1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass{emulateapj}
3:
4: \newcommand{\ms}{\mbox{m s$^{-1}$}}
5: \newcommand{\degrees}{\mbox{$^\mathrm{o}$}}
6:
7: \begin{document}
8:
9: \title{Hot Jupiter Variability in Eclipse Depth}
10:
11: \author{Emily Rauscher,\altaffilmark{1} Kristen Menou,\altaffilmark{1}
12: James Y-K. Cho\altaffilmark{2}, Sara Seager\altaffilmark{3} \& Bradley
13: M. S. Hansen\altaffilmark{4}}
14:
15: \altaffiltext{1}{Department of Astronomy, Columbia University, 550
16: West 120th Street, New York, NY 10027, USA}
17:
18: \altaffiltext{2}{Astronomy Unit, School of Mathematical Sciences,
19: Queen Mary, University of London, Mile End Road, London E1 4NS, UK}
20:
21: \altaffiltext{3}{Dept. of Earth, Atmospheric, and Planetary sciences,
22: and Dept. of Physics, Massachusetts Institute of Technology, 54-1626,
23: 77 Massachusetts Ave., Cambridge, MA, 02139}
24:
25: \altaffiltext{4}{Department of Physics and Astronomy and Institute for
26: Geophysics and Planetary Physics, University of California, 475
27: Portola Plaza, Box 951547, Los Angeles, CA 90095, USA}
28:
29: \begin{abstract}
30: Physical conditions in the atmospheres of tidally locked,
31: slowly rotating hot Jupiters correspond to dynamical circulation
32: regimes with Rhines scales and Rossby deformation radii comparable to
33: the planetary radii. Consequently, the large spatial scales of moving
34: atmospheric structures could generate significant photospheric
35: variability. Here, we estimate the level of thermal infrared
36: variability expected in successive secondary eclipse depths, according
37: to hot Jupiter turbulent ``shallow-layer'' models. The variability, at
38: the few percent level or more in models with strong enough winds, is
39: within the reach of Spitzer measurements. Eclipse depth variability is
40: thus a valuable tool to constrain the circulation regime and global
41: wind speeds in hot Jupiter atmospheres.
42: \end{abstract}
43:
44: \section{Introduction}
45:
46: The regime of circulation in hot Jupiter atmospheres may be unlike any
47: of the familiar cases in the Solar System. Hot Jupiters are gaseous
48: giant planets found in close, circular orbits around their parent
49: stars, with periods on the order of a few days
50: \citep[e.g.,][]{Butler06}. General arguments suggest that these
51: planets should be tidally locked \citep{Guillot1996, Rasio1996, Lubow1997,
52: Ogilvie2004}: their permanent day-sides are then continuously subject
53: to intense stellar irradiation, while night-sides are only subject to
54: modest internal energy fluxes. Given such an uneven energetic
55: forcing, atmospheric winds would tend to redistribute heat around the
56: planet. The nature and efficiency of this redistribution process is
57: important in determining a variety of hot Jupiter observational
58: properties \citep[e.g.,][]{Seager2005,
59: Burrows2005,Fortney2005,Barman2005,Iro,Burrows2006,Fortney2006}.
60:
61:
62: Two groups have attempted to address the global atmospheric
63: circulation problem in hot Jupiter atmospheres, using different
64: approaches \citep{ShowGui,Cho2003,Menou03,C&S}. On dynamical grounds,
65: it was argued in \citet[][see also Showman \& Guillot 2002]{Menou03},
66: and explicitly shown via turbulent shallow-layer simulations in
67: \citet{Cho2003,Cho2006}, that tidally locked hot Jupiters occupy a
68: regime of circulation that is qualitatively different from that of
69: Solar System giant planets. The few bands and prominent circumpolar
70: vortices emerging in these simulations can be understood in terms of
71: the large Rhines scale and Rossby deformation radius in these
72: atmospheres
73: \citep[e.g.,][]{ChoPolvani96a,ChoPolvani96b}. \citet{Menou03}
74: suggested that the large resulting spatial scales of moving
75: circulation structures could lead to detectable hot Jupiter
76: variability.
77:
78: Soon, interesting observational constraints will be placed on these
79: circulation regime arguments. Three hot Jupiters have been detected
80: through infrared secondary eclipses with the Spitzer Space Telescope:
81: HD189733b \citep{Deming2006}, HD209458b \citep{Deming2005}, and TrES-1
82: \citep{Charbonneau2005}. The planetary day-side flux is deduced from
83: the eclipse depth measured when the planet is hidden behind its
84: star. Repeated eclipse measurements could thus reveal detectable
85: levels of variability of the planetary day-side flux in these three
86: systems. In this Letter, we quantify the level of thermal infrared
87: variability expected in secondary eclipse depths, according to the
88: shallow-layer models of \citet{Cho2003,Cho2006}.
89:
90:
91: \section{Shallow-Layer Circulation Models}
92:
93: As explained in detail in \citet{Cho2003,Cho2006}, turbulent
94: equivalent-barotropic models published to date greatly emphasize
95: dynamical aspects of hot Jupiter atmospheric circulation. We adopt the
96: same notation and adiabatic models as in \citet{Cho2006}: the global
97: wind strength, $\bar U$, and the net amplitude of radiative forcing,
98: $\eta$, are parameterized, not predicted. However, given these two
99: (and other relevant global planetary) parameters, the turbulent
100: atmospheric circulation is consistently found to develop a broad
101: equatorial wind and two large circumpolar vortices revolving around
102: the poles in several planetary days (= orbits). The dynamically
103: modified layer thickness in these models is a proxy for the planet's
104: photospheric temperature field. We refer the reader to \citet{Cho2006}
105: for details on the models, as well as a vast exploration of their
106: parameter space.
107:
108:
109: Our limited goal here is to show that the thermal variability expected
110: in at least some of these models is sufficiently large to be
111: detectable via repeated Spitzer secondary eclipse
112: measurements. Consequently, we focus on a limited set of four models,
113: with global wind speed values from $\bar U=100$ to $800$~\ms~and a
114: moderate amplitude of radiative forcing, $\eta =0.05$ (allowing the
115: weak thermal contrast of features in slow wind models to remain
116: apparent).
117:
118:
119:
120: The models are explicitly calculated for HD209458b
121: parameters, at moderate T63 ($192 \times 96$ grid) resolution, over a
122: hundred planetary days or more. Resolution tests (up to T170) show
123: that T63 is sufficient to capture atmospheric temperature features
124: well enough for our present purpose. Daily outputs from the
125: simulations are used to generate model temperature maps of the
126: day-side thermal emission in our variability study (see
127: below). Table~1 summarizes the range of photospheric temperatures
128: derived from these four models, for the two brightest planets in our
129: study.
130:
131: Figures~\ref{fig:models1} and~\ref{fig:models2} show snapshots of
132: orthographically projected, day-side temperature maps in HD209458b
133: models with $\bar U = 100$ and $400$~\ms, respectively, for two
134: successive eclipses (i.e. after one HD209458b day). These projections
135: illustrate how thermal variability in total eclipse depth is expected,
136: from one eclipse to the next, if large, high contrast temperature
137: features, associated with moving circulation structures, are present
138: (in particular, the cyclonic circumpolar vortices most obviously
139: visible in Fig.~\ref{fig:models2}). Each temperature map in
140: Figs.~\ref{fig:models1} and~\ref{fig:models2} is shown partially
141: eclipsed, for the specific geometry of the HD209458 system. According
142: to these circulation models, cold polar vortices have relatively small
143: (apparent) areas, so that the magnitude of their contribution to a
144: variable eclipsed day-side flux is unclear without a detailed
145: calculation.
146:
147:
148: \section{Thermal Variability in Eclipse Depth}
149:
150:
151:
152: Our method to calculate eclipse depths in models like the ones shown
153: in Figs.~\ref{fig:models1} and~\ref{fig:models2} follows very closely
154: that used in \citet{Rauscher06} to study partial eclipse diagnostics.
155: Planet-daily outputs from the circulation simulations provide the
156: temperature fields used to model successive eclipses. Accounting for
157: system specific inclination and geometry, these temperature fields are
158: orthographically projected onto a 2D disk discretized with ($100$,
159: $200$) resolution elements in ($r$,$\theta$). To calculate spectra,
160: we assume that the vertical temperature profile follows radiative
161: equilibrium according to the cloudless models of \citet{Seager2005},
162: under the assumption that the local flow temperature from the
163: circulation model equals the effective temperature in the radiative
164: model. We then integrate the spectral emission contributed by each
165: apparent surface element on the planetary disk, in the global range of
166: effective temperatures from $700$ to $2000$~K.\footnote{For
167: temperatures slightly below $700$~K in the $\bar{U}=$800 \ms~models, a
168: simple linear extrapolation of the spectra is performed.}
169: Monochromatic fluxes at Earth are then integrated over Spitzer
170: spectral bands (for IRAC, IRS, and MIPS)\footnote{{\tt
171: http://ssc.spitzer.caltech.edu/obs/}} to predict the corresponding
172: successive eclipse depths. {In the future, it will be important to
173: improve upon this simple treatment by fully incorporating radiative
174: transfer in circulation models.}
175:
176:
177: As in \citet{Rauscher06}, we also perform an idealized, bolometric
178: blackbody analysis, to avoid relying exclusively on model specific
179: features of the cloudless spectral emission models used. In these
180: simpler blackbody models, the bolometric flux contributed by each
181: planetary disk surface element is scaled as the fourth power of the
182: local photospheric temperature.
183:
184:
185:
186: Figure~\ref{fig:model_curves}a shows relative variations in successive
187: eclipse depths predicted by the four HD209458b circulation models,
188: assuming simple bolometric blackbody emission. Variations are
189: semi-periodic, as expected from the quasi-periodic motion of the
190: circumpolar vortices around the planet. As the magnitude of the global
191: wind speed, $\bar U$, is increased from $100$ to $800$~\ms, leading to
192: higher contrast motion-induced temperature structures, the amplitude
193: of thermal infrared variability also increases, from a few \% to
194: $\sim60$-$70$\%. Figure~\ref{fig:model_curves}b shows eclipse depth
195: variations for the HD209458b model with $\bar U=400$~\ms~as before,
196: but this time using the detailed spectral emission models to calculate
197: contributions in various Spitzer bands. While the overall variability
198: scale is similar, it is clear that, by using Spitzer bands, one
199: preferentially filters emission from a selective range of temperatures
200: on the planetary disk, which contribute to varying levels of
201: variability. The effect is substantial and we find that variability is
202: the strongest in shortest wavelength Spitzer bands, where thermal
203: contrast from the cold circumpolar vortices is the highest.
204:
205:
206: We calculated explicit circulation models only for HD209458b but we
207: can use the general dynamical similarity of hot Jupiter atmospheres
208: \citep{Menou03} to rescale simply our results for HD189733b and
209: TrES-1. Assuming identical dynamics but allowing for different
210: average atmospheric temperatures, we linearly rescale our model
211: temperature maps proportionally to the zero albedo,
212: fully redistributed equilibrium temperature, $T_p$, of the other two
213: planets \citep[as in][]{Rauscher06}. We find that the eclipse depth
214: variability properties for HD189733b and TrES-1 do not deviate from
215: those of HD209458b by more than $\sim10$\%. Increasing planetary
216: albedos also has little effects on the variability properties, as long
217: as albedos remain $\ll 1$. Finally, as a matter of generality, we have
218: checked that arbitrarily varying any system's orbital inclination in
219: the range 80-90\degrees~ has little effect on its variability
220: properties.
221:
222: \section{Detecting Eclipse Depth Variability}
223:
224: We now address the feasibility of detecting variability in eclipse
225: depth, at the level predicted by the above models, with Spitzer. Let
226: us define $\sigma_{ed}$ as the fractional error on the eclipse depth,
227: that is the ratio of the full (1-$\sigma$) error on the eclipsed flux
228: to the eclipsed flux itself. Values of $\sigma_{ed}$ for the
229: existing secondary eclipse measurements \citep[taken
230: from][]{Charbonneau2005,Deming2005,Deming2006} are listed in bold in
231: Table~2. These numbers already suggest that eclipse depth variability
232: at the level of $5$-$20$\% could be detected in these systems.
233:
234: Our variability models indicate, however, that specific Spitzer bands
235: may be much more useful than others for eclipse depth variability
236: detections. We perform simple estimates of $\sigma_{ed}$ errors for
237: any combination of Spitzer instrument and hot Jupiter system as
238: follows. We use the same set of system parameters as in Table~1 of
239: \citet{Rauscher06}. Assuming that errors on the non-eclipsed flux are
240: comparatively very small, we write $\sigma_{ed}=\sigma_1 / \sqrt{N}$,
241: where $\sigma_1$ is the noise per data point in units of the eclipsed
242: flux and $N$ is the number of single data points collected during a
243: full eclipse period. $N$ is the ratio of the secondary eclipse
244: duration to the instrumental cadence (taken from existing
245: measurements). The eclipse duration time is calculated as $t_{\rm
246: ec}={2}\sqrt{(R_*+R_p)^2-a^2\sin^2(90^\mathrm{o}-i)}/v$, where
247: $v=2\pi a/P$ is the planet's orbital velocity, $P$ its orbital period,
248: $a$ its orbital semi-major axis, $i$ is the orbital inclination, and
249: $R_*$ and $R_p$ are the stellar and planetary radii, respectively
250: \citep[see Table~1 of][]{Rauscher06}. We obtain values of $t_{\rm ec}=
251: 1.76$, $3.23$ and $2.55$~hrs for HD189733b, HD209458b and TrES-1,
252: respectively.
253:
254:
255: Finally, we extrapolate instrument-specific $\sigma_{ed}$ errors known
256: for one system to the other two systems of interest by assuming
257: blackbody emission for both the star and the planet. This results in
258: the following instrument-specific scaling between systems A and B:
259: \begin{eqnarray}
260: \frac{\sigma_{ed}^B}{\sigma_{ed}^A} & = &
261: \sqrt{\frac{t_{ec}^A}{t_{ec}^B}} \sqrt{\frac{F_{*}^B}{F_{*}^A}}
262: \left(\frac{F_{p}^A}{F_{p}^B}\right) \nonumber \\ & = &
263: \sqrt{\frac{t_{ec}^A}{t_{ec}^B}} \left(\frac{d_B}{d_A}\right)
264: \left(\frac{R_{p}^A}{R_{p}^B}\right)^2
265: \left(\frac{R_{*}^B}{R_{*}^A}\right) %\times \nonumber \\ & &
266: \left(\frac{B_\lambda(T_{p}^A)}{B_\lambda(T_{p}^B)}\right)
267: \sqrt{\frac{B_\lambda(T_{*}^B)}{B_\lambda(T_{*}^A)}}, \nonumber
268: \end{eqnarray}
269: \noindent where $d$ is the distance to the system and $B_\lambda$ is
270: the Planck function evaluated at the central wavelength of the
271: instrumental band under consideration. $T_*$ is the stellar effective
272: temperature and $T_p=T_*\sqrt{R_*/2a}$ is the fully redistributed
273: planetary equilibrium temperature calculated in the small albedo
274: limit, exactly as in \citet{Rauscher06}. The resulting extrapolated
275: values of $\sigma_{ed}$ for the three systems of interest are listed
276: in Table~2.
277:
278:
279:
280: Repeated eclipse depth measurements with IRAC for HD189733b and
281: HD209458b appear to be the most likely to succeed in detecting
282: atmospheric variability, at the few percent level or more. Such
283: variability measurements would be difficult for
284: TrES-1. Figure~\ref{fig:model_curves} shows that any pair of
285: successive eclipses will generally not display the full range of
286: variability amplitude: more than two eclipse measurements are needed
287: to sample variability properties adequately. This requirement can be
288: quantified by calculating distributions of fractional variations in
289: eclipse depth over series of 2, 3, 4 or more successive eclipse
290: measurements. Comparing these distributions of eclipse depth
291: variations to the (1-$\sigma_{ed}$) errors listed in Table~2, we find
292: that detecting the eclipse depth variability predicted by the
293: $\bar{U}=$400 or 800 \ms~models at the 2-3 $\sigma$ level requires 3-4
294: IRAC eclipses (at 4.5 or, slightly better, at 8\micron). A generally
295: larger number of eclipses is needed to detect the variability
296: predicted by models with smaller global wind speeds. Eclipse depth
297: variations at the level predicted by the $\bar{U}=$100 \ms~model would
298: be systematically masked by eclipse depth measurement uncertainties,
299: according to our estimates. Finally, we note that a sufficiently large
300: number of successive eclipse measurements could reveal the
301: quasi-periodicity apparent in Fig.~\ref{fig:model_curves}.
302:
303:
304: \section{Discussion and Conclusion}
305:
306: We have illustrated, using simple turbulent shallow-layer
307: circulation models, how eclipse depth variability can be used to
308: constrain the circulation regime and global wind speeds in hot Jupiter
309: atmospheres.
310:
311: Clearly, our circulation models and variability predictions are highly
312: idealized. We have focused on simple thermal diagnostics in models
313: describing an atmosphere as a single horizontal layer of turbulent
314: fluid. Issues related to detailed radiative transfer \citep[e.g.,
315: variation of photospheric height with
316: wavelength;][]{Iro,Seager2005,HarrUps06}, the presence of
317: high-altitude haze or the possible existence of clouds could all
318: seriously affect our conclusions. {Even strictly within the framework
319: of our shallow-layer models, we know that the parameterized amplitude
320: of net radiative forcing, $\eta$, can affect the thermal contrast, and
321: therefore the detectability, of moving atmospheric structures
322: \citep[see][for details]{Cho2006}. We have recalculated all our
323: circulation and variability models with increased values of $\eta=0.1$
324: and $0.2$.\footnote{In these models with stronger radiative forcing,
325: an additional $5$-$10$\% contribution to the flow kinetic energy
326: results from conversion of available potential energy.} We find that
327: the variability amplitude is reduced by up to a factor of two from the
328: case with $\eta=0.05$ shown in Fig.~\ref{fig:model_curves}. Values of
329: $\eta > 0.2$ would further reduce the variability amplitude. In the
330: future, multi-wavelength phase curve data \citep[e.g.,][]{HarrUps06}
331: may provide useful observational diagnostics on the relevant value of
332: $\eta$ for a given planet.}
333:
334:
335: {Despite these shortcomings}, our results are promising in showing
336: that eclipse depth variability is a new and potentially powerful tool
337: to diagnose circulation and wind speeds in hot Jupiter atmospheres. In
338: the future, as more refined atmospheric models are developed and more
339: data becomes available, this tool should become increasingly useful in
340: characterizing hot Jupiter atmospheres.
341:
342: We thank an anonymous referee for comments that helped improve the
343: manuscript. This work was supported by NASA contract NNG06GF55G, NASA
344: Astrobiology Institute contract NNA04CC09A and a Spitzer Theory
345: grant. q
346:
347:
348: \begin{thebibliography}
349:
350: %\bibitem[Alonso et al.(2004)]{Alonso2004} Alonso, R., et al.\ 2004, \apjl, 613, L153
351:
352: \bibitem[Barman et al.(2005)]{Barman2005} Barman, T.~S., Hauschildt, P.~H., \& Allard, F.\ 2005, \apj, 632, 1132
353:
354: %\bibitem[Bouchy et al.(2005)]{Bouchy2005} Bouchy, F., et al.\ 2005, \aap, 444, L15
355:
356: %\bibitem[Brown et al.(2001)]{Brown2001} Brown, T.~M., Charbonneau, D., Gilliland, R.~L., Noyes, R.~W., \& Burrows, A.\ 2001, \apj, 552, 699
357:
358: \bibitem[Burrows et al.(2005)]{Burrows2005} Burrows, A., Hubeny, I., \& Sudarsky, D.\ 2005, \apjl, 625, L135
359:
360: \bibitem[Burrows et al.(2006)]{Burrows2006} Burrows, A., Sudarsky, D., \& Hubeny, I.\ 2006, ApJ, 650, 1140
361:
362: \bibitem[Butler et al.(2006)]{Butler06} Butler, R.~P. et al. 2006,
363: ApJ, 646, 505
364:
365: \bibitem[Charbonneau et al.(2005)]{Charbonneau2005} Charbonneau, D., et al.\ 2005, \apj, 626, 523
366:
367: %\bibitem[Charbonneau et al.(2000)]{Charbonneau2000} Charbonneau, D., Brown, T.~M., Latham, D.~W., \& Mayor, M.\ 2000, \apjl, 529, L45
368:
369: \bibitem[Cho et al.(2003)]{Cho2003} Cho, J.~Y.-K., Menou, K., Hansen, B.~M.~S., \& Seager, S.\ 2003, \apjl, 587, L117
370:
371: \bibitem[Cho et al.(2006)]{Cho2006} Cho, J.~Y.-K., Menou, K., Hansen,
372: B., \& Seager, S.\ 2006, ApJ submitted, astro-ph/0607338
373:
374: \bibitem[Cho \& Polvani(1996a)]{ChoPolvani96a} Cho, J. Y.-K. \&
375: Polvani, L. M. 1996a, Phys. Fluids, 8, 1531
376:
377: \bibitem[Cho \& Polvani(1996b)]{ChoPolvani96b} Cho, J. Y.-K. \&
378: Polvani, L. M. 1996b, Science, 273, 335
379:
380: \bibitem[Cooper \& Showman(2005)]{C&S} Cooper, C.~S., \& Showman, A.~P.\ 2005, \apjl, 629, L45
381:
382: \bibitem[Deming et al.(2006)]{Deming2006} Deming, D., Harrington, J., Seager, S., \& Richardson, L.~J.\ 2006, \apj, 644, 560
383:
384: \bibitem[Deming et al.(2005)]{Deming2005} Deming, D., Seager, S., Richardson, L.~J., \& Harrington, J.\ 2005, \nat, 434, 740
385:
386: \bibitem[Guillot et al.(1996)]{Guillot1996} Guillot, T., Burrows, A., Hubbard, W.~B., Lunine, J.~I., \& Saumon, D.\ 1996, \apjl, 459, L35
387:
388: \bibitem[Harrington et al.(2006)]{HarrUps06} Harrington, J. et
389: al. 2006, Science, 314, 623
390:
391: \bibitem[Fortney et al.(2005)]{Fortney2005} Fortney, J.~J., Marley, M.~S., Lodders, K., Saumon, D., \& Freedman, R.\ 2005, \apjl, 627, L69
392:
393: \bibitem[Fortney et al.(2006)]{Fortney2006} Fortney, J.~J., Cooper,
394: C.~S., Showman, A.~P., Marley, M.~S.\& Freedman, R.\ 2006, \apj in
395: press, astro-ph/0608235
396:
397: \bibitem[Iro et al.(2005)]{Iro} Iro, N., Bezard, B. \& Guillot, T
398: 2005, A\&A, 436, 719
399:
400: \bibitem[Lubow et al.(1997)]{Lubow1997} Lubow, S.~H., Tout, C.~A., \&
401: Livio, M.\ 1997, \apj, 484, 866
402:
403: %\bibitem[Mazeh et al.(2000)]{Mazeh2000} Mazeh, T., et al.\ 2000, \apjl, 532, L55
404: \bibitem[Menou et al.(2003)]{Menou03} Menou, K. Cho, J.~Y.-K., Seager,
405: S. \& Hansen, B.~M.~S.\ 2003, \apjl, 587, L113
406:
407: \bibitem[Ogilvie \& Lin(2004)]{Ogilvie2004} Ogilvie, G.~I., \& Lin,
408: D.~N.~C.\ 2004, \apj, 610, 477
409:
410: %\bibitem[Perryman et al.(1997)]{Hipp} Perryman, M.~A.~C., et al.\ 1997, \aap, 323, L49
411:
412: \bibitem[Rasio et al.(1996)]{Rasio1996} Rasio, F.~A., Tout, C.~A.,
413: Lubow, S.~H., \& Livio, M.\ 1996, \apj, 470, 1187
414:
415: \bibitem[Rauscher et al.(2007)]{Rauscher06} Rauscher, E., Menou, K.,
416: Seager, S., Deming, D., Cho, J.~Y.-K., \& Hansen, B.~M.~S.\ 2007, ApJ submitted
417:
418: \bibitem[Seager et al.(2005)]{Seager2005} Seager, S., Richardson,
419: L.~J., Hansen, B.~M.~S., Menou, K., Cho, J.~Y.-K., \& Deming, D.\
420: 2005, \apj, 632, 1122
421:
422: \bibitem[Showman \& Guillot(2002)]{ShowGui} Showman, A.~P. \& Guillot,
423: T. 2002, A\&A, 385, 166
424:
425: %\bibitem[Sozzetti et al.(2004)]{Sozzetti2004} Sozzetti, A., et al.\ 2004, \apjl, 616, L167
426:
427: \end{thebibliography}
428:
429: \clearpage
430:
431: \begin{deluxetable}{lccc}
432: \tablewidth{0pt}
433: \tablecaption{\citet{Cho2006} models under consideration}
434: \tablehead{
435: \colhead{} & \colhead{} & \multicolumn{2}{c}{Min-Max Temperature (K)} \\
436: \colhead{} & \colhead{$\bar{U}$ (\ms)} & \colhead{HD209458b} & \colhead{HD189733b}
437: }
438: \startdata
439: Model 1 & 100 & 1147-1275 & 956-1062 \\
440: Model 2 & 200 & 1112-1287 & 926-1072 \\
441: Model 3 & 400 & 1006-1308 & 838-1088 \\
442: Model 4 & 800 & 669-1372 & 557-1141 \\
443: \enddata
444: \end{deluxetable}
445:
446: \clearpage
447:
448: \begin{deluxetable}{lccc}
449: \tablewidth{0pt}
450: \tablecaption{Measured (in bold) and estimated values of $\sigma_{ed}$.}
451: \tablehead{
452: \colhead{Instrument, wavelength} & \colhead{HD 189733b} & \colhead{HD 209458b} & \colhead{TrES-1}
453: }
454: \startdata
455: IRAC, 4.5 \micron & 0.018 & 0.039 & \textbf{0.20} \\
456: IRAC, 8 \micron & 0.014 & 0.026 & \textbf{0.16} \\
457: IRS, 16 \micron & \textbf{0.055} & 0.11 & 0.62 \\
458: MIPS, 24 \micron & 0.086 & \textbf{0.18} & 0.96 \\
459: \enddata
460: \end{deluxetable}
461:
462: \clearpage
463:
464: \begin{figure}
465: \begin{center}
466: \includegraphics[width=0.35\textwidth]{f1a.eps}
467: \includegraphics[width=0.35\textwidth]{f1b.eps}
468: \caption{Partially eclipsed temperature maps (in K) in a HD209458b
469: model with a $100$ \ms~ global wind speed, for two successive
470: eclipses. Little thermal infrared variability is expected.}
471: \label{fig:models1}
472: \end{center}
473: \end{figure}
474:
475: \clearpage
476:
477: \begin{figure}
478: \begin{center}
479: \includegraphics[width=0.35\textwidth]{f2a.eps}
480: \includegraphics[width=0.35\textwidth]{f2b.eps}
481: \caption{Same as Fig.~\ref{fig:models1} in a model with a
482: $400$~\ms~global wind speed. Significant thermal infrared variability
483: is expected in this case.} \label{fig:models2}
484: \end{center}
485: \end{figure}
486:
487: \clearpage
488:
489: \begin{figure}
490: \begin{center}
491: %\includegraphics[width=0.45\textwidth]{f3a.eps}
492: %\includegraphics[width=0.45\textwidth]{f3b.eps}
493: \includegraphics[width=0.45\textwidth]{f3.eps}
494: \caption{Relative flux variations at full eclipse according to
495: HD209458b circulation models. Discrete points are connected for
496: clarity and fluxes are normalized to the minimum value in each case.
497: (a) Variations for simple bolometric blackbody emission, in models
498: with $\bar{U}$=100 (purple line), 200 (red), 400 (blue), and 800
499: \ms~(black). (b) Variations according to detailed spectral emission
500: models, in various Spitzer bands, for the $\bar{U}$=400 \ms~model of
501: HD209458b. From top to bottom, the curves correspond to
502: 3.6\micron~(\emph{black}), 4.5\micron~(\emph{blue}),
503: 8\micron~(\emph{red}), 6\micron~(\emph{green}),
504: 16\micron~(\emph{orange}), and 24\micron~(\emph{yellow}).}
505: \label{fig:model_curves}
506: \end{center}
507: \end{figure}
508:
509:
510: \end{document}