The CO2 Abundance in Comets C/2012 K1 (PanSTARRS), C/2012 K5 (LINEAR), and 290P/Jager as Measured with Spitzer
Adam J. McKay, Michael S.P. Kelley, Anita L. Cochran, Dennis Bodewits, Michael A. DiSanti, Neil Dello Russo, Carey M. Lisse
aa r X i v : . [ a s t r o - ph . E P ] N ov The CO Abundance in Comets C/2012 K1(PanSTARRS), C/2012 K5 (LINEAR), and290P/J¨ager as Measured with
Spitzer
Adam J. McKay a , Michael S. P. Kelley b , Anita L. Cochran a ,Dennis Bodewits b , Michael A. DiSanti c , d , Neil Dello Russo e ,Carey M. Lisse e a Univerisity of Texas Austin/McDonald Observatory, 1 University Station, Austin,TX 78712, (U.S.A);[email protected], [email protected] b Department of Astronomy, University of Maryland, College Park, MD20742-2421 (U.S.A.); [email protected], [email protected] c NASA Goddard Center for Astrobiology, NASA GSFC, Mail Stop 690, Greenbelt,MD 20771 (U.S.A.); [email protected] d Solar System Exploration Division, Mail Stop 690, Greenbelt, MD 20771 (U.S.A) e Johns Hopkins University Applied Physics Laboratory, 11100 Johns Hopkins Rd.,Laurel, MD, 20723 (U.S.A.); [email protected], [email protected]
Copyright c (cid:13)
Number of pages: 31Number of tables: 6Number of figures: 7
Preprint submitted to Icarus 11 September 2018 roposed Running Head: CO in Comets PanSTARRS, LINEAR, and J¨ager Please send Editorial Correspondence to:
Adam J. McKayUniversity of Texas Austin2512 Speedway, Stop C1402Austin, TX 78712, USA.Email: [email protected]: (512) 471-6493 2
BSTRACT
Carbon dioxide is one of the most abundant ices present in comets and istherefore important for understanding cometary composition and activity. Wepresent analysis of observations of CO and [O I ] emission in three cometsto measure the CO abundance and evaluate the possibility of employing ob-servations of [O I ] emission in comets as a proxy for CO . We obtained NIRimaging sensitive to CO of comets C/2012 K1 (PanSTARRS), C/2012 K5(LINEAR), and 290P/J¨ager with the IRAC instrument on Spitzer . We ac-quired observations of [O I ] emission in these comets with the ARCES echellespectrometer mounted on the 3.5-meter telescope at Apache Point Observa-tory and observations of OH with the Swift observatory (PanSTARRS) andwith Keck HIRES (J¨ager). The CO /H O ratios derived from the
Spitzer im-ages are 12.6 ± ± ± abundance for PanSTARRSis close to the average abundance measured in comets at similar heliocentricdistance to date, while the abundances measured for LINEAR and J¨ager aresignificantly larger than the average abundance. From the coma morphologyobserved in PanSTARRS and the assumed gas expansion velocity, we derive arotation period for the nucleus of about 9.2 hours. Comparison of H O produc-tion rates derived from ARCES and
Swift data, as well as other observations,suggest the possibility of sublimation from icy grains in the inner coma. Weevaluate the possibility that the [O I ] emission can be employed as a proxy forCO by comparing CO /H O ratios inferred from the [O I ] lines to those mea-sured directly by Spitzer . We find that for PanSTARRS we can reproduce theobserved CO abundance to an accuracy of ∼ abundance inferred fromthe [O I ] lines. These upper limits are consistent with the CO abundancesmeasured by Spitzer . Keywords:
Comets; Comets, Coma; Comets, Composition4
Introduction
The abundances of CO and CO in comets may either reflect thermal evolu-tion (Belton and Melosh, 2009) or formation conditions (A’Hearn et al., 2012),with the distinct possibility that both evolution and formation conditions playa major role. The formation of CO likely occurs via grain surface interac-tions of OH and CO, though this reaction is not completely understood (e.g.Garrod and Pauly, 2011; Noble et al., 2011). Another possible pathway is di-rect oxidation of CO on grain surfaces (Minissale et al., 2013). In either case,this would imply that CO forms from destruction of CO and hence, if thesereactions are efficient (i.e. most CO in the protosolar disk is converted to CO via these reactions), on average CO should be more abundant than CO incomets. This abundance pattern could also be caused by thermal evolutiondue to CO being more volatile than CO or the protosolar disk inherentlyhaving a CO /CO ratio greater than unity in the comet forming region thatwas inherited from the ISM. In the case of evolution there should be observedtrends in CO/CO ratios as a function of the dynamical history of the comet.Definitive evidence for any such trend has not been observed (A’Hearn et al.,2012), though it is possible that not enough comets have been observed forany trend that is present to become apparent. Therefore, knowledge of the COand CO abundances in comets is paramount for creating a complete pictureof cometary composition and differentiating between the effect of formationconditions and subsequent thermal evolution on cometary composition.However, the lack of a permanent dipole moment for CO means it is bestobserved directly through its vibrational transitions at infrared wavelengths5CO also has electronic transitions, but these are very weak and have neverbeen observed astronomically). The only successful direct observations of CO in comets have been of its ν vibrational band at 4.26 µ m, which is heavily ob-scured by the presence of telluric CO and therefore cannot be observed fromthe ground. Before 2004, the CO abundance had been measured for only a fewcomets (Combes et al., 1988; Crovisier, 1997), with many more observationsbecoming available over the last 10 years thanks to space-borne assets. Ob-servations of CO in comets by Spitzer (Pittichov´a et al., 2008; Reach et al.,2009, 2013), AKARI (Ootsubo et al., 2010, 2012),
WISE (Bauer et al., 2011,2012; Stevenson et al., 2015; Bauer et al., 2015), and the Deep Impact space-craft (Feaga et al., 2007; A’Hearn et al., 2011; Feaga et al., 2014), have re-vealed that CO is the second most abundant gas present in most cometarycomae (behind H O). This may favor a mechanism where the CO presentin comets was formed via reactions that destroy CO and also possibly favorthe idea that the measured abundances are indeed primordial. Observations ofCameron band emission of CO with HST have also been employed as a proxyfor CO (Weaver et al., 1994). However, Cameron band emission has a signif-icant contribution from electron impact excitation of CO, which complicatesderivation of the CO abundance (Bhardwaj and Raghuram, 2011).The only facilities that can currently observe CO in comets are the Spitzer
Space Telescope and the
WISE spacecraft (directly in the IR), as well as
HST (through Cameron band emission.) However, as these are all space-borne fa-cilities, the observing time available is limited. In addition,
Spitzer and
WISE all have very stringent elongation requirements, meaning many objects gounobserved (even though
WISE/NEOWISE is a survey, it only observes at90 ◦ elongation, meaning observations of comets are serendipitous and can-6ot be planned for detailed study of a particular comet). The James WebbSpace Telescope (JWST) is expected to supercede the capabilities providedby Spitzer and
WISE for cometary science (Kelley et al., 2015), but JWSTobserving time for comets may be limited. Ground-based observations are ingeneral more accessible than space-borne assets, allowing for more detailedstudy of a larger number of objects. Therefore, establishment of an indirect,ground-based measure of CO abundances in comets is vital in order to pro-vide the number of measurements needed for further interpretation of cometorigin and evolution.As atomic oxygen is a photodissociation product of CO , observations of theforbidden oxygen lines at 5577, 6300, and 6364 ˚A can serve as a viable proxy.These forbidden lines are fairly bright features in cometary spectra and can bereadily observed in moderately bright comets (V=10) with medium aperturetelescopes (2-3 meter class) (Capria et al., 2005; Cochran, 2008; Decock et al.,2013; McKay et al., 2015, and references therein). However, the photochem-istry of O I release from CO photodissociation, as well as from its otherprimary parents H O and CO, is still poorly understood, and limits the use-fulness of O I as a reliable proxy (McKay et al., 2013; Decock et al., 2013).However, if independent, contemporaneous measurements of H O, CO , CO,and O I are available, it is possible to employ comets as a “laboratory” toconstrain the relevant photochemistry.We present analysis of Spitzer
Infrared Array Camera (IRAC) imaging ofcomets C/2012 K1 (PanSTARRS), C/2012 K5 (LINEAR), and 290P/J¨ager(hereafter PanSTARRS, LINEAR, and J¨ager, respectively), which we employto measure the CO production rate in each comet. We also present high7esolution optical spectroscopy of these comets in an effort to observe the[O I ] emission and therefore constrain the photochemistry responsible for therelease of O I into the coma. In section 2 we describe our observations andreduction and analysis procedures. Section 3 presents our CO productionrates and [O I ] line measurements, and a comparison of the CO abundancesinferred from the [O I ] emission and the abundances measured with Spitzer .In section 4 we discuss how the measured CO production rates fit in with thegrowing sample of CO observations in comets, as well as the implications ofour results for the photochemistry of O I release and the ability to use [O I ]observations as a proxy for CO . Section 5 summarizes our conclusions. We obtained NIR images at 3.6 and 4.5 µ m for studying CO using the IRACinstrument on Spitzer (Werner et al., 2004; Fazio et al., 2004), while we ob-tained optical spectra for studying atomic oxygen with the ARCES echellespectrometer mounted on the Astrophysical Research Consortium 3.5-m tele-scope at Apache Point Observatory (APO) in Sunspot, New Mexico. We alsoobtained optical spectra of J¨ager with the HIRES instrument mounted onKeck I and imaging of PanSTARRS with the
Swift spacecraft to measure theOH production rate, which gives us a measure of the H O production rate.8 .1.1 CO - Spitzer IRAC As Spitzer is well into its post-cryogenic mission, IRAC presently observesin two pass bands: one centered at 3.6 µ m and the other at 4.5 µ m. Bothfilters have broad wavelength coverage, with bandwidths of 0.8 and 1.0 µ m,respectively. The 4.5 µ m band is very useful for measuring the CO abun-dances in comets, as this pass band includes the ν transition at 4.26 µ m. Italso contains the ν (1-0) band of CO at 4.7 µ m, but in 15 out of 17 cometsin the AKARI survey (Ootsubo et al., 2012), the CO feature was at least 10times brighter than the CO feature, and so CO is typically the dominant gasemission feature in the IRAC 4.5 µ m band. This is due to the fluorescenceefficiency of CO being approximately an order of magnitude larger than thatfor CO, while the CO abundance in comets is typically equal to or less thanthe CO abundance. While this is true for most comets, there are examples,such as C/2006 W3 (Christensen) and 29P/Schwassman-Wachmann 1, whereCO emission contributes significantly (more than 20%) to the 4.5 µ m bandflux (Ootsubo et al., 2012; Reach et al., 2013).We supply details of our observations in Table 1. The IRAC array is a 256x 256 pixel InSb array, covering a 5’ x 5’ region on the sky. We performedobservations of each comet field several days after each cometary observationin order to image the field without the comet in it. These observations aretermed “shadow observations” and provide a measurement of the backgroundto be subtracted from the cometary images. We observed each comet in highdynamic range mode. This entailed obtaining exposures with both short andlong exposure times in order to avoid saturation of the inner coma, while stillkeeping high signal-to-noise ratio (SNR) in the fainter outer coma (details of9he exposure times used are given in Table 1). Observing in high dynamicrange mode also helps protect against saturation due to bright field stars.For these observations no pixels were saturated in the longest exposure times,therefore we performed analysis on the longest exposure time images for op-timal SNR.For each comet, we combined all images of the same exposure time using theMOPEX software (Makovoz and Khan, 2005). This process creates a mosaicin the rest frame of the comet from the individual images, averaging over-lapping data together, but ignoring cosmic rays and bad pixels. Two mosaicsare created: one for the comet data, the other for the shadow (background)data. We subtracted the shadow mosaic from the comet mosaic to remove thebackground. This includes zodiacal light and celestial sources.After the mosaic images were created and the sky background was separated,the next step was to remove the dust contribution from the 4.5 µ m band flux,isolating the gas emission. We accomplished this via the following method.First we split the 4.5 µ m band image into wedges centered on the optocenterof the comet (for PanSTARRS, this consituted 20 wedges, while for J¨ager andLINEAR the best results were derived assuming spherical symmetry, i.e. nosplitting of the images into wedges was applied). We then fit the 4.5 µ m bandimage morphology in each wedge (or in the case of LINEAR and J¨ager, thewhole image) to a model consisting of a 1/ ρ profile (to approximate the gas)plus the 3.6 µ m band image (indicative of the dust). Both the 1/ ρ profile (i.e.gas model) and 3.6 µ m band fluxes were allowed to be multiplied by a scalefactor. The gas scale factor was allowed to vary from wedge to wedge, but the10caling factor for the 3.6 µ m band was forced to be the same for all wedges.The key parameter we retrieved from this modelling is the 3.6 µ m band scalefactor. Lastly, we multiplied the 3.6 µ m band image by the retrieved scalefactor and subtracted the scaled 3.6 µ m band image from the 4.5 µ m bandimage to obtain a dust-subtracted image.From the dust-subtracted image, we measured the flux for apertures rangingfrom 6-60 pixels (7-70”) in radius. We converted the broadband photometryto CO line fluxes following the IRAC data handbook (Laine, 2015). The linefluxes were then used to calculate average column densities inside each aper-ture employing fluorescence efficiencies from Crovisier and Encrenaz (1983).Then the production rate Q is given by Q = < N > vd (1)where < N > is the average column density in the photometric aperture, v is the expansion velocity, and d is the projected diameter of the photometricaperture. We assume an expansion velocity of the coma following Tseng et al.(2007): v = 0 . R − . h (2)where R h denotes the heliocentric distance in AU of the comet and the derivedvelocity is in km s − . This approach assumes a negligible effect of photodisso-ciation on the spatial profile in the photometric aperture, but as our aperturesare <
10% of the CO scale length, this approximation is justified. We calcu-lated production rates for a variety of aperture sizes to quantify any trends in11erived production rates with aperture size. We find that small trends withaperture size are present, which may be due to residual dust, structure in thegas coma (i.e., the assumed 1/ ρ relation for the gas surface brightness profileis violated), and/or the difference between the 3.6 and 4.5 µ m point-spreadfunctions. Residual dust may be present due to color variations in the coma.We prefer to keep our approach simple, and refrain from using additionalfree-parameters and assumptions to model these deviations. The uncertain-ties adopted for the CO production rates are either derived from the 3.6 µ mmodel scale factor uncertainties (LINEAR and J¨ager), or from the standarddeviation of the derived production rates for all apertures (PanSTARRS),whichever is larger.To enhance structures in the coma morphology, we divided each image by a1/ ρ profile, where ρ is the projected distance from the optocenter. A 1/ ρ profileis what is expected for a coma in steady state expansion. Any deviations fromthis theoretical spatial profile are enhanced in the resulting image, allowingstudies of the coma morphology to be performed. I - ARCES ARCES is a cross-dispersed echelle spectrometer, providing a spectral reso-lution of R ≡ λ ∆ λ = 31,500 and a spectral range of 3500-10,000 ˚A with nointerorder gaps. This large, uninterrupted spectral range allows for simultane-ous observations of all three oxygen lines. More specifics for this instrumentare discussed elsewhere (Wang et al., 2003).12he observation dates and geometries are described in Table 2. All nightswere photometric, meaning absolute flux calibration of the spectra was pos-sible. We used an ephemeris generated from JPL Horizons for non-siderealtracking. Guiding was accomplished using a boresight technique, which uti-lizes optocenter flux that falls outside the slit to keep the slit on the optocenter.We observed a G2V star, a fast rotating (vsin(i) >
150 km s − ) O, B, or Astar, and a flux standard for calibration of the comet spectra. The calibrationstars used for each observation date are given in Table 2. We obtained spectraof a quartz lamp for flat fielding and acquired spectra of a ThAr lamp forwavelength calibration.Spectra were extracted and calibrated using Image Reduction and AnalysisFacility (IRAF) scripts that perform bias subtraction, cosmic ray removal, flatfielding, and wavelength calibration. We employed the fast-rotator spectrumto remove telluric features, the flux standard spectrum to convert from countsto physical units, and the solar analog spectrum to remove Fraunhofer lines.We assumed an exponential extinction law and extinction coefficients for APOwhen flux calibrating the cometary spectra. More details of our reduction pro-cedures can be found in McKay et al. (2015) and references therein.Because of the small size of the ARCES slit, it is necessary to obtain an es-timate of the slit losses to achieve an accurate flux calibration. We find thetransmittance through the slit by performing aperture photometry on the slitviewer images as described in McKay et al. (2014). This introduces a 10% un-13ertainty in our absolute flux calibration.The O I lines are also present as a telluric emission feature, meaning a combi-nation of high spectral resolution and large geocentric velocity (and thereforelarge Doppler shift) is needed to separate the cometary line from the telluricfeature. For all observations the telluric and cometary lines are well sepa-rated. We fit the line profiles using the Gaussian-fitting method describedin McKay et al. (2012). Emission from the C ∆v=-1 Swan band can alsocontaminate the cometary 5577 ˚A feature (e.g. Decock et al., 2013). How-ever, there is no trace of C emission in any of the observed comets in thewavelength region surrounding the 5577 ˚A feature. Therefore we consider anycontamination from C negligible. The 6300 ˚A and 6364 ˚A lines are both tran-sitions from the D to the P ground state, therefore the flux ratio reflects thebranching ratio for these transitions of 3.0. This means that the flux ratioof these lines is independent of the coma physics, and the expected value of3.0 is well established by both theory and observation (Sharpee and Slanger,2006; Cochran and Cochran, 2001; Cochran, 2008; McKay et al., 2012, 2013;Decock et al., 2013). As a check of our analysis procedures, we confirmed thatwe reproduced this ratio before proceeding with further analysis.With the measured line fluxes, we calculate the oxygen line ratio, defined as R = I I + I (3)where I y denotes the flux of line y . The CO /H O ratio can be inferredfrom the oxygen line ratio using the following relation (McKay et al., 2012;14ecock et al., 2013): N CO N H O = RW redH O − W greenH O − W greenCO N CO N H O + RW redCO N CO N H O W greenCO − RW redCO (4)where N is column density and R is the oxygen line ratio. The release rate W is defined as W ≡ τ − αβ (5)where τ represents the photodissociative lifetime of the parent molecule, α isthe yield into the excited state of interest, and β represents the branching ratiofor a given line out of a certain excited state. This relation is derived by notingthat the line flux contributed from each species is given by the product ofcolumn density N and release rate W , substituting this into Eq. 3 and solvingfor N CO N H O (see McKay et al. (2012) for more details). We ignore the contributionof more complex oxygen-bearing molecules like H CO and CH OH as thesespecies are less abundant than H O, CO , and CO and release oxygen througha multi-step process, making them very inefficient at contributing to the O I population. If the contribution of CO photodissociation to the O I populationis also considered negligible (Raghuram and Bhardwaj, 2014), Eq. 4 simplifiesto (McKay et al., 2013): N CO N H O = RW redH O − W greenH O W greenCO − RW redCO (6)The results of Eq. 4 and 6 are independent of heliocentric distance. For smallfields of view, the column density ratio reflects the production rate ratio (seeMcKay et al., 2015, and references therein for more details).We performed additional analysis accounting for preferential collisional quench-15ng of D atoms (responsible for the 6300 ˚A and 6364 ˚A lines) as comparedto S atoms (responsible for the 5577 ˚A line), which can be important forsmall fields of view or high production rates (Bhardwaj and Raghuram, 2012;Raghuram and Bhardwaj, 2014; Decock et al., 2015). The oxygen line ratioemployed in Eqs. 4 and 6 assumes the ratio was calculated using 6300 ˚A and6364 ˚A line intensities that are unaffected by collisional quenching. Since thismay not be the case, the observed 6300 ˚A and 6364 ˚A line intensities need tobe increased to account for the D atoms that were de-excited through colli-sions and so do not contribute to the 6300 ˚A and 6364 ˚A line intensities. Inorder to account for this, we need to model the number density of the dom-inant collisional partner, H O. Therefore an estimate of the H O productionrate is needed.We determined H O production rates from our [O I ]6300 ˚A line observationsby employing algorithms based on those used in Morgenthaler et al. (2007)and McKay et al. (2012), which involves a Haser model modified to emulatethe more physical vectorial model. With an H O production rate in hand, weestimate the percentage of atoms lost to collisional quenching by employing thealgorithms mentioned above to estimate the expected [O I ]6300 ˚A flux withoutcollisional quenching. The correction factor is then simply the expected fluxwithout quenching divided by the observed flux. More details concerning thismethod are presented in McKay et al. (2015).16 .3 OH-Swift and HIRES The
Swift telescope (Gehrels et al., 2004) observed PanSTARRS on May 6and 7, 2014 at R=2.04 AU from the Sun. We employed the UVOT instru-ment (Mason et al., 2004; Roming et al., 2005) to obtain photometry of thecomet. UVOT’s broadband filters provide a measure of the comet’s water anddust production rates (Bodewits et al., 2014). We obtained photometry usingbroadband V (central λ λ Swan band emission in the V-band filter. Cor-recting for the filter transmission at the relevant wavelengths, the measuredfluxes can be converted into column densities using heliocentric distance andvelocity dependent fluorescent efficiencies (Schleicher and A’Hearn, 1988). Toderive water production rates, we compare the measured OH content of thecoma with an OH distribution calculated using the vectorial model (Festou,1981; Combi et al., 2004). Most of the uncertainty in the derived productionrates is introduced from the modeling, with a negligible contribution comingfrom photon noise. We measured fluxes in several aperture sizes, and adopt thestandard deviation of the derived production rates as our 1-sigma uncertainty.For J¨ager, we obtained observations of OH with the HIRES instrument (Vogt et al.,1994) on Keck I in January 2014. The HIRESb configuration provides obser-vations of the OH ∆ v =0 band at 3080 ˚A . We utilized the 0.86 × O production rate using a Haser model that has been modified to em-ulate the vectorial model (see McKay et al. (2014) for more details) and thefluorescence efficiency from Schleicher and A’Hearn (1988). The scale lengthsfor the Haser model are adopted from Cochran and Schleicher (1993).
We provide measured fluxes for CO from our Spitzer observations (including3.6 µ m image scale factors, see section 2.1.1), [O I ]6300 ˚A emission from ourARCES observations, and OH from our Swift and Keck HIRES observationsin Table 3. All uncertainties are 1-sigma. O Production Rates and Collisional Quenching Factors
We show a spectrum of J¨ager showing the OH lines (from which we derivedthe H O production rate) in Fig. 1. As discussed in Section 2.3, the noisy back-ground is largely due to scattered sunlight from the Full Moon. We presentour H O production rates and collisional quenching correction factors in Ta-ble 4. A small collisional quenching factor is required for the PanSTARRSdata, while for LINEAR and J¨ager the effect is negligible due to their much18maller H O production rates. As we are using [O I ]6300 emission to deriveH O production rates, it is desirable to have independent production ratesdetermined via other methods to confirm that there is no systematic errorbeing introduced by employing [O I ] emission.PanSTARRS was by far the brightest of these comets, and therefore severalother measurements of the H O production rate are available. Gibb et al. (pri-vate communication) measured the H O production rate with NIRSPEC, andtheir value is consistent with our value of (4.35 ± × molecules s − derived from [O I ]6300 emission to within ∼ Swift /UVOT ob-servations we derived a water production rate of (9.5 ± × moleculess − in apertures between 50-200 arcsec (5.3 × - 2.1 × km at the comet).Analysis of OH observations by Knight and Schleicher (2014) and of Lyman- α emission by Combi et al. (2014) derive similar production rates. One possibil-ity for the discrepency is that the H O production rate depends on rotationalphase of the nucleus. However, as our [O I ]6300 observations and the NIR-SPEC observations occured on completely different nights, it is unlikely thatboth would have sampled the same part of the rotational variation. In addition,this would imply that the observations of Schleicher, Combi et al. (2014), andour Swift observations (which also occured on different dates) would have allsampled the same part of the rotational variation that was also distinct fromthat sampled by our [O I ]6300 observations and the NIRSPEC observations.A more likely possibility is related to the fields of view (FOV) of the differenttelescope/instrument combinations employed. Our [O I ]6300 observations andthe NIRSPEC observations of Gibb et al. both employed narrow slits (pro-jected FOV at the comet on the order of several thousand km), while the Swift , Combi et al. (2014), and Schleicher observations all used much larger19OV, on the order of tens to hundreds of thousands of kilometers. A similar de-pendence of derived H O production rates with FOV was observed for C/2009P1 (Garradd) (Combi et al., 2013; Bodewits et al., 2014; DiSanti et al., 2014;Feaga et al., 2014; McKay et al., 2015). This was interpreted as an extendedsource of icy grains that sublimated outside the FOV of slit-based spectro-scopic measurements, but within the FOV of narrow band imaging observa-tions. A similar phenomenon could be applicable to PanSTARRS. This raisesthe question of which H O production rate is appropriate to adopt for com-parison to our [O I ] observations. As our O I observations have a small FOV,we will employ the H O production rate derived from small FOV observationsfor the analysis throughout the rest of this paper (but see Section 4.1 for morediscussion on how this affects comparison to other observations).There are no other sources of H O production rates available for J¨ager or LIN-EAR. We have observations of OH for J¨ager, but these are two months afterthe [O I ] observations. However, as the collisional quenching was determinedto be negligible for LINEAR and J¨ager, any systematic uncertainties in ourH O production rate due to employing [O I ]6300 emission to obtain an H Oproduction rate will have a negligible effect on our CO /H O ratios inferredfrom the oxygen line ratio.It is possible that systematic uncertainties in the H O production rate willalso affect the
Spitzer -derived CO /H O ratios. However, as discussed above,independent direct observations of H O using NIRSPEC are consistent withour adopted H O production rate for PanSTARRS. For J¨ager, our preferredH O production rate for comparison to the
Spitzer measurement of CO is20rom HIRES using OH due to this observation being more contemporaneouswith the Spitzer observations than the [O I ] observations. LINEAR has noindependent measure of the H O production rate, but the agreement of H Oproduction rates derived for PanSTARRS and J¨ager to other methods givesus confidence that our derived H O production rate for LINEAR is accurateto the quoted uncertainties. Production Rates and Coma Morphology
In Fig. 2 we show the
Spitzer
IRAC images of PanSTARRS, LINEAR, andJ¨ager. PanSTARRS was by far the brightest of the three comets observed, as isevident in the quality of the images. For PanSTARRS even in the raw mosaics(i.e. no image enhancement or dust subtraction), it is evident that in the 4.5 µ m image there is a diffuse, extended emission that is not present in the 3.6 µ m image and is likely due to CO or CO gas. We present the dust-subtractedimages in Fig. 3.We present the derived CO production rates and CO /H O ratios under theassumption of negligible CO emission in Table 4. The quoted uncertaintiesinclude only stochastic noise and uncertainties with the modeling used to iso-late the gaseous emission, and do not include any systematic error associatedwith any possible CO emission. While in principal there is both emission fromCO and CO in the 4.5 µ m image, NIRSPEC observations by Gibbs et al.(private communication) constrain the CO/H O ratio for PanSTARRS at ∼ µ m flux isminimal (on the order of 3%) and we can assume, within our uncertainties,21hat all the gas emission we observe is due to CO in this case. There areno independent measurements of CO available in J¨ager or LINEAR, mean-ing in these cases our CO production rates could in fact be upper limits.However, for most comets in the AKARI survey (Ootsubo et al., 2012), theCO emission was much weaker than that from CO . Therefore in general it islikely that our CO production rates for J¨ager and LINEAR are not contam-inated by CO emission, but without direct, independent observations of COwe cannot be certain. However, even with a CO/H O ratio as high as ∼ /H Oabundances from the
Spitzer observations of LINEAR and J¨ager only dropto about 23% and 27%, respectively. Therefore, for our derived CO abun-dances to change significantly, LINEAR and J¨ager would have to have ex-tremely abnormal CO/H O ratios ( > µ m images was not obvious and the detection of CO is sensitive to modelassumptions employed to isolate the gas emission. Therefore in this case ourderived CO production rate may be better interpreted as an upper limit.In Fig. 4, the top row shows (from left to right) the 3.6 µ m, 4.5 µ m, anddust-subtracted images of PanSTARRS. The bottom row is the same, except22hese images have been divided by a 1/ ρ profile to show coma features. Figs. 5and 6 show the analogous figures for LINEAR and J¨ager, respectively. All theimages show a clear tail excess. Even the gas images show some residual tails,suggesting that the dust subtraction is not perfect. For PanSTARRS, a spiralshape is visible in the 4.5 µ m image and the gas image, but is not present inthe 3.6 µ m image. This is likely the manifestation of a CO jet. Observationsof the CN morphology also show this spiral structure, while observations ofthe dust through R-band imaging do not (Knight and Schleicher, 2014). The Swift imaging of OH also does not show any discernible morphology. This mayindicate that the OH (and its parent H O) is released from the nucleus in amanner that is different from the CN parent and CO , but this could also bedue to any morphology that is present being blurred by the random directionof the velocity that OH receives after photodissociation of H O.We can use the separation between arcs of the spiral morphology to obtainan estimate of the rotation period. To determine the positions of the arcs,we measured the total flux in concentric annuli centered on the comet photo-center. Annuli containing the arcs will have higher flux than adjacent annuli.To increase the accuracy of the derived arc positions, we then fit the spatialdistribution of flux within the annuli containing an arc with a Gaussian func-tion plus constant background. We derive mean peak centers at 25.1 ± ± ±
300 and51200 ± ± I Line Ratios and Inferred CO Abundances
We present our oxygen line ratio measurements and 3-sigma upper limits inTable 5. Unfortunately, LINEAR and J¨ager were not bright enough for de-tection of the [O I ]5577 ˚A line, and the upper limits are not particularlyconstraining. However, PanSTARRS was much brighter and we have a firmdetection of the [O I ]5577 ˚A line, as shown in Fig. 7.We derive CO /H O ratios from our oxygen line ratios (or 3-sigma upper limitsin the case of LINEAR and J¨ager) using release rates from Bhardwaj and Raghuram(2012) and McKay et al. (2015). We summarize our CO abundances directlymeasured by Spitzer and our inferred CO abundances from our oxygen line24bservations in Table 5. The specific values for the release rates are given in Ta-ble 6. The rates from Bhardwaj and Raghuram (2012) are derived from a pho-tochemical model of a cometary coma, while the empirical rates from McKay et al.(2015) are rates that are able to reproduce the CO /H O ratio determinedby Feaga et al. (2014) for comet C/2009 P1 (Garradd). The difference be-tween empirical release rates A and B from McKay et al. (2015) is a factor of1.5 in the CO release rates that accounts for differences in the CO/H O abun-dance in Garradd measured by McKay et al. (2015) and Feaga et al. (2014).For PanSTARRS, we used Eq. 4, which includes the contribution of CO, witha CO/H O ratio of ∼
3% (Gibbs et al. private communication). As no inde-pendent measure of the CO abundance is available for J¨ager or LINEAR, weapplied Eq. 6, which assumes no contribution to the O I population from CO.If the contribution of CO is significant, this would not affect our upper limit,since including a contribution from CO only lowers the inferred upper limiton CO . Therefore our derived upper limits are true upper limits.Using release rates from Bhardwaj and Raghuram (2012), we infer a CO /H Oratio of ∼
4% for PanSTARRS, while the abundance measured by
Spitzer isapproximately 12%. The empirical release rates from McKay et al. (2015) re-produce the CO /H O ratio to better accuracy, predicting a CO /H O ratioof ∼
10% (release rates A) or ∼
14% (release rates B). The upper limits in-ferred for J¨ager and LINEAR using the McKay et al. (2015) release rates areconsistent with the values measured by
Spitzer , but do not provide furtherconstraints on their accuracy. The upper limit inferred for LINEAR using re-lease rates from Bhardwaj and Raghuram (2012) may be inconsistent withthe
Spitzer result (see section 4.2), but the J¨ager results are consistent with25he Bhardwaj and Raghuram (2012) release rates. Abundances
The CO /H O ratio of 12% measured for PanSTARRS is slightly lower thanthe mean of the AKARI survey of comets measured at heliocentric distancesof less than 2.5 AU (i.e. inside the canonical water sublimation line where sub-limation rates of H O, CO , and CO do not vary much with respect to eachother (e.g. Meech and Svoren, 2004)), which is approximately 17% (Ootsubo et al.,2012), but is well within the spread of values observed in comets at similarheliocentric distance to date. The observed CO abundances of 29% and 31%for LINEAR and J¨ager, respectively, are higher than any comet observed byAKARI within 2.5 AU from the Sun. However, these values are close to themean value of 30% found by Reach et al. (2013), although this data set hasmuch more scatter than the AKARI survey.As both LINEAR (1.6 AU) and J¨ager (2.2 AU) were observed at heliocen-tric distances less than 3 AU, sublimation effects are not likely responsiblefor these high abundances (Meech and Svoren, 2004). One possibility is thatsince AKARI observed over a much larger FOV (1’ × O that may have been present around the comets in their survey. Ifan extended source of water was present in J¨ager and LINEAR, this would26ave resulted in larger derived H O production rates and therefore smallerCO /H O ratios than would have been measured using H O prodution ratesfrom narrow slit observations such as ours. Therefore if we had used an obser-vational set up similar to AKARI with a large FOV to derive H O productionrates, our derived CO /H O ratios might have been be lower, bringing themeasured abundances for LINEAR and J¨ager closer to the mean value de-rived from AKARI. A similar effect is expected for the Reach et al. (2013)sample, as they adopted H O production rates from wide field OH imaging.However, without any data indicating the magnitude of an extended source ofH O production around LINEAR and J¨ager, we cannot evaluate this possibil-ity further. Another caveat to consider is the possibility that CO emission iscontributing to
Spitzer ’s 4.5 µ m filter, meaning the CO production rates arein fact lower than presented here. However, as discussed in Section 3.2, thisis not likely, as an abnormally large CO abundance is required to change thederived CO /H O ratio significantly. As mentioned in section 3.2, we cannotrule out the possibility that our detection of CO in LINEAR is better inter-preted as an upper limit, in which case its CO abundance would be moretypical.This study, while only adding three new comets to the sample, is consistentwith the findings of previous CO surveys in comets that the average CO abundance in comets is about 15-30%, higher than previously thought (Ootsubo et al.,2012; Reach et al., 2013). Only one of our comets has a measurement of theCO abundance (PanSTARRS), and the preliminary CO abundance in thiscomet derived by Gibbs et al. (private communication) is much less than theCO abundance, consistent with the idea that the formation of CO via grain-27urface reactions involving CO is a viable pathway for CO formation in theprotosolar disk. With no observations of CO in LINEAR or J¨ager available,we cannot reach any conclusions on the CO/CO ratio in those comets. I as a Proxy for CO Using our
Spitzer observations, we were able to compare actual CO abun-dances for these comets to CO abundances inferred using observations of theoxygen line ratio. For PanSTARRS, the Bhardwaj and Raghuram (2012) re-lease rates underestimate the CO abundance by about a factor of three. Asimilar discrepency was found for C/2009 P1 (Garradd) (McKay et al., 2015).The upper limits inferred from the oxygen line ratio for J¨ager are not par-ticularly constraining, as all three sets of release rates provide upper limitsconsistent with the Spitzer measurements. However, for LINEAR the upperlimit on CO /H O using the Bhardwaj and Raghuram (2012) release rates issimilar to the value measured by
Spitzer . While not conclusive, this suggeststhat we should have been able to detect the [O I ]5577 ˚A line in this comet,which we did not. However, a lack of knowledge of the CO abundance and thequality of the data prevent us from making a firm conclusion. Therefore thereis suggestive (but not conclusive) evidence that the Bhardwaj and Raghuram(2012) release rates do not reproduce the LINEAR observations.The empirical release rates from McKay et al. (2015) also reproduce the CO abundance observed in comet Garradd (by definition, as this was a requirementin the derivation of these release rates; the two sets of release rates correspondto different values of the CO/H O ratio used in Eq. 4). The ability of these28elease rates to reproduce the CO abundance in PanSTARRS to within anaccuracy of 20% is encouraging, but the lack of detections of the [O I ]5577line in J¨ager and LINEAR prevent further evaluation. The empirical releaserates B seem to reproduce the CO abundance in PanSTARRS more accu-rately than the release rates A. More simultaneous observations of CO andthe oxygen line ratios in comets are needed to further evaluate this method,and specifically the release rates proposed by McKay et al. (2015).It is important to stress that the release rates from McKay et al. (2015) arestrictly empirical. They seem to satisfactorily reproduce current observations,but there is no physical explanation for why they are different from those de-rived using photochemical models, such as those presented in Bhardwaj and Raghuram(2012). Laboratory measurements of the O I release rates are required to helpsettle this discrepancy. It may be possible that the release rates from McKay et al.(2015) are simply effective release rates. The release rates derived using pho-tochemical models might be correct in the strict sense, but perhaps otherphysical processes occur in the coma (collisional processes, radiative transfereffects, etc.) that modify the [O I ] emission so that applying those release ratesto remote sensing observations does not reproduce the measured CO abun-dance of the comet. A more detailed understanding of the coma environmentand its effect on [O I ] emission is needed.29 Conclusions
We have presented near-contemporaneous observations of CO using Spitzer
IRAC and observations of the forbidden oxygen lines in three comets: C/2012K1 (PanSTARRS), C/2012 K5 (LINEAR), and 290P/J¨ager, as well as obser-vations of OH in PanSTARRS and J¨ager. Our measured CO abundances arewithin the spread of values previously observed, corroborating previous obser-vations that the typical CO /H O ratio in comets is in the range 15-30% andconsistent with the theory that CO forms via grain surface reactions involvingCO. We find evidence for a possible extended source for H O sublimation inPanSTARRS, which we interpret as an icy grain halo, similar to that observedfor C/2009 P1 (Garradd). We detected all three forbidden oxygen lines onlyfor PanSTARRS; for the other two comets only the 6300 ˚A and 6364 ˚A lineswere detected. Therefore for LINEAR and J¨ager we only obtained an upperlimit on the oxygen line ratio. We compared the CO abundance inferredfrom the oxygen line ratios to the CO abundance observed by Spitzer toevaluate our understanding of the photochemistry responsible for the releaseof O I into the coma. The upper limits derived for LINEAR and J¨ager are notparticularly constraining, but we determined that the empirical release ratesfrom McKay et al. (2015) reproduced the CO abundance in PanSTARRSmore accurately than the release rates from Bhardwaj and Raghuram (2012).The reason why the empirical release rates seem to reproduce the CO abun-dance more accurately than those determined from photochemical modelslike Bhardwaj and Raghuram (2012) is unclear. More work is needed on allfronts, observational, laboratory, and theoretical, to fully understand O I emis-sion in comets and employ it as a reliable proxy for CO .30 cknowledgements We thank two anonymous reviewers whose comments improved the qualityof this manuscript. This work was supported by the NASA Planetary Atmo-spheres Program through grant number NNX08A052G. This work is partiallybased on observations made with the
Spitzer
Space Telescope, which is op-erated by the Jet Propulsion Laboratory, California Institute of Technologyunder a contract with NASA. We thank the APO and Keck observing stafffor their invaluable help in conducting the observations. We are thankful toMatthew Knight for productive discussions concerning the coma morphologyseen at optical wavelengths for C/2012 K1 (PanSTARRS), as well as DavidSchleicher, Michael Combi, and Erika Gibb for sharing their unpublished pro-duction rates. We thank John Barentine, Jurek Krzesinski, Chris Churchill,Pey Lian Lim, Paul Strycker, and Doug Hoffman for developing and opti-mizing the ARCES IRAF reduction script used to reduce the ARCES data.We would also like to acknowledge the JPL Horizons System, which was usedto generate ephemerides for nonsidereal tracking of the comets during theARCES observations, and the SIMBAD database, which was used for selec-tion of reference stars. The authors wish to recognize and acknowledge thevery significant cultural role and reverence that the summit of Maunakea hasalways had within the indigenous Hawaiian community. We are most fortunateto have the opportunity to conduct observations from this mountain.31 eferences
M. F. A’Hearn, M. J. S. Belton, W. A. Delamere, L. M. Feaga, D. Hampton,J. Kissel, K. P. Klaasen, L. A. McFadden, K. J. Meech, H. J. Melosh, P. H.Schultz, J. M. Sunshine, P. C. Thomas, J. Veverka, D. D. Wellnitz, D. K.Yeomans, S. Besse, D. Bodewits, T. J. Bowling, B. T. Carcich, S. M. Collins,T. L. Farnham, O. Groussin, B. Hermalyn, M. S. Kelley, M. S. Kelley, J.-Y.Li, D. J. Lindler, C. M. Lisse, S. A. McLaughlin, F. Merlin, S. Protopapa,J. E. Richardson, and J. L. Williams. EPOXI at Comet Hartley 2.
Science ,332:1396–, June 2011. doi: 10.1126/science.1204054.M. F. A’Hearn, L. M. Feaga, H. U. Keller, H. Kawakita, D. L. Hampton,J. Kissel, K. P. Klaasen, L. A. McFadden, K. J. Meech, P. H. Schultz,J. M. Sunshine, P. C. Thomas, J. Veverka, D. K. Yeomans, S. Besse,D. Bodewits, T. L. Farnham, O. Groussin, M. S. Kelley, C. M. Lisse,F. Merlin, S. Protopapa, and D. D. Wellnitz. Cometary Volatiles andthe Origin of Comets.
Astrophysical Journal , 758:29, October 2012. doi:10.1088/0004-637X/758/1/29.J. M. Bauer, R. G. Walker, A. K. Mainzer, J. R. Masiero, T. Grav, J. W. Dai-ley, R. S. McMillan, C. M. Lisse, Y. R. Fern´andez, K. J. Meech, J. Pittichova,E. K. Blauvelt, F. J. Masci, M. F. A’Hearn, R. M. Cutri, J. V. Scotti, D. J.Tholen, E. DeBaun, A. Wilkins, E. Hand, E. L. Wright, and WISE Team.WISE/NEOWISE Observations of Comet 103P/Hartley 2.
AstrophysicalJournal , 738:171, September 2011. doi: 10.1088/0004-637X/738/2/171.J. M. Bauer, E. Kramer, A. K. Mainzer, R. Stevenson, T. Grav, J. R.Masiero, R. G. Walker, Y. R. Fern´andez, K. J. Meech, C. M. Lisse,P. R. Weissman, R. M. Cutri, J. W. Dailey, F. J. Masci, D. J. Tholen,G. Pearman, E. L. Wright, and WISE Team. WISE/NEOWISE Prelim-32nary Analysis and Highlights of the 67p/Churyumov-Gerasimenko nearNucleus Environs.
Astrophysical Journal , 758:18, October 2012. doi:10.1088/0004-637X/758/1/18.J. M. Bauer, R. Stevenson, E. Kramer, A. K. Mainzer, T. Grav, J. R. Masiero,Y. R. Fern´andez, R. M. Cutri, J. W. Dailey, F. J. Masci, K. J. Meech,R. Walker, C. M. Lisse, P. R. Weissman, C. R. Nugent, S. Sonnett, N. Blair,A. Lucas, R. S. McMillan, E. L. Wright, t. WISE, and NEOWISE Teams.The NEOWISE-Discovered Comet Population and the CO+CO2 produc-tion rates.
ArXiv e-prints , September 2015.M. J. S. Belton and J. Melosh. Fluidization and multiphase transport ofparticulate cometary material as an explanation of the smooth terrains andrepetitive outbursts on 9P/Tempel 1.
Icarus , 200:280–291, March 2009. doi:10.1016/j.icarus.2008.11.012.A. Bhardwaj and S. Raghuram. Model for Cameron-band emission in comets:a case for the EPOXI mission target comet 103P/Hartley 2.
Monthly Noticesof the Royal Astronomical Society , 412:L25–L29, March 2011. doi: 10.1111/j.1745-3933.2010.00998.x.A. Bhardwaj and S. Raghuram. A Coupled Chemistry-emission Model forAtomic Oxygen Green and Red-doublet Emissions in the Comet C/1996B2 Hyakutake.
Astrophysical Journal , 748:13, March 2012. doi: 10.1088/0004-637X/748/1/13.N. Biver, D. Bockel´ee-Morvan, P. Colom, J. Crovisier, F. Henry, E. Lel-louch, A. Winnberg, L.E.B. Johansson, M. Gunnarsson, H. Rickman,F. Rantakyr¨o, Davies. J.K., W.R.F. Dent, G. Paubert, R. Moreno, J. Wink,D. Despois, D. Benford, M. Gardner, D.C. Lis, D. Mehringer, T.G. Phillips,and H. Rauer. The 1995-2002 long-term monitoring of comet C/1995 O1(Hale-Bopp) at radio wavelength.
Earth, Moon, and Planets , 90(1):5–14,33002.D. Bodewits, T. L. Farnham, M. F. A’Hearn, L. M. Feaga, A. McKay, D. G.Schleicher, and J. M. Sunshine. The Evolving Activity of the DynamicallyYoung Comet C/2009 P1 (Garradd).
Astrophysical Journal , 786:48, May2014. doi: 10.1088/0004-637X/786/1/48.M.T. Capria, G. Cremonese, A. Bhardwaj, and M.C. de Sanctis. O S and O Demission lines in the spectrum of 153P/2002 C1 (Ikeya-Zhang).
Astronomyand Astrophysics , 442(3):1121–1126, 2005.A. L. Cochran and D. G. Schleicher. Observational Constraints on the Lifetimeof Cometary H O. Icarus , 105:235–253, September 1993. doi: 10.1006/icar.1993.1121.A.L. Cochran. Atomic oxygen in the comae of comets.
Icarus , 198(1):181–188,2008.A.L. Cochran and W.D. Cochran. Observations of O ( S) and O ( D) inspectra of C/1999 S4 (LINEAR).
Icarus , 154(2):381–390, 2001.M. Combes, J. Crovisier, T. Encrenaz, V. I. Moroz, and J.-P. Bibring. The2.5-12 micron spectrum of Comet Halley from the IKS-VEGA Experiment.
Icarus , 76:404–436, December 1988. doi: 10.1016/0019-1035(88)90013-9.M. Combi, W. Harris, and W. Smyth. Gas dynamics and kinetics in thecometary coma: Theory and observations. In
Comets II , pages 523–552,Tucson, AZ, USA, 2004. University of Arizona Press.M. R. Combi, J. T. T. M¨akinen, J.-L. Bertaux, E. Qu´emerais, S. Ferron,and N. Fougere. Water production rate of Comet C/2009 P1 (Garradd)throughout the 2011-2012 apparition: Evidence for an icy grain halo.
Icarus ,225:740–748, July 2013. doi: 10.1016/j.icarus.2013.04.030.M. R. Combi, J. T. M¨akinen, J. L. Bertaux, E. Qu´emerais, and S. Ferron. TheWater Production Rate of Recent Comets (2013-2014) by SOHO/SWAN:34P/Encke (2013), C/2013 R1 (Lovejoy), and C/2013 A1 (Siding Spring).In
AAS/Division for Planetary Sciences Meeting Abstracts , volume 46of
AAS/Division for Planetary Sciences Meeting Abstracts , page 110.09,November 2014.J. Crovisier. Infrared observations of volatile molecules in comet Hale-Bopp.
Earth, Moon, and Planets , 79(1/3):125–143, 1997.J. Crovisier and T. Encrenaz. Infrared fluorescence of molecules in comets -The general synthetic spectrum.
Astronomy and Astrophysics , 126:170–182,September 1983.A. Decock, E. Jehin, D. Hutsem´ekers, and J. Manfroid. Forbidden oxygen linesin comets at various heliocentric distances.
Astronomy and Astrophysics ,555:A34, July 2013. doi: 10.1051/0004-6361/201220414.A. Decock, E. Jehin, P. Rousselot, D. Hutsem´ekers, J. Manfroid, S. Raghuram,A. Bhardwaj, and B. Hubert. Forbidden oxygen lines at various nucleocen-tric distances in comets.
Astronomy and Astrophysics , 573:A1, January2015. doi: 10.1051/0004-6361/201424403.M. A. DiSanti, G. L. Villanueva, L. Paganini, B. P. Bonev, J. V. Keane, K. J.Meech, and M. J. Mumma. Pre- and post-perihelion observations of C/2009P1 (Garradd): Evidence for an oxygen-rich heritage?
Icarus , 228:167–180,January 2014. doi: 10.1016/j.icarus.2013.09.001.G. G. Fazio, J. L. Hora, L. E. Allen, M. L. N. Ashby, P. Barmby, L. K. Deutsch,J.-S. Huang, S. Kleiner, M. Marengo, S. T. Megeath, G. J. Melnick, M. A.Pahre, B. M. Patten, J. Polizotti, H. A. Smith, R. S. Taylor, Z. Wang, S. P.Willner, W. F. Hoffmann, J. L. Pipher, W. J. Forrest, C. W. McMurty,C. R. McCreight, M. E. McKelvey, R. E. McMurray, D. G. Koch, S. H.Moseley, R. G. Arendt, J. E. Mentzell, C. T. Marx, P. Losch, P. Mayman,W. Eichhorn, D. Krebs, M. Jhabvala, D. Y. Gezari, D. J. Fixsen, J. Flores,35. Shakoorzadeh, R. Jungo, C. Hakun, L. Workman, G. Karpati, R. Kichak,R. Whitley, S. Mann, E. V. Tollestrup, P. Eisenhardt, D. Stern, V. Gorjian,B. Bhattacharya, S. Carey, B. O. Nelson, W. J. Glaccum, M. Lacy, P. J.Lowrance, S. Laine, W. T. Reach, J. A. Stauffer, J. A. Surace, G. Wilson,E. L. Wright, A. Hoffman, G. Domingo, and M. Cohen. The Infrared ArrayCamera (IRAC) for the Spitzer Space Telescope.
Astrophysical JournalSupplement Series , 154:10–17, September 2004. doi: 10.1086/422843.L. M. Feaga, M. F. A’Hearn, T. L. Farnham, D. Bodewits, J. M. Sun-shine, A. M. Gersch, S. Protopapa, B. Yang, M. Drahus, and D. G. Schle-icher. Uncorrelated Volatile Behavior during the 2011 Apparition of CometC/2009 P1 Garradd.
Astronomical Journal , 147:24, January 2014. doi:10.1088/0004-6256/147/1/24.L.M. Feaga, M.F. A’Hearn, J.M. Sunshine, O. Groussin, and T.L. Farnham.Asymmetries in the distribution of H O and CO in the inner coma of comet9P/Tempel 1 as observed by Deep Impact. Icarus , 190(2):345–356, 2007.M.C. Festou. The density distribution of neutral compounds in cometaryatmospheres I - models and equations.
Astronomy and Astrophysics , 95(1):69–79, 1981.R. T. Garrod and T. Pauly. On the Formation of CO and Other InterstellarIces. Astrophysical Journal , 735:15, July 2011. doi: 10.1088/0004-637X/735/1/15.N. Gehrels, G. Chincarini, P. Giommi, K. O. Mason, J. A. Nousek, A. A. Wells,N. E. White, S. D. Barthelmy, D. N. Burrows, L. R. Cominsky, K. C. Hurley,F. E. Marshall, P. M´esz´aros, P. W. A. Roming, L. Angelini, L. M. Barbier,T. Belloni, S. Campana, P. A. Caraveo, M. M. Chester, O. Citterio, T. L.Cline, M. S. Cropper, J. R. Cummings, A. J. Dean, E. D. Feigelson, E. E.Fenimore, D. A. Frail, A. S. Fruchter, G. P. Garmire, K. Gendreau, G. Ghis-36llini, J. Greiner, J. E. Hill, S. D. Hunsberger, H. A. Krimm, S. R. Kulka-rni, P. Kumar, F. Lebrun, N. M. Lloyd-Ronning, C. B. Markwardt, B. J.Mattson, R. F. Mushotzky, J. P. Norris, J. Osborne, B. Paczynski, D. M.Palmer, H.-S. Park, A. M. Parsons, J. Paul, M. J. Rees, C. S. Reynolds,J. E. Rhoads, T. P. Sasseen, B. E. Schaefer, A. T. Short, A. P. Smale, I. A.Smith, L. Stella, G. Tagliaferri, T. Takahashi, M. Tashiro, L. K. Townsley,J. Tueller, M. J. L. Turner, M. Vietri, W. Voges, M. J. Ward, R. Willingale,F. M. Zerbi, and W. W. Zhang. The Swift Gamma-Ray Burst Mission.
Astrophysical Journal , 611:1005–1020, August 2004. doi: 10.1086/422091.M. S. P. Kelley, C. E. Woodward, D. Bodewits, T. L. Farnham, M. Gudipati,D. E. Harker, D. C. Hines, M. M. Knight, L. Kolokolova, A. Li, I. de Pater,S. Protopapa, R. W. Russell, and Wooden D. W. Sitko, M. L. CometaryScience with the James Webb Space Telescope. submitted, 2015.M. M. Knight and D. G. Schleicher. Coma Morphology of Recent Comets:C/ISON (2012 S1), C/Pan-STARRS (2012 K1), C/Jacques (2014 E2), andC/Siding Spring (2013 A1). In
AAS/Division for Planetary Sciences Meet-ing Abstracts , volume 46 of
AAS/Division for Planetary Sciences MeetingAbstracts , page 209.20, November 2014.S. Laine, editor.
IRAC Instrument Handbook .Spitzer Science Center, Pasadena, 2015. URL http://irsa.ipac.caltech.edu/data/SPITZER/docs/irac/iracinstrumenthandbook/ .D. Makovoz and I. Khan. Mosaicking with MOPEX. In P. Shopbell, M. Brit-ton, and R. Ebert, editors,
Astronomical Data Analysis Software and Sys-tems XIV , volume 347 of
Astronomical Society of the Pacific ConferenceSeries , page 81, December 2005.K. O. Mason, A. Breeveld, S. D. Hunsberger, C. James, T. E. Kennedy,P. W. A. Roming, and J. Stock. Performance of the UV/Optical Telescope37UVOT) on SWIFT. In K. A. Flanagan and O. H. W. Siegmund, editors,
X-Ray and Gamma-Ray Instrumentation for Astronomy XIII , volume 5165of
Society of Photo-Optical Instrumentation Engineers (SPIE) ConferenceSeries , pages 277–286, February 2004. doi: 10.1117/12.503713.A. J. McKay, N. J. Chanover, J. P. Morgenthaler, A. L. Cochran, W. M.Harris, and N. D. Russo. Forbidden oxygen lines in Comets C/2006 W3Christensen and C/2007 Q3 Siding Spring at large heliocentric distance:Implications for the sublimation of volatile ices.
Icarus , 220:277–285, July2012. doi: 10.1016/j.icarus.2012.04.030.A. J. McKay, N. J. Chanover, J. P. Morgenthaler, A. L. Cochran, W. M.Harris, and N. D. Russo. Observations of the forbidden oxygen lines inDIXI target Comet 103P/Hartley.
Icarus , 222:684–690, February 2013. doi:10.1016/j.icarus.2012.06.020.A. J. McKay, N. J. Chanover, M. A. DiSanti, J. P. Morgenthaler, A. L.Cochran, W. M. Harris, and N. D. Russo. Rotational variation of daugh-ter species production rates in Comet 103P/Hartley: Implications for theprogeny of daughter species and the degree of chemical heterogeneity.
Icarus ,231:193–205, March 2014. doi: 10.1016/j.icarus.2013.11.029.A. J. McKay, A. L. Cochran, M. A. DiSanti, G. Villanueva, N. D. Russo, R. J.Vervack, J. P. Morgenthaler, W. M. Harris, and N. J. Chanover. Evolutionof H O, CO, and CO production in Comet C/2009 P1 Garradd duringthe 2011-2012 apparition. Icarus , 250:504–515, April 2015. doi: 10.1016/j.icarus.2014.12.023.K. Meech and J. Svoren. Physical and chemical evolution of cometary nuclei.In
Comets II , pages 317–335, Tucson, AZ, USA, 2004. University of ArizonaPress.M. Minissale, E. Congiu, G. Manic`o, V. Pirronello, and F. Dulieu. CO for-38ation on interstellar dust grains: a detailed study of the barrier of the CO+ O channel. Astronomy and Astrophysics , 559:A49, November 2013. doi:10.1051/0004-6361/201321453.J.P. Morgenthaler, W.M. Harris, and M.R. Combi. Large Aperture O I 6300˚A Observations of Comet Hyakutake: Implications for the Photochemistryof OH and O I Production in Comet Hale-Bopp.
Astrophysical Journal , 657:1162–1171, March 2007. doi: 10.1086/511062.M. J. Mumma and S. B. Charnley. The Chemical Composition ofComets - Emerging Taxonomies and Natal Heritage.
Annual Review ofAstonomy and Astrophysics , 49:471–524, September 2011. doi: 10.1146/annurev-astro-081309-130811.J. A. Noble, F. Dulieu, E. Congiu, and H. J. Fraser. CO Formation in Qui-escent Clouds: An Experimental Study of the CO + OH Pathway.
Astro-physical Journal , 735:121, July 2011. doi: 10.1088/0004-637X/735/2/121.T. Ootsubo, F. Usui, H. Kawakita, M. Ishiguro, R. Furusho, S. Hasegawa,M. Ueno, J.-i. Watanabe, T. Sekiguchi, T. Wada, Y. Ohyama, S. Oyabu,H. Matsuhara, T. Onaka, T. Nakagawa, and H. Murakami. Detection ofParent H O and CO Molecules in the 2.5-5 µ m Spectrum of Comet C/2007N3 (Lulin) Observed with AKARI. Astrophysical Journal Letters , 717:L66–L70, July 2010. doi: 10.1088/2041-8205/717/1/L66.T. Ootsubo, H. Kawakita, S. Hamada, H. Kobayashi, M. Yamaguchi, F. Usui,T. Nakagawa, M. Ueno, M. Ishiguro, T. Sekiguchi, J.-i. Watanabe, I. Sakon,T. Shimonishi, and T. Onaka. AKARI Near-infrared Spectroscopic Surveyfor CO in 18 Comets. Astrophysical Journal , 752:15, June 2012. doi:10.1088/0004-637X/752/1/15.J. Pittichov´a, C. E. Woodward, M. S. Kelley, and W. T. Reach. Ground-Based Optical and Spitzer Infrared Imaging Observations of Comet391P/GIACOBINI-ZINNER.
Astronomical Journal , 136:1127–1136, Septem-ber 2008. doi: 10.1088/0004-6256/136/3/1127.S. Raghuram and A. Bhardwaj. Photochemistry of atomic oxygen green andred-doublet emissions in comets at larger heliocentric distances.
Astronomyand Astrophysics , 566:A134, June 2014. doi: 10.1051/0004-6361/201321921.W. T. Reach, J. Vaubaillon, M. S. Kelley, C. M. Lisse, and M. V. Sykes.Distribution and properties of fragments and debris from the split Comet73P/Schwassmann-Wachmann 3 as revealed by Spitzer Space Telescope.
Icarus , 203:571–588, October 2009. doi: 10.1016/j.icarus.2009.05.027.W. T. Reach, M. S. Kelley, and J. Vaubaillon. Survey of cometary CO , CO,and particulate emissions using the Spitzer Space Telescope. Icarus , 226:777–797, September 2013. doi: 10.1016/j.icarus.2013.06.011.P. W. A. Roming, T. E. Kennedy, K. O. Mason, J. A. Nousek, L. Ahr, R. E.Bingham, P. S. Broos, M. J. Carter, B. K. Hancock, H. E. Huckle, S. D.Hunsberger, H. Kawakami, R. Killough, T. S. Koch, M. K. McLelland,K. Smith, P. J. Smith, J. C. Soto, P. T. Boyd, A. A. Breeveld, S. T. Hol-land, M. Ivanushkina, M. S. Pryzby, M. D. Still, and J. Stock. The SwiftUltra-Violet/Optical Telescope.
Space Science Reviews , 120:95–142, Octo-ber 2005. doi: 10.1007/s11214-005-5095-4.D. G. Schleicher and M. F. A’Hearn. The fluorescence of cometary OH.
As-trophysical Journal , 331:1058–1077, August 1988. doi: 10.1086/166622.BD Sharpee and TG Slanger. O (1d2-3p2, 1, 0) 630.0, 636.4, and 639.2 nmforbidden emission line intensity ratios measured in the terrestrial nightglow.
The Journal of Physical Chemistry A , 110(21):6707–6710, 2006.R. Stevenson, J. M. Bauer, R. M. Cutri, A. K. Mainzer, and F. J. Masci.NEOWISE Observations of Comet C/2013 A1 (Siding Spring) as It Ap-proaches Mars.
Astrophysical Journal Letters , 798:L31, January 2015. doi:400.1088/2041-8205/798/2/L31.W.-L. Tseng, D. Bockel´ee-Morvan, J. Crovisier, P. Colom, and W.-H. Ip.Cometary water expansion velocity from OH line shapes.
Astronomy andAstrophysics , 467:729–735, May 2007. doi: 10.1051/0004-6361:20066666.S. S. Vogt, S. L. Allen, B. C. Bigelow, L. Bresee, B. Brown, T. Cantrall,A. Conrad, M. Couture, C. Delaney, H. W. Epps, D. Hilyard, D. F. Hil-yard, E. Horn, N. Jern, D. Kanto, M. J. Keane, R. I. Kibrick, J. W. Lewis,J. Osborne, G. H. Pardeilhan, T. Pfister, T. Ricketts, L. B. Robinson, R. J.Stover, D. Tucker, J. Ward, and M. Z. Wei. HIRES: the high-resolutionechelle spectrometer on the Keck 10-m Telescope. In D. L. Crawford andE. R. Craine, editors,
Instrumentation in Astronomy VIII , volume 2198of
Society of Photo-Optical Instrumentation Engineers (SPIE) ConferenceSeries , page 362, June 1994.S.-i. Wang, R. H. Hildebrand, L. M. Hobbs, S. J. Heimsath, G. Kelderhouse,R. F. Loewenstein, S. Lucero, C. M. Rockosi, D. Sandford, J. L. Sundwall,J. A. Thorburn, and D. G. York. ARCES: an echelle spectrograph forthe Astrophysical Research Consortium (ARC) 3.5m telescope. In M. Iyeand A. F. M. Moorwood, editors,
Instrument Design and Performance forOptical/Infrared Ground-based Telescopes , volume 4841 of
Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series , pages 1145–1156, March 2003. doi: 10.1117/12.461447.H.A. Weaver, P.D. Feldman, J.B. McPhate, M.F. A’Hearn, C. Arpigny, andT.E. Smith. Detection of CO Cameron band emission in comet P/Hartley2 (1991 XV.
Astrophysical Journal , 422(1):374–380, 1994.M. W. Werner, T. L. Roellig, F. J. Low, G. H. Rieke, M. Rieke, W. F. Hoff-mann, E. Young, J. R. Houck, B. Brandl, G. G. Fazio, J. L. Hora, R. D.Gehrz, G. Helou, B. T. Soifer, J. Stauffer, J. Keene, P. Eisenhardt, D. Gal-41agher, T. N. Gautier, W. Irace, C. R. Lawrence, L. Simmons, J. E. VanCleve, M. Jura, E. L. Wright, and D. P. Cruikshank. The Spitzer Space Tele-scope Mission.
Astrophysical Journal Supplement Series , 154:1–9, Septem-ber 2004. doi: 10.1086/422992. 42 able 1
Observation Log-
Spitzer
Comet Date (UT) R (AU) ∆ a (AU) Exp. Times (s) Effective On-Source Exp. Time (s)LINEAR 1/31/2013 1.51 1.03 1.2 and 30 236LINEAR 2/15/2013 1.66 1.04 0.6, 12, and 100 936J¨ager 2/3/2014 2.18 1.81 0.6, 12, and 100 562PanSTARRS 5/25/2014 1.83 1.24 0.6 and 6 26.4 a Distance from
Spitzer able 2 Observation Log-APO/Keck
Comet Date (UT) r (AU) ∆ (AU) ˙∆ (km s − ) G2V Fast Rot. Flux CalLINEAR 2/7/2013 1.57 1.11 47.0 HD 25370 HD 27660 HR 1544LINEAR 2/15/2013 1.66 1.35 46.2 HD 25370 HD 27660 HR 1544J¨ager 11/6/2013 2.45 1.78 -23.7 HD 259216 HR 2532 Hilt 600J¨ager 11/15/2013 2.42 1.66 -21.7 HD 259516 HR 2532 Hilt 600J¨ager a a Obtained with Keck HIRES able 3 Observed Fluxes
Comet Date (UT) Species Flux a µ m Scale Factor b LINEAR 1/31/2013 CO ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
60 - a Fluxes are in 10 − ergs s − cm − . For CO Spitzer observations, fluxes are for a33-pixel aperture. For
Swift
OH observations, flux is given for a 50-pixel aperture.For [O I] and OH from ARCES and HIRES, fluxes are integrated over the entireslit. b Only applicable to
Spitzer observations of CO . able 4 Production Rates, CO /H O Ratios, and Collisional Quenching Factors Q (10 mol s − )Comet R (AU) CO H O CO /H O (%) Coll. Quench. FactorPanSTARRS 1.83 54.6 ± ± a ± b ±
80 - -LINEAR 1.51 1.12 ± c ± ± c d ± ± ± e ± a H O Production from ARCES observations of [O I] emission. b H O Production from
Swift observations of OH. c Due to uncertainties associated with the model-dependent dust subtraction, thesevalues may be better interpreted as upper limits. d Q H O from January Keck HIRES observations of OH, no collisional quenchingfactor is given due to no O I observation being obtained at this epoch. e Q H O from November ARCES observations of [O I]6300 emission. Q CO is notprovided as the Spitzer observation was in early February, therefore a comparisonto the H O production rate from November is not necessarily meaningful. able 5 Inferred vs. Measured CO /H O Ratio
Comet O I Ratio CO /H O (%)BR12 McKay2015A McKay2015B
Spitzer
PanSTARRS 0.054 ± ± ± ± ± < < < <
64 28.9 ± < < < <
116 31.3 ± able 6 O I Release Rates
Parent O I State a Release Rates (10 − s − )McKay2015A McKay2015B BR2012H O S 0.64 0.64 2.6H O D 84.4 84.4 84.4CO
S 50.0 33.0 72.0CO
D 75.0 49.5 120.0CO S 4.0 4.0 4.0CO D 5.1 5.1 5.1 a These rates are for a given electron state, not the line. Therefore if not all linescoming from that state are observed, the branching ratio needs to be accounted for.For D, both the 6300 ˚A and 6364 ˚A lines are usually observed, so no correction isneeded. However, for S, typically only the 5577 ˚A line is observed (as is the casein this work), so the above rates need to be multiplied by a branching ratio of 0.9to get the yield for S atoms that will decay through the 5577 ˚A line.
Spitzer
IRAC images at 3.6 (left column) and 4.5 (right column) µ m ofPanSTARRS (top row), LINEAR (middle row), and J¨ager (bottom row). Thesolar and velocity directions are indicated by the arrows labeled “Sun” and“v”, respectively, while celestial north is depicted by the arrow labeled “N”. Itis apparent in the 4.5 µ m image of PanSTARRS that there is diffuse emissionnot present in the 3.6 µ m image, which is likely due to CO . There appearsto be some diffuse emission in the 4.5 µ m image of J¨ager as well, though it isnot as obvious as for PanSTARRS. The gas emission is not obvious in the 4.5 µ m image of LINEAR.Fig 3: Dust-subtracted images of PanSTARRS (top), LINEAR (middle), andJ¨ager (bottom). The solar and velocity directions are indicated by the arrowslabeled “Sun” and “v”, respectively, while celestial north is depicted by thearrow labeled “N”.Fig 4: Spitzer images of PanSTARRS before (top row) and after (bottomrow) division by a 1/ ρ profile. Left to right is 3.6 µ m, 4.5 µ m, and the dust-subtracted image. The tail is obvious in the 3.6 µ m and 4.5 µ m images, and afaint residual is still evident in the dust-subtracted image. The dust-subtractedand 4.5 µ m images show a spiral structure that is not evident at 3.6 µ m,which is likely the manifestation of a CO jet. Each subpanel has dimensionsof 220,000 km on a side. 49ig 5: Spitzer images of LINEAR before (top row) and after (bottom row) divi-sion by a 1/ ρ profile. Left to right is 3.6 µ m, 4.5 µ m, and the dust-subtractedimage. The tail is obvious in the 3.6 µ m and 4.5 µ m images, and a faintresidual is still evident in the dust-subtracted image. In all images the comamorphology is symmetric, showing no obvious structures. Each subpanel hasdimensions of 140,000 km on a side.Fig 6: Spitzer images of J¨ager before (top row) and after (bottom row) divisionby a 1/ ρ profile. Left to right is 3.6 µ m, 4.5 µ m, and the dust-subtracted image.The tail is obvious in the 3.6 µ m and 4.5 µ m images, and a faint residual is stillevident in the dust-subtracted image. There is some possible extension of fluxtoward the bottom of the frame in the dust-subtracted image, but otherwiseno coma features are present. Each subpanel has dimensions of 240,000 km ona side.Fig 7: Spectrum of PanSTARRS depicting the [O I ]5577 ˚A line. The cometaryline is redshifted compared to the telluric line and is significantly weaker inintensity. 50 ig. 1. ig. 2. ig. 3. ig. 4.Fig. 5. ig. 6. ig. 7.ig. 7.