Staff Profiles

Ricky Nilsson

Research Associate

Download full CV
  • Education

      • Ph.D. in Astronomy at the Department of Astronomy, Stockholm University, Sweden, 2012
      • M.Sc. in Engineering Physics at Lund Institute of Technology, Lund University, Sweden, 2005
  • Research Interests

    Research Interests

      The main topic of my research up to now can be summarized as characterization of planetary systems, involving investigations of exo-Kuiper belts and extrasolar planets (exoplanets), using observational studies of thermal emission and scattering/polarization properties of dust grains, as well as direct imaging and recovery of spectra of young Jupiter-mass planets and brown dwarfs. Below, I briefly describe current research involvements.


      The last decade of observations of exoplanets and brown dwarfs has revealed planetary systems showing a great diversity in properties, with planetary parameters and configurations quite unlike that of the Solar System. However, important regions of parameter space remain unexplored due to technical limitations. Two of my principal research involvements are on projects that aim to reach these uncharted regions with groundbreaking observing techniques and data processing methods.


      Gas-giant planets of a few Jupiter-masses on orbits of 5–50 astronomical units (AU) around young stars in the solar vicinity are now directly observable with modern high-contrast imaging techniques, probing a new region of planet mass vs. separation parameter space. My main research at the moment concerns direct imaging and spectroscopy of exoplanets with Project-1640 (P1640; e.g., Oppenheimer et al. 2012) at Palomar Observatory’s 5-m Hale telescope. The immediate aim of P1640 is to obtain images and low-resolution spectra of young (<1 Gyr) gas-giant planets (and brown dwarfs) around A- and F type stars within 75 parsec (pc) of the Sun, in order to characterize planetary systems around nearby stars. A methodical survey of system characteristics derived from direct imaging and spectroscopy of exoplanets is essential to obtain more complete statistics on system frequencies and properties, and a better understanding of planet formation and evolution. This is also a key step to future imaging of Earth-like exoplanets.

      Our 3-year survey was recently completed, and although some important new results on companions in benchmark systems like, e.g., HR 8799 (Oppenheimer et al. 2013; Pueyo et al. 2015), κ And (Hinkley et al. 2013), HD 19467 (Crepp et al. 2015), and GJ 758 (Nilsson et al. 2017) were obtained, confirmed discoveries have been few. This is in line with early results from GPI and SPHERE, leading to our preliminary conclusion that gas-giant planets at probed masses and orbital distances are less numerous than anticipated.

      In the meantime, I have gained significant experience in performing coronagraphic high-contrast hyperspectral imaging observations, and learned valuable lessons about the intricacies of storing and processing large amounts of survey data. Approximately 100 nights have been spent observing with P1640, with additional daytime setup and calibration of the coronagraphic system. As P1640 has a relatively small core team, I have been fortunate to touch on all aspects of the project; from strategic target selection, detailed planning of observations, instrument setup, and calibration, to tuning of the PALM-3000 (P3k Dekany et al. 2013) AO and CAL (e.g., Vasisht et al. 2014; Zhai et al. 2012) systems, and subsequent data acquisition. I have made extensive use of PCXP (Zimmerman et al. 2011) for extraction of tightly packed microspectra from images into data cubes, as well as the PCA-based algorithms S4 (Fergus et al. 2014) and KLIP (Soummer et al. 2012) for speckle subtraction. Another important step in the data processing pipeline is correction of atmospheric and instrument induced dispersion, for which I wrote the cube alignment code (CACS; Nilsson et al., in prep.) implementing automatic astrometric grid-spot identification, tracking, radial scaling, and fine-alignment through cross-correlation. I also initiated the P1640 database for organizing target data, data products, and instrument telemetry, which is of paramount importance for the ongoing statistical analysis in the final survey paper.

      During the course of our key science program at Palomar, I have gradually grown into a role as project manager, and been able to efficiently coordinate our team’s work on everything from survey strategy to calibration issues, data analysis, software upgrades, and publications.


      Since my move to Caltech in 2015, about half of my time has been spent on an exciting new project that aims to upgrade the Wide-Field Infrared Camera (WIRC; Wilson et al. 2003) at the prime focus of the Hale telescope with spectro-polarimetric capabilities (WIRC+Pol), with the goal of studying the atmospheric composition and cloud structure of many nearby brown dwarfs and self-luminous exoplanets. Clouds induce polarization in the near-infrared radiation from these targets, and non-uniform emission over the observed projected disk, from patchy clouds, hot-spots, or rotation-induced oblateness will create a measurable polarization signal (on the order of 0.1–0.5%). Although cloudy atmospheres have been suggested (e.g., Wakeford & Sing 2015) by SED modeling of many brown dwarfs, a detected polarization signal would constitute a first solid confirmation of clouds, and would also provide an independent diagnostic of cloud properties (de Kok et al. 2011). WIRC-POL will combine polarimetry with time-resolved photometry and spectroscopy to greatly enhance our understanding of cloudy atmospheres. Additional science goals include high-precision photometry of transiting exoplanets to improve the ephemerides and cloud characterization of super-Earths detected by the K2 mission, retrieving masses from transit timing variations in multi-planet systems, and characterizing hot Jupiter compositions using secondary eclipse observations.

      The results from the WIRC-POL key science program in 2017-2020 will be able to better inform the atmospheric models that go into predicting observations of atmospheres of gas giants around nearby Sun-like stars in reflected visible light and low-resolution spectra with future space and ground based telescopes.

      My role on this project is as project scientist, co-managing all instrument component upgrades and science program definitions together with PI, Dimitri Mawet.


      The focus of my previous research projects have been optical- and submillimeter observations of debris disks. In one study we observed optical light scattered from dust grains in nearby debris disks (chosen from IRAS, ISO, and Spitzer IR-excess samples) using the polarimetric coronagraph PolCor-2, developed at Stockholm Observatory. The instrument used a high-speed CCD in conjunction with a coronagraphic technique, where the direct light from the star is blocked out and the stellar point-spread-function wings are suppressed, making the faint scattered optical light from the disk detectable. By also inserting a polarizing filter in different orientations, the angle and degree of polarization can be calculated and used to further enhance the contrast. Our high-contrast imaging of low surface brightness features successfully captured the AU Mic debris disk and produced interesting results for the controversial circumbinary disk around BD+31643 (Olofsson et al. 2012).

      In another project we performed a submillimeter survey of a sample of IR-excess main-sequence stars with the LABOCA bolometer array, observing at 870 μm on the APEX telescope, to determine the existence, extent, mass, and evolution of Kuiper Belt-like disk structures. Studies at mid- to far infrared wavelengths have shown excess emission in the spectral energy distribution (SED) of many young main-sequence stars, indicating warm circumstellar dust extending out to some tens of astronomical units (AU). Probing more extended, cold dust components, in Kuiper-Belt analogues reaching out to some hundreds of AU, requires sensitive submillimeter observations. Results from a precursor study of the β Pictoris Moving Group were presented in Nilsson et al. (2009). Out of the observed sample of stars in the larger survey, we detected 10 submillimeter disks with at least a 3-σ significance, increasing the number of currently known exo-Kuiper-Belts (inferred from cold extended dust disks detected in the submillimeter) from 27 to 32 (Nilsson et al. 2010).

      A third project involved spectroscopic observations of the disk around the young (∼ 12 Myr old) star β Pictoris. We imaged the disk in three emission lines using integral-field spectroscopy at the Very Large Telescope, and obtained the first complete image of Fe i emission (probing the neutral gas in the disk plane) and Ca ii (probing vertically more extended gas) (Nilsson et al. 2012). We concluded that the paucity of mid-plane Ca ii emission can be explained by optical thickness, and does not require the Ca gas to be spatially separated from the Fe gas, as previously suggested.


      In the following paragraphs, I briefly describe previous research and minor projects, which have led to publications.

      In 2008–2009 I worked together with an international team of engineers, biologists, and physicists constructing a concept for a mission to Jupiter’s moon Europa, involving everything from detailed evaluation of scientific instrumentation, mission sequences, basic spacecraft design, technology requirements, and cost estimations (Böttcher et al. 2009). The payload included an orbiter for investigating surface features and characterizing Europas internal structure, a lander for survey of the local environment, and a deployable cryobot for ice excavation and subsurface sampling in search for biosignatures. My main responsibility was the evaluation and conceptual design of a multispectral remote-sensing device envisioned for the orbiter.


      The mass-transfer onto a white dwarf (WD) in a cataclysmic variable (CV) system can occasionally excite non-radial pulsation modes in the WD, making it a so called ZZ Ceti star. In 2006 we performed photometric observations of five CVs with the Nordic Optical Telescope, and identified two new CV/ZZ Ceti hybrids (Nilsson et al. 2006). The results of our observations showed that the main pulsation frequencies agree with those found in previous CV/ZZ Ceti hybrids, and that data for the small sample of pulsating WDs in CV systems found so far, seems to indicate that the r − i color could be a good marker for the instability strip of this class of pulsating WDs.


      Future extremely large ground-based optical telescopes (ELTs), with apertures of around 40 m and light-collecting areas approaching 1000 m2, will enable measurements of photon-stream statistics and observations of astrophysical events on nanosecond time-scales. During 06/2004–06/2006 I participated in a concept study of an instrument on the planned European 100-m Overwhelmingly Large Telescope (now down-scaled to a 40-m Extremely Large Telescope), that can meet the required demands of rapid photon counting and handle extreme data rates (see, e.g., Barbieri et al. 2007). By using avalanche photodiode (APD) detectors in conjunction with digital hardware-correlators, we could demonstrate the advantages of real-time data reduction to statistical functions (e.g., 2:nd order correlation functions) for these demanding measurements, compared to traditional photometric methods, both in astronomical observations and laboratory settings (Nilsson 2005).


  • Publications