Meteor showers of comet C/1917 F1 Mellish
aa r X i v : . [ a s t r o - ph . E P ] O c t Mon. Not. R. Astron. Soc. , 1– ?? (2010) Printed 21 November 2018 (MN LaTEX style file v2.2) Meteor showers of comet C/1917 F1 Mellish
P. Vereˇs, L. Kornoˇs and J. T´oth
Faculty of Mathematics, Physics, and Informatics, Comenius University, 842 48 Bratislava, Slovakia Mlynsk´a dolina
Released 2010 Xxxxx XX
ABSTRACT
December Monocerotids and November Orionids are weak but established annualmeteor showers active throughout November and December. Analysis of a high qualityorbits subset of the SonotaCo video meteor database shows that the distribution oforbital elements, geocentric velocity and also the orbital evolution of the meteorsand potential parent body may imply a common origin for these meteors comingfrom the parent comet C/1917 F1 Mellish. This is also confirmed by the physicalproperties and activity of these shower meteors. An assumed release of meteoroidsat the perihelion of the comet in the past and the sky-plane radiant distributionreveal that the December Monocerotid stream might be younger than the NovemberOrionids. A meteoroid transversal component of ejection velocity at the perihelionmust be larger than 100 m/s. A few authors have also associated December CanisMinorids with the comet C/1917 F1 Mellish. However, we did not find any connection.
Key words: comets, individual: C/1917 F1 Mellish – meteors, meteoroids – celestialmechanics – catalogues
The comet C/1917 F1 Mellish, formerly designated as1917 a ( Mellish ) or
Mellish I , was discovered by J.E.Mellish on March 20, 1917 and was observed for 96 days(Askl¨of 1923, 1932) from many places on the Earth. In thesouthern hemisphere, the comet reached up to +1 magni-tude. Astronomer J.F. Skjellerup noted that the brightnessof the cometary head was about +3 magnitude, with thediffuse coma and narrow tail about 10 ◦ long on April 19,1917 (Orchiston & Skjellerup 2003). The comet is a Halley-type comet, with a relatively low inclination, and has oneof the smallest perihelion distances. It was observed onlyat one apparition. Despite the relatively long observationalarc, the precision of the orbital elements is questionable.Askl¨of (1932) published a slightly modified orbit of thecomet and noted that the orbit is given with a period of145 ± . e ∼ − . .
19 AU). Several authors (Porter 1952; Hasegawa 1962) de-termined that the comet–Earth distance is close enough to observe a meteor shower and predicted the radiant positionsand activity of the shower on Dec. 15 (Dec. 20 respectively)and the geocentric velocity of meteors ∼
40 km/s. The firstfew meteors associated with the comet C/1917 F1 Mellishwere obtained by the Harvard Super Schmidt photographicsurvey (Whipple 1954; McCrosky & Posen 1961). Severalcandidates of this meteor stream, later designated as theDecember Monocerotids (MON), were also detected and dis-tinguished by radar surveys (Nilsson 1964; Sekanina 1973).Another study connecting the December Monocerotids withthe comet C/1917 F1 Mellish was made by Drummond(1981); Olsson-Steel (1987). Surprisingly, the radar datapublished by Nilsson (1964) and Sekanina (1973) revealedthat the meteors having similar radiant positions, activityand geocentric velocities appear to have 10 ◦ lower inclina-tions. Kres´akov´a (1974) noted that December Monocerotidsseem to have 2 components. The author also speculates thatthe stream may have a common origin with the Geminidmeteor stream. Moreover, Harvard radio data revealed apossible meteor stream with low inclined orbits but almostthe same orbital elements as the December Monocerotids,active between November 27 - December 7 (Nilsson 1964;Sekanina 1973). The possible genetic connection betweenthe comet and the December Monocerotids was studied byFox & Williams (1985).Various photographic searches confirmed the existenceof a weak stream at RA = 90 . ◦ , DEC = 15 . ◦ onNovember 27, with v g = 43 . c (cid:13) P. Vereˇs, L. Kornoˇs and J. T´oth stream was named as ξ -Orionids (Xi-Orionids, ω − Orionids),currently recognized as established meteor shower Novem-ber Orionids (NOO) within the IAU Meteor Data Cen-ter (IAU MDC) catalogue (Ohtsuka 1989) and later byLindblad (1999). Moreover, other photographic DecemberMonocerotids were published (Ohtsuka 1989) and complexanalysis of December Monocerotids and ξ -Orionids doneby Lindblad & Olsson-Steel (1990). Even some historicalrecords of fireballs might confirm that December Mono-cerotids were active in past centuries (Fox & Williams 1985;Hasegawa 1999).In 1969, Hindley published his telescopic meteor obser-vation from Dec. 11, 1964 and assigned 5 meteors to the newstream called 11 Canis Minorids. Hindley (1969) computedthat these meteors have parabolic orbits and much higherinclinations (over 100 ◦ ). A year later, the author suggesteda connection between the shower and comet C/1917 F1 Mel-lish (Hindley 1970) and determined the activity during De-cember 9-14. Kres´akov´a (1974) revealed that 9 meteors thatcreate the second component of the December Monocerotidsmight be 11 Canis Minorids activity within December 4-15.Their inclination was determined as i = 29 . ◦ and the peri-helion distance as q = 0 . v g ∼
40 km/s.The maximum activity of 11 Canis Minorids is expected at L ⊙ = 252 . ◦ (December 3).Now, the December Monocetorids and November Ori-onids are weak (few meteors per hour at maximum) butannual established meteor showers. The shower 11 Canis Mi-norids, (December Canis Minorid according to IAU nomen-clature) is classified as a ”working” shower. Despite severalinvestigations, past publications analyzed only a small num-ber of orbits and provided disperse data on the mean orbit,the position of radiant, the activity and precision of orbits,and did not reveal the orbital evolution of the meteoroid par-ticles released from the parent comet, in order to explain thecurrent state of these meteor showers. A significant numberof the analyzed orbits are hyperbolic or parabolic.This work uses recent and precise video multi-stationorbits, obtained by the SonotaCo video network in Japan,which provide the highest number of relatively precise me-teor orbits detected continually between the years 2007 − The freely accessible database of the meteors detected bythe SonotaCo network contains 64540 multi-station me-
Figure 1.
Orbits of December Monocerotids and November Ori-onids selected from the SonotaCo database of video orbits.
Table 1.
Orbital elements of the comet C/1917 F1 Mellish inequinox J2000 reference frame (JPL).element valuea 27.6473325 AUe 0.993121q 0.190186 AUi 32.6828 ◦ ω ◦ Ω 88.6683 ◦ M 0.0259325 ◦ epoch (JD) 2421334.0heliocentric distance of Ascending Node 0.783556 AUheliocentric distance of Descending Node 0.250005 AU teor orbits with additional parameters such as the be-ginning and terminal height, absolute magnitude, equato-real and ecliptical coordinates of the radiant, the streamassignment, etc. Although the SonotaCo method usesits own shower assignment algorithm, we have demon-strated (Vereˇs & T´oth 2010) that its results are consistentwith the widely-used Southworth-Hawkins D-criterion D SH (Southworth & Hawkins 1963) for meteor stream identifica-tion. In accordance with several restriction criteria devel-oped in our previous work (Vereˇs & T´oth 2010), we selecteda high quality subset of orbits for further analysis. The cri-teria fulfilled 111 of 250 December Monocerotids and 110 of333 November Orionids detected in the years 2007-2009.Of 111 MON meteors, only 8 have hyperbolic orbits andamong 110 NOO meteors, 14 have semimajor axes largerthan 100 AU or eccentricity higher than 1 .
0. We also em-ployed D SH to distinguish possible rogue sporadic meteorsamong MON and NOO assigned to the showers by Sono-taCo. In comparison with the nominal orbit of the cometC/1917 F1 Mellish, all MON and NOO meteors fall within c (cid:13) , 1– ?? eteor showers of comet C/1917 F1 Mellish C oun t a (AU) December Monocerotids November OrionidsC/1917 F1 Mellish 0.87 0.90 0.93 0.96 0.99 1.02048121620 December Monocerotids November Orionids C oun t e C/1917 F1 Mellish10 20 30 40 50048121620 December Monocerotids November Orionids C oun t i (deg) C/1917 F1 Mellish 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35048121620 C/1917 F1 Mellish
December Monocerotids November Orionids C oun t q (AU)
36 38 40 42 44 46048121620
December Monocerotids November Orionids C oun t Geocentric velocity (km/s)
204 208 212 216048121620 C/1917 F1 Mellish
December Monocerotids November Orionids C oun t Longitude of perihelion (deg)
Figure 2.
Distributions of the of orbital elements and geocentric velocity (using b–spline) of the December Monocerotids and NovemberOrionids. The arrow shows the orbital evolution of the comet C/1917 F1 Mellish over the last 5000 yr, the dot shows its nominal orbitalparameters. D SH < .
4. Even within stricter D SH < .
15, 105 MONand 97 NOO are found (according to SonotaCo shower as-signment). Independently, we selected the MON and NOOshower members by using the iteration procedure accord-ing to Porubˇcan & Gavajdov´a (1994). In the iteration forthe D SH =0.15, 105 MON and 97 NOO were identified. Thisresult is almost identical to SonotaCo assignment of bothshower members. The showers appear very narrow in theorbital element space. The mean orbits are presented in Ta-ble 2, in comparison with the mean orbits by other authors.Each orbital element was calculated as the median, withgiven standard deviation. The photometric mass was com- puted for each meteor according to Betlem et al. (1999) andits mean value for the showers is mentioned in Table 2. Theorbits of the meteor showers are given in Figure 1. Figure2 depicts selected orbital elements of both meteor showers.Figure 3 shows the q − i phase space of individual meteorsand nominal orbit of the comet C/1917 F1 Mellish.The semimajor axes of both shower meteors seem to bemuch smaller than the nominal semimajor axis of the cometor even the range of semimajor axes derived from the 5000 yrorbital evolution of cometary clones (the orbital evolutionis described in chapter 3). The eccentricity, the periheliondistance and the inclination of MON are very close to the c (cid:13) , 1– ?? P. Vereˇs, L. Kornoˇs and J. T´oth i ( deg ) q Figure 3.
December Monocerotids, November Orionids and thecomet C/1917 F1 Mellish in the phase space of perihelion distanceand inclination. cometary orbit: the NOO meteors exhibit slightly lower butstill high eccentricities, notably lower inclinations and lowerperihelion distances. Generally, the orbits of MON and NOOare very similar. In e and q the values are the same withinthe standard deviation: a notable difference is seen only inthe inclination. The gap between the MON and NOO is alsovisible in Figure 1. The q − i phase space and the dispersionof orbits from the mean orbit of each shower imply that theNOO exhibit wider dispersion. The cometary orbit is locatedwithin MON elements and, therefore, the NOO meteoroidsseem to be older than MON. The inclination exhibits in-teresting behavior. The nominal orbit of the parent cometlies within the MON orbits, while the NOO orbits are lessinclined and are apparently well separated from the MONorbits. There seems to be no overlapping in the inclinationbetween the two clumps representing the MON and NOO inFigure 2.As seen in Table 2 and Figures 1-3, the orbital distri-bution of the December Monocerotids and November Ori-onids have common features. There is good agreement inthe geocentric velocity and minimal difference in the angu-lar elements and eccentricity. The perihelion distance of theNovember Orionids is generally lower. It is worth pointingout that the perihelion distance of the MON and NOO show-ers is one of the lowest observed. Only a few meteor show-ers, among them the prominent Geminids, have such closeperihelion distances. The dispersion of the MON and NOOorbital elements is relatively low, which gives the chance forgood definition and differentiation of these showers.The cumulative absolute magnitude distribution in Fig-ure 4 shows that the population index of both showers is al-most identical. However, NOO contains smaller particles, ac-cording to Figure 4. This finding is consistent with the radardata, where NOO are more significant than MON (Nilsson1964, Sekanina 1973).We determined the maximum activity of the MON forthe longitude of the Sun L ⊙ = 259 . ◦ (December 11) withthe duration of the shower from November 26 to December21. The maximum of MON occurs one day earlier in the -8 -6 -4 -2 0 2 4020406080100 December Monocerotids November Orionids R e l a t i v e c oun t ( % ) Cumulative distribution (absolute magnitude)
Figure 4.
Cumulative distribution of absolute magnitudes of De-cember Monocerotids and November Orionids.
SonotaCo data than in the IAU MDC catalogue. The radiantposition during the maximum activity was determined as RA = 98 . ◦ , DEC = 8 . ◦ and the daily motion is givenby the following equations in the equatoreal and eclipticalcoordinates: RA = (101 . ± . ◦ + (0 . ± .
01) ( L ⊙ − . ◦ ) DC = ( 8 . ± . ◦ − (0 . ± .
02) ( L ⊙ − . ◦ ) (1) λ = (101 . ± .
1) + (0 . ± .
01) ( L ⊙ − . ◦ ) β = ( − . ± . − (0 . ± .
01) ( L ⊙ − . ◦ ) , (2)where 259.5 ◦ represents the solar longitude of the DecemberMonocerotid’s maximum, derived from the SonotaCo data(eq. 2000.0).According to SonotaCo data, the maximum activity ofthe NOO occurs on L ⊙ = 249 . ◦ (December 1) and it isactive from November 16 until December 16. The maximumoccurrs 4 days after the maximum predicted by the IAUMDC catalogue ( L ⊙ = 245 . ◦ ). The motion of the radiantis given by equations (3) and (4): RA = (92 . ± . ◦ + (0 . ± .
03) ( L ⊙ − . ◦ ) DC = (15 . ± . ◦ − (0 . ± .
02) ( L ⊙ − . ◦ ) (3) λ = (92 . ± . ◦ + (0 . ± .
02) ( L ⊙ − . ◦ ) β = ( − . ± . ◦ − (0 . ± .
02) ( L ⊙ − . ◦ ) . (4)The sky-plane positions of the radiants in the equatoreal andecliptical grid are depicted in Figure 5. The radiant positionscalculated using the nominal orbit of the comet C/1917 F1Mellish and computed by several methods by DOSMETHsoftware (Nesluˇsan, Svoreˇn & Porubˇcan 1998) are shown aswell.A possible common origin of both meteor showers canalso be inferred from Figure 6 depicting the heliocentric dis-tance of the ascending and descending nodes of the MONand NOO as a function of solar longitude. As expected, theascending node lies very close to the value of 1 AU but thedescending node gradually rises with the solar longitude andboth meteor showers overlap without any gap or visible dif-ference in the descending node.Other common properties might be derived from the be-ginning and terminal heights of individual meteors. Figure 7 c (cid:13) , 1– ?? eteor showers of comet C/1917 F1 Mellish Table 2.
The mean orbital elements, geocentric velocity, photometric mass of meteoroids and activity intervals of high quality orbits ofDecember Monocerotids and November Orionids from the SonotaCo database, compared with other authors. No – number of meteors.Authors: OH89 – Ohtsuka (1989), LIND90 – Lindblad & Olsson-Steel (1990), LIND99 – Lindblad (1999), JEN06 – Jenniskens (2006).elements a e q i ω Ω v g [km/s] mass [g] activity No author MON
NOO
80 90 100 1105101520 80 90 100 110
C/1917 F1 Mellish radiants C/1917 F1 Mellish radiants C/1917 F1 Mellish radiants December Monocerotids November Orionids December Monocerotids November Orionids December Monocerotids November Orionids December Monocerotids November Orionids C/1917 F1 Mellish radiants RA DC
80 90 100 110 DC
80 90 100 110-20-15-10-50 E c li p t i ca l l a t i t ud e Ecliptical longitude
Ecliptical longitude
Figure 5.
Equatoreal and ecliptical radiants (Eq. 2000.0) of De-cember Monocerotids and November Orionids. Upper and lowerright graphs show the radiants corrected for daily motion andreduced to the solar longitude of the activity maximum. Theoret-ical geocentric radiants derived by DOSMETH from the nominalorbit of C/1917 F1 Mellish are displayed as well (open circles). shows clearly that heights of the MON and NOO are practi-cally identical. The geocentric velocity and entry geometryis almost the same for both meteor showers; therefore, thebeginning and terminal height would mostly depend on thephysical properties of meteoroids, such as the mass, the bulkdensity and internal structure. The heights are given in Ta-ble 3 and the heights as a function of photometric mass inFigure 7. In Table, 3 we also compare the heights of MONand NOO with the high quality Geminids orbits from theSonotaCo database. Geminids from the SonotaCo databasehave beginning and terminal heights 5 km lower than othervideo observations made by similar techniques (Koten et al.2004). MON and NOO meteors have beginning heights 6–7 km higher than Geminids, which could indicate that me-teors from C/1917 F1 Mellish have lower bulk densities andare more fragile. On the other, hand Geminids with similar
Table 3.
The beginning and terminal heights of the DecemberMonocerotids and November Orionids in comparison with theGeminids using the high quality SonotaCo data.shower beginning height [km] terminal height [km]MON 102 . ± . . ± . . ± . . ± . . ± . . ± .
230 240 250 260 270 2800.00.20.40.60.81.01.2
December MonocerotidsNovember Orionids C/1917 F1 Mellish H e li o c en t r i c d i s t an c e ( A U ) Longitude of the Sun (deg)
MONNOO Descending nodeAscending node
Figure 6.
The heliocentric distances of the ascending and de-scending nodes of the December Monocerotids and November Ori-onids as a function of the longitude of the Sun, compared withthe ascending and descending nodes of the comet C/1917 F1 Mel-lish. Range of the theoretical radiants computed by DOSMETHfor the parent comet is displayed with a heavy line. Vertical linesrepresent the maxima of the meteor showers. geocentric velocities belong to the densest and most rigidmeteors observed (Rendtel 2004).The database does not contain any December Ca-nis Minorids, yet we tried to find some representativesamong the high quality data set. The published mean or-bit of December Canis Minorids has a very low semi-major axis, ω similar to a previously published value,and i and Ω similar to MON. According to SonotaCo c (cid:13) , 1– ?? P. Vereˇs, L. Kornoˇs and J. T´oth
Table 4.
The mean orbital elements of 6 December Canis Minorids candidates. With respect to the parent comet, the subset hasrelatively low D-criterion, D SH ∼ . ± .
08a [AU] e q [AU] i ω Ω v g L ⊙ RA DC1 . ± . . ± .
02 0 . ± .
02 30 . ± . ± . . ± . . ± . ± ± H e i gh t ( k m ) Photometric mass (g)
Figure 7.
The beginning and terminal heights of the DecemberMonocerotids and November Orionids as a function of the pho-tometric mass. (http://sonotaco.jp/doc/J5/index.html), the shower mightbe active from November 30 until December 9, with the max-imum on December 4. Nevertheless, the database does notcontain any meteors of this stream. Because of the little in-formation there is about the stream, we tried to select candi-dates from the high quality subset of orbits. The subset wasselected using the iteration method (Porubˇcan & Gavajdov´a1994) with respect to the IAU MDC catalogue shower pa-rameters. Only 6 meteors fulfilled our criteria (Table 4). Me-teors were detected during the activity of both MON andNOO. The D-criterion of meteors was on average greaterthan 0.3 with respect to the assumed parent comet. It iseven doubtful if the December Canis Minorids is a regularshower or if it is only an occasional shower observed whenthe Earth crosses a narrow filament of the meteoroid par-ticles, or these meteors are just scattered meteors of theMON-NOO complex; or that even these meteors belong tothe sporadic background.
The first orbital evolution analysis of the comet C/1917F1 Mellish 800 yr to the past (Carusi et al. 1984) revealedthat its orbit evolves slowly: notably, the nodes evolve veryslowly. The inclined orbit avoids giant planet encounters andthere is only a little chance of substantial gravitational in-teraction with the terrestrial planets near the perihelion.Fox & Williams (1985) and Hasegawa (1999) studied the op-tion that the ancient fireballs observed between December6 and 18 apparently emanated from the same radiant. Theyworked out that these bolides could not be connected to the -3000 -2000 -1000 0 1000 20000.20.40.60.81.0 H e li o c en t r i c d i s t an c e ( A U ) Year
Ascending nodeDescending node -3000 -2000 -1000 0 1000 20000.20.40.60.81.0
Descending nodeAscending node H e li o c en t r i c d i s t an c e ( A U ) Year
Figure 8.
The orbital evolution of the heliocentric distance ofthe ascending and descending nodes of the nominal orbit of thecomet C/1917 F1 Mellish (solid line) and its clones (points). Up- clones generated within the 0.8 yr orbital period error, down -clones altered in order to put ascending node close to 1 AU at thepresent day.
Geminid meteor stream because of its rapid orbital evolu-tion but might belong to the MON. Fox & Williams (1985)confirmed the slow evolution of the ascending node in the2400 yr integration of the cometary orbit to the past. Theirwork also confirmed that the heliocentric distance of theMON ascending node is stable as well and is close to 1 AUone thousand years to the past or to the future.As mentioned previously, the orbit of the parent cometwas determined with a low precision. In our study, we setout to calculate the orbital evolution of nominal and clonedorbits of the parent comet. We created 100 clones within the0.8 yr uncertainty of the orbital period (Askl¨of 1932), with c (cid:13) , 1– ?? eteor showers of comet C/1917 F1 Mellish -3000 -2000 -1000 0 1000 20002025303540 a ( A U ) Year -3000 -2000 -1000 0 1000 20000.100.150.200.250.30 q ( A U ) Year -3000 -2000 -1000 0 1000 2000262830323436 i ( deg ) Year -3000 -2000 -1000 0 1000 2000210212214216218220
Long i t ude o f pe r i he li on ( deg ) Year
Figure 9.
The orbital evolution of clones (grey dots) and the nominal orbit of the comet C/1917 F1 Mellish (solid line). fixed perihelion distance and altered semimajor axis and ec-centricity accordingly. We could not modify other orbitalelements because their uncertainties are unknown. Anotherset of 100 clones was made in order to create orbits with theheliocentric distance of ascending node close to 1 AU whileneither nominal orbit nor first 100 clones have the ascend-ing node close enough to the orbit of the Earth to createan observed meteor shower. In this case, the eccentricity,the semimajor axis and the perihelion distance were altered.The beginning of the integration was set at the epoch of theperihelion passage of C/1917 F1 Mellish (JD 2421334.0, Eq.2000.0). The multistep Adams-Bashforth-Moulton type upto 12th order numerical integrator, with variable step-width,was used. All planets were considered as perturbing bodiesand the Earth and Moon were treated separately.The integration shows that both sets of clones behavein a similar way. Figure 8 depicts the heliocentric distance ofthe ascending and descending node of the nominal orbit andthe clones integrated 5000 years to the past. The ascendingnode of the nominal orbit gradually falls and retreats fromthe orbit of the Earth to ∼ . − . ∼ .
25 AU)and rises slowly to ( ∼ . ± . σ ) of the semimajor axes of cloned orbits is in the range of a ⊂ (25; 30) AU, while the nominal semimajor axis rises upto 40 AU 2500 BC. The inclination of both clones and thenominal orbit falls from initial 32 ◦ down to i ⊂ (26; 30) ◦ .Even at 3000 yr BC, the inclination is not low enough toexplain the low inclination of the NOO ( i ∼ . q ⊂ (0 .
14; 0 .
23) AU) and the summation of angu-lar elements (longitude of perihelion – π ) rises graduallyfrom 210 ◦ to π ⊂ (212; 218) ◦ after 5000 yr integration to thepast. Clones derived from the uncertain orbital period didnot encounter the Earth but had close flybies within the Hillsphere of Venus (2%) and Mercury (0 . a <
100 AU),the integration uses a beta parameter representing the so-lar radiation pressure (Klaˇcka 2004) for each particle ( β =2 · − ) derived from an assumed low bulk density ( ̺ =750 kg/m ) and the typical photometric mass of observedmeteors (0.5 g) and starts for the epoch and the orbitalelements valid for the moment of the meteor observation.The numerical integration computed the orbital evolutionfor 5000 yr to the past (Figure 10 and Figure 11).Unlike Fox & Williams (1985), perturbed orbits of theMON and NOO reduced their heliocentric distance of theascending nodes as time goes to the past. It seems that it c (cid:13) , 1– ?? P. Vereˇs, L. Kornoˇs and J. T´oth -3000 -2000 -1000 0 1000 2000051015202530 December Monocerotids C/1917 F1 Mellish a ( A U ) Year -3000 -2000 -1000 0 1000 2000051015202530 November Orionids C/1917 F1 Mellish a ( A U ) Year -3000 -2000 -1000 0 1000 20000.80.91.01.1 December Monocerotids C/1917 F1 Mellish e Year -3000 -2000 -1000 0 1000 20000.80.91.01.1 November Orionids C/1917 F1 Mellish e Year -3000 -2000 -1000 0 1000 20000.00.20.4
December Monocerotids C/1917 F1 Mellish q ( A U ) Years -3000 -2000 -1000 0 1000 20000.00.20.4
November Orionids C/1917 F1 Mellish q ( A U ) Years -3000 -2000 -1000 0 1000 20001020304050 December Monocerotids C/1917 F1 Mellish i ( deg ) Year -3000 -2000 -1000 0 1000 20001020304050 November Orionids C/1917 F1 Mellish i ( deg ) Year
Figure 10.
The orbital evolution in a , e , q and i of the December Monocerotids, the November Orionids and the nominal orbit of thecomet C/1917 F1 Mellish, with the possible variation shown with grey dashed lines, derived from the comet clones orbital evolution.c (cid:13) , 1– ?? eteor showers of comet C/1917 F1 Mellish -3000 -2000 -1000 0 1000 2000200220240260280 December Monocerotids C/1917 F1 Mellish
Long i t ude o f pe r i he li on ( deg ) Year -3000 -2000 -1000 0 1000 2000200220240260280
November Orionids C/1917 F1 Mellish
Long i t ude o f pe r i he li on ( deg ) Year -3000 -2000 -1000 0 1000 20000.40.60.81.01.2 H e li o c en t r i c d i s t an c e o f a sc . node ( A U ) Year
December Monocerotids C/1917 F1 Mellish -3000 -2000 -1000 0 1000 20000.40.60.81.01.2 H e li o c en t r i c d i s t an c e o f a sc . node ( A U ) Year
November Orionids C/1917 F1 Mellish-3000 -2000 -1000 0 1000 20000.00.10.20.30.40.5
December Monocerotids C/1917 F1 Mellish H e li o c en t r i c d i s t an c e o f de sc . node ( A U ) Year -3000 -2000 -1000 0 1000 20000.00.10.20.30.40.5
November Orionids C/1917 F1 Mellish H e li o c en t r i c d i s t an c e o f de sc . node ( A U ) Year
Figure 11.
The orbital evolution in π , the heliocentric distance of the ascending and descending nodes of the December Monocerotids,the November Orionids and the nominal orbit of the comet C/1917 F1 Mellish, with the possible variation shown with grey dashed lines,derived from the clones orbital evolution. takes only two or three hundreds years until the ascendingnode reaches the nominal orbit of the parent comet. On thecontrary, the semimajor axes of both the MON and NOOare generally constant over 5000 yr, which might be due tolow perturbations of giant planets and higher inclinations.Even with the beta parameter, the semimajor axes remainfar away from the nominal or even cloned orbits of the parentcomet. Meteoroids could be injected to these orbits directlyafter the ejection from the cometary core. The inclination ofthe MON is quite consistent with the current orbital inclina-tion of the comet C/1917 F1 Mellish. The orbital evolutionin inclination implies that MON were released recently, gen-erally hundreds, or at most, 3000 years in the past. On the contrary, the NOO meteoroids have lower inclinations in thepresent day. The orbital evolution reveals that NOO mete-oroids could have departed from the comet mostly 4000 yearsprior and almost all low inclined NOO could be explained byan orbital evolution within the last 5000 years – most of theNOO inclinations intersected the comet evolution path ofits inclination. The longitudes of the perihelia of the MONattain the same values as the nominal cometary orbit in therecent centuries and then disperse. On the other hand, thelongitudes of the perihelia of the NOO remain much longeralong the evolved nominal orbit of the comet and disperseslowly after thousands of years, which could support theyounger age of MON as well. A similar feature is visible for c (cid:13) , 1– ?? P. Vereˇs, L. Kornoˇs and J. T´oth the perihelion distance of both showers. Currently the per-ihelion distance of MON fits well with the current orbit ofthe comet but gets more dispersed around 500 B.C. The per-ihelion distance of NOO is slightly different during the last2000 years but generally intercepts the perihelion distanceof the comet earlier than 1000 B.C. The eccentricity of MONis dispersed much more in the past than in the case of NOO.The heliocentric distance of the ascending node of MON liesclose to the current ascending node of the comet but dis-perses fast in the past. This distance is currently lower forNOO but could be explained by the orbital evolution (Fig-ure 11). These implications of the orbital evolution suggestthat the NOO shower is older than the MON shower andboth streams may originate from the same parent comet.Resulting from the nodal distances of the comet andthe relatively fast evolution of the ascending nodes of theshowers (centuries), there is a possibility that we observethe outer edge of a widely evolved complex stream. Thestreams (MON and NOO) might be observed as two streamsas a result of a geometric selection effect. Meteoroids withinclinations between the MON and NOO might have nodeson non-Earth crossing orbits.
The relatively large heliocentric distance of the ascendingnode of the parent comet and the much lower and stablesemimajor axes of the meteors indicate that these particleswere injected into these orbits immediately after ejectionfrom the comet, while the comet might not have been onthe same orbit at that time. The derivation of the ejec-tion velocity depends on the model used. If we assumea spherical cometary nucleus with an albedo of 0 .
04, ac-tive surface 0 .
15% (Ma, Williams & Chen 2002, bulk den-sity 750 kg/m , a radius of the nucleus 3.1 km (Jenniskens2006), ejection at the perihelion q ∼ .
19 AU; and if themeteoroids are escaping only from the Sun-facing hemi-sphere with the Gaussian distribution of velocities with thecenter on the subsolar point, the maximum ejection veloc-ity might range from 5 m/s (Imponente & Sigismondi 2001)to 112 m/s (Crifo & Rodionov 1997; Ma, Williams & Chen2002). The escape velocity of the meteoroid particle changesits orbital elements so that we may calculate according toPecina & ˇSimek (1997).On the other hand, if we know the orbital elements ofthe comet (before the ejection of the meteoroid) and the me-teoroid after the ejection (assuming that the observed mete-oroid escaped the comet recently and did not undergo rapidorbital changes due to gravitational and non-gravitationalperturbations), we may directly derive the ejection veloci-ties from equations by Pecina & ˇSimek (1997) as a low esti-mate. The ejection velocity affects, at most, the semimajoraxis. In the perihelion, the semimajor axis change is ruledby the transversal component of the velocity vector ∆ v t .The range of ejection velocities derived according to mean,the peak and minimum-maximum values of the semimajoraxis (Figure 2) of each meteor shower, in comparison withthe range of semimajor axes of the cometary clones, inte-grated 2000 yr to the past is presented in Table 5. The meanand peak values of the transversal ejection velocity com- Table 5.
The transversal component of the meteoroid ejectionvelocity derived from the the difference of the semimajor axis ofmeteor showers and the parent comet. The range of cometarysemimajor axis lies within a ⊂ (26; 29) AU.shower method a [AU] − ∆ v t [m/s]MON mean 8.8 104 ± ± − ± ± − ponents are in good agreement with the ejection model byMa, Williams & Chen (2002). While the eccentricity, the in-clination and angular elements of the MON lie within thecometary clones’ range, the derivation of the radial and nor-mal components of the velocity vector are ambiguous, fol-lowing equations by Pecina & ˇSimek (1997). On the otherhand, a difference in the inclination, the eccentricity andangular elements of the NOO could not be explained by thedirect ejection of the particles into current orbits, but onlyby the following orbital evolution. We demonstrated that the December Monocerotids andNovember Orionids obtained from the SonotaCo database of3 year observations (2007-2009) have most likely a commonorigin and come from the comet C/1917 F1 Mellish. Thecommon origin is supported by their similar orbital charac-teristics, the activity, physical properties assumed from thebeginning, and the terminal heights of the meteors, the de-scending nodes of both showers as a function of the solarlongitude, and the narrow Southworth-Hawkins D-criterionfor most December Monocerotids and November Orionidswith respect to the parent comet ( D SH < . ◦ lower inclinations than December Monocerotids. Thereis a non-zero chance of a close encounter of the comet withVenus or Mercury which could cause a sudden change inthe inclination of the parent comet. Furthermore, a closeencounter with a planet might cause the tidal breakup ofthe comet and create a significant release of matter. Thescenario of the cometary core disintegration might also besupported by the extremely low perihelion distance of theparent comet and both meteor showers. c (cid:13) , 1– ?? eteor showers of comet C/1917 F1 Mellish Another option is a gradual shift in the inclination,demonstrated in the simulation. But a change in the inclina-tion of more than 10 ◦ of the parent comet would be solvedonly through a longer orbital evolution. This option is alsoobscured while we do not observe any orbits between rel-atively well defined clumps of the December Monocerotidsand November Orionids in the q − i phase space and radiantsky-plane distributions. Eventually there is a wide and mas-sive stream of the meteoroids but only some of them haveascending nodes close to the Earth’s orbit; and due to se-lectional effects, we may observe two distinguished streamsand only the distant edge of the stream. The nodal distanceof the comet is currently more than 0.2 AU from the Earth’sorbit and it retreated in the past increasingly, as well asthe ascending nodes of the observed meteors. The observedshower meteors might have left the cometary nucleus a fewcenturies ago but, due to the stable orbit of the comet, bothstreams might be replenished regularly and weak shower ac-tivity might be observed each year. The semimajor axes ofboth meteor streams are much lower than the nominal orbitof the comet or its clones’ evolution 5000 yr to the past. Al-most no change of the semimajor axes of meteoroids withinthe orbital evolution suggests that these particles were in-jected directly into these orbits right after ejection fromthe cometary nucleus. We determined that transversal com-ponent of the ejection velocity would be about 100 m/s ifthe ejection occurred at the perihelion. Further precise or-bits and physical data of the December Monocerotids andNovember Orionids are needed for additional research.The key question is the accuracy of the C/1917 F1 Mel-lish orbit. A new measurements of the photographic platesof the comet might reveal a more precise orbit and bring newlight onto the orbital evolution of the comet and its meteors.Our work did not confirm any December Canis Minorids me-teors in the SonotaCo database; however, 6 candidates wereselected. Their connection to the comet C/1917 F1 Mellishis uncertain. ACKNOWLEDGMENTS
This work was supported by Slovak Grant Agency VEGA,No. 1/0636/09 and by a grant of Comenius Uviversity, No.UK/245/2010.
REFERENCES
Askl¨of S., 1923, Arkiv f¨or Mat. Astron. och Fys., 18, No. 7Askl¨of S., 1932, Arkiv f¨or Mat. Astron. och Fys., 23A, No. 11Betlem H., Jenniskens P., van´t Leven J., ter Kuile C., JohanninkC., Zhao H., Lei Ch., Li G., Zhu J., Evans S., Spurn´y P., 1999,Meteorit. Planet. Sci., 34, 979Carusi A., Kres´ak L., Perozzi E. & Valsecchi G.B., 1984, IASInternal Report no. 12, RomeChamberlin A.B., Yeomans D.K., Chodas P.W., Giorgini J.D.,Jacobson R.A., Keesey M.S., Lieske J.H., Ostro S.J., StandishE.M. & Wimberly R.N., 1997, Bull. Amer. Astron. Soc. 29,1014.Crifo J.F. & Rodionov V., 1997, Icarus, 127, 319Drummond, J., 1981, Icarus, 45, 545Fox K. & Williams I.P., 1985, MNRAS 217, 407Hasegawa I., 1962, General Index List of Theoretical Radiant Points of Meteors Associated with Comets, 3rd Ed., Doc. desObservateurs, Paris 14Hasegawa I., 1999, in W.G. Baggaley & V. Porubˇcan, eds., Me-teoroids 1998, Proc. Int. Conf. held at Tatransk´a Lomnica,Slovakia, Astron. Inst. Slov. Acad. Sci., p. 177Hindley K.B. & Houlden M.A., 1970, Nature 225, 1232Hindley K.B., 1969, J. British Astron. Assoc., 79, 138Hoffmeister C., 1937, Die Meteore. Akademische Verlagsge-sellschaft, LeipzigImponente G. & Sigismondi C., 2001, WGN, 29, 176Jenniskens P., Meteor Showers and Their Parent Comets, Cam-bridge University Press, Cambridge, UKJones J., 1995, MNRAS, 275, 773Klaˇcka J., 2004, Celest. Mech. & Dyn. Astron., 89, 1Koten P., Boroviˇcka J., Spurn´y P., Betlem H. & Evans S., 2004,Astron. Astrophys., 428, 683Kres´akov´a M., 1974, Bull. Astron. Inst. Czechosl., 9, 88Lindblad B.A., 1971, Space Research, 11, 287Lindblad B.A., 1987, in M.Fulchignoni & L. Kres´ak, eds., TheEvolution of the Small Bodies of the Solar System, NorthHolland, p. 229Lindblad, B.A., Olsson-Steel, D., 1990, Bull. Astron. Inst.Czechosl., 41, 193Lindblad B.A., 1999, in W.G. Baggaley & V. Porubˇcan, eds.,Meteoroids 1998, Proc. Int. Conf. held at Tatransk´a Lomnica,Slovakia, Astron. Inst. Slov. Acad. Sci., p. 227.Ma Y., Williams I.P. & Chen W., 2002, MNRAS, 337, 1081McCrosky R.E. & Posen A., 1961, Smithson. Contr. Astrophys.,4, No. 2., 15Nilsson C.S., 1964, Austral. J. Phys., 17, 205Nesluˇsan L., Svoreˇn J. & Porubˇcan V., 1998, Astron. Astrophys.,331, 411Ohtsuka K., 1989, WGN 17, No. 3, 93Olsson-Steel, D., 1987, Austral. J. Astron, 2, 21Orchiston W. & Skjellerup J.F., 2003, MNSSA, 62, 54Pecina P. & ˇSimek M., 1997, Astron. & Astrophys., 317, 594Porter J.K., 1952, Comets and Meteor Streams, Chapman andHall, London (p. 123)Porubˇcan V. & Gavajdov´a M., 1994, Planet. Space Sci., 42, 151Rendtel J., 2004, Earth, Moon & Planets, 95, 27Sekanina Z., 1973, Icarus, 18, 253SonotaCo, 2009, WGN 37, 55Southworth R.R. & Hawkins G.S., 1963, Smithson. Contr. Astro-phys., 7, 261Vereˇs P. & T´oth J., 2010, WGN 38:2, 54Whipple F.L., 1954, Astron. J., 69, 201Whipple F.L. & Hakwins G.S., 1959, Meteors, in Fl¨uge (Ed.)Handbuch der Physik, Bd LII, Astrophysik III., SpringerPubl. House, Berlinc (cid:13) , 1– ?? S e m i m a j o r a x i s ( A U ) Year I n c li na t i on ( deg ) Year
December Monocerotids November Orionids C/1917 F1 Mellish
Long i t ude o f pe r i he li on Year
December MonocerotidsNovember Orionids C/1917 F1 Mellish H e li o c en t r i c d i s t an c e o f node s ( A U ))