Results 1 to 8 of 8

Thread: The search for companions of GJ 687 (a short history)

  1. #1
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912

    Post The search for companions of GJ 687 (a short history)

    The Search for Companions of GJ 687

    In hopes it will be of interest, here is the chronology of a long-running astronomical debate, with the wrong turns it took and its current (and assuredly not final) resolution. It is presented as it is written, in multiple parts. Comments, corrections, & criticisms are welcome.

    There is a nearby red dwarf generally known as GJ 687 or BD+68°946. In 2014 it was found to unambiguously possess a Neptune-mass planet, but that was not the first time a companion had been reported for that star. Many astronomy papers published prior to 1995 indicate that GJ 687 was a spectroscopic or astrometric binary, or that it had a substellar companion massive enough to be a brown dwarf. These were false positive detections from faulty instruments and various other errors. A history of the search for a companion to this star is needed to clarify what happened, but the sequence of events is difficult to follow because of the ever-changing names used for that star. Here, any star name equivalent to GJ 687 is underlined.

    ===

    1949: van de Kamp & Lippincott (Sproul Observatory), in "Parallaxes and Proper Motions of Twelve Nearby Stars", make a few notes regarding Cin. 18, 2354: "Strong series. Indications of trend in y." This is the star that van de Kamp calls AOe 17415-6 in his 1930 paper, "List of Stars Nearer than Five Parsecs", and in many subsequent revisions of this material (1940, 1945, 1953, 1955, 1969).

    1957: Krön, Gasciogne & White (1957: "Red and Infrared...") append a footnote to the entry on Yale 4029 that "Van de Kamp suspects AsB [astrometric binary—REM]".

    1967: Wilson, in "Radial Velocities of dK and dM Stars", gives Vyssotsky 322 (a.k.a. +68°946) a radial velocity range of 16.1 km sec^-2 in Table I, taken from measurements made with Palomar's 8-inch telescope. In Table II and subsequent comments, Wilson states Vyssotsky 322 is a "probable" spectroscopic binary.

    1967: In "Parallax and orbital motion of the astrometric binary CI 18,2354 from photographs taken with 24-inch Sproul refractor," Lippincott states: "Measurements of photographs taken over the interval 1938-1966 yield the binary characteristics of this nearby red dwarf." She works out a period of 24.50 years for the "unseen companion," a striking eccentricity of 0.9, and 5.65 AU for the semi-major axis a. "Photometric observations of Kron et al. (1957) show no infrared excess which would be indicative of a detectable very red companion." With a mass of 0.3 M☉ for the primary star, the mass of "Ci 18,2354 B" is found to be 0.026 ± 0.010 M☉.

    1969: van de Kamp repeats elements of Lippincott's 1967 study in his article, "Stars nearer than Five Parsecs," but he makes a change, writing that the unseen companion of AOe 17415-6 has a "Major Semi-axis of Perturbation" of .48 AU.

    1971: In "The Nearby Stars," van de Kamp again refers to Lippincott's 1967 study but adopts Gliese 687 and BD+68°946 as names for the star.

    1973: Abt sees publication of his "Catalog of Individual Radial Velocities, 12h-24h, Measured by Astronomers of the Mount Wilson Observatory." He gives four velocity measurements of 18C2354, a.k.a. W 10198, that become important later: -24.0, -4.8, -17.8, and -22.0 km sec^-2. All velocity measures in his catalog were made prior to 1952, then converted to universal time at the midpoint of film exposure. The anomalous reading for July 21, 1916, is unexplained.

    Date (UT) ... Date (normal) ... Radial Velocity ... Probable Error*
    14 Jun 14.413 ... June 14, 1914 ... -24.0 km sec^-2 ... 3.8 km sec^-2
    16 Jul 21.350 ... July 21, 1916 ... -4.8 km sec^-2 ... 0.4 km sec^-2
    16 Aug 10.210 ... Aug. 10, 1916 ... -17.8 km sec^-2 ... 2.6 km sec^-2
    16 Sep 12.196 ... Sept. 12, 1916 ... -22.0 km sec^-2 ... [blank]
    * "[P]robable errors of the velocities computed for those spectra measured by two or more persons" (Abt, 1973).


    1974: Two papers come out with relevant comments. "The Luminosity law for Late-Type Main Sequence Stars in the Solar Neighborhood" (Eggen, 1974) lists V322 (Vyssotsky number) as a spectroscopic binary ("Sp.B.") in the "Notes to Table XVI". However, in "Luminosities and temperatures of M dwarf stars from infrared photometry" (Veeder, 1974), G240-63 is called an astrometric binary ("AB").


    end of part I
    more to come
    Last edited by Roger E. Moore; 2019-Mar-06 at 07:27 PM.
    Do good work. —Virgil Ivan "Gus" Grissom

  2. #2
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    The Search for Companions of GJ 687 (part II)


    1975: Possibly in response to papers critical of the work at Sproul (see next entry, 1976), van de Kamp reviews his and others work in detecting unseen companions to nearby stars. Of BD+68°946, he again discusses Lippincott's 1967 paper and says the perturbation was first discovered on plates taken at Sproul in 1938-1948. Lippincott, he writes, used 702 plates taken by Sproul's 24-inch reflector from 1938-1966. He gives that star's mass as 0.274 M☉ but makes no comment as to whether the change in stellar mass from the value in Lippincott's 1967 paper would affect the companion's determined size.

    1976: Heintz publishes "Systematic Trends in the Motions of Suspected Stellar Companions," which summarizes new research showing the nonexistence of many previously reported "unseen companions" of single stars and "invisible third bodies" in visual double stars. He blames "systematic errors" for the majority of these false positives.

    He also mentions the results of several earlier papers that found anomalies in research relying on Sproul's 24-inch telescope (Hershey, 1973: "Astrometric analysis of the field of AC +65°6955 from plates taken with the Sproul 24-inch refractor"; Gatewood & Eichhorn, 1973: "An unsuccessful search for a planetary companion of Barnard’s star (BD+4°3561)"; Heintz, 1974: "On Inconsistencies Found in Long-Term Parallax Series" (abstract)). Heintz concludes that many reported astrometric binaries and giant planets found by van de Kamp, Lippincott, and others are the result of "well-known instrument errors" without specifying the issue. Among the stars no longer believed to have unseen companions is Gliese 687.

    1977: In response to Heinz and others, Lippincott publishes "Astrometric analyses of the unseen companions to CI 18, 2354 and Wolf 1062 from plates taken with the 61-cm Sproul refractor". Many steps are taken to improve the accuracy of earlier measurements of CI 18, 2354, including a reanalysis of Sproul plates 1937-1976 using a Grant two-coordinate measuring machine, which Lippincott describes as "impersonal" (i.e., not subject to human error). Earlier plate measurements that she regards as questionable are discarded, all resulting in "an improved photocentric orbit." The period of the companion is now 26.37 years, and its most likely mass is 0.009 M☉ (a great reduction from her 1967 paper) but not over 0.06 M☉—in short, it is substellar.

    In a surprising turn, Lippincott then mentions that Heinz (of the 1976 paper) "reminded the author of four radial velocity determinations in 1914-1916 (Abt 1973), one of which differs by 16 km/sec from the average of the others" (see previous note for 1973). This point brings up the possibility that CI 18, 2354 is a short-period, close spectroscopic binary in addition to being an astrometric binary. "The mass of an additional late-type dwarf to the visible component," she writes, "would not raise the minimum mass of the unseen astrometric companion above 0.01 M☉." Lippincott ends saying it is premature to consider the "spectroscopic binary" angle further.

    1978: "Astrometric search for unseen stellar and sub-stellar companions to nearby stars and the possibility of their detection" is Lippincott's review of Sproul's unseen-companion research involving stars closer than five parsecs (per van de Kamp's earlier work). A photo of Sproul's 24-inch (61-cm) refractor is included, with the comment that the telescope was not used from May 1966 to January 1967, when it was "renovated mechanically and electronically". The paper mentions two further adjustments made to the telescope in the 1940s, the consequences of which she says are now accounted for. Plates are studied using the Grant two-coordinate machine, which Lippincott writes "yields greater accuracy" in measurements once it began being used in 1972.

    BD+68°946 is noted to have "unseen components" in Table I. Further detailing the system, Lippincott says the star is probably a spectroscopic binary in a manner unrelated to its being an astrometric binary ("a candidate for [a] sub-stellar mass companion"). In Table III and subsequent notes, the unseen companion of BD+68°946 (a.k.a. Ci 2354 & Ci 18, 2354) is given a mass range of 0.006 to 0.07 M☉, a period of about 26 years, eccentricity of 0.9 (as before), and a perturbation "confined almost entirely to the Y coordinate." Periastron is predicted for the early 1990s. She identifies Wilson, 1967 (not Abt, 1973) as the source of the possibility that the star is a spectroscopic binary, adding the "radial velocity and astrometric interpretations of additional bodies are not mutually exclusive (Harrington, 1975)."

    1978: Also this year, "On the origin of close binary and planetary systems" appears. The author, Fleck, focuses on four "unresolved astrometric binaries having planetary companions... and reliable orbits (van de Kamp 1975)." Among these is BD +68°946. He equivocates on the "planetary" label, sometimes putting the word in parentheses, and points out that the "reliability of van de Kamp’s solutions has been questioned by others (e.g., Gatewood and Eichhorn 1973)."


    end of part II
    more to come
    Last edited by Roger E. Moore; 2019-Mar-06 at 07:29 PM.
    Do good work. —Virgil Ivan "Gus" Grissom

  3. #3
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    The Search for Companions of GJ 687 (part III)


    1984: The use of the term "black dwarf" for substellar bodies of less than or equal to 0.08 M☉ (i.e., what we today call a brown dwarf) dates back to Kumar's 1962 paper, "Study of Degeneracy in Very Light Stars". The term reappears in Herczeg's 1984 review work, "Duplicity on the main sequence." On page 37-38, Herczeg is briefly but highly critical of the work of van de Kamp and Lippincott, though he also strives to give an "impartial review" of past multiple-star research. In the end, he allows that BD 68° 946 might be one of several "good candidates" for having a black dwarf companion, though Herczeg also dislikes the term "black dwarf" as being incorrect.

    1986: Winglee, Dulk, and Bastian publish "A search for cyclotron maser radiation from substellar and planet-like companions of nearby stars," which includes Gl 687 (a.k.a. BD 68° 946). The parameters for the alleged companion are from Lippincott, 1977. The authors fail to detect cyclotron maser radiation from this star, which would have been indicative of the presence of a substellar or planetary companion. The authors comment that "unambiguous detections and derivation of the companions’ masses and orbits have been difficult because the proper motions of the stars are close to those arising from random and systematic errors in the observations (e.g., Gatewood 1976; Heintz 1978)." This is the first in a long series of negative detections for a companion of GJ 687--except for the following one.

    1987: In a startling turn, SAO 17568 is claimed to be an astrometric binary using a speckle camera. In "ICCD speckle observations of binary stars. II. Measurements during 1982-1985 from the Kitt Peak 4 m telescope," authors McAlister, Hartkopf, Hutter, & Franz present a long table of stars newly identified as binaries. SAO 17568 is given a new catalog designation as CHARA 62 Aa (for Center for High Angular Resolution Astronomy at Georgia State University). See Table III.

    Disastrously, that same year, Blazit, Bonneau, and Foy publish "Speckle interferometric measurements of binary stars: IV". Therein, GL 687 (a.k.a. CHARA 62 Aa) is named as a supposed astrometric binary (24.5-year period), but this binary status is not detected nor confirmed using speckle interferometric observations. The CHARA designation for this star appears to have been dropped hereafter.

    1989: In "MAP determinations of the parallaxes of stars in the regions of HD 2665, BD +68.946 deg, and Lambda Ophiuchi," Gatewood writes that the motion of BD +68°946 is highly linear (and thus not perturbed by significant companions) over the course of two years of investigation, though he says that was not long enough to detect the companion proposed by Lippincott, 1977, with a 25-year period. BD +68°946, he says, is the subject of a search for "infrared 'brown' dwarfs and Jovian planets."

    1990: "A systematic search for brown dwarfs orbiting nearby stars," by Henry and McCarthy, reports GL 687 is "an astrometric binary (Lippincott 1977) and an SB [spectroscopic binary—REM] with velocity range 16 km s^-1 Gliese (1969), although the spectroscopic component is believed to be unrelated to the astrometric one (Lippincott 1978). The astrometric orbit is not convincing... The secondary mass is estimated to be 10 Jupiters... Kenyon reports no velocity variations in three observations over three years at the 0.7 km s^-1 level... If the companion is real, it must be of very low mass at the limit of detectability for infrared speckle, astrometric, and radial velocity techniques."

    1991: McMillan & Herbst, "in H-alpha and broadband photometric monitoring of ten dwarfs: flares and spots," discover something new about Gl 687: it might be a (slightly) variable star. "An unexpected result, which we note in passing and for which we have no explanation, is the large value of a(V) found for the bright, absorption line star G1 687. It appears to be a variable star of small amplitude. A periodogram analysis of its data revealed no evidence for periodicity; the star deserves some additional photometric attention."

    1992: Two papers this year repeat and seem to accept past claims of binary status for this star. Tokovinin writes that Gl 687 is a long-period (9,000 days) spectroscopic binary and also an "astrometric and speckle binary" in "The frequency of low-mass companions to K and M stars in the solar neighbourhood." Weis, in "Photometry of dwarf K and M stars", says in Table 1 that VYSS. 322AB is multiple (by using the AB designation), and table data came from "combined photometry".

    1993: "Activity in late-type stars. VIII. The nature of the dM(e) or "zero" H-alpha stars," by Byrne, makes reference to Gl 687AB.


    end of part III
    more to come
    Last edited by Roger E. Moore; 2019-Mar-06 at 09:09 PM.
    Do good work. —Virgil Ivan "Gus" Grissom

  4. #4
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    The Search for Companions of GJ 687 (part IV)


    The following are brief descriptions of failed detections of companions to GJ 687, from 1999 to 2014. These studies looked at other M-dwarf stars as well, with varied results.

    1999: "Binary star speckle measurements during 1992-1997 from the SAO 6-m and 1-m telescopes in Zelenchuk," by Balega, Balega, Maksimov, et al. No companion star found.

    2001: "A coronagraphic survey for companions of stars within 8 parsecs," by Oppenheimer, Golimowski, Kulkarni, et al. G240-063: no companion faint stars or brown dwarfs found.

    2002: "A near-infrared, wide-field, proper-motion search for brown dwarfs," by Hinz, McCarthy, Simons, et al. No unreported stellar or brown dwarf companions at wide separation (∼100-1400 AU).

    2003: "A dedicated M dwarf planet search using the Hobby-Eberly telescope," by Endl, Cochran, Tull, et al. No planets found.

    2006: "Exploring the frequency of close-in Jovian planets around M dwarfs," by Endl, Cochran, Kurster, et al. No "variability indicative of a giant planet in a short-period orbit."

    2009: "Search for cold debris disks around M-dwarfs. II.", by Lestrade, Wyatt, Bertold, et al. No cold debris disk found.

    2011: "A Spitzer IRAC imaging survey for T dwarf companions around M, L, and T dwarfs: observations, results, and Monte Carlo population analyses," by Carson, Marengo, Patten, et al. No substellar companion found.

    2012: "The solar neighborhood. XXVIII. The multiplicity fraction of nearby stars from 5 to 70 AU and the brown dwarf desert around M dwarfs," by Dieterich, Henry, Golimowski, et al. No new companions found.

    2012: "The nearby population of M-dwarfs with WISE: a search for warm circumstellar dust," by Avenhaus, Schmid, & Meyer. No infrared excesses detected.

    2014: "The TRENDS high-contrast imaging survey. IV. The occurrence rate of giant planets around M dwarfs," by Montet, Crepp, Johnson, et al. No giant planets found.


    An interesting study:

    2013: "The most common habitable planets – atmospheric characterization of the subgroup of fast rotators," by Pinotti. "In order to study the effectiveness of the method for the most common main-sequence class of stars, I have chosen the nearby Gliese 687 as a representative of the M dwarf population of the solar neighbourhood (see Table 2). At a distance of 14.77 light-years, it is close enough for searches using future observatories with direct imaging capabilities. Its high metallicity (Berger et al. 2006) of +0.11 indicates a good chance that rocky planets have formed around the star during the protoplanetary disc lifetime." The paper details a hypothetical planet of Gliese 687 and the effects the star might have upon it.


    end of part IV
    more to come
    Do good work. —Virgil Ivan "Gus" Grissom

  5. #5
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    The Search for Companions of GJ 687 (part V)


    Then, completely by surprise...

    https://arxiv.org/abs/1405.2929

    2014: "The Lick-Carnegie exoplanet survey: Gliese 687 b: A Neptune-mass planet orbiting a nearby red dwarf," by Burt, Vogt, Butler, et al. "Precision radial velocities from the Automated Planet Finder (APF) and Keck/HIRES reveal an Msin (i) = 18 ± M⊕ planet orbiting the nearby M3V star GJ 687. This planet has an orbital period P = 38.14 days and a low orbital eccentricity. Our Strömgren b and y photometry of the host star suggests a stellar rotation signature with a period of P = 60 days. The star is somewhat chromospherically active, with a spot filling factor estimated to be several percent. The rotationally induced 60 day signal, however, is well separated from the period of the radial velocity variations, instilling confidence in the interpretation of a Keplerian origin for the observed velocity variations. Although GJ 687 b produces relatively little specific interest in connection with its individual properties, a compelling case can be argued that it is worthy of remark as an eminently typical, yet at a distance of 4.52 pc, a very nearby representative of the galactic planetary census. The detection of GJ 687 b indicates that the APF telescope is well suited to the discovery of low-mass planets orbiting low-mass stars in the as yet relatively un-surveyed region of the sky near the north celestial pole."


    end of part V
    more to come, but need to research again
    Do good work. —Virgil Ivan "Gus" Grissom

  6. #6
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    More will be said about research involving GJ 687 b, but first, my view on the matter so far.

    I've long been fascinated by attempts to detect unseen companions (planets, etc.) of nearby stars, and as a teenager I avidly read anything about it. For a short time, I was pen-pals with Dr. Peter van de Kamp at Sproul, and he kindly sent copies of his work to me, including some of his notes. I have them in one of my book-boxes somewhere, hope to find them again one day.

    What I have learned from all the misfired attempts to find those companions is that IT WAS NEVER A SIMPLE THING. Mistakes are never limited to "the telescope was broken" or "they just made it all up". I took pains to be very exacting in my look through all the papers named above, and it is clear that MANY people contributed to the false positives that were detected for this one star. This is not even the most complicated scenario; the search for companions of 70 Ophiuchi extends back several centuries.

    In this particular case, false positives came in from every direction, from a variety of observational methods, and some of these erroneous "discoveries" even agreed with one another. No wonder mistaken information was so often repeated. Worse still is the use of multiple names for this star, and the depth to which information can be buried in long tables that will confound any but the most compulsive searchers. I was reduced to hunting for stars by 1900 Ra & Dec, or putting information together from bits in multiple papers. It's ridiculous.

    Even worse, the planet found wasn't big enough to have been discovered by many earlier techniques. Only when telescopes and computers and other methods became as refined as they are today could any progress have been made toward discovery of the real planet. And maybe some new bombshell is waiting to pop out and undo everything we thought we knew.

    It's never a simple story.

    Looking forward to comments, corrections, and critical thinking.
    Do good work. —Virgil Ivan "Gus" Grissom

  7. #7
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    Quote Originally Posted by Roger E. Moore View Post
    The Search for Companions of GJ 687 (part II)

    1975: Possibly in response to papers critical of the work at Sproul (see next entry, 1976), van de Kamp reviews his and others work in detecting unseen companions to nearby stars. Of BD+68°946, he again discusses Lippincott's 1967 paper and says the perturbation was first discovered on plates taken at Sproul in 1938-1948. Lippincott, he writes, used 702 plates taken by Sproul's 24-inch reflector from 1938-1966. He gives that star's mass as 0.274 M☉ but makes no comment as to whether the change in stellar mass from the value in Lippincott's 1967 paper would affect the companion's determined size.
    This entry has errors, please replace it with this entry. My apologies.

    1975: Possibly in response to papers critical of the work at Sproul (see next entry, 1976), van de Kamp reviews his and others work in detecting unseen companions to nearby stars ("Unseen Astrometric Companions of Stars"). Of BD+68°946, he again discusses Lippincott's 1967 paper and says the perturbation was first discovered on plates taken at Sproul in 1938-1948. Lippincott, he writes, used 702 plates taken by Sproul's 24-inch reflector from 1938-1966. He gives that star's mass as 0.274 M☉ and reworks the companion's mass to be 0.026 M☉ (which, oddly, is the same figure given in Lippincott, 1967, using a stellar mass of 0.3 M☉). He is puzzled by the companion's high eccentricity (0.9) and allows for the chance that more than one companion exists, though he adds that the figure is "possibly spurious". Page 299 of the paper describes a number of adjustments made over the years to the 24-inch refractor and the attempts made to correct for them in plates taken with the telescope.

    .
    Last edited by Roger E. Moore; 2019-Mar-08 at 05:02 PM.
    Do good work. —Virgil Ivan "Gus" Grissom

  8. #8
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    5,912
    New annotations have been made to SIMBAD for this star:

    http://cdsannotations.u-strasbg.fr/a...dObject/326850
    Do good work. —Virgil Ivan "Gus" Grissom

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •