
In order to understand the origin and evolution of planetary systems, we must first understand the characteristics of the gaseous component of protoplanetary disks. Observations and models of extra-solar planets and protoplanetary disks demonstrate that the final characteristics of a planetary system depend sensitively on factors directly associated with the thick gaseous nebula that forms the bulk of the mass during the early stages of star and planet formation: disk structure, dust dynamics and composition, and dynamical interactions with the gas disk such as planet migration and orbital eccentricity enhancement. To accurately trace the path from raw interstellar dust and gas to fully formed planets, we must improve our understanding of the dynamic, thermodynamic, and chemical structure of the gaseous disk during each stage of planet formation.
During my first two years as an NPP Fellow at NASA GSFC, working with Dr. Michael Mumma (GSFC) and Dr. Geoff Blake (CalTech), I initiated a ground-based IR survey of molecular gas emission from the planet-forming region of circumstellar disks in order to constrain the temperature and density of proto-planetary material during the initial stages of planet formation. By combining advanced data reduction algorithms with accurate modeling of the terrestrial atmospheric spectrum and a novel double-differencing data analysis technique, I can achieve the photon-noise limit (S/N ~ 2000) on bright nearby stars with disks using state-of-the-art spectrographs such as NIRSPEC on the Keck telescope and CRIRES at ESO's Very Large Telescope facility (Paranal, Chile).
We first used this approach to discover OH ro-vibrational emission in the planet-forming (1-10 AU) region of high-mass Herbig Ae stars. OH is a sensitive tracer of the dissociation and recombination of H2 and H2O, and thus depends on the local UV and IR radiation fields, and the radial density distribution. Our current sample of four stars suggests correlations in temperature and abundance with disk mass that may signal grain growth (i.e. planetesimal formation) and/or water vapor depletion. In disks near low-mass stars, we have preliminary evidence of several new carbon-chain volatiles along with new detections of H2O, and we are now planning and proposing new bservations of additional disks. By testing the temperature and composition of the protoplanetary disk in the region where planets are forming, these measurements will provide critical data for further improving planet formation models, and will lay the groundwork for observations over a wider range of temperature and density, for more molecules, with the next generation of ultra-sensitive high-resolution NIR imagers and spectrographs such as the NIRCam and NIRSpec instruments on the James Webb Space Telescope.

Mandell et al. 2008, Figure 1: Examples of line detections for both stars. The top traces show the stellar spectrum with terrestrial lines, as measured (black). A synthetic atmospheric spectrum is overlaid for comparison (green). The very bottom traces show individual residuals for the science target (blue) and the comparison star (black). The middle trace shows the difference of the science and comparison star residuals (blue) bounded by 3σ photon-noise uncertainty limits (red). Individual residuals (bottom) are multiplied by a factor of 10, their difference (middle) by a factor of 50.

Mandell et al. 2008, Figure 3:Predictions of a thin wedge fluorescence model for AB Aurigae. For a specific absorbing column density, a lower-state rotational temperature (Tlow) and wedge thickness (H/R) are derived to match the observed component-averaged fluxes for each transition; the Δχ2 value (compared to the minimum value of 1.52) and standard confidence intervals for the fit are also shown.
Mandell, A. M., Mumma, M. J., Blake, G. A., Bonev, B. P., Villanueva, G. L., & Salyk, C. "Discovery of OH in Circumstellar Disks around Young Intermediate-Mass Stars". 2008, ApJ, 681, L25
NPP Fellow
