A search for star-planet interactions in chromospheric lines

Abstract

Massive planets in very close orbits around their central stars can induce so-called star-planet interactions (SPI), which may be of magnetic or gravitational nature. In both cases, SPI can potentially cause recurring chromospheric emission on the host star visible in Ca II H & K and/or H$\alpha$. The emission would be bound to the planetary orbit, not to the rotation period of the star. We searched for SPI in a sample of 7 stars with massive close-in planets using high-resolution spectroscopic data taken at HRS (HET) and FEROS (La Silla). We find no periodically recurring emission in the planet-hosting stars. In the case of HD 41004 AB, a binary system consisting of a K dwarf and an M dwarf, where the M dwarf is orbited by a brown dwarf companion, we find signs of cyclic variation in the Ca II K and H$\alpha$ emission lines that could be associated to interactions between the M dwarf and its companion. We present our first results of this interesting system that may become an important system for the understanding of SPI. Comment: 5 pages, 9 figures, proceeding of the Cool Stars 16 Conference (Seattle 2010)

Full text

Page 1

arXiv:1012.1720v1 [astro-ph.SR] 8 Dec 2010
**Volume Title**
ASP Conference Series, Vol. **Volume Number**
**Author**
c ?**Copyright Year** Astronomical Society of the Pacific
A search for star-planet interactions in chromospheric lines
Lenz, L. F.1, Reiners, A.1, and K¨ urster, M.2
1Universit¨ at G¨ ottingen, Institut f¨ ur Astrophysik, Friedrich-Hund-Platz 1,
37077 Goettingen, Germany
2Max Planck Institut f¨ ur Astronomie, K¨ onigstuhl 17 69117 Heidelberg,
Germamy
Abstract.
so-called star-planet interactions (SPI), which may be of magnetic or gravitational na-
ture. In both cases, SPI can potentially cause recurring chromospheric emission on the
host star visible in Ca ii H & K and/or Hα. The emission would be bound to the plan-
etary orbit, not to the rotation period of the star. We searched for SPI in a sample of
7 stars with massive close-in planets using high-resolution spectroscopic data taken at
HRS (HET) and FEROS (La Silla). We find no periodically recurring emission in the
planet-hosting stars. In the case of HD 41004 AB, a binary system consisting of a K
dwarf and an M dwarf, where the M dwarf is orbited by a brown dwarf companion, we
find signs of cyclic variation in the Ca ii K and Hα emission lines that could be associ-
ated to interactionsbetweenthe M dwarfand its companion. We presentour first results
of this interesting system that may become an important system for the understanding
of SPI.
Massive planets in very close orbits around their central stars can induce
1.Introduction
Star-planet interactions (SPI) in stellar chromospheres could occur as recurring en-
hanced Ca ii H & K flux following the periodicity of the planet orbiting its host star.
So far signs for SPI were found only in few systems and the empirical evidence is
still heavyly debated (Poppenhaeger et al. 2010b; Shkolnik et al. 2008; Scharf 2010).
The possible physical scenarios for SPI are gravitational and/or magnetic interactions.
Gravitational interaction in the system could lead to tidal bulges on the surface of the
star, changing the local geometry. It is perhaps conceivable that this may in turn favor
magnetic reconnection thereby enhancing the observed stellar activity. Tidal interac-
tion would lead to two enhancement peaks over the planetary phase. In the case of
magnetic interaction, the reconnection of field lines from the planet and the star could
lead to enhanced stellar activity peaking once per orbit (Cuntz et al. 2000). Recent ob-
servations for HD 179949 even suggest an “On/Off” behavior: after some years the
periodicity of the Ca ii K emission changed from the planetary to the stellar rotation
period (Shkolnik et al. 2008).
1

Page 2

2
Lenz, L. F., Reiners, A., K¨ urster, M.
2.Observations
We obtained FEROS (La Silla) and HRS (HET) spectra for a sample of 7 stars, each of
which has a verified massive close-in planet. The spectra have a S/N of ca. 30 spectral
pixel−1(FEROS) and 90 spectral pixel−1(HRS) in the Ca ii K core. The parameters of
the systems are listed in Table 1. A typical spectrum is shown in Fig.1.
Figure 1.
ESO/MPG Telescope, La Silla
Ca ii K line in a spectrum of HD 75289observedwith FEROS at the 2.2
Table 1. Parameters of the program stars
StarSpectral-Type
dMPorb
[d]
Spectrograph#
[AU] [MJup] spectra
υAnd
τBoo
HD 217101
HD 212301
HD 75289
Gl 876
HD 41004AB
F8 V
F7 V
G8 IV
F8 V
G0 V
M4 V
M2
0.059
0.046
0.073
0.036
0.046
0.021
0.018
0.69
3.9
1.33
0.45
0.42
0.02
18.4
4.62
3.31
7.12
2.457
3.51
1.94
1.33
HRS
HRS
FEROS
FEROS
FEROS
FEROS
FEROS
11
12
7
6
7
4
9
FEROS spectra were reduced with the FEROS reduction pipeline. A continuum
normalization was applied to the reduced spectra, using two normalization regions in
the pseudo continuum left and right of the Ca ii K line core.

Page 3

Star-Planet Interactions
3
The HRS data reduction was performed with the MIDAS software following stan-
dard reduction techniques for optical spectra except for the normalization with flat-
fields. The HRS flatfields cannot be used for the Ca ii K region since the separate
calibration fibre is wider than the science fiber and the flatfield orders overlap in the
region around 4000Å. Thus we reduced the data without flatfielding. For our differen-
tial analysis this is not crucial. Normalization of the HRS spectra was performed by
fixing the flux of each spectrum to unity at 3929.5Å. To remove the underlying flux
distribution each spectrum was divided by a reference spectrum which was the first ob-
served spectrum of all spectra from one target. A polynomial was fitted to this quotient
representing the missing blaze function. Finally each spectrum was divided by its fit
for normalization.
3. Results
To analyse the spectra for variations in the Ca ii K and Hα lines we cross correlate all
spectra of one target in order to correct for spectral shifts due to the barycentric Earth
movement and spectrograph instabilities and compute the respective mean spectrum for
each dataset. We subtract these mean spectra binwise from each spectrum to receive
the residual fluxes. Variable emission in the line core due to chromospheric activity
would result in a bulge in the residual spectrum. For most of our target stars we find no
variability in the line residuals that could be associated with SPI. Planetary masses are
between 0.02 and 3.9 MJup, all planets orbit their hosts at distances closer than 0.08 AU,
orbital periods are shorter than 5 d. The data quality of both spectrographs should allow
us to detect variations in the residuals on the order of those reported in Shkolnik et al.
(2008). We typically took 6 or more observations per star at a cadence of one spectrum
per night for the FEROS data and one spectrum every 6 nights for the HET data. Fig.2
shows two typical residual plots of the target stars HD 75289 and HD 217107.
Figure 2.Ca ii residuals for two of the program stars.
The υAnd spectra observed with the HRS at HET show variations in the Ca ii K
flux. Theintegrated fluxresiduals plotted overtheorbital phase oftheplanet (porb=4.6d)
give no significant variation that would indicate a dependence between the chromo-
spheric activity and the orbiting planet. Combining our data with FOCES data of υ
And (Poppenhaeger et al. 2010a) indicates a period of the variability of 9.5 days which
is close to the rotation period of the star. Fig.3 gives an overview of the total residual
flux plotted over both periods.

Page 4

4
Lenz, L. F., Reiners, A., K¨ urster, M.
Figure 3.
period (porb=4.6d) and the estimated rotation period of 9.5d
υAnd: total residual Ca ii K line flux folded with the planetary rotation
3.1.A brown dwarf companion to HD 41004B
HD 41004AB is a binary system of a K dwarf and a M dwarf with a separation of
0.5”. We could not separate the two stars so that our spectra contain light of both
components. Furthermore the B component (the M dwarf) is orbited by a brown dwarf
companion (MBD=18.4 MJup, Porb=1.33 d). The Ca ii K and H α lines show variation
in the residual plots. For the residual plots for HD 41004AB we subtract a minimal
spectrum calculated from all spectra instead of a mean spectrum. The residual plots are
shown in Fig4.
We analyze nine spectra of this system, which where recorded in the nights be-
tween Nov 28 until Dec 10. Two of the spectra for HD 41004 AB have cosmics close
to the Ca ii K line. Thus they are only used for the analysis of the Hα line and excluded
from the Ca ii K line analysis. The residuals show variability in both the Ca ii K and the
Hα lines. At this point it is not clear whether the variability is caused by interactions
of the M dwarf and its brown dwarf companion. Alternative scenarios include intrinsic
variability of the A and/or B component and gravitational forces of the binary system.
Nevertheless, the short timescale of the variation is consistent with the brown dwarf’s
orbital period and renders Interaction of the brown dwarf and its host star a very likely
explanation. The integrated line flux of the Ca ii K and Hα residuals folded with the
orbital period of the brown dwarf is shown in Fig. 5. So far, our data set is too small to
present a meaningful period analysis. However, our data are consistent with a variation
of both lines with the 1.33d period and in phase with each other. It is therefore quite
likely that we have found star-companion interactions in this system.
4. Conclusions
We find no flux variations in phase with the planetary orbit in the chromospheric Ca ii K
and Hα lines in the spectra of our sample stars that have close-in planets. The detected
variability of υAnd is probably due to the rotation of the star and not bound to the
planet.
HD41004 ABpotentially isavery good candidate to observe interactions since the
mass-ratio of the M dwarf and its brown dwarf companion is between that of interacting
binaries and star-planet systems. More spectra will be obtained in the future to search
for periodicities of the flux variations.

Page 5

Star-Planet Interactions
5
Figure 4.
data. upperpanel: minimumspectrumcomputedfromthe normalizedspectra; lower
panel: residual flux; right: variability in Hα line cores of the HD 41004 AB FEROS
data. upperpanel: minimumspectrumcomputedfromthe normalizedspectra; lower
panel: residual flux
left: Variability in Ca ii K line cores of the HD 41004 AB FEROS
Figure 5.
orbital period of the brown dwarf (porb=1.33d); right: Total residual Hα line flux
folded with the same period
HD 41004 AB: left: Total residual Ca ii K line flux folded with the
Acknowledgments.
1664/4-1. The Hobby-Eberly Telescope (HET) is a joint project of the University
of Texas at Austin, the Pennsylvania State University, Stanford University, Ludwig-
Maximilians-Universit¨ at M¨ unchen, and Georg-August-Universit¨ at G¨ ottingen. TheHET
is named in honor of its principal benefactors, William P. Hobby and Robert E. Eberly.
A.R. acknowledges research funding from the DFG under RE
References
Cuntz, M., Saar, S. H., & Musielak, Z. E. 2000, ApJ, 533, 151
Poppenhaeger, K., Lenz, L. F., Reiners, A., & Schmitt, J. H. M. M. 2010a, ArXiv e-prints.
1010.5632
Poppenhaeger,K., Robrade, J., & Schmitt, J. H. M. M. 2010b, A&A, 515, 98
Scharf, C. A. 2010, ApJ, 722, 1547
Shkolnik, E., Bohlender, D. A., Walker, G. A. H., & Cameron, A. C. 2008, ApJ, 676, 628