Astronomical Science

HARPS Observations of the 2012 Transit of

Paolo Molaro1 ... Thy return Lorenzo Monaco 2 Posterity shall witness; years must roll Mauro Barbieri 3 Away, but then at length the splendid sight Simone Zaggia 4 Again shall greet our distant children’s eyes.

1 INAF–Osservatorio Astronomico di Since then only six other transits have ­Trieste, Italy taken place in three epochs, namely 2 ESO 1762–1769, 1874–1882 and 2004–2012; 3 Dipartimento di Fisica e Astronomia, there will not be another Venus transit Universitá degli Studi di Padova, Italy until December 2117. As shown in Fig- 4 INAF–Osservatorio Astronomico di ure 1, taken from an old book by Proctor Padova, Italy (1874), the cycle of the transits of Venus is precisely 243 years, so that of 2012 was similar to the one observed by On 6 June 2012 the black disc of Venus James Cook from Tahiti in 1769 during passed across the Solar disc, taking his first voyage around the world, which nearly eight hours to complete the led to the discovery of transit. The event was followed by mil- and the Cook Islands. The observations Figure 1. Sketch of the transits of Venus by lions of people worldwide. The transit by Cook and the astronomer Charles Richard Antony Proctor (Proctor, 1874). of Venus is one of the rarest astronomi- Green were recorded in a paper (Cook & cal events, occurring approximately Green, 1771) from which a figure is repro­ (± 96 076) kilometres, thanks to photo­ every 120 years. By means of HARPS duced in Figure 2. In 1716 the Royal graphic recording (Harkness, 1888). A spectroscopic observations, and using Astronomer Sir Edmund Halley, in an arti­ stunning video of the 2012 transit of the as a mirror, we detected the cle entitled “A new Method of determining Venus was captured by the Solar Dynam­ ­Rossiter–McLaughlin effect due to the the of the , or his Distance ics Observatory and can be seen on the eclipse by Venus of the Solar disc with from the ”, published in the Philo­ NASA website1. a precision of few cm s–1. The obser­ sophical Transactions (Halley, 1716), sug­ vation demonstrates that this effect can gested the use of observations of the be measured even for transits of exo- to find a value for the dis­ Science with the transit planets of Earth size, or even smaller, tance of the Earth from the Sun, i.e. the provided enough photons can be col- (AU): In 1874 the young Kingdom of Italy organ­ lected by a very high resolution and ised an astronomical expedition to extremely stable spectrograph, such as We therefore recommend again and ­Madhapur, India, with the aim of observ­ the planned HIRES instrument for the again, to the curious investigators of the ing the transit of Venus spectroscopically E-ELT. stars to whom, when our lives are over, for the first . In 1875 the Italian these observations are entrusted, that astronomer Tacchini sent a telegram from they, mindful of our advice, apply them- Calcutta: “First observations disturbed A bit of history selves to the undertaking of these obser- by small clouds – Good results spectro­ vations vigorously. And for them we scopic and ordinary – Spectrum of Venus In 1627, Kepler, in one of the first applica­ desire and pray for all good luck, espe- observed, details probably related to its tions of the Copernican view of the cos­ cially that they be not deprived of this atmosphere.” (Pigatto & Zanini, 2001). We mos, first predicted that there would coveted spectacle by the unfortunate know today that this was not possible, be transits of the inner planets and the obscuration of cloudy heavens, and that transit of Venus in 1631 in particular. the immensities of the celestial spheres, Figure 2. Drawing of the observations of the 1769 However, Kepler had died in 1630 and compelled to more precise boundaries, transit of Venus from Tahiti, taken from Cook & Green (1771). ­Gassendi, who was the first to document may at last yield to their glory and eternal a transit of Mercury, also missed the fame. ­predicted transit since this transit of Venus could not be observed from Astronomers did indeed organise major . However the young (22 years expeditions to the remotest parts of the old) British astronomer Jeremiah world to obtain an estimate of the magni­ ­Horrocks realised that transits of Venus tude of the astronomical unit. From an occur in pairs separated by eight years, ­initial value of about 10 million kilometres, and in 1639 he and his friend William as set by the ancient Greeks, the AU was Crabtree were the humans to observe the increased to 120–155 million kilometres phenomenon. Horrocks (1618–1641) at the end of the 18th century and further wrote a poem commemorating the event: refined by measurements during the tran­ sits of the 19th century to 149 341924

22 The Messenger 153 – September 2013 since the is too tiny we submitted a Director’s Discretionary passage, and were used to fix the refer­ to be detected in this way. However, after Time (DDT) proposal to observe the RM ence of the Solar radial velocity. 138 years, on the occasion of the last effect from Chile at night. The purpose transit, we found a new way to exploit was to detect the RM effect caused by The radial velocities were obtained by spectroscopic observations of the transit the transit of Venus using the almost full the HARPS pipeline. Simultaneous spec­ of Venus by detecting the very small Moon as a mirror (DDT 289.D-5015). tra were collected with a reference ThAr ­Rossiter–McLaughlin (RM) effect. When lamp and used to correct overnight a body passes in front of a star the con­ There are a few differences that have to instrumental drifts, which were about sequent occultation of a small area of be considered. Due to the different spa­ 40 cm s–1 at the beginning and undetect­ the rotating stellar surface produces a tial location, the transit of Venus as seen able, i.e. less than 20 cm s–1, after a distortion of the stellar line profiles, which from the Moon has a slightly different ­couple of hours of observations. The rela­ can be measured as a drift of the radial ­timing and projection on the Solar disc tive of the Moon with respect velocity. The phenomenon was observed from that which is seen from Earth. The to the Sun and the usual one due to the in eclipsing binaries by McLaughlin (1924) Moon was about eight degrees ahead observer were accounted for. The bar­ and Rossiter (1924) and in the Jupiter- of the Earth and Venus reached the Sun– ycentric radial velocities were then meas­ like planets (Queloz et al., 2000), but Moon alignment with a delay of about ured relative to the out-of-transit Solar becomes increasingly difficult to observe two hours. The transit was also slightly radial velocity, which was measured as when the eclipsing body is as small as longer than from Earth since the Moon the mean value of all the out-of-transit a planet and, in particular, for an Earth- was above the rotation plane of the observations and had a value of 102.53 ± sized planet such as Venus. The RM Earth–Sun. 0.10 m s–1. This latter quantity comprises effect has been observed in about the zero offset of the mask used in the 60 extrasolar planets, providing impor­ Our observations with HARPS began at cross-correlation by the HARPS pipeline, tant information on the angle between 2 h:44 m UT on 6 June 2012 when the plus specific motions of the Sun on that the sky projection of the orbital axis and Moon reached about 40° above the hori­ day. The resulting radial velocities and the stellar rotation axis, and showing zon at about mid-transit and continued their temporal evolution in phase with that several have tilted orbits. until the end of the transit, and for some the transit are shown in Figure 4. The val­ time afterwards. The observations com­ ues show clearly the Solar 5-minute Most surprisingly, the integrated light of prised a series of 245 spectra, each with oscillations of the p-modes, but also the Sun at high spectral resolution, which an integration time of 60 s with 22 s of a clear trend in phase with the passage is needed to reveal the RM effect, is readout, and delivered a signal-to-noise of Venus in front of the receding hemi­ extremely difficult to obtain with direct ratio of ~ 400, each at 550 nm at a sphere, with a half amplitude modulation observations. The simplest workaround is resolving power of R ~ 115 000. The first of ~ 80 cm s–1. to collect the Solar light as reflected by 227 observations cover the phases the Moon or by other minor bodies of between about mid-transit to the end and For the configuration of Venus, Sun, the Solar System. Chile was out of the for about two hours after the passage. Moon and Earth it is possible to make a visibility strip for the transit of Venus (c.f., Eighteen additional observations were very accurate model for the RM effect, Figure 3); a few weeks before the event also taken at twilight a few hours after the which considers Solar differential rota­ maps.com - eclipse M. Zeiler,

Figure 3. Visibility of the transit of Venus on 6 June 2012 as seen from the Earth. From the Moon it is slightly differ­ ent, being longer and delayed by a couple of hours (see text).

The Messenger 153 – September 2013 23 Astronomical Science Molaro P. et al., HARPS Observations of the 2012 Transit of Venus

tion, limb darkening and Solar axis incli­ Figure 4. The Solar disc from the Solar  2NK@QQNS@SHNM@WHR nation. The theoretical model is com­ 5DKNBHSX:LR< Dynamics Observatory image of the 5DMTRO@SGbS Sun on 6 June 2012 with the path of pared with the observations in Figure 4  the transit of Venus as seen from the as a continuous blue line. Note that this centre of the Moon drawn in. Radial (MFQDRR is not a fit to the data but an independent  velocity measurements (the red points theoretical model of the RM effect. Once connected by the red line) from the 227 Solar spectra during the transit  $FQDRR the p-mode oscillations have been filtered are shown (x-axis is time). The thin 2NK@QDPT@SNQ out, the radial velocity difference be- continuous blue line shows the theo­ tween the model and the observations is  retical RM effect. The radial veloci- –4 cm s–1. This offset can be entirely ties show clearly the 5-minute Solar l ­oscillations as well as a decrease in ascribed to our ability to establish the the second part of the transit, up to –1 out-of-transit Solar radial velocity needed l 80 cm s , due to the partial coverage for the normalisation; this latter is known of the receding Solar hemisphere by –1 Venus. with an uncertainty of ~ 10 cm s . l

2TM This is the smallest radial velocity effect @QBLHM l ever detected with HARPS and it dem­ onstrates that the RM effect can be  Figure 5. The tracks and dates of the detected despite the fact that the radial Earth’s transits across the Solar disc as seen from Jupiter are shown, velocity change due to Venus is compa­  adapted from a figure in Meeus (1989). rable to that of the Solar oscillation. The For the transit in 2014, the Moon, at RM effect is one of the most promising a distance of 1(47.4? from Earth, is also ways by which astronomers plan to study shown. The Moon will produce its  own eclipse of a small portion of the exoplanets and new high resolution Solar disc, with an estimated RM ­spectrographs at the E-ELT are also pro­ ,NNM effect of only ~ 2 cm s–1.  posed for this purpose. The present $@QSG observations show that an RM effect as small as those caused by Earth-like plan­  ets eclipsing their host star could be detected even in the presence of a com­ 2084. More interestingly the Earth will be among exoplanets and this will represent parable stellar jitter. seen transiting the Sun from Jupiter on a sort of unique test-bench experiment. 5 January 2014, and then again in 2026. As for all unique experiments, a certain As shown in Figure 5, where all the pas­ dose of good luck is required in order Goodbye until 2117 (or is it?) sages are drawn, the transit of 2026 will to have a clear sky. This is something that be a grazing one, quite unfavourable for Jeremiah Horrocks ­certainly had when The next transit of Venus will occur in any kind of observation. So effectively he first observed the transit of Venus December 2117 so the observations the transit occurring next year is a unique through the cloudy ­English sky. described here cannot be repeated or event, providing an opportunity to repeat improved by any other kind of instrument. the experiment, but with an Earth transit The only other transit visible directly instead of that of Venus. References from the Earth in the next few years will Cook, J. & Green, C. 1771, RSTP, 61, 397 be the one of Mercury on 9 May 2016, The predicted RM on this occasion will Halley, E. 1716, Phil. Trans. Royal Society, 29, 454 which occurs during the day in South also be even smaller. From Jupiter the Harkness, W. 1888, AJ, 8, 108 America, between 11:12 and 18:42. angular size of Sun is 369 arcseconds McLaughlin, D. B. 1924, ApJ, 60, 22 ­However, we may not need to wait for and of the Earth 4.2 arcseconds, so that Molaro, P. et al. 2013, MNRAS, 429, L79 Meeus, J. 1989, Transits, (Richmond, Virginia: 105 years for another similar opportunity. the predicted modulation of the RM effect Willmann-Bell) In the Solar System other transits can is only about about 20 cm s–1. Interest­ Pigatto, L. & Zanini, V. 2001, Journal of Astronomical be seen from the other planets too, with ingly, together with the Earth, the Moon History and Heritage, 4, 43 only the exception of the innermost will also produce a transit on the Solar Proctor, R. A. 1874, Transits of Venus: A Popular Account of Past and Coming Transits, from the planet Mercury. More interestingly, the surface. The transit of the Moon will be First Observed by Horrocks, A.D. 1639, to the Earth too is seen transiting in front of delayed by about four fours from the Transit of A.D. 2012, (London: Longmans, Green the Sun from the outer planets. A transit transit of the Earth.. The RM effect due and Co.) occurs every time the heliocentric con­ to the Moon will only be about 2 cm s–1 Queloz, D. et al. 2000, A&A, 359, L13 Rossiter, R. A. 1924, ApJ, 60 junctions take place near one of the and is probably beyond the limit of our nodes of their orbits. Accurate computa­ technique. This is a unique con­figuration tions of the transits of all planets of where we can possibly detect the RM Links the Solar System were made by Meeus effect of an Earth-size planet together 1 Solar Dynamics Observatory Venus transit video: (1989). He found that the Earth will be with its moon. The presence of http://www.nasa.gov/mission_pages/sunearth/ seen transiting the Sun from Mars in could be quite a common con­figuration multimedia/venus-transit-2012.html

24 The Messenger 153 – September 2013