In this thesis I investigate the use of numerical modeling
techniques applied to the study of extrasolar planets. In the first
part (Chapters 2-4) I discuss the algorithms and applications of
the systemic code in the detection and characterization of
exoplanets through radial velocity (RV) and transit timing
observations. The second part (Chapters 5-6) deals with
hydrodynamic and N-body simulations applied to the study of
planet formation. For each chapter, I provide a detailed review of
the numerical techniques involved in the respective introductions.
Chapter 2 discusses several aspects related to the dynamical
fitting of RV observations. I introduce the systemic package I
developed, and describe several applications of the numerical
algorithms developed for the code. As a case study, I investigate
the dynamical fitting of HD128311 and the characterization of the
2:1 mean motion resonance (MMR) through radial velocities and a
small number of central transit times. I present an updated Keck RV
dataset and show that the addition of three years of new RV
coverage yields only a modest improvement in the characterization
of the system.
In Chapter 3, I study planet detection through transit timing
variations (TTV), deviations from linear transit ephemeris that can
be caused by additional planets exerting gravitational
perturbations on a transiting planet. I created synthetic RV and
TTV datasets for several planetary configurations, with the intent
of modeling timing observations from the Kepler mission. I use
the algorithms described in Chapter 2 to solve the so-called
"inverse problem", the task of characterizing additional,
non-transiting planets through their signatures in the RV and TTV
datasets of transiting. I show that the space of the best-fitting
solutions may be remarkably degenerate if the perturbing planet is
not observed directly (e.g. as in the case of Kepler 19-c), and
that more extensive RV coverage can be used to break the
degeneracy.
In Chapter 4, I present the discovery of four new exoplanet
candidates characterized with Keck/HIRES RV observations. The new
exoplanets discovered around the host stars HD31253, HD218566,
HD177830 and HD99492 comprise masses between Msini ~
27 M_earth to Msini ~ 8 M_jupiter. Of particular
interest for the scope of this thesis, HD177830 is currently the
only multiple-planet system orbiting a binary with a_B < 100 AU.
This separation is slightly below the limit at which the binarity
of the system influences planet formation. Finally, we strengthen
the case for the non-detection of HD74156d, the detection of which
was claimed to be in accordance to the "Packed Planetary System"
hypothesis.
Chapter 5 explores a class of self-gravitating instabilities
driven by features in the surface density of protoplanetary disks
("groove modes"). The emergence of these instabilities is
studied via a generalized eigenvalue code and full two-dimensional
hydrodynamical simulations. I find that gaps in the surface
density, such as those naturally carved in response to the
formation of a giant planet, can excite a global two-armed mode at
comparatively lower disk masses than in absence of such gaps.
Chapter 6 describes a new code, SPHIGA, used to explore the
issue of forming planets in circumstellar (CS) or circumbinary (CB)
orbits during the planetesimal accretion phase and its feasibility
within the core accretion framework. I investigate the balance
between accreting and erosive impacts for the circumbinary planet
Kepler 16-b and the feasibility of planet formation in situ
as opposed to migration of an embryo formed at or outside the ice
line.
