THE ASTROPHYSICAL JOURNAL, 555 : 410È417, 2001 July 1
( 2001. The American Astronomical Society. All rights reserved. Printed in U.S.A.
TWO NEW PLANETS FROM THE ANGLO-AUSTRALIAN PLANET SEARCH1
R. PAUL BUTLER,2,3 C. G. TINNEY,3 GEOFFREY W. MARCY,4 HUGH R. A. JONES,5 ALAN J. PENNY,6 AND KEVIN APPS7
Received 2000 December 25 ; accepted 2001 March 7
ABSTRACT
Precise Doppler measurements from the Anglo-Australian Telescope (AAT) UCLES spectrometer
reveal periodic Keplerian velocity variations in the stars HD 160691 and HD 27442. HD 160691 has a
period of 743 days, a semiamplitude of 54 m s~1, and a high eccentricity, e \ 0.62, typical of extrasolar
planets orbiting beyond 0.2 AU. The minimum (M sin i) mass of the companion is 1.97 M , and the
J
semimajor axis is 1.65 AU. HD 27442 has a 415 day period, a semiamplitude of 32 m s~1, and an eccentricity of 0.058. The minimum mass is 1.43 M , and the semimajor axis is 1.18 AU. This is the Ðrst
J
extrasolar planet orbiting beyond 0.2 AU that is in a circular orbit similar to solar system planets. The
photon-limited precision of AAT/UCLES measurements is 3 m s~1 as demonstrated by stable stars and
Keplerian Ðts to planet-bearing stars. In addition, we present conÐrmation of four previously announced
planets.
Subject headings : planetary systems È stars : individual (HD 160691, HD 27442)
1.
systems is richer and more complicated than previously
imagined.
Future breakthroughs will require greater measurement
sensitivity. In particular, the most pressing questions at the
moment are does the planet mass function continue to rise
through the Jupiter to the Neptune-mass range, and what
fraction of stars have ““ solar systemÈlike ÏÏ planets, i.e.,
Jupiter and Saturn analogs in circular orbits beyond 4 AU ?
Doppler precision of 3 m s~1 or better is required to
address these questions.
The Anglo-Australian Planet Search, therefore, has the
speciÐc goal of achieving a long-term precision of 3 m s~1.
In ° 2 we show that this goal has been reached. Section 3
reports the stellar characteristics and Doppler velocities of
the host stars for two giant planets that have emerged from
this survey. Section 4 presents velocities and orbital solutions from the Anglo-Australian Telescope (AAT) data for
four planets previously announced from the Keck,
CORALIE, and ESO planet surveys and is followed by a
discussion of our results.
INTRODUCTION
All D50 extrasolar planets discovered over the last 5 yr
have come from precision Doppler surveys of nearby dwarf
stars ranging in spectral type from late F through M4
(Mayor & Queloz 1995 ; Marcy & Butler 1998, 2000 ; Noyes
et al. 1997 ; Cochran et al. 1997 ; Vogt et al. 2000 ; Kurster et
al. 2000 ; Udry et al. 2000 ; Fischer et al. 2001). With one
exception (Marcy, Butler, & Vogt 2000), all the published
planets have Doppler velocity amplitudes greater than 30 m
s~1. With a measurement error of D10 m s~1, Doppler
velocity amplitudes of 30 m s~1 are the smallest that can be
easily detected. Detection of planets with smaller amplitudes, such as solar system analogs and Neptune-to-Saturn
mass planets in short period orbits, requires measurement
precision of 3 m s~1 or better (Butler & Marcy 1997 ; Butler
et al. 2001).
For the case of massive planets orbiting within 3 AU, the
D50 known planets constitute a useful statistical sample.
The substellar companion mass function abruptly rises at
5 M and continues to rise down to the detection limit near
J
1 M . About 7% of nearby stars have massive planets with
J periods less than 5 yr, including D0.75% with ““ 51
orbital
PegÈlike ÏÏ planets in 3È5 day circular orbits (Cumming,
Marcy, & Butler 1999 ; Butler et al. 2001). Although existing
planet hunting techniques are still in their infancy, two
primary results have emerged over the last 5 yr. Extrasolar
planets are common, and the architecture of planetary
2.
THE ANGLO-AUSTRALIAN DOPPLER SURVEY
The Anglo-Australian Planet Search began observations
in 1998 January and is currently surveying 200 stars. Initial
results from this work have been published by Tinney et al.
(2001).
The AAT program stars have been chosen to be among
the brightest, chromospherically inactive dwarf and subgiant stars ranging in spectral type from late F through
early M. Most of the stars are south of declination [20¡ to
prevent overlap with the Keck planet survey (Vogt et al.
2000), and most of the stars are brighter than V \ 7.5, consistent with achieving signal-to-noise ratio (S/N) Z 200 with
exposure times of 10 minutes or less. This S/N is required to
achieve photon-limited precision of 3 m s~1 (Butler et al.
1996).
High-resolution spectra, R D 45,000, are taken with the
UCLES echelle spectrometer (Diego et al. 1990) on the 3.9 m
AAT. These spectra span the wavelength range from 4820
to 8550 A . An iodine absorption cell (Marcy & Butler 1992)
provides wavelength calibration from 5000 to 6000 A . The
spectrometer point-spread function is derived from the
1 Based on observations obtained at the Anglo-Australian Telescope,
Siding Spring, Australia.
2 Department of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road, NW, Washington, DC 20015-1305 ;
paul=dtm.ciw.edu.
3 Anglo-Australian Observatory, P.O. Box 296, Epping, NSW 1710,
Australia.
4 Department of Astronomy, University of California, Berkeley, CA
94720 ; and Department of Physics and Astronomy, San Francisco State
University, San Francisco, CA 94132.
5 Astrophysics Research Institute, Liverpool John Moores University,
Twelve Quays House, Egerton Wharf, Birkenhead CH41 1LD, UK.
6 Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX,
UK.
7 Physics and Astronomy, University of Sussex, Falmer, Brighton BN1
9QJ, UK.
410
TWO NEW PLANETS FROM PLANET SEARCH
411
detailed shapes of the embedded iodine lines (Valenti,
Butler, & Marcy 1995 ; Butler et al. 1996). Similar systems
on the Lick 3 m and the Keck 10 m telescopes currently
provide photon-limited precision of 3 m s~1 (Butler et al.
1996 ; Butler & Marcy 1997 ; Vogt et al. 2000). No nightly
corrections (Walker et al. 1995) are applied to the Lick,
Keck, or AAT systems.
Figure 1 shows AAT velocities of four stable dwarf stars
with spectral types ranging from late F to early G. The rms
of the measured velocities for these stars range from 3.8 to
4.9 m s~1, consistent with measurement uncertainties of 3È4
m s~1 and intrinsic stellar variability of 2È3 m s~1 (Saar,
Butler, & Marcy 1998). Of these stars, the F8 dwarf HD
196378 (HR 7875) was previously reported to be a velocity
variable. Kurster et al. (1999) report, ““ Our best candidate
so far for having an orbiting planet is the F8V star /2 Pav
(\HR 7875).ÏÏ They report a velocity amplitude of 39.5 m
s~1 and a probable period of 42.5 days. Our velocities
would have revealed this planet, but we have failed to detect
it.
Figure 2 shows AAT velocities of four stable mid-G
dwarfs, while Figure 3 shows four stable stars ranging from
late G to early K. These stars exhibit a velocity rms ranging
from 2.3 to 5.0 m s~1, consistent with measurement uncertainty of D3 m s~1. Of these stars, the G8 V star q Ceti (HD
FIG. 2.ÈAAT Doppler velocities of stable G dwarfs
FIG. 1.ÈAAT Doppler velocities of stable late F and early G dwarfs.
These observations span the 2.8 yr of the AAT Planet Search Project.
10700, HR 509) is frequently used as a velocity standard by
precision velocity groups because it is bright, chromospherically inactive, and visible from both the northern and
southern hemispheres. By the late 1980s the CanadaFrance-Hawaii Telescope precision velocity survey had
shown an rms of 13 m s~1 for this star (Campbell, Walker,
& Yang 1988 ; Walker et al. 1995). Measurements taken
since 1994 November from the Lick precision velocity
program show an rms of 4.6 m s~1 (Butler et al. 1996), while
the ESO precision velocity program has reported an rms of
14 m s~1 for this star (Kurster et al. 2000). After 3 yr, the
AAT velocities for this star have an rms of 3.9 m s~1, as
shown in Figure 3.
Slowly rotating, chromospherically inactive, mainsequence dwarf stars ranging in spectral type from early G
to mid-M have been shown to be intrinsically stable at the
3 m s~1 level (Butler & Marcy 1997 ; Saar et al. 1998 ; Vogt
et al. 2000 ; this paper). In addition, Fischer et al. (2001) have
shown several class IV subgiants are stable at the 6 m s~1
level. In Figure 4 we present four stable subgiants from the
AAT survey. These stars are stable at the 3È4 m s~1 level.
With intrinsic stability at this level, it is possible to search
for planets around subgiants using the precision Doppler
technique. In addition, the stability of such stars makes
these stars excellent targets for stellar seismology campaigns, especially as they are predicted to have larger ampli-
412
BUTLER ET AL.
FIG. 3.ÈAAT Doppler velocities of stable late G and K dwarfs
tudes than main-sequence dwarfs. Bedding et al. (2001) have
recently detected a 17 minute oscillation in HD 2151 using
the AAT Planet Search hardware and software. They took
1200 velocity measurements over Ðve nights with an rms of
3.3 m s~1. On a timescale of hours the precision was 2.2 m
s~1.
3.
STELLAR CHARACTERISTICS AND ORBITAL SOLUTIONS
3.1. Stellar Characteristics of HD 160691
A total of 140 observations of HD 160691 (HR 6585, HIP
86796, GL 691, k Ara) have been made by Hipparcos
(Perryman et al. 1997), yielding a distance of 15.3 pc and a V
magnitude of 5.20. The resulting absolute magnitude is
M \ 4.28. The star is photometrically stable within HipV
parcos
measurement error, with photometric scatter of
0.002 mag. The Bright Star Catalog (Hoffleit & Jaschek
1982) assigns a spectral type of G3 IVÈV, in reasonable
agreement with the Hipparcos spectral type of G5 V. The
star is chromospherically inactive, with log R@(HK) \
[5.02 (Henry et al. 1996). Its chromospherically inferred
age is D6 Gyr. Combining Hipparcos astrometry of HD
160691 with the SIMBAD radial velocity ([9.0 km s~1)
yields an extremely low space velocity with respect to the
local standard of rest : U, V , W \ [4, ]3, ]3 km s~1.
Vol. 555
FIG. 4.ÈAAT Doppler velocities of stable G subgiants
Like many of the planet-bearing stars, HD 160691 is
extremely metal-rich. The [Fe/H] derived from highresolution spectroscopy is 0.28 ^ 0.04 (Favata, Micela, &
Sciortino 1997), in good agreement with our photometric
estimate of ]0.29. The lithium line at 6707.8 A was not
detected in high-resolution (R \ 100,000) spectra (Favata,
Micela, & Sciortino 1996). The mass of HD 160691 estimated from B[V , M , and [Fe/H] is 1.08 ^ 0.05 M .
Bol
_
3.2. Doppler V elocities and Orbital Fit for HD 160691
The 21 Doppler velocity measurements of HD 160691
obtained between 1998 November and 2000 November are
listed in Table 1 and shown graphically in Figure 5. The
best-Ðt Keplerian yields an orbital period of 743.5 days, a
velocity amplitude of 53.6 m s~1, and an eccentricity of 0.62.
The minimum (M sin i) mass of the planet is 1.97 M , and
J
the semimajor axis is 1.65 AU. The rms to the Keplerian Ðt
is 2.98 m s~1, yielding s2 \ 0.98.
l
3.3. Stellar Characteristics of HD 27442
A total of 107 observations of HD 27442 (HR 1355, HIP
19921, v Ret) have been made by Hipparcos, yielding a distance of 18.2 pc and a V magnitude of 4.55. The resulting
absolute magnitude is M \ 3.25. The star is photoV
metrically stable within Hipparcos
measurement error, with
photometric scatter of 0.003 mag. The Bright Star Catalog
assigns this star a spectral type of K2 IVa.
No. 1, 2001
TWO NEW PLANETS FROM PLANET SEARCH
413
TABLE 2
VELOCITIES FOR HD 27442
FIG. 5.ÈDoppler velocities for HD 160691. The rms to the best-Ðt
Keplerian (solid line) is 3 m s~1. The period is 743.5 days, and the semiamplitude is 53.6 m s~1. Assuming the mass of HD 160691 is 1.08 M , the
_
minimum (M sin i) mass of the companion is 1.97 M , and the semimajor
J
axis is 1.65 AU. The orbital eccentricity is 0.62, similar to other planets
orbiting beyond 1 AU.
On the basis of measured equivalent widths of Fe lines
observed at high resolution (R D 60,000), Randich et al.
(1999) determined the metallicity of HD 27442 to be [Fe/
H] \ 0.22. This is somewhat more metal-rich than earlier
estimates, including Elgaroy, Engvold, & Lund (1999), who
report [Fe/H] \ 0.00 based on averaging several previous
estimates, and the photometric estimate of Eggen (1993),
who found [Fe/H] \ 0.06. On the basis of evolutionary
tracks, Randich et al. (1999) Ðnd the mass of HD 27442 to
be 1.2 ^ 0.1 M and the age to be 10 Gyr, consistent with
_
subgiant status. They also report an upper limit for the
equivalent width of Li at 3 mA .
JD
([2,450,000)
Radial Velocity
(m s ~1)
Error
(m s~1)
831.0816 . . . . . . . .
1,118.1404 . . . . . .
1,525.9634 . . . . . .
1,527.0333 . . . . . .
1,630.9035 . . . . . .
1,745.3341 . . . . . .
1,767.3337 . . . . . .
1,768.3132 . . . . . .
1,828.1498 . . . . . .
1,830.0084 . . . . . .
1,856.1407 . . . . . .
1,857.0932 . . . . . .
1,919.0796 . . . . . .
1,921.1039 . . . . . .
[47.5
[11.5
2.1
[7.5
[44.4
[23.1
[13.5
[17.3
13.2
11.8
19.6
20.1
0.0
3.6
2.4
2.2
2.0
2.4
2.2
2.3
2.8
2.5
2.5
3.2
2.6
3.9
2.7
2.3
January. These measurements are graphically shown in
Figure 6. The best-Ðt Keplerian yields an orbital period of
415.2 days, a velocity amplitude of 32.5 m s~1, and an
eccentricity of 0.058. The minimum (M sin i) mass of the
planet is 1.35 M , and the semimajor axis is 1.16 AU. The
J
rms to the Keplerian
Ðt is 2.96, yielding s2 \ 1.65. This is
l
3.4. Doppler V elocities and Orbital Fit for HD 27442
Table 2 lists the 14 Doppler velocity measurements of
HD 27442 obtained between 1998 January and 2001
TABLE 1
VELOCITIES FOR HD 160691
JD
([2,450,000)
Radial Velocity
(m s ~1)
Error
(m s~1)
1,118.8874 . . . . . .
1,119.9022 . . . . . .
1,120.8870 . . . . . .
1,121.8928 . . . . . .
1,236.2864 . . . . . .
1,410.8977 . . . . . .
1,412.9773 . . . . . .
1,413.8981 . . . . . .
1,630.3042 . . . . . .
1,683.0926 . . . . . .
1,684.1320 . . . . . .
1,718.1184 . . . . . .
1,742.9096 . . . . . .
1,743.9240 . . . . . .
1,745.0440 . . . . . .
1,766.9330 . . . . . .
1,767.9689 . . . . . .
1,827.8973 . . . . . .
1,828.8866 . . . . . .
1,829.8890 . . . . . .
1,855.9058 . . . . . .
[4.3
[4.8
[4.7
[4.3
[20.1
[40.8
[41.2
[33.8
46.2
49.0
46.2
33.8
21.1
30.7
19.4
11.3
13.4
0.6
6.4
0.0
[1.1
3.3
2.9
2.8
2.9
4.0
2.8
5.5
2.8
3.3
3.8
4.0
3.6
3.2
3.8
3.2
3.1
3.2
3.0
3.8
3.5
4.6
FIG. 6.ÈDoppler velocities for HD 27442. The rms to the best-Ðt
Keplerian (solid line) is 2.9 m s~1. The period is 415.2 days, and the semiamplitude is 32.5 m s~1. Assuming the mass of HD 27442 is 1.2 M , the
_ this
minimum (M sin i) mass of the companion is 1.35 M . The orbit of
planet is similar to the Earth in both semimajor axis,J a \ 1.16 AU, and
eccentricity, e \ 0.058. This is the only known planet orbiting beyond 0.15
AU that is in a circular orbit, similar to solar system planets.
TABLE 3
ORBITAL PARAMETERS
Parameter
HD 160691
HD 27442
Orbital period P (days) . . . . . . . . . . . .
Velocity amplitudeK (m s~1) . . . . . .
Eccentricity e . . . . . . . . . . . . . . . . . . . . . . . .
u (deg) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Periastron time (JD) . . . . . . . . . . . . . . . .
M sin i (M ) . . . . . . . . . . . . . . . . . . . . . . . . .
J
a (AU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
rms (m s~1) . . . . . . . . . . . . . . . . . . . . . . . . .
743 (10)
53.6 (2)
0.62 (0.05)
305 (5)
2,451,626.8 (5)
1.97 (0.14)
1.65 (0.12)
2.98
415.2 (5)
32.5 (2)
0.058 (0.05)
347 (20)
2,451,849.3 (4)
1.35 (0.11)
1.16 (0.11)
2.96
414
BUTLER ET AL.
the only planet found to date orbiting beyond 0.15 AU in an
orbit as circular as solar system planets. Orbital parameters
for both HD 160691 and HD 27442 are listed in Table 3.
4.
ORBITAL SOLUTIONS FOR HD 134987, HD 13445,
HD 75289, AND HD 17051
Keplerian orbital parameters for four previously
announced planet candidates have been conÐrmed by our
AAT observations. The high precision of these data will be
a powerful tool in the search for additional planetary companions to these stars.
4.1. HD 134987
The planet orbiting HD 134987 (HR 5657, G5 V) was
announced from the Keck survey in 1999. As outlined in
Vogt et al. (2000), this star is similar to 51 Pegasi in spectral
type, enhanced metallicity, and low chromospheric activity.
A total of 11 observations of this star have been made by
the AAT between 1998 April and 2000 July. These observations are listed in Table 4 and shown graphically in
Figure 7. The AAT derived orbital parameters are consistent with Vogt et al. (2000).
4.2. HD 13445
A planet orbiting the nearby star HD 13445 (GL 86, HIP
10138, K1 V) was announced by the CORALIE team in
TABLE 4
VELOCITIES FOR HD 134987
JD
([2,450,000)
Radial Velocity
(m s ~1)
Error
(m s~1)
917.2282 . . . . . . . .
1,003.0032 . . . . . .
1,213.2775 . . . . . .
1,276.0475 . . . . . .
1,382.9573 . . . . . .
1,413.8811 . . . . . .
1,630.2677 . . . . . .
1,683.0609 . . . . . .
1,706.0960 . . . . . .
1,717.9564 . . . . . .
1,742.9340 . . . . . .
0.0
[14.6
[12.1
[19.2
75.2
30.9
89.3
25.5
[0.6
[4.8
[8.2
5.6
7.2
4.7
6.5
5.4
2.5
4.1
5.0
5.9
4.3
3.5
FIG. 7.ÈDoppler velocities for HD 134987 (G5 V). The solid line is a
Keplerian orbital Ðt with a period of 264.6 days, a semiamplitude of 53.7 m
s~1, and an eccentricity of 0.37, yielding a minimum (M sin i) of 1.63 M
for the companion. The rms of the Keplerian Ðt is 2.7 m s~1. These resultsJ
are consistent with Vogt et al. (2000).
Vol. 555
1999 (Queloz et al. 2000). In addition to the short-period
Keplerian orbit, they report a long-term linear trend of
[0.36 m s~1 day~1. The rms to their Keplerian Ðt plus
linear trend is 7 m s~1.
This star has been observed 23 times as part of the AngloAustralian Survey. These observations are listed in Table 5.
The Ðrst observation was made in 1998 January, and the
observations span 3 yr. As shown in Figure 8, these observations conÐrm the CORALIE result (Queloz et al. 2000).
The rms to our Ðt of a Keplerian plus linear trend is 3.66 m
s~1, yielding s2 \ 0.86, slightly better than expected based
l
on our estimated measurement error.
TABLE 5
VELOCITIES FOR HD 13445
JD
([2,450,000)
Radial Velocity
(m s ~1)
Error
(m s~1)
831.0350 . . . . . . . .
1,211.9651 . . . . . .
1,213.9815 . . . . . .
1,214.9298 . . . . . .
1,235.9312 . . . . . .
1,236.9078 . . . . . .
1,383.2736 . . . . . .
1,387.3139 . . . . . .
1,411.2467 . . . . . .
1,413.2313 . . . . . .
1,414.3164 . . . . . .
1,473.0974 . . . . . .
1,525.9320 . . . . . .
1,526.9613 . . . . . .
1,743.3292 . . . . . .
1,745.2853 . . . . . .
1,828.1337 . . . . . .
1,829.0121 . . . . . .
1,829.9880 . . . . . .
1,856.1052 . . . . . .
1,918.9660 . . . . . .
1,919.9811 . . . . . .
1,921.0019 . . . . . .
83.5
338.2
398.5
349.9
[277.2
[333.5
0.0
356.6
[406.7
[255.3
[88.1
[403.8
79.6
207.6
[459.4
[204.5
228.2
213.2
150.5
[160.9
[209.3
[41.7
97.0
4.0
5.7
5.3
4.7
5.2
5.5
4.8
4.0
4.9
4.0
4.2
4.3
4.9
4.8
6.5
5.3
5.2
5.4
6.3
6.2
4.6
4.8
5.0
FIG. 8.ÈPhased Doppler velocities for HD 13445 (K1 V). The solid line
is a Keplerian orbital Ðt with a period of 15.764 days, a semiamplitude of
379 m s~1, and an eccentricity of 0.046, yielding a minimum (M sin i) of
4.04 M for the companion. The rms of the Keplerian Ðt is 3.7 m s~1. A
J
linear trend
of [108.1 m s~1 yr~1 has been removed from these velocities.
These results are consistent with Queloz et al. (2000), with the possible
exception of the linear trend.
No. 1, 2001
TWO NEW PLANETS FROM PLANET SEARCH
415
TABLE 7
VELOCITIES FOR HD 17051
FIG. 9.ÈPhased Doppler velocities for HD 75289 (G0 V). The solid line
is a Keplerian orbital Ðt with a period of 3.508 days, a semiamplitude of 56
m s~1, and an eccentricity of 0.01, yielding a minimum (M sin i) of 0.45 M
for the companion. The rms of the Keplerian Ðt is 4.8 m s~1. These resultsJ
are consistent with Udry et al. (2000).
Our derived orbital parameters agree with the
CORALIE results within measurement error, with the possible exception of the linear trend. The simultaneous best Ðt
to the AAT velocities gives a linear trend of [0.296 ^ 0.005
m s~1 day~1, about 20% smaller than that, [0.36 m s~1
day~1, found by Queloz et al. (2000) from their CORALIE
data. With their lower precision CORAVEL data, they Ðnd
a steeper linear trend of [0.5 m s~1 day~1 spanning 1980
to the present. The trend for HD 13445 implies the presence
of an additional companion with a period much longer than
10 yr and an amplitude greater than 1 km s~1. It could be a
low-mass stellar companion (Queloz et al. 2000).
4.3. HD 75289
The CORALIE team (Udry et al. 2000) have announced a
““ 51 PegÈlike ÏÏ planet orbiting the star HD 75289 (HR 3497,
HIP 43177, G0 V). Their observations were carried out
between 1998 November and 1999 October and yield an
rms to a Keplerian Ðt of 7.5 m s~1. This star has been
observed 13 times as part of the Anglo-Australian Planet
Search. The Ðrst observation was made in 1998 January,
and the observations span 3 yr. These velocities are listed in
Table 6. As shown in Figure 9, these observations conÐrm
the CORALIE result. The rms velocity residual to our
Keplerian Ðt is 4.82 m s~1, yielding s2 \ 1.13.
l
JD
([2,450,000)
Radial Velocity
(m s ~1)
Error
(m s~1)
1,118.1146 . . . . . .
1,235.9437 . . . . . .
1,383.2779 . . . . . .
1,413.2245 . . . . . .
1,473.1000 . . . . . .
1,525.9348 . . . . . .
1,526.9763 . . . . . .
1,743.3364 . . . . . .
1,745.2936 . . . . . .
1,828.1295 . . . . . .
1,856.1093 . . . . . .
1,856.9355 . . . . . .
1,918.9703 . . . . . .
[55.1
39.5
[71.5
[72.3
0.0
47.9
76.1
[32.3
[50.1
66.7
62.5
44.5
[2.7
4.4
5.0
7.4
6.2
4.8
4.6
5.9
5.9
5.0
5.6
5.3
8.6
5.4
4.4. HD 17051
In 1998 June the ESO Precise Radial Velocity Survey
announced a planet orbiting HD 17051 (• Hor, HR 810,
HIP 12653, G0 V) with a 600 day orbit (Kurster et al. 1999 ;
see also Glanz 1998). On the basis of the same data, they
later announced the planet had a period of 320 days
(Kurster et al. 2000). Their data set consists of 95 measurements taken between late 1992 and early 1998. They report
a Keplerian semiamplitude of 67 m s~1. The rms velocity
residual to their Keplerian Ðt is 27 m s~1, while they estimate their measurement error to be 17 m s~1. They attribute the di†erence to stellar activity. A stellar Doppler
““ jitter ÏÏ of 20 m s~1 would account for the di†erence
between the internal measurement error and the observed
scatter to a Keplerian Ðt.
The Anglo-Australian Planet Search began observing
HD 17051 in 1998 November. A total of 13 observations
TABLE 6
VELOCITIES FOR HD 75289
JD
([2,450,000)
Radial Velocity
(m s ~1)
Error
(m s~1)
830.1656 . . . . . . . .
914.9334 . . . . . . . .
1,212.1495 . . . . . .
1,213.1426 . . . . . .
1,214.2518 . . . . . .
1,236.9418 . . . . . .
1,274.0100 . . . . . .
1,275.9947 . . . . . .
1,631.0085 . . . . . .
1,717.9152 . . . . . .
1,856.2491 . . . . . .
1,919.1969 . . . . . .
1,920.1472 . . . . . .
[0.8
[39.8
40.3
[39.9
33.8
15.6
31.7
[15.2
[25.4
[7.1
27.1
0.0
77.9
4.6
4.3
6.1
5.1
6.3
5.8
8.2
4.1
4.9
5.5
6.3
5.4
5.2
FIG. 10.ÈDoppler velocities for HD 17051 (G0 V). The solid line is a
Keplerian orbital Ðt with a period of 312 days, a semiamplitude of 63 m
s~1, and an eccentricity of 0.15, yielding a minimum (M sin i) mass of 2.13
M for the companion. The rms of the Keplerian Ðt, 10.4 m s~1, is twice the
J
internal
measurement error, consistent with the observed chromospheric
activity and youth of this star. These results are consistent with Kurster et
al. (2000).
416
BUTLER ET AL.
Vol. 555
TABLE 8
ORBITAL PARAMETERS
Parameter
HD 134987
HD 13445a
HD 75289
HD 17051
Orbital period P (days) . . . . . . . . . . . . .
Velocity amplitude K (m s~1) . . . . . .
Eccentricity e . . . . . . . . . . . . . . . . . . . . . . . .
u (deg) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Periastron time (JD) . . . . . . . . . . . . . . . .
M
(M )b . . . . . . . . . . . . . . . . . . . . . . . . . .
Star
_
M sin i (M ) . . . . . . . . . . . . . . . . . . . . . . . . . .
J
a (AU) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
rms (m s~1) . . . . . . . . . . . . . . . . . . . . . . . . . .
264.6 (5)
53.7 (4)
0.37 (0.12)
345 (10)
2,451,628.8 (4)
1.05
1.63
0.82
2.73
15.764 (0.005)
379 (1)
0.046 (0.002)
260 (3)
2,451,225.1 (0.2)
0.8
4.04
0.114
3.66
3.508 (0.001)
56 (1)
0.014 (0.005)
0
2,451,214.5 (0.001)
1.15
0.46
0.047
4.82
312 (5)
63 (4)
0.15 (0.05)
309 (5)
2,451,492.8 (4)
1.03
2.13
0.91
10.4
a Additional slope is [108.1 ^ 2 m s~1 yr~1.
b Stellar masses from discovery papers.
have been made through 2001 January, as listed in Table 7
and shown in Figure 10. The solid line in Figure 10 is the
best-Ðt Keplerian to the measured Doppler velocities.
Within measurement error, we conÐrm the orbital parameters of Kurster et al. (2000). The rms to the Keplerian Ðt to
the AAT data is 10.4 m s~1, about twice the estimated
measurement error.
As Kurster et al. (2000) note, HD 17051 is a young,
chromospherically active G0 V star, consistent with the
measured log R@(HK) \ [4.65 (Henry et al. 1996), and a
rotation period of 8 days (Saar & Osten 1997 ; Saar et al.
1997). Saar et al. (1998) Ðnd the Doppler ““ jitter ÏÏ corresponding to this level of activity to be 10 m s~1, consistent
with the rms to the Keplerian Ðt to the AAT data. Table 8
lists the orbital parameters for these four stars based on the
AAT measurements.
5.
DISCUSSION
The two newly detected planets announced here both
orbit beyond 1 AU. HD 160691 is typical of the examples of
such planets discovered to date, moving on a highly elliptical orbit. In contrast, HD 27442 is in an Earth-like orbit,
with a semimajor axis of 1.16 AU and an eccentricity of
0.058 ^ 0.05 (consistent with circular). This is the Ðrst
planet to be discovered orbiting beyond 0.15 AU in a circular orbit like the planets of our solar system.
The AAT data provide the Ðrst conÐrmation of orbital
parameters of four planets recently announced from the
Keck, CORALIE, and ESO planet searches. In addition,
the AAT data appear to rule out the claimed planet around
HD 196378.
Two broad strategies are being pursued by the various
groups carrying out precision velocity surveys. Several
groups are carrying out surveys with precision of 10È20 m
s~1. There are several advantages to this strategy. Relatively low S/N (D70) spectra, or small wavelength coverage
(D50 A ), are required to achieve this level of precision, and
data reduction strategies are simpliÐed. A precision of 10 m
s~1 allows the reliable detection of planets that induce
amplitudes of 30 m s~1 or larger. Since only these largeamplitude signals are detectable, chromospherically active
stars with associated Doppler ““ jitter ÏÏ of 10È20 m s~1, such
as HD 17051, remain viable candidates.
The other strategy is to pursue much higher precision,
3 m s~1 or better. This strategy carries several penalties,
including the need for large wavelength coverage (Z1000
A ), high S/N (Z200), and complex data analysis. The payo†
for such a strategy is the ability to detect lower mass planets
in short-period orbits and Jupiter-like planets in distant
([4 AU) orbits.
To illustrate this, consider the problem of detecting a
Jupiter analog. Jupiter induces a Doppler velocity variation
in the Sun with an amplitude of 12.5 m s~1. However, the
mean expectation value for sin i of n/4 reduces this to a
typical amplitude of 10 m s~1 in a Doppler velocity survey.
The top panel of Figure 11 shows synthetic observations of
a Jupiter analog with a measurement uncertainty of 5 m
s~1. The solid line is a best-Ðt Keplerian to this data. The
result is an unconvincing 2 p detection with no constraint
on orbital eccentricity. The lower panel shows the same
situation but for 2 m s~1 measurement uncertainties. In this
case a solid 5 p detection is made, and the eccentricity is
determined to within ^0.05. The eccentricity of a 3 p detection is poorly constrained to within ^0.2. Without knowledge of the orbital eccentricity, it is not possible to
categorize a Jupiter-mass companion at 5 AU as a solar
system analog. True solar system analogs must reside in
circular orbits.
The Anglo-Australian Planet Search has been surveying
the 200 brightest dwarf and subgiant stars ranging in spectral type from late F to early M and south of declination
[20¡ since 1998 January. Long-term photon-limited precision of 3 m s~1 has been achieved, unique among the
southern hemisphere precision velocity surveys. The long-
FIG. 11.ÈSimulated Jupiter signal observed with a precision of 5 and
2 m s~1. Solid lines are best-Ðt Keplerians to the simulated data sets. With
measurement precision of 5 m s~1, an unreliable 2 p detection is obtained
with no information on the orbital eccentricity. With precision of 2 m s~1,
a solid 5 p detection is made, and the eccentricity is determined to within
^0.05.
No. 1, 2001
TWO NEW PLANETS FROM PLANET SEARCH
term goal of this project is to maintain and improve this
precision for another decade to allow for the detection of
true solar system analogs, Jupiter-mass planets orbiting
beyond 4 AU.
We gratefully acknowledge the support and encouragement of the director of the Anglo-Australian Observatory, Brian Boyle. The superb technical support at the
Anglo-Australian Telescope has been critical to the success
of this projectÈin particular we acknowledge E. Penny,
417
R. Paterson, D. Sta†ord, F. Freeman, S. Lee, J. Pogson, and
G. Scha†er. We acknowledge support by NSF grant AST
99-88087 and travel support from the Carnegie Institution
of Washington (to R. P. B.), by NASA grant NAG5-8299
and NSF grant AST 95-20443 (to G. W. M.), and by Sun
Microsystems. We thank the Australian and UK Telescope
assignment committees (ATAC and PATT) for allocations
of telescope time. This research has made use of the
SIMBAD database, operated at the Centre de Donnees de
Strasbourg, France.
REFERENCES
Kurster, M., Endl, M., Els, S., Hatzes, A. P., Cochran, W. D., Dobereiner,
Bedding, T. R., et al. 2001, ApJ, 549, L105
S., & Dennerl, K. 2000, A&A, 353, L33
Butler, R. P., & Marcy, G. W. 1997, in ASP Conf. Ser. 134, Brown Dwarfs
Kurster, M., Hatzes, A. P., Cochran, W. D., Dennerl, K. Dobereiner, S., &
and Extrasolar Planets, ed. R. Rebolo, E. L. Martin, & M. R. Zapatero
Endl, M. 1999, in ASP Conf. Ser. 185, Precise Stellar Radial Velocities,
Osorio (San Francisco : ASP), 162
ed. J. B. Hearnshaw & C. D. Scarfe (San Francisco : ASP), 154
Butler, R. P., Marcy, G. W., Fischer, D. A., Vogt, S. S., Tinney, C. G., Jones,
Marcy, G. W., & Butler, R. P. 1992, PASP, 104, 270
H. R. A., Penny, A. J., & Apps, K. 2001, in ASP Conf. Ser., Planetary
ÈÈÈ. 1998, ARA&A, 36, 57
Systems in the UniverseÈObservation, Formation and Evolution, ed.
ÈÈÈ. 2000, PASP, 112, 137
A. Penny, P. Artymowicz, A. M. Lagrange, & S. Russell (San Francisco :
Marcy, G. W., Butler, R. P., & Vogt, S. S. 2000, ApJ, 536, L43
ASP), in press
Mayor, M., & Queloz, D. 1995, Nature, 378, 355
Butler, R. P., Marcy, G. W., Williams, E., McCarthy, C., Dosanjh, P., &
Noyes, R. W., Jha, S., Korzennik, S. G., Krockenberger, M., Nisenson, P.,
Vogt, S. S. 1996, PASP, 108, 500
Brown, T. M., Kennelly, E. J., & Horner, S. D. 1997, ApJ, 483, L111
Campbell, B., Walker, G. A. H., & Yang, S. 1988, ApJ, 331, 902
Perryman, M. A. C., et al. 1997, A&A, 323, L49
Cochran, W. D., Hatzes, A. P., Butler, R. P., & Marcy, G. W. 1997, ApJ,
Queloz, D., et al. 2000, A&A, 354, 99
483, 457
Randich, S., Gratton, R., Pallavicini, R., Pasquini, L., & Carretta, E. 1999,
Cumming, A., Marcy, G. W., & Butler, R. P. 1999, ApJ, 526, 890
A&A, 348, 487
Diego, F., Charalambous, A., Fish, A. C., & Walker, D. D. 1990, Proc.
Saar, S. H., Butler, R. P., & Marcy, G. W. 1998, ApJ, 498, L153
SPIE, 1235, 562
Saar, S. H., Huovelin, R. A., Osten, R. A., & Shcherbakov, A. G. 1997,
Eggen, O. J. 1993, AJ, 106, 80
A&A, 326, 741
Elgaroy, O., Engvold, O., & Lund, N. 1999, A&A, 343, 222
Saar, S. H., & Osten, R. A. 1997, MNRAS, 284, 803
Favata, F., Micela, G., & Sciortino, S. 1996, A&A, 311, 951
Tinney, C. G., Butler, R. P., Marcy, G. W., Jones, H. R. A., Penny, A. J.,
ÈÈÈ. 1997, A&A, 323, 809
Vogt, S. S., Apps, K., & Henry, G. W. 2001, ApJ, 551, 507
Fischer, D. A., Marcy, G. W., Butler, R. P., Vogt, S. S., Frink, S., & Apps,
Udry, S., et al. 2000, A&A, 356, 590
K. 2001, ApJ, 551, 1107
Valenti, J., Butler, R. P., & Marcy, G. W. 1995, PASP, 107, 966
Glanz, J. 1998, Science, 280, 2037
Vogt, S. S., Marcy, G. W., Butler, R. P., & Apps, K. 2000, ApJ, 536, 902
Henry, T. J., Soderblom, D. R., Donahue, R. A., & Baliunas, S. L. 1996, AJ,
Walker, G. A. H., Walker, A. R., Irwin, A. W., Larson, A. M., Yang, S. L. S.,
111, 439
& Richardson, D. C. 1995, Icarus, 116, 359
Hoffleit, D., & Jaschek, C. 1982, Yale Bright Star Catalog (4th ed. ; New
Haven : Yale Univ. Press)