GJ 581 update: Additional evidence for a Super‐Earth in the habitable zone

Abstract

We present an analysis of the significantly expanded HARPS 2011 radial velocity data set for GJ 581 that was presented by Forveille et al. (2011). Our analysis reaches substantially different conclusions regarding the evidence for a SuperEarth-mass planet in the star's Habitable Zone. We were able to reproduce their reported χ 2 ν and RMS values only after removing some outliers from their models and refitting the trimmed down RV set. A suite of 4000 N-body simulations of their Keplerian model all resulted in unstable systems and revealed that their reported 3.6σ detection of e=0.32 for the eccentricity of GJ 581e is manifestly incompatible with the system's dynamical stability. Furthermore, their Keplerian model, when integrated only over the time baseline of the observations, significantly increases the χ 2 ν and demonstrates the need for including non-Keplerian orbital precession when modeling this system. We find that a four-planet model with all of the planets on circular or nearly circular orbits provides both an excellent self-consistent fit to their RV data and also results in a very stable configuration. The periodogram of the residuals to a 4-planet all-circular-orbit model reveals significant peaks that suggest one or more additional planets in this system. We conclude that the present 240-point HARPS data set, when analyzed in its entirety, and modeled with fully self-consistent stable orbits, by and of itself does offer significant support for a fifth signal in the data with a period near 32 days. This signal has a False Alarm Probability of < 4% and is consistent with a planet of minimum mass 2.2M⊕, orbiting squarely in the star's Habitable Zone at 0.13 AU, where liquid water on planetary surfaces is a distinct possibility.

Hα ACTIVITY OF OLD M DWARFS: STELLAR CYCLES AND MEAN ACTIVITY LEVELS FOR 93 LOW-MASS STARS IN THE SOLAR NEIGHBORHOOD

“…We take particular note of GJ 581, both because of the considerable interest in its planetary system (e.g., Bonfils et al 2005b;Mayor et al 2009;Vogt et al 2012), and because our observed periodicity confirms the discovery of Gomes da Silva et al (2012). It is important to note that our periodogram analysis (Figure 3(c)) shows a peak at 448 days with higher .…”

Stellar activity masquerading as planets in the habitable zone of the M dwarf Gliese 581

“…Striking similarities exist with our Planets b, c Removed Figure 2: Periodograms for the HARPS RVs before (blue) and after (red) correcting for stellar activity, with the planets successively subtracted. In the bottom panel we also show (pink) the periodogram after subtracting four circular Keplerian signals, illustrating that the signal interpreted as 581 g (7,9) was caused by fitting a sinusoidal signal to the 581d signal and performing, in essence, an incomplete correction for stellar activity. for RV versus I Hα suggests the stellar lines used for RV determination are emitted from a region of convective overshoot.…”

“…This spectrum has anomalously high emission in both Hα and sodium, leading us to conclude it might be associated with a flare event. We note that this was not one of the spectra omitted from the most recent HARPS analysis (9).…”

Radial-velocity fitting challenge

“…There is several examples in the literature, where by analyzing the same RV measurements different teams detected different planetary configurations. This is the case of the famous planetary system GJ581, for which the number of planet detected is ranging between 3 and 6 (Hatzes 2016;Anglada-Escudé & Tuomi 2015;Robertson et al 2014;Baluev 2013;Vogt et al 2012;Gregory 2011;Vogt et al 2010;Mayor et al 2009), of HD40307, for which 4 to 6 planets have been announced (Díaz et al 2016;Tuomi et al 2013), and GJ667C, for which 3 to 7 planets have been detected (Feroz & Hobson 2014;Gregory 2012). All those systems are affected by stellar signals, and therefore depending on the model used to analyze the data, different teams arrives to different conclusions.…”