The olivine-dominated composition of the Eureka family of Mars Trojan asteroids

0.35 MB·

by  G. Borisov

MNRAS000,1–7(2017) Preprint27January2017 CompiledusingMNRASLATEXstylefilev3.0
The olivine-dominated composition of the Eureka family of
Mars Trojan asteroids
G. Borisov,1,2⋆ A. Christou,1† S. Bagnulo,1 A. Cellino,3 T. Kwiatkowski4, A. Dell’Oro5
1Armagh Observatoryand Planetarium, College Hill, Armagh BT61 9DG, UK
2Institute of Astronomy and National Astronomical Observatory,72, Tsarigradsko Chauss´ee Blvd., BG-1784 Sofia, Bulgaria
3INAF - OsservatorioAstrofisico di Torino, viaOsservatorio 20, I-10025 Pino Torinese(TO), Italy
7 4Astronomical Observatory,Adam MickiewiczUniversity,ul. S loneczna 36, PL-60-286 Poznan´, Poland
1 5INAF - OsservatorioAstrofisico di Arcetri, Largo E. Fermi 5, I-50125, Firenze, Italy
0
2

n
a Accepted2016November23.Received206November22;inoriginalform2016October18
J

6
2 ABSTRACT
We have used the XSHOOTER echelle spectrograph on the European South Obser-
]
vatory (ESO) Very Large Telescope (VLT) to obtain UVB-VIS-NIR (ultraviolet-blue
P
(UVB), visible (VIS) and near-infrared(NIR)) reflectance spectra of two members of
E
theEurekafamilyofL5MarsTrojansinordertotestageneticrelationshiptoEureka.
.
h In addition to obtaining spectra, we also carried out VRI photometry of one of the
p VLT targets using the 2-m telescope at the Bulgarian National Astronomical Obser-
-
vatory – Rozhen and the two-channel focal reducer. We found that these asteroids
o
r belong to the olivine-dominated A, or Sa, taxonomic class. As Eureka itself is also an
t olivine-dominatedasteroid,itis likelythatall family asteroidsshareacommonorigin
s
a and composition. We discuss the significance of these results in terms of the origin of
[ the martian Trojan population.

1 Key words: planets andsatellites:individual:Mars–minor planets,asteroids:indi-
v
vidual:Trojanasteroids–techniques:imagingspectroscopy–techniques:photometric
5
2
7
7
0 1 INTRODUCTION oid family are also stable Trojans. For this reason, the dis-
1. covery that asteroid (5261) Eureka belongs to the rare A
The so-called Mars Trojans are asteroids located in the L4
0 taxonomicclass(Rivkinet al.2007;Lim et al.2011)isvery
orL5Lagrangian pointsofMars. Theyarethoughttohave
7 interesting.ThespectralreflectancepropertiesofA-classas-
beentheresincetheveryearly phasesoftheSolarSystem’s
1 teroidshavebeeninterpretedasdiagnosticofanoverallcom-
v: hasistoofry2.0T15h,eroefwwehriechnienieghctonwfierrmeeadtMLa5rtainandT1roajtanLs4.inOtcoctua-l position rich in the mineral olivine [(Mg++, Al++)2SiO4], a
i magnesium iron silicate thatis aprimarycomponentof the
X pants of the Mars Trojan clouds might represent a small Earth’s mantle and – supposedly – of the other terrestrial
setofsurvivorsofanearlygenerationofplanetesimalsfrom
r planets. The low abundance of olivine-rich objects among
a which theinnerSolar System was built.The long-term sta-
asteroids is an old conundrum. In particular, the existence
bilityof theMarsTrojan cloudshasbeen extensivelyinves-
of metal meteorites suggests that during the early phases
tigated by Schollet al. (2005). According to these authors,
of the Solar System’s history, several planetesimals reached
intheL4andL5regionsofMars,therearecombinationsof
sizes sufficient for complete thermal differentiation due to
orbitaleccentricityandinclinationthatarestableovertime-
theheat produced by thedecay of short-lived isotopes, pri-
scalescomparabletotheSolarSystem’sage.OftheeightL5 marily26Al.Metalmeteoritesareinterpretedinthisscenario
Trojans, seven (including Eureka) form the Eureka family,
as fragments of the metal-rich cores of such differentiated
whoseexistencehasbeenpointedoutbyChristou(2013)and
bodies, exposed after their complete collisional disruption.
dela FuenteMarcos & dela FuenteMarcos (2013). Due to
In this context, one should expect that olivine-rich mete-
itscompactness,thisisprobablyageneticfamilyratherthan
orites and asteroids should be common because olivine is
arandomgroupingoforbits.Itformedsometimeinthelast
a primary component of the mantles of differentiated bod-
Gyr (C´uk et al. 2015). The members of the Eureka aster-
ies. However, this prediction is not borne out of the ob-
servations. Asteroids belonging to the olivine-rich A class,
oreventothefairly similar Sa sub-classoftheStaxonomic
⋆ E-mail:[email protected](GB) complex(Bus & Binzel2002;DeMeo et al.2009b),arequite
† E-mail:[email protected](AC)
©2017TheAuthors
2 G. Borisov et al.
rare.Attheendofthe1990s,severalauthorsproposedthat 2.1 Spectroscopy
the original olivine-rich asteroids were ’battered to bits’ by
2.1.1 XSHOOTER
collisionsandhadthusdisappearedlongago(Burbineet al.
1996; Chapman 1997). Whatever the cause of such disap- This instrument is an echelle spectrograph with an almost
pearance,collisional evolutionor,perhapsmorelikely,mas- fixedspectralsetup.TheobservercanchoosebetweenSLIT
sive removal of the original planetesimals accreted in the (andslitwidth)andIFU(integralfieldunit)modes.Herewe
mainbelt,aspredictedbytheGrandTackandNicemodels used the SLIT mode. A detailed description of the instru-
of early evolution of the Solar System (Walsh et al. 2011, ment is available at Vernetet al. (2011) and ESO’s web-
2012), only a small number of survivors belonging to the page1. This spectrograph has the ability to simultaneously
A class can be found among the current asteroid popula- obtain data over the entire 300–2480nm spectral range by
tion.Asa consequence,thediscoveryof an A-class asteroid splitting the incoming light from the telescope into three
confinedinaspecialdynamicalenvironmentwithintheter- beams,eachsenttoadifferentarm:ultraviolet–blue(UVB),
restrialplanetregionsuggeststhattheothermembersofthe visible (VIS), and near-infrared (NIR). Using two dichroic
Eurekafamily alsodeserveacarefulinvestigation.Based on filters, the light is first sent to the UVB arm, then to the
these considerations, we began an observing campaign to VISarm,and finally theremaininglight arrives at theNIR
obtain information on the spectral reflectance properties of arm. The disadvantage of this choice of optical light path
these objects. This is a challenging task because, in spite is the high thermal background in the K-region of the NIR
of their relative proximity to the Earth, Mars Trojans are spectrum.TheobservationsarepresentedinTable1.Obser-
faint. Their observation therefore requires the use of large- vationswere donein noddingmode to facilitate subsequent
apertureinstruments. sky signal estimation and subtraction.
In our investigation, we used XSHOOTER, the first
European South Observatory (ESO) second-generation in-
strument developed for the Very Large Telescope (VLT; 2.2 VRI Photometry
Vernetet al.2011),whichiscurrentlymountedonVLTUnit
2.2.1 FoReRo2
2 Kueyen. XSHOOTER is a spectrograph capable of ob-
tainingspectrafrom 300to2480nminasingleshotathigh The instrument is a two-channel focal reducer that adapts
spectralresolution(from3000upto10000).Inthefollowing the imaging elements of the detector to the characteristic
sectionswepresentthespectraweobtainedfortwomembers size of the object or the seeing disk. FoReRo2 was built
of the Eureka family, and compare them with the available mainly for observations of cometary plasma but has been
reflectancespectra for Eurekaitself. proved suitable for many other tasks (Jockers et al. 2000).
We also obtained some limited spectrophotometric in- Behind the RC (Cassegrain) focus, the light beam is rec-
formation by carrying out multiband VRI photometry. We ollimated by a lens collimator. A dichroic mirror reflects
used the two-channel focal reducer – Rozhen (FoReRo2) of the blue part of the spectrum and transmits the red part.
the 2m Ritchey-Chr´etien-Coud´e telescope at the Bulgar- Foreachchannel,cameralensesform reducedimages ofthe
ian National Astronomical Observatory (BNAO) to obtain Cassegrainfocalplane,whicharerecordedbytwoCCDsys-
(V−R) and (V−I) colours of a member of the Eureka fam- tems. Filters are placed into the parallel beam after colour
ily,also one of ourtwotargets observed with XSHOOTER. separation. Forthisinvestigation we used VRIstandard fil-
Thecolourindiceswerecomparedwiththeresultsobtained ters; therefore, only the red channel of the instrument was
by Neese (2010), and with the average values of the colour in operation. Theobservations are presented in Table 2.
indices for different asteroid taxonomic classes.
This paper is organised as follows. In Section 2, we
2.2.2 ACAM
present our VLT and Rozhen observations. The adopted
procedures for data reduction are described in Section 3, TheACAMinstrumentofthe4.2-mWilliamHerschelTele-
andtheresultsinSection4,separatelyforspectroscopyand scope(WHT) at theObservatoriodelRoquedelos Mucha-
multibandphotometry.Ourmainconclusionsandageneral chos(LaPalma,Spain)wasusedtoobtainSDSSphotometry
discussion of our results is given in Section 5. ofasteroid(3819)Robinson.Theobservationsarepresented
in Table 3.
2 OBSERVATIONS 3 DATA REDUCTION
3.1 XSHOOTER
We used XSHOOTER during two nights, FoReRo2 dur-
ing two nights as well and ACAM for 1 night. The ob-
To reduce the XSHOOTER spectra we first processed the
jects that were observed are as follows: XSHOOTER: data through the ESO esoreflex pipeline version 2.6.8.
(385250) 2001 DH47, (311999) 2007 NS2 and spectral
All spectra were reduced under the assumption of point-
solar analogue star HD 67010; FoReRo2: (385250) 2001
likesourcesandhadtheirinstrumentsignatureremoved,i.e.
DH47, (289) Nenetta [an asteroid with olivine-dominated
de-biased,flat-fielded,wavelength-calibrated,order-merged,
surface (Sanchezet al. 2014)] and Stetson standard field
extracted,sky-subtractedand,finally,flux-calibrated.
L101; ACAM: (3819) Robinson [an asteroid with olivine-
To achieve a signal-to-noise ratio (S/N) sufficient for
dominated surface (Sanchezet al. 2014)] and Stetson stan-
dardfieldL104.Detailsofourobservationsarepresentedin
Tables 1–3. 1 http://www.eso.org/sci/facilities/paranal/instruments/xshooter.html
MNRAS000,1–7(2017)
The composition of Mars Trojan asteroids 3
Table 1.Observations-XSHOOTER@VLT.
Apparenet Spectral Exposuretime
Object Date UT Airmass
magnitude arm (s)
03:04 1.174 UVB 3040
(385250) 2001DH47 2016February02 | | 19.00 VIS 3280
04:07 1.377 NIR 3600
UVB 40
HD67010 2016February02 05:15 1.158 8.69 VIS 30
NIR 100
01:37 1.022 UVB 3040
(311999) 2007NS2 2016March02 | | 19.05 VIS 3280
02:21 1.367 NIR 3600
UVB 40
HD67010 2016March02 03:30 1.168 8.69 VIS 30
NIR 100
Table 2.Observations-FoReRo2@2mRCC.
Apparenet Exposuretime
Object Date UT Airmass Filter
magnitude (s)
2016February06 22:00 1.30 I 10×300
(385250) 2001DH47 | | | 18.55 R 10×300
2016February07 01:00 1.50 V 10×300
20:00 1.15 I 3×300
(289)Nenetta 2016February06 | | 14.13 R 3×300
21:00 1.25 V 3×300
19:50 2.10 12.64 V 3×(5×60)
StetsonL101 2016February06 21:30 1.50 | R 3×(5×60)
23:00 1.30 20.64 I 3×(×60)
Table 3.Observations-ACAM@WHT.
Apparenet Exposuretime
Object Date UT Airmass Filter
magnitude (s)
06:10 1.94 g' 3×80
(3819)Robinson 2016March11 | 2.02 16.36 r' 3×60
06:25 2.12 i' 3×60
02:30 1.145 12.75 g' (4×5)and(3×15)
StetsonL104 2016March11 | r' (4×5)and(3×10)
06:00 1.853 21.12 i' (4×5)and(3×15)
scientific analysis the reduced 1D spectra for both the as- 3.2 FoReRo2
teroidandsolaranaloguestandardstarswerethenrebinned
in 20nm steps. Subsequently, the spectrum of the asteroid All imaging data had their instrument signature removed
was dividedbythesolar analogue spectrum,theresult nor- aswell byde-biasingand flat-fielding.The images in I were
malisedtounityat550nmandagainsmoothedusingafloat- de-fringedusingthemedianIimagecombinedfromallIim-
ing average with different wavelength steps in the VIS and agesforthenight.Standarddaophotaperturephotometry
NIRregions. with aperture size 2×full width at half-maximum (FWHM)
Finally, we carried out running sigma clipping of the was performed. Aperture correction, which was measured
data with a 20nm window in wavelength and a 3σ crite- through a large aperture using the growth-curve method
rion. During this stage we excluded sections of the spectra (Stetson 1990), was applied in order to place the instru-
affected by high telluric line contamination in thefollowing mental magnitudes of the observed point source objects,
wavelengthintervals:0.90–1.00,1.35–1.50and1.80–1.90µm, whichweremeasuredthroughasmallaperture(2×FWHM),
as suggested by Moehler et al. (2014) and Gourgeot et al. on the same system as those of the standard stars. Stetson
(2015). standard field L101, observed on three different airmasses,
MNRAS000,1–7(2017)
4 G. Borisov et al.
was used for obtaining extinction coefficients in each filter.
Linearregression fitsformagnitude–magnitudeandcolour–
colourcalibration wereperformed aswell. Then,theinstru-
mental magnitudes of the asteroid targets were absolutely
calibratedusingcoefficientsfromthosefits,and,finally,their
true(V−R) and (V−I) colours were computed.
3.3 ACAM
All imaging data had their instrument signature removed
as well by de-biasing and flat-fielding. The same calibra-
tionprocedureastheoneforFoReRo2wasperformedusing
Stetson standard field L104. The photometry performed in
SDSS-g'r'i'filterswasconvertedtoVRIusingrelationspre-
sented byJordi et al. (2006).
Figure 1. (Colour online) Reflectance spectrum of the asteroid
(385250)2001DH47(blackdotted)comparedwithaveragespec-
4 RESULTS traforA(red),S(green)andSa (purple)taxonomicclasses.The
spectrumof(5261)Eurekaisoverplotted inblue.
4.1 Spectroscopy
TheresultingXSHOOTERspectraarerepresentedbyblack
lines in Figs 1–4.
On Figs 1 and 2, we show, apart from the re-
flectance spectra of our two targets, a spectrum
of (5261) Eureka (blue) obtained on 2015May 19
(Rayneret al. 2003, spectrum available online at
http://smass.mit.edu/minus.html - sp41) and aver-
age spectra for the A (red), S (green) and Sa (purple)
classes in the DeMeo et al. (2009b) taxonomy produced
from data available in the PDS data base (EAR-A-
VARGBDET-5-BUSDEMEOTAX-V1.0; DeMeo et al.
2009a) as point-by-point means of 6, 144 and 2 asteroids
respectively. We find a 1µm absorption feature for both
Eureka family members where the reflectance minimum is
located slightly longward of 1µm. This is also the case for
the A- and Sa-type spectra and a diagnostic feature of an
olivine-rich surface. Figure 2. (Colour online) Reflectance spectrum of the asteroid
The regions with high telluric lines contamination, ex- (311999) 2007 NS2 (black dotted line) compared with average
cluded from our analysis, (see Section 3), do not affect dis- spectraforA(red),S(green)andSa (purple)taxonomicclasses.
tinguishing between the S, A and Sa classes, as the 1µm Thespectrumof(5261)Eurekaisoverplotted inblue.
feature in S-class spectra is below 0.9µm. We also note the
higher and flatterreflectance of theS-class spectrum in the
Seeking to strengthen the case for the presence of
region of 1.0-1.5µm, which distinguishes it from the other
olivine, we compared our spectra with a number of labora-
spectra.
tory olivinespectra from therelab database. Exposureof
Overall, the two asteroid spectra are in better agree-
asteroidsurfacestothespaceenvironmentchangestheirop-
ment with an Sa than an A taxonomy, both in the visible ticalproperties.Thereforeweappliedasimplespaceweath-
andtheinfrared(IR).Eureka,alsoanSa-typeasteroidinthe ering correction tothelaboratory spectra prior tothecom-
DeMeo et al.classificationscheme2,appearstobesomewhat
parison by dividing the olivine relab spectra with a first
more reflective than 311999 and 385250 from ∼0.6 up to at
orderpolynomial.Throughvisualinspection,wefoundthat
least 1.3µm. Eureka’s 1µm feature is unique among A/Sa- spectra MS-CMP-042-A and MS-CMP-014 (green lines in
types.Itistheonlyobjectwiththatparticularbandshape.
Figs3and4alteredbyourspace-weatheringcorrection,red
Thespectraofthetwonewobjects,whichweclassify asSa, lines) are reasonable matches to the asteroid spectra. The
thoughtheyhavealower S/Nratio,appearconsistentwith
first laboratory spectrum corresponds to unprocessed pure
Eureka’s. At 2µm the reflectivities of the three objects are
olivine; the second, also of pure olivine, is altered by laser
indistinguishable from each other; although possibly a con-
irradiationtosimulatemicrometeoroidimpacts.Althoughit
sequence of the lower S/N ratio, we believe it to be a real
isdifficulttoobtainagoodmatchovertheentirewavelength
feature of thespectrum.
interval covered by our spectra, by restricting ourselves to
thevisiblepart.andthesimplespaceweatheringmodel,we
2 TheobservedcoincidencebetweentheSa spectrumandthatof find that our best match (red lines) fits the visible part of
Eurekais not accidental; this asteroid isone of onlytwo objects ourspectraquitewell,particularlytheminorabsorptionfea-
whosespectradefinethistaxonomicclass turesaround0.63and0.8µm.Progressingfurtheralongthis
MNRAS000,1–7(2017)
The composition of Mars Trojan asteroids 5
10 exposures, acquired in I, R, and V filters with FoReRo2
atBNAO.Assomeimageswereofinferiorquality,onlyeight
exposuresineachfilterwereused,fromwhichaveragevalues
were computed.
Because of the faintness of the asteroid target, we
used relatively long exposures of 300s. This resulted in
a duty cycle of 318s and extended the observing run to
3h. On 2016 January 8 we carried out photometric ob-
servations of (385250) 2001 DH47. A rotation period of
P≈4.0±0.8handapeak-to-peakamplitudeof0.6magwere
derived (Borisov et al. 2016). The asteroid light variation
could potentially introduce systematic effects in the colour
indices.
We used those parameters to understand the effect of
the asteroid lightcurves in determining the colour indices
obtained from the Rozhen data. Due to the relatively high
Figure3.Reflectancespectrumofasteroid(385250)2001DH47
uncertainty in the rotation period P and a long time in-
(black linewith points) compared withtwo olivinespectra from
relab(greendashedline)andthesamespectradividedbyaslope terval between the January 8 and February 6 observations,
tosimulatespace-weathering(redline). wecouldnotdetermineaccuraterotation phasesforourex-
posures. In addition, the January 8 observations were per-
formed at the solar phase angle of 30.◦4, while on February
6thephaseanglewas8.◦2.Assumingtheclassicamplitude–
phaserelationship withm=0.02magdeg−1 (Guti´errez et al.
2006), we estimated the lightcurve amplitude of (385250)
2001 DH47on February 6 to be0.4mag.
A simulated lightcurve of (385250) 2001 DH47, with a
simplesinusoidalshape,rotationperiodP=4h,andpeak-to-
peakamplitude A=0.4mag, ispresentedin Fig.5,with the
relative times of the I, R, and V exposures superimposed
on the lightcurve. The rotation phase of the first point is
arbitrary, but the intervals between consecutive points re-
flect the actual timings of the exposures. In this model,
the average brightness in the I, R, and V filters would be
0.084,−0.080, and 0.102, respectively, and thecolour indices
would be (V−I)=0.018mag, (V−R)=0.183mag. In reality,
the rotation phase of the first point of the observing se-
quence is unknown, so we repeated the computations with
Figure 4. Reflectance spectrum of the asteroid (311999) 2007
phase shifts from 0.02 to 1 in steps of 0.025. Also, for each
NS2 (black line with points) compared with two olivine spectra
fromrelab(greendashedline)andthesamespectradividedby phase shift, we computed colour indices while changing the
aslopetosimulatespace-weathering(redline). rotation period from 3.2h to 4.8h, in steps of 0.1h. In this
way, we obtain 39×17=663 pairs of colour indices (V−I),
and(V−R),whichallowedustoestimatethesystematicun-
lineofinvestigationprobablyrequiresamorerefinedmodel certainty of the colour indices computed from the Rozhen
ofspace-weatheringeffects,whichisbeyondthescopeofthis observationsdueto asteroid brightness variation.
paper. To estimate the statistical uncertainties resulting from
the random scatter of the measurements we used formal
sigma valuesgiven by thedaophot package for each single
4.2 VRI Photometry
V,R,Iinstrumentalmagnitude.Thesewerethenpropagated
Visiblecolourphotometrycanbeusedasaconsistencycheck to final estimates of the (V−I) and (V−R) colours, result-
andalsotoeliminatecandidatespectroscopicclasses.Itsad- ing in sigma values of 0.05 and 0.04mag, respectively. The
vantage is that it is generally applicable to fainter objects former is slightly greater than the latter due to the greater
than spectroscopy. On the other hand, unambiguous tax- noise of theI-filtermeasurements.
onomic classification and mineralogical interpretation typi- Fig. 6 presents the (V−R) and (V−I) colour indices
callyrequiresdetailedknowledgeofthereflectanceasafunc- of (385250) 2001 DH47, compared with the average colours
tion of wavelength as well as extending observations into of all taxonomic classes in the Tholen (1984) classification
theIR.Ourpurposehereistoidentify,throughdirectmea- scheme as given in Dandyet al. (2003) as well as the Sa
surements,thedomainthatEureka,itsfamilymembersand andSr classesintheBus & Binzel(2002)classification.The
asteroids of a similar mineralogical composition occupy in systematicuncertaintydueto thelightcurveisshown asan
colour space with a view to possible future observations of ellipse.Therandommeasurementuncertaintiesofthecolour
thefainter Eurekafamily asteroids. indices are represented bytheerror bars.
The (V−R) and (V−I) colour indices for one of theob- Sa and Sr classes in the Bus & Binzel (2002) taxon-
jects,(385250)2001DH47,werederivedfromthreeseriesof omy – distinct from the classes of the same name in the
MNRAS000,1–7(2017)
6 G. Borisov et al.
Figure5.Asimulatedlightcurveof(385250)2001DH47withan
assumedsinusoidalbrightnessvariation,rotationperiodof4hand
apeak-to-peakamplitudeof0.4mag.Whiletherotationphaseof
the first point is arbitrary, data intervals reflect actual times of
theexposures intheI,R,andVfilters.
DeMeo et al.(2009b)taxonomy–areintermediatebetween
Figure 6. (V−R)/(V−I) colour diagram of the Eureka family
S and A, and S and R, respectively. The visible (0.44–
member (385250) 2001 DH47 as well as A-class asteroids (289)
0.92µm) spectra that define these classes have a very steep
Nenettaand(3819)Robinson,versusmeancoloursofthedifferent
ultraviolet slope shortward of 0.7µm. The 1µm absorp- taxonomicclassesdiscussedinthetext.Thethreeellipsesshown
tion feature while deep and clearly visible in Sr, is shallow intheplotrepresentthesystematicuncertaintyinthecoloursof
and not well defined in Sa, which shows that it might be (385250)2001DH47duetothelightcurveeffect.Theseareshown
shifted above 1µm and can be associated with a presence fortherotationperiodsolution(P=4h;greenline),aswellasfor
of olivine. To compute colours representing the Sa and Sr two extreme cases: P=3.2h (blue dotted line) and P=4.8h (red
classeswesearchedforasteroidsbelongingtothoseclassesin dashed line). If our measurements were free from the random
error,ourestimatedcoloursforthisasteroidshouldbelocatedon
theSMASSIIdatabase.SDSScoloursforfiveSa andthree
one of the eclipses. Note that the random uncertainties (shown
Sr objectsincludedintheSDSSdatabaseofmovingobjects byerrorbars)aresmallerthanthissystematiceffect.
(EAR-A-I0035-5-SDSSTAX-V1.1;Carvano et al.2010)were
convertedintotheVRIsystemusingthesameprocedureas
for theACAM measurements. Trojan cloud – including the previously-observed Eureka –
Our colours for the Main Belt asteroids (289) Nenetta exhibit properties that are best interpreted in terms of a
and (3819) Robinson are included as well. Both of highsurfaceabundanceofolivine.Objectssharingthesame
these asteroids have been mineralogically classified as property,taxonomically classified as members of the A and
olivine-dominated [defined as S(I)-type in the Gaffey et al. Sa classes in the DeMeo et al. (2009b) taxonomy, are quite
(1993) classification] based on their 0.5–2.5 µm spectra unusualamongtheasteroidpopulation.Asmentionedinthe
(Sanchezet al.2014).Basedonvisible(0.44–0.92 µm)spec- Introduction, there are reasons to believe that olivine-rich
traalone,Bus& Binzel(2002)classified(289)Nenettaasan bodies were common among planetesimals accreted in the
A-typeasteroidand(3819)RobinsonasanSr-typeasteroid. innerregionsoftheSolarSystemduringtheearlyphasesof
Interestingly, Eureka was also classified as Sr in their work. planetary formation. The current underabundance of such
Our (V−R) colour of (3819) Robinson is lower than that of objectsmaywellindicatethatmostofthosethatwereorig-
(289)Nenetta,consistentwiththesomewhatshallowerslope inally present have been lost – so called ”missing mantle
shortward of the reflectance peak for that spectral class as problem”(DeMeo et al. 2015). If this interpretation is cor-
compared totheA type. rect,thenitisinterestingthatwefindasmallgroupofthese
Wethereforeconcludethatthecolourof(385250) 2001 bodiesin oneof theMars Trojan clouds. Duetothepartic-
DH47, after taking into account rotation-related photomet- ular dynamical environment that ensures orbital stability
ric effects, is consistent with the spectroscopic determina- over long time-scales, the Martian Trojan clouds are one of
tion of its taxonomy. Taking into account the asteroids’ ro- the few places where one would expect to find samples of
tational brightness changes will be important in our future thefirst generation of planetesimals accreted in thisregion.
attempts to constrain the taxonomic class through colour Inotherwords,theseasteroidsmightwellbesamplesofthe
photometry. original building blocks that came together to form Mars
and of the other terrestrial planets. The common fate of
thesebodieselsewhereseemstohavebeenanearlycomplete
removal, possibly the result of intense collisional evolution
5 DISCUSSION AND CONCLUSIONS
(Burbineet al. 1996; Chapman 1997; Sanchezet al. 2014).
Our spectroscopic and spectro-photometric data confirm It is also possible that the dynamical excitation of plan-
that three members of the Eureka family in the L5 Mars etesimals in the current main belt during an early phase of
MNRAS000,1–7(2017)
The composition of Mars Trojan asteroids 7
migration of the giant planets (Walsh et al. 2011, 2012) is REFERENCES
implicated in thedisappearance of theseobjects.
Borisov G., Christou A., Unda-Sanzana E., 2016, Minor Planet
The fact that these olivine-rich asteroids belong to a
Bulletin,43,216
group (the Eureka family) of objects that may share a
Burbine T. H., Meibom A., Binzel R. P., 1996,
common origin is also interesting. Recent numerical mod- MeteoriticsandPlanetaryScience,31,607
elling of the family’s evolution under planetary gravita- BusS.J.,BinzelR.P.,2002,Icarus,158,146
tionalperturbationsandtheYarkovskyeffectledC´uk et al. Carvano J. M., Hasselmann P. H., Lazzaro D., Moth´e-Diniz T.,
(2015) to conclude that the group is likely a genetic family 2010,A&A,510,A43
formedroughlyinthelastGyroftheSolarSystem’shistory. ChapmanC.R.,1997,Nature,385,293
Whetheror not thefamily sprungoff a common parent – a ChristouA.A.,2013,Icarus,224,144
C´ukM.,ChristouA.A.,HamiltonD.P.,2015,Icarus,252,339
proto-Eureka–hasanobviousbearingontherelativeabun-
Dandy C. L., Fitzsimmons A., Collander-Brown S. J., 2003,
dance of olivine-rich material near Mars in the early Solar
Icarus,163,363
System.Itisalsoimportanttoviewthisinthecontextofthe
DeMeo F., Binzel R. P., Slivan S. M., Bus S. J., 2009a, NASA
apparentcompositionaldiversityoftheMartianTrojanpop-
PlanetaryDataSystem,114
ulationoverall.Rivkinet al.(2003)obtainedvisiblespectra DeMeoF.E.,BinzelR.P.,SlivanS.M.,BusS.J.,2009b,Icarus,
ofEureka,(101429)1998VF31atL5and(121514)1999UJ7 202,160
atL4,latercomplementedbyNIRspectralcoverageforthe DeMeo F. E., Alexander C. M. O., Walsh K. J., Chap-
firsttwoasteroids(Rivkinet al.2007).Theyfoundthatthe man C. R., Binzel R. P., 2015, The Composi-
latter two asteroids do not share Eureka’s taxonomy and tional Structure of the Asteroid Belt. pp 13–41,
concluded that all these objects were once parts of larger doi:10.2458/azu uapress 9780816532131-ch002
delaFuenteMarcosC.,delaFuenteMarcosR.,2013,MNRAS,
bodies that formed separately in different locations of the
432,31
Solar System.
Gaffey M. J., Burbine T. H., Piatek J. L., Reed K. L., Chaky
Futureinvestigations of objects in Mars’ Trojan clouds
D.A.,BellJ.F.,BrownR.H.,1993,Icarus,106,573
shouldincludehigh-S/NspectrainthevisualandNIRspec-
GourgeotF.,BarucciM.A.,Alvarez-CandalA.,MerlinF.,Perna
tral regions to help us better understandthe compositional D.,LazzaroD.,2015, A&A,582,A13
relationships.ExtendingtheMarsTrojaninventorydownto Guti´errezP.J.,DavidssonB.J.R.,OrtizJ.L.,RodrigoR.,Vidal-
smallersizesanddeterminingtheirrotationalcharacteristics Nun˜ezM.J.,2006,A&A,454,367
wouldalsohelpevaluatethestabilityoftheseobjectsinthe JockersK.,etal.,2000,KinematikaiFizikaNebesnykhTelSup-
size regime where non-gravitational perturbations such as plement,3,13
theYarkovskyeffect become important. JordiK.,GrebelE.K.,AmmonK.,2006,A&A,460,339
Lim L. F., Emery J. P., Mueller M., Rivkin A. S., Trilling D.,
BurtB.J.,2011, p.1199
MoehlerS.,etal.,2014,A&A,568,A9
NeeseC.,2010,NASAPlanetaryDataSystem,123
Rayner J. T., Toomey D. W., Onaka P. M., Denault A. J.,
Stahlberger W. E., Vacca W. D., Cushing M. C., Wang S.,
ACKNOWLEDGEMENTS
2003,PASP,115,362
Thisworkwassupportedviaagrant(ST/M000834/1) from Rivkin A. S., Binzel R. P., Howell E. S., Bus S. J., Grier J. A.,
the UK Science and Technology Facilities Council. Based 2003,Icarus,165,349
Rivkin A. S., Trilling D. E., Thomas C. A., DeMeo F., Spahr
on observations collected at the European Organisation for
T.B.,BinzelR.P.,2007,Icarus,192,434
Astronomical Research in the Southern Hemisphere under
SanchezJ.A.,etal.,2014,Icarus,228,288
ESO programme 296.C-5030 (PI: A.Christou).
SchollH.,MarzariF.,TricaricoP.,2005,Icarus,175,397
Wegratefullyacknowledgeobservinggrantsupportfromthe
StetsonP.B.,1990,PASP,102,932
Institute of Astronomy and Rozhen National Astronomical TholenD.J.,1984,PhDthesis,UniversityofArizona,Tucson
Observatory,Bulgarian Academy of Sciences. VernetJ.,etal.,2011,A&A,536,A105
WethankColinSnodgrassforkindlyagreeingtoobserveas- WalshK.J.,MorbidelliA.,RaymondS.N.,O’BrienD.P.,Man-
teroid 3819 with ACAM@WHT during program ITP6, and dellA.M.,2011,Nature,475,206
MaximeDevogeleforhissuggestiontocompareourasteroid WalshK.J.,MorbidelliA.,RaymondS.N.,O’BrienD.P.,Man-
spectra with laboratory olivine spectra. dellA.M.,2012,MeteoriticsandPlanetaryScience,47,1941
Astronomical research at the Armagh Observatory and
Planetariumisgrant-aidedbytheNorthernIrelandDepart- Thispaper has been typeset fromaTEX/LATEX fileprepared by
ment for Communities (DfC). theauthor.
This research utilises spectra acquired from the NASA
textscrelab facility at Brown University.
Eureka spectral data utilised in this publication were ob-
tained andmadeavailable bytheTheMIT-UH-IRTFJoint
Campaign for NEO Reconnaissance. The IRTF is operated
by the University of Hawaii under Cooperative Agreement
no.NCC5-538withtheNationalAeronauticsandSpaceAd-
ministration, Office of Space Science, Planetary Astronomy
Program.TheMITcomponentofthisworkissupportedby
NASA grant 09-NEOO009-0001, and by the National Sci-
enceFoundation underGrantsNos. 0506716 and 0907766.
MNRAS000,1–7(2017)