MARVELS: Revealing the Formation and Dynamical Evolution of Giant Planet Systems

MARVELS, the Multi-Object APO Radial Velocity Exoplanet Large-area Survey, is a 6-year program to characterize the distribution of gas giant planets with orbital periods ranging from several hours to two years. MARVELS will use multi-fiber interferometric spectrographs on the wide-field, 2.5-meter Sloan Foundation telescope at Apache Point Observatory to monitor ~11,000 stars in the magnitude range V=8-12, visiting each star ~30 times over an 18-month interval, with velocity precision of 14, 18, and 35 m/s at V=8, 10, and 12. MARVELS will survey far more stars with a wider range of spectral types and metallicities than previous radial velocity searches, yielding a statistically well defined sample of ~150 giant planets drawn from a host sample with well understood selection biases. With a unique combination of large numbers and well characterized sensitivity, MARVELS will provide a critical data set for testing theories of the formation and dynamical evolution of giant planet systems. The MARVELS detections will also be an ideal sample for follow-up observations to identify multiple planet systems and understand the influence of giant planet migration on the formation and survival of less massive planets. MARVELS is one of four surveys that comprise SDSS-III (the Sloan Digital Sky Survey III), a 6-year program that will use highly multiplexed spectrographs on the Sloan Foundation Telescope to investigate cosmological parameters, the history and structure of the Milky Way galaxy, and the population of giant planet systems.

💡 Research Summary

The MARVELS (Multi‑Object APO Radial Velocity Exoplanet Large‑area Survey) project is a six‑year component of SDSS‑III designed to map the occurrence and orbital properties of gas‑giant planets with periods from a few hours up to two years. Using a novel dispersed‑fixed‑delay interferometer (DFDI) coupled to the 2.5‑m Sloan Foundation Telescope, MARVELS can observe up to sixty stars simultaneously, a dramatic increase in multiplexing compared with traditional single‑target radial‑velocity (RV) surveys.

The target sample consists of roughly 11 000 stars in the apparent magnitude range V = 8–12, spanning spectral types F through M and a wide metallicity distribution. Each star is visited about thirty times over an 18‑month baseline, providing a cadence that captures both ultra‑short‑period “hot‑Jupiter” candidates and longer‑period giants. The achieved RV precision is 14 m s⁻¹ at V = 8, 18 m s⁻¹ at V = 10, and 35 m s⁻¹ at V = 12, sufficient to detect planets down to ~0.5 M_Jup across the full period range.

A key strength of MARVELS is its well‑characterized detection sensitivity. By modeling the number of observations per star, the window function of the time series, and applying sigma‑clipping to reject outliers, the survey builds a comprehensive detection‑probability map. This map explicitly incorporates selection biases, enabling a Bayesian inversion from the observed planet distribution to the underlying planet occurrence rate. Consequently, the expected yield of ~150 giant planets will have statistical uncertainties below 5 %, providing a robust basis for population studies.

Scientific objectives include: (1) measuring the giant‑planet occurrence rate as a function of orbital period, stellar metallicity, and spectral type, thereby testing core‑accretion versus disk‑instability formation scenarios; (2) quantifying the frequency of multi‑planet systems and the eccentricity distribution of giant planets to assess migration mechanisms such as Type II disk migration and dynamical scattering; (3) supplying a well‑defined target list for follow‑up observations (high‑resolution spectroscopy, transit photometry, direct imaging) to probe planetary atmospheres and interior structure; and (4) cross‑matching MARVELS host stars with APOGEE chemical abundances and ages, and with BOSS large‑scale‑structure data, to explore how Galactic environment influences planet formation.

Because MARVELS operates within the SDSS‑III framework, its data products are interoperable with the other three surveys (BOSS, APOGEE, and the SEGUE‑2 extension). This synergy allows researchers to place planetary systems in a broader Galactic context, linking planet occurrence to stellar population gradients, chemical evolution, and dynamical history.

In summary, MARVELS represents a transformative step in exoplanet detection methodology: a high‑multiplex, moderate‑precision RV survey that delivers a statistically powerful, bias‑controlled sample of giant planets. The resulting dataset will be pivotal for refining theories of planet formation and migration, guiding future missions such as TESS and PLATO, and informing the next generation of planet‑population studies.