David Sudarsky

We have created a general methodology for calculating the wavelength-dependent light curves of close-in extrasolar giant planets (EGPs) as they traverse their orbits. Focussing on the transiting EGPs HD189733b, TrES-1, and HD209458b, we calculate planet/star flux ratios during secondary eclipse and compare them with the Spitzer data points obtained so far in the mid-infrared. We introduce a simple parametrization for the redistribution of heat to the planet's nightside, derive constraints on this parameter (P_n), and provide a general set of predictions for planet/star contrast ratios as a function of wavelength, model, and phase. Moreover, we calculate average dayside and nightside atmospheric temperature/pressure profiles for each transiting planet/P_n pair with which existing and anticipated Spitzer data can be used to probe the atmospheric thermal structure of severely irradiated EGPs. We find that the baseline models do a good job of fitting the current secondary eclipse dataset, but that the Spitzer error bars are not yet small enough to discriminate cleanly between all the various possibilities.

We have created a general methodology for calculating the wavelength-dependent light curves of close-in extrasolar giant planets (EGPs) as they traverse their orbits. Focusing on the transiting EGPs HD 189733b, TrES-1, and HD 209458b, we calculate planet/star flux ratios during secondary eclipse and compare them with the Spitzer data points obtained so far in the mid-infrared. We introduce a simple parameterization for the redistribution of heat to the planet's night side, derive constraints on this parameter (P_n), and provide a general set of predictions for planet/star contrast ratios as a function of wavelength, model, and phase. Moreover, we calculate average dayside and nightside atmospheric temperature/pressure profiles for each transiting planet/P_n pair with which existing and anticipated Spitzer data can be used to probe the atmospheric thermal structure of severely irradiated EGPs. We find that the baseline models do a good job of fitting the current secondary eclipse data set, but that the Spitzer error bars are not yet small enough to discriminate cleanly among all the various possibilities.

We calculate the planet-star flux-density ratios as a function of wavelength from 0.5 microns to 25 microns for the transiting extrasolar giant planets TrES-1 and HD209458b and compare them with the recent Spitzer/IRAC-MIPS secondary eclipse data in the 4.5, 8.0, and 24 micron bands. With only three data points and generic calibration issues, detailed conclusions are difficult, but inferences regarding atmospheric composition, temperature, and global circulation can be made. Our models reproduce the observations reasonably well, but not perfectly, and we speculate on the theoretical consequences of variations around our baseline models. One preliminary conclusion is that we may be seeing in the data indications that the day side of a close-in extrasolar giant planet is brighter in the mid-infrared than its night side, unlike Jupiter and Saturn. This correspondence will be further tested when the data anticipated in other Spitzer bands are acquired, and we make predictions for what those data may show.Comment: 15 pages, including 3 color figures, submitted to the Astrophysical Journa

Using a full frequency-dependent atmosphere code that can incorporate irradiation by a central primary star, we calculate self-consistent boundary conditions for the evolution of the radius of the transiting planet HD 209458b. Using a well-tested extrasolar giant planet evolutionary code, we then calculate the behavior of this planet's radius with age. The measured radius is in fact a transit radius that resides high in HD 209458b's inflated atmosphere. Using our derived atmospheric and interior structures, we find that irradiation plus the proper interpretation of the transit radius can yield a theoretical radius that is within the measured error bars. We conclude that if HD 209458b's true transit radius is at the lower end of the measured range, an extra source of core heating power is not necessary to explain the transit observations.

Using a self-consistent atmosphere code, we construct a new model of the atmosphere of the transiting extrasolar giant planet HD 209458b to investigate the disparity between the observed strength of the sodium absorption feature at 589 nm and the predictions of previous models. For the atmospheric temperature-pressure profile we derive, silicate and iron clouds reside at a pressure of several mbar in the planet's atmosphere. These clouds have significant vertical extent and optical depth due to our slant viewing geometry and lead to increased absorption in bands directly adjacent to the sodium line core. Using a non-LTE sodium ionization model that includes photoionization by stellar UV flux, collisional processes with H_2, and radiative recombination, we show that the ionization depth in the planet's atmosphere reaches ~1/2 mbar at the day/night terminator. Ionization leads to a slight weakening of the sodium feature. We present our baseline model, including ionization and clouds, which falls near the observational error bars. The sensitivity of our conclusions to the derived atmospheric temperature-pressure profile is discussed.

Because the opacity of clouds in substellar mass object (SMO) atmospheres depends on the composition and distribution of particle sizes within the cloud, a credible cloud model is essential for accurately modeling SMO spectra and colors. We present a one--dimensional model of cloud particle formation and subsequent growth based on a consideration of basic cloud microphysics. We apply this microphysical cloud model to a set of synthetic brown dwarf atmospheres spanning a broad range of surface gravities and effective temperatures (g_surf = 1.78 * 10^3 -- 3 * 10^5 cm/s^2 and T_eff = 600 -- 1600 K) to obtain plausible particle sizes for several abundant species (Fe, Mg2SiO4, and Ca2Al2SiO7). At the base of the clouds, where the particles are largest, the particle sizes thus computed range from ~5 microns to over 300 microns in radius over the full range of atmospheric conditions considered. We show that average particle sizes decrease significantly with increasing brown dwarf surface gravity. We also find that brown dwarfs with higher effective temperatures have characteristically larger cloud particles than those with lower effective temperatures. We therefore conclude that it is unrealistic when modeling SMO spectra to apply a single particle size distribution to the entire class of objects.

We have generated new, self-consistent spectral and atmosphere models for the effective temperature range 600-1300 K thought to encompass the known T dwarfs. For the first time, theoretical models are compared with a family of measured T dwarf spectra at wavelengths shortward of ~1.0 μm. By defining spectral indices and standard colors in the optical and very near-infrared, we explore the theoretical systematics with T_eff, gravity, and metallicity. We conclude that the short-wavelength range is rich in diagnostics that complement those in the near-infrared now used for spectral subtyping. We also conclude that the wings of the Na D and K I (7700 Å) resonance lines and aggressive rainout of heavy metals (with the resulting enhancement of the sodium and potassium abundances at altitude) are required to fit the new data shortward of 1.0 μm. Furthermore, we find that the water bands weaken with increasing gravity, that modest decreases in metallicity enhance the effect in the optical of the sodium and potassium lines, and that at low values of T_eff, in a reversal of the normal pattern, optical spectra become bluer with further decreases in T_eff. Moreover, we conclude that T dwarf subtype is not a function of T_eff alone but that it is a nontrivial function of gravity and metallicity as well. As do Marley and coworkers in their 2002 work, we see evidence in early T dwarf atmospheres of a residual effect of clouds. With cloudless models, we obtain spectral fits to the two late T dwarfs with known parallaxes, but a residual effect of clouds on the emergent spectra of even late T dwarfs cannot yet be discounted. However, our focus is not on detailed fits to individual objects but on the interpretation of the overall spectral and color trends of the entire class of T dwarfs, as seen at shorter wavelengths.