NASA’s Kepler and TESS transit-survey space missions have led to the identification of thousands of candidate planets orbiting other Sun-like stars with radii which in some cases are as small as our own Earth. The radial-velocity method of mass determination, pioneered by 2019 Nobel prizewinners Michel Mayor (Geneva) and Didier Queloz (Cambridge), offers direct measurement of an extra-solar planet’s mass. At Earth-like distances from Sun-like stars, where a planet might conceivably harbour liquid water, further characterisation has been hampered by stray signals from sunspots and turbulence in the host stars’ outer layers. During the summer of 2019, Cameron has developed a novel mathematical technique that strips away these confusing signals. The present proposal is for a PhD project to exploit the opportunities arising from this development.
The radial-velocity technique works by dispersing the light of a star into its constituent colours, revealing the narrow, dark absorption lines arising from chemical elements present in the star’s surface layers. As the star “wobbles” about its centre of mass with the planet, the wavelengths of the spectral lines oscillate by a tiny amount that is nonetheless detectable by radial-velocity spectrometers such as the Geneva-built HARPS, HARPS-N and ESPRESSO. The exquisite precision and stability of these instruments would allow detection of Earth analogues but for the fact that stars rotate and do not have immaculate surfaces.
Former PhD student Raphaëlle Haywood (St Andrews 2011-2015), working with Cameron and Queloz, used solar spectra to identify the sources of these confusing signals. Dark starspots cause time-varying distortions in the shapes of all the spectral lines simultaneously as the spots rotate across the face of the star. Millions of turbulent cells of hot gas, rising and falling at speeds of 1 or 2 km per second, produce anomalous velocity patterns in regions permeated by short-lived magnetic fields.
Cameron’s novel signal-separation technique passes the recorded starlight through a mathematical “filter” that is sensitive to the changes in line shape caused by stellar activity, but not to the shifts caused by genuine dynamical motion. This partitions the velocity signal into a shape-sensitive component driven by activity, and a shift-sensitive part in which planet signals are preserved with high fidelity. Cameron’s paper, currently in pre-submission revision after review by co-authors, includes studies of solar spectra and a previously-known exoplanetary system orbiting the star HD219134. These demonstrate that habitable-zone terrestrial planets orbiting stars only slightly less massive then the Sun may now be within reach for mass determination by existing instruments.
In 80 nights/year of guaranteed observing time since mid-2012, the HARPS-N “rocky-planet search” campaign has amassed hundreds of observations apiece on several dozen bright stars. Planets with rock/iron compositions should induce “wobbles” with signal amplitudes of a few tens of cm per second, which were previously masked by stellar activity but should now be within our grasp.
The student will participate in HARPS-N guaranteed-time observing campaigns, with the aim of gaining familiarity with the instrumentation and searching for planet candidates with orbital periods ranging from tens to hundreds of days. The first year of study will involve developing software tools that apply the new signal-separation technique to data products from the HARPS-N data pipeline. The next stage will be searching the “cleaned” velocity-shift signals for evidence of periodic wobbles, measuring the planet masses and analysing their statistical significances. For transiting planets, such mass determinations also reveal the density, and hence bulk composition, of the planet. HARPS-N encourages students to “champion” the planets they analyse, ensuring lead authorship on the discovery papers resulting from their work. The outcomes of the thesis are likely to include significant improvements in the mass determinations of known planets, and discoveries of new low-mass planets whose velocity signals were previously below the detection threshold.