiv1\pmod\ell$.  If $d$ is a product of powers of primes from

819: $S$, then the rank of $E(\Fpbar(t^{1/d}))$ is zero.

820: \end{thm}

821: 

822: The proof of the theorem will will be given in the rest of this section.

823: 

824: \subsection{}

825: We use the notation of Subsection~\ref{ss:JacobiSums} on Jacobi sums.

826: Given $p$, $d$ prime to $p$, and $\a=(a_0,\dots,a_3)\in A_{d,2}$, we

827: say that $\a$ is ``supersingular'' (some authors would say ``pure'')

828: if for one (and thus every) $q=p^f\equiv1\pmod{d}$ and all

829: $s\in(\Z/d\Z)^\times$ we have

830: $$\sum_{i=0}^3\sum_{j=0}^{f-1}\left\langle\frac{sp^ja_i}{d}\right\rangle=2f.$$

831: If $\a$ is supersingular, then for every prime $\wp$ of $\Q(\mu_d)$

832: over $p$, the valuation $\ord_{\wp}J_q(\a)$ is $f$ and this implies

833: that $J_q(\a)$ is a root of unity times $q$; this is the motivation

834: for the terminology ``supersingular''.

835: 

836: \subsection{}

837: By \cite{UlmerR1}*{6.4 and 7.7}, if $(d,6p)=1$, then the rank of

838: $E(\Fpbar(t^{1/d}))$ is equal to the number of elements

839: $t\in\Z/d\Z\setminus\{0\}$ such that $\a=(t,-6t,2t,3t)$ is

840: supersingular.  We are going to show that for suitable $d$ there are

841: no supersingular $\a$ of this form by using a descending induction

842: based on the following elementary identity.  Suppose that

843: $a\in\Z/d\Z$, $\ell$ is a prime such that $\ell^2|d$ and $\ell\nodiv

844: a$.  Let $H$ be the cyclic subgroup of $(\Z/d\Z)^\times$ generated by

845: $1+d/\ell$.  Then we have

846: $$\sum_{s\in H}\left\langle\frac{sa}{d}\right\rangle

847: =\left\langle\frac{a}{d/\ell}\right\rangle+\frac{\ell-1}{2}.$$ 

848: 

849: It follows that if $\a=(a_0,\dots,a_3)\in A_{d,2}$, $\ell^2|d$,

850: $\ell\nodiv a_i$ for all $i$, and if $\a$ is supersingular, then its

851: image in $A_{d/\ell}$ is also supersingular.  Indeed, we have

852: $$2f\ell=\sum_{s\in H}\sum_{i=0}^3

853: \sum_{j=0}^{f-1}\left\langle\frac{sp^ja_i}{d}\right\rangle

854: =\sum_{i=0}^3\sum_{j=0}^{f-1}

855: \left\langle\frac{p^ja_i}{d/\ell}\right\rangle+2f(\ell-1)$$

856: and similarly if $a$ is replaced by $ta$ with $t\in(\Z/d\Z)^\times$.  

857: 

858: 

859: \subsection{}

860: We can now prove the theorem.  Suppose given $p$ and $d$ which is a

861: product of primes in $S$.  If the rank of $E(\Fpbar(t^{1/d}))$ were

862: positive, then we would have a $t\in\Z/d\Z$ such that

863: $\a=(t,-6t,2t,3t)$ is supersingular.  Without loss of generality we

864: may assume that $t\in(\Z/d\Z)^\times$ and then that $\a=(1,-6,2,3)$.

865: Applying the observation of the previous subsection repeatedly, we may

866: ``reduce the level'' and find a $d'$ which is a product of distinct

867: primes from $S$ such that $(1,-6,2,3)\in A_{d',2}$ is supersingular.

868: But for such a $d'$ we have $f=1$, i.e., $p\equiv1\pmod{d'}$ and with

869: this one easily checks that

870: $$\sum_{i=0}^3\sum_{j=0}^{f-1}\left\langle\frac{p^ja'_i}{d'}\right\rangle=

871: \sum_{i=0}^3\left\langle\frac{a'_i}{d'}\right\rangle=1\neq

872: 2f$$ 

873: and so we arrive at a contradiction to the assumption the

874: $E(\Fpbar(t^{1/d}))$ has positive rank.  This completes the proof of

875: the Theorem.

876: 

877: \subsection{}

878: The theorem shows that for any prime $p$ such that $p-1$ is not a

879: power of $2$ times a power of $3$, there is an elliptic curve over

880: $\Fp(t)$ with bounded rank in a $\Zl$ tower $\Fpbar(t^{1/\ell^n})$ for

881: suitable $\ell$.  We will prove a stronger result for certain small

882: $p$ not of this type (namely $p=2, 3, 5, 7$) in the next section and

883: so it seems likely that this kind of statement holds for all $p$.

884: 

885: In the same direction, it seems quite likely that a more refined

886: analysis would show that given $p$, and for $E$ as above, the rank of

887: $E(\Fpbar(t^{1/d}))$ is bounded as $d$ runs through all integers which 

888: are products of powers of primes $\ell$ such that  no power of $p$ is 

889: congruent to $-1$ modulo $\ell$.

890: 

891: Generalizing in another direction, a geometric analysis as in

892: \cite{UlmerR1}*{\S5} applied to the curves in \cite{UlmerR2}*{\S7}

893: might allow one to prove a version of Theorem~\ref{thm:nonisol} for

894: higher dimensional abelian varieties.

895: 

896: Finally, we note that it is not hard to deduce from

897: Theorem~\ref{thm:nonisol} that the curve defined over $\Q(t)$ by the

898: equation $y^2+xy=x^3-t$ has bounded rank over $\Qbar(t^{1/d})$ as $d$

899: ranges over all positive integers.  We omit the details since similar

900: results were shown by Shioda \cite{Shioda}*{Cor.~9} using closely

901: related techniques.

902: 

903: 

904: \section{Non-isotrivial elliptic curves with bounded ranks in $\Zhatp$-towers}

905: \subsection{}

906: We will use completely different techniques, unrelated to Fermat

907: varieties, to give a few examples of non-isotrivial elliptic curves

908: with bounded ranks in towers $\Fpbar(t^{1/d})$ as $d$ ranges over all

909: integers prime to $p$.

910:  

911: \begin{thm}\label{thm:nonisod}

912: If $p\in\{2,3,5,7,11\}$ then there exists an elliptic curve $E$ over

913: $\Fp(t)$ with $j(E)\not\in\Fp$ such that the rank of

914: $E(\Fpbar(t^{1/d}))$ is zero for all positive integers $d$ prime to

915: $p$.

916: \end{thm}

917:  

918: The proof of the theorem, which uses ideas from \cite{Ulmerpd}, will

919: be given in the rest of this section.

920:  

921: \subsection{}

922: Given an elliptic curve $E$ over $\Fp(t)$ with $j(E)\not\in\Fp$,

923: choose a non-zero invariant differential $\omega$ on $E$ and let

924: $\Delta=\Delta(E,\omega)$ and $A=A(E,\omega)$ be the discriminant and

925: Hasse invariant of $E$; the definition of the latter is reviewed in

926: \cite{Ulmerpd}*{\S2}.  Our assumptions imply that $\Delta$ and $A$ are

927: non-zero elements of $\Fp(t)$.

928:  

929: Consider the following conditions on $E$:

930: \begin{itemize}

931: \item $E$ has good or multiplicative reduction at $t=0$ and $t=\infty$.

932: \item At every finite non-zero place of $\Fp(t)$, $E$ obtains good

933:   reduction over a tamely ramified extension.

934: \item At every finite non-zero place $v$ of $\Fp(t)$, we have 

935: $$\frac{\ord_v(A)}{p-1}-\frac{\ord_v(\Delta)}{12} <\frac{1}{p}.$$

936: \end{itemize}

937: Note that the third condition is automatic at places where $E$ has

938: good ordinary reduction, in particular at places where $A$ and

939: $\Delta$ are units.  Note also that if $E$ satisfies these conditions

940: then it continues to satisfy them over the extensions $\Fq(t^{1/d})$

941: for any power $q$ of $p$ and any $d$ prime to $p$.

942: 

943: \subsection{}

944: It follows from \cite{Ulmerpd}*{Section~3 and the first sentence of

945: Section~6} that an elliptic curve over $\Fp(t)$ satisfying the

946: conditions of the previous subsection has rank 0 or 1 over any

947: extension $K=\Fq(t^{1/d})$. To see this, we consider the Frobenius and

948: Verschiebung isogenies

949: \begin{equation*}

950: \xymatrix{E\ar[r]^{Fr}&E^{(p)}\ar[r]^{V}&E}

951: \end{equation*}

952: whose composition is multiplication by $p$.  Section~3 of

953: \cite{Ulmerpd} computes the Selmer groups for $Fr$ and $V$ in terms of

954: the reduction types of $E$, $A$, and $\Delta$.  Under the conditions

955: of the previous subsection, the results are that $\sel(K,V)=0$ and

956: $\sel(K,Fr)$ is zero if $E$ has good reduction at 0 or $\infty$ and has

957: order $p$ if $E$ has multiplicative reduction at both 0 and

958: $\infty$.

959: 

960: We have an exact sequence

961: $$E^{(p)}(K)\to\sel(K,Fr)\to\sel(K,p)\to\sel(K,V)$$ 

962: and so the Selmer group for $p$ is either trivial or of order $p$.  In

963: the examples we give below, when $\sel(K,Fr)$ is non-trivial, there is

964: a point of order $p$ in $E^{(p)}(K)$ mapping to a generator of

965: $\sel(K,Fr)$ and so $\sel(K,p)=0$ and $E(K)$ has rank 0.

966: 

967: \subsection{}

968: We now give explicit examples of elliptic curves satisfying our conditions.  

969: 

970: Suppose $p=2$ and let $E$ be defined by

971: $$y^2+(t-1)xy+(t-1)^2y=x^3.$$ 

972: If $\omega=dx/((t-1)x+(t-1)^2)$, then $A=(t-1)$, $\Delta=t(t-1)^8$,

973: and $j=(t-1)^4/t$.  Standard methods show that $E$ has good, ordinary

974: reduction away from 0, 1, and $\infty$; that $E$ has multiplicative

975: reduction at 0 and $\infty$; and that at $t=1$, $E$ obtains good

976: reduction over an extension with ramification index 3 and the

977: inequality involving $A$ and $\Delta$ is satisfied.  The point

978: $(x,y)=((t-1)^2,(t-1)^3)$ on $E^{(2)}$ has order 2 and maps

979: non-trivially to $\sel(K,Fr)$ and so $\sel(K,2)=0$ for all

980: $K=\Fq(t^{1/d})$.

981: 

982: If $p=3$, let $E$ be defined by 

983: $$y^2=x^3+(t-1)^2x^2+t(t-1)^3x.$$

984: If $\omega=dx/2y$, then $A=(t-1)^2$, $\Delta=-t^2(t-1)^9$, and

985: $j=-(t-1)^3/t^2$.  Standard methods show that $E$ has good, ordinary

986: reduction away from 0, 1, and $\infty$; that $E$ has multiplicative

987: reduction at 0 and $\infty$; and that at $t=1$, $E$ obtains good

988: reduction over an extension with ramification index 4 and the

989: inequality involving $A$ and $\Delta$ is satisfied.  The points

990: $(x,y)=(t^2(t-1)^4,\pm t^2(t-1)^6)$ on $E^{(3)}$ have order 3 and map

991: non-trivially to $\sel(K,Fr)$ and so $\sel(K,3)=0$ for all

992: $K=\Fq(t^{1/d})$.

993: 

994: If $p=5$, let $E$ be defined by 

995: $$y^2=x^3+3(t-1)^4x+(t+1)(t-1)^5.$$

996: If $\omega=dx/2y$, then $A=(t-1)^4$, $\Delta=2t(t-1)^{10}$, and

997: $j=(t-1)^2/2t$.  Standard methods show that $E$ has good, ordinary

998: reduction away from 0, 1, and $\infty$; that $E$ has multiplicative

999: reduction at 0 and $\infty$; and that at $t=1$, $E$ obtains good

1000: reduction over an extension with ramification index 6 and the

1001: inequality involving $A$ and $\Delta$ is satisfied.  The points with

1002: $x$ coordinate $2(t-1)^8(t^2\pm2t-1)$ on $E^{(5)}$ have order 5 and

1003: map non-trivially to $\sel(K,Fr)$ and so $\sel(K,5)=0$ for all

1004: $K=\Fq(t^{1/d})$.

1005: 

1006: If $p=7$, let $E$ be defined by 

1007: $$y^2=x^3+(t-1)(t+1)^3x+5(t-1)(t+1)^5.$$

1008: If $\omega=dx/2y$, then $A=(t-1)(t+1)^5$, $\Delta=2(t-1)^2(t+1)^9$,

1009: and $j=4(t-1)$.  Standard methods show that $E$ has good, ordinary

1010: reduction away from $\pm1$ and $\infty$; that $E$ has multiplicative

1011: reduction at $\infty$; that at $t=1$, $E$ obtains good reduction over

1012: an extension with ramification index 6 and the inequality involving

1013: $A$ and $\Delta$ is satisfied; and that at $t=-1$, $E$ obtains good

1014: reduction over an extension with ramification index 4 and the

1015: inequality involving $A$ and $\Delta$ is satisfied.  It follows that

1016: $\sel(K,7)=0$ for all $K=\Fq(t^{1/d})$.

1017: 

1018: If $p=11$, let $E$ be defined by 

1019: $$y^2=x^3+8(t-1)(t+1)^3x+2(t-1)(t+1)^5.$$

1020: If $\omega=dx/2y$, then $A=(t-1)^2(t+1)^8$, $\Delta=9(t-1)^2(t+1)^9$,

1021: and $j=5(t-1)$.  Standard methods show that $E$ has good, ordinary

1022: reduction away from $\pm1$ and $\infty$; that $E$ has multiplicative

1023: reduction at $\infty$; that at $t=1$, $E$ obtains good reduction over

1024: an extension with ramification index 6 and the inequality involving

1025: $A$ and $\Delta$ is satisfied; and that at $t=-1$, $E$ obtains good

1026: reduction over an extension with ramification index 4 and the

1027: inequality involving $A$ and $\Delta$ is satisfied.  It follows that

1028: $\sel(K,11)=0$ for all $K=\Fq(t^{1/d})$.

1029: 

1030: 

1031: \subsection{}

1032: %The reader familiar with 

1033: The theory of modular forms modulo $p$ suggests

1034: %have no trouble seeing 

1035: that the strategy employed in this section will

1036: not work for large $p$.  Nevertheless, I conjecture that for all $p$

1037: there are elliptic curves (indeed, absolutely simple abelian varieties

1038: of any dimension) over $\Fp(t)$ which have bounded Mordell-Weil ranks

1039: in the tower $\Fq(t^{1/d})$.

1040: 

1041: \begin{bibdiv}

1042: \begin{biblist}

1043: [\resetbiblist{Ulm05}]

1044: 

1045: \bib{Aoki}{article}{

1046:     author={Aoki, N.},

1047:      title={Simple factors of the Jacobian of a Fermat curve and the Picard

1048:             number of a product of Fermat curves},

1049:    journal={Amer. J. Math.},

1050:     volume={113},

1051:       date={1991},

1052:      pages={779\ndash 833},

1053: }

1054: 

1055: \bib{BayerGonzalez}{article}{

1056:    author={Bayer, P.},

1057:    author={Gonz{\'a}lez, J.},

1058:    title={On the Hasse-Witt invariants of modular curves},

1059:    journal={Experiment. Math.},

1060:    volume={6},

1061:    date={1997},

1062:    pages={57--76},

1063: }

1064: 

1065: 

1066: \bib{Conrad}{article}{

1067:     author={Conrad, B.},

1068:      title={Chow's $K/k$-image and $K/k$-trace, and the Lang-N\'eron theorem},

1069:    journal={Enseign. Math. (2)},

1070:     volume={52},

1071:       date={2006},

1072:      pages={37\ndash 108},

1073: }

1074: 

1075: \bib{Davenport}{book}{

1076:     author={Davenport, H.},

1077:      title={Multiplicative number theory},

1078:     series={Graduate Texts in Mathematics},

1079:     volume={74},

1080:  publisher={Springer-Verlag},

1081:      place={New York},

1082:       date={2000},

1083:      pages={xiv+177},

1084: }

1085: 

1086: 

1087: \bib{Ellenberg}{article}{

1088:     author={Ellenberg, J. S.},

1089:      title={Selmer groups and Mordell-Weil groups of elliptic curves over towers of function fields},

1090:      date={2005},

1091:     status={Preprint, to appear in {\it Compositio Mathematica\/}},

1092:     label={Ellen}

1093: }   

1094: 

1095: 

1096: 

1097: \bib{Fastenberg}{article}{

1098:     author={Fastenberg, L. A.},

1099:      title={Mordell-Weil groups in procyclic extensions of a function field},

1100:    journal={Duke Math. J.},

1101:     volume={89},

1102:       date={1997},

1103:      pages={217\ndash 224},

1104: }

1105: 

1106: \bib{HardyWright}{book}{

1107:     author={Hardy, G. H.},

1108:     author={Wright, E. M.},

1109:      title={An introduction to the theory of numbers},

1110:  publisher={The Clarendon Press Oxford University Press},

1111:      place={New York},

1112:       date={1979},

1113:      pages={xvi+426},

1114: }

1115: 

1116: \bib{Ogus}{article}{

1117:     author={Ogus, A.},

1118: title={Hodge cycles and crystalline cohomology},

1119: pages={357--414},

1120: book={

1121:      title={Hodge cycles, motives, and Shimura varieties},

1122:     author={Deligne, Pierre},

1123:     author={Milne, James S.},

1124:     author={Ogus, Arthur},

1125:     author={Shih, Kuang-yen},

1126:     series={Lecture Notes in Mathematics},

1127:     volume={900},

1128:  publisher={Springer-Verlag},

1129:      place={Berlin},

1130: },

1131:       date={1982},

1132: }

1133: 

1134: \bib{Shioda}{article}{

1135:     author={Shioda, T.},

1136:      title={An explicit algorithm for computing the Picard number of certain

1137:             algebraic surfaces},

1138:    journal={Amer. J. Math.},

1139:     volume={108},

1140:       date={1986},

1141:      pages={415\ndash 432},

1142: }

1143: 

1144: \bib{Silverman1}{article}{

1145:     author={Silverman, J. H.},

1146:      title={A bound for the Mordell-Weil rank of an elliptic surface after a

1147:             cyclic base extension},

1148:    journal={J. Algebraic Geom.},

1149:     volume={9},

1150:       date={2000},

1151:      pages={301\ndash 308},

1152: }

1153: 

1154: \bib{Silverman2}{article}{

1155:     author={Silverman, J. H.},

1156:      title={The rank of elliptic surfaces in unramified abelian towers},

1157:    journal={J. Reine Angew. Math.},

1158:     volume={577},

1159:       date={2004},

1160:      pages={153\ndash 169},

1161: }

1162: 

1163: \bib{Stiller}{article}{

1164:     author={Stiller, P. F.},

1165:      title={The Picard numbers of elliptic surfaces with many symmetries},

1166:    journal={Pacific J. Math.},

1167:     volume={128},

1168:       date={1987},

1169:      pages={157\ndash 189},

1170: }

1171: 

1172: \bib{TS}{article}{

1173:     author={Tate, J. T.},

1174:     author={Shafarevitch, I. R.},

1175:      title={The rank of elliptic curves},

1176:   language={Russian},

1177:    journal={Dokl. Akad. Nauk SSSR},

1178:     volume={175},

1179:       date={1967},

1180:      pages={770\ndash 773},

1181: }

1182: 

1183: \bib{TateHT}{article}{

1184: author={Tate, J. T.},

1185: title={Classes d'isog\'enie des vari\'et\'es ab\'eliennes sur un corps

1186:   fini (d'apr\`es T.~Honda)},

1187: book={

1188:      title={S\'eminaire Bourbaki. Vol. 1968/69: Expos\'es 347--363},

1189:     series={Lecture Notes in Mathematics, Vol. 179},

1190:  publisher={Springer-Verlag},

1191:      place={Berlin},

1192:       date={1971},

1193:       },

1194: pages={95--110},

1195: label={Tat68}

1196: }

1197: 

1198: 

1199: \bib{Ulmerpd}{article}{

1200:     author={Ulmer, D.},

1201:      title={$p$-descent in characteristic $p$},

1202:    journal={Duke Math. J.},

1203:     volume={62},

1204:       date={1991},

1205:      pages={237\ndash 265},

1206: }

1207: 

1208: \bib{UlmerR1}{article}{

1209:     author={Ulmer, D.},

1210:      title={Elliptic curves with large rank over function fields},

1211:    journal={Ann. of Math. (2)},

1212:     volume={155},

1213:       date={2002},

1214:      pages={295\ndash 315},

1215: }

1216: 

1217: \bib{UlmerR2}{article}{

1218:     author={Ulmer, D.},

1219:      title={$L$-functions with large analytic

1220: rank and abelian varieties with large algebraic rank over function fields},

1221: date={2005},

1222: label={Ulmer},

1223:     status={Preprint, to appear in {\it Inventiones Mathematicae\/}},

1224: }

1225: 

1226: 

1227: \bib{Washington}{book}{

1228:     author={Washington, L. C.},

1229:      title={Introduction to cyclotomic fields},

1230:     series={Graduate Texts in Mathematics},

1231:     volume={83},

1232:  publisher={Springer-Verlag},

1233:      place={New York},

1234:       date={1997},

1235:      pages={xiv+487},

1236: }

1237: 

1238: \bib{WeilNS}{article}{

1239:     author={Weil, A.},

1240:      title={Numbers of solutions of equations in finite fields},

1241:    journal={Bull. Amer. Math. Soc.},

1242:     volume={55},

1243:       date={1949},

1244:      pages={497\ndash 508},

1245: }

1246: 

1247: \end{biblist}

1248: \end{bibdiv}

1249: 

1250: 

1251: \end{document}

1252:  

1253: