GJ 436b is a prime target for understanding warm Neptune exoplanet atmospheres and a target for multiple James Webb Space Telescope (JWST) Guaranteed Time Observation programs. Here, we report the first space-based optical transmission spectrum of the planet using two Hubble Space Telescope (HST) Space Telescope Imaging Spectrograph (STIS) transit observations from 0.53 to 1.03 mu m. We find no evidence for alkali absorption features, nor evidence of a scattering slope longward of 0.53 mu m. The spectrum is indicative of moderate to high metallicity (similar to 100-1000x solar), while moderate-metallicity scenarios (similar to 100x. solar) require aerosol opacity. The optical spectrum also rules out some highly scattering haze models. We find an increase in transit depth around 0.8 mu m in the transmission spectra of three different sub-Jovian exoplanets (GJ 436b, HAT-P-26b, and GJ 1214b). While most of the data come from STIS, data from three other instruments may indicate this is not an instrumental effect. Only the transit spectrum of GJ 1214b is well fit by a model with stellar plages on the photosphere of the host star. Our photometric monitoring of the host star reveals a stellar rotation rate of 44.1 days and an activity cycle of 7.4 years. Intriguingly, GJ 436 does not become redder as it gets dimmer, which is expected if star spots were dominating the variability. These insights into the nature of the GJ 436 system help refine our expectations for future observations in the era of JWST, whose higher precision and broader wavelength coverage will shed light on the composition and structure of GJ 436b's atmosphere.

This dissertation investigates two stellar systems known to contain extrasolar planets. It is comprised of five chapters that are readily divided into three independent but related analyses. Chapter 1 reports on the analysis of low signal-to-noise secondary-eclipse observations of the Neptune-sized exoplanet GJ 436b using the Spitzer Space Telescope in multiple infrared channels. The measured wavelength-dependent eclipse depths provide constraints on the planet’s dayside atmospheric composition and thermal profile. The analysis indicates that GJ 436b’s atmosphere is abundant in carbon monoxide and deficient in methane relative to thermochemical equilibrium models for the predicted hydrogen-dominated atmosphere. Chapter 2 discusses the techniques used to analyze GJ 436b, introduces the Least Asymmetry centering method and compares its effectiveness to two existing techniques, and describes the functions used to model Spitzer’s position- and time-dependent systematics. Additionally, it includes best-fit parameters with uncertainties, histograms of the free parameters, and correlation plots between free parameters. Chapter 3 reports on the analysis of eleven HD 149026b secondary-eclipse observations at five Spitzer wavelengths plus three primary-transit observations at 8.0 µm. Chemical-equilibrium models find no indication of a temperature inversion in the dayside atmosphere of HD 149026b. The best-fit model favors large amounts of CO and CO2 , moderate heat redistribution (f = 0.5), and a strongly eniii hanced metallicity. These analyses use BiLinearly-Interpolated Subpixel Sensitivity (BLISS) mapping and parameter orthogonalization. The former is a new technique to model two position-dependent systematics, intrapixel variability and pixelation. The latter is a technique that accelerates the convergence of Markov chains that employ the Metropolis random walk sampler. Chapter 4 reports on the detection of GJ 436c, a 0.65 ± 0.04 R⊕ exoplanet transiting a nearby M-dwarf star with a period of 1.365862 ± 8×10−6 days. It also presents evidence for a similarly sized exoplanet candidate (currently labeled UCF-1.02) orbiting the same star with an undetermined period. Assuming an Earth-like density of 5.515 g/cm3 , GJ 436c has a predicted mass of 0.28 Earth-masses (M⊕, 2.6 Mars-masses) and a surface gravity of 0.65 g (where g is the gravity on Earth). Its weak gravitational field and close proximity to its host star imply that GJ 436c is unlikely to have retained its original atmosphere; however, a transient atmosphere is possible if recent impacts or tidal heating were to supply volatiles to the surface. Chapter 5 presents numerical simulations of the GJ 436 system using the Mercury N-body integrator and detailed calculations used to constrain the atmospheric composition of the sub-Earth-sized planet GJ 436c. The simulations find a ∼35-year periodic trend in the osculating elements wherein GJ 436c’s eccentricity varies between 0 and 0.21, its peak-to-trough inclination amplitude is 3.2◦ , and transit-timing variations range from ±200 to ±3 minutes.