- Email: [email protected]
A population of small refractory meteoroids in asteroidal orbits
Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Planetary and Space Science journal homepage: www.elsevier.com/locate/pss
A population of small refractory meteoroids in asteroidal orbits M. Campbell-Brown n University of Western Ontario, Department of Physics and Astronomy, London, ON, Canada N6A 3K7
art ic l e i nf o
a b s t r a c t
Article history: Received 8 January 2015 Received in revised form 2 March 2015 Accepted 25 March 2015
More than 7000 two-station meteors observed with two different video systems, both parts of the Canadian Automated Meteor Observatory, have been analysed. The more sensitive (limiting magnitude þ6.5) influx system shows a significant population of slow meteors with begin heights under 86 km, while the less sensitive (limiting magnitude þ 4) tracking system shows many more fast meteors ablating at high altitudes. The low, slow population has asteroidal orbits with low inclinations and moderate eccentricities, and radiants which are not, in general, associated with the sporadic sources. In spite of their low begin heights, which imply that they are strong and refractory, the meteors have early peaked light curves which are not predicted by classical ablation theory for non-fragmenting objects. & 2015 Elsevier Ltd. All rights reserved.
Keywords: Meteor Interplanetary dust
1. Introduction The ablation height and speed distributions of meteors in the Earth's atmosphere are of great interest because of what they reveal about the origin and composition of asteroidal and cometary material striking the Earth. The height at which a meteoroid begins to ablate depends on its speed, the angle at which it collides with the atmosphere, and its physical strength; in general, bodies moving faster, entering at shallower angles and with weaker or more volatile composition ablate higher. The effect of entry angle is the weakest effect: Štork et al. (2002) show the effect of speed and entry angle on begin and end heights; the speed has a strong effect, but the effect of zenith angle is only apparent in meteors with radiants less than 10° from the horizon. Begin heights are not strongly affected by the photometric mass, but there is a significant association between end height and mass (with more massive meteoroids having lower end heights). Ceplecha (1967) introduced the kB parameter to characterize the point at which the meteoroid had received enough energy to begin intensive ablation. He defined it as
kB = log10ρB + 2.5 log10 (v∞ ) − 0.5 log10 cos(zR )
(1)
where ρB is the density of the atmosphere at the height at which the meteoroid becomes visible, in g cm 3, v∞ is the decelerationcorrected speed of the meteoroid, in cm s 1, and zR is the zenith distance of the radiant. kB should depend only on the thermal properties of the meteoroid, notably the thermal conductivity, the specific heat, and the density. Later Ceplecha (1988) used the kB n
Tel.: þ 1 5196612111. E-mail address: [email protected]
parameter and orbital parameters to divide meteoroids into five strength categories, corresponding to asteroidal, ordinary chondrites, carbonaceous chondritic material, and strong and weak cometary material. The boundaries between the categories must be adjusted with detector sensitivity, since the beginning height of a meteor is higher in more sensitive systems (Kikwaya et al., 2011). Among the populations of meteors which can be identified based on begin height and speed, there is an interesting population with low speeds and begin heights which are low even among slow meteors. Borovička et al. (2005) examined spectra and orbits of 97 meteors, mostly between þ 3 and 1 magnitude. Of these, 14 were identified as iron-rich, with spectra almost exclusively composed of iron lines. This group had low speeds and low begin heights, and all of them had asteroidal orbits, with Tisserand parameters with respect to Jupiter of more than 3; all but two had inclinations of 10° or less. Most were very faint: all but one were fainter than þ2.5 and most peaked between þ 3 and þ 4. All of the meteors had light curves which peaked in the early part of the trajectory, contrary to what would be expected from classical single body theory, and most had sudden onsets, which started with the brightest part of the trail. The authors speculated that this might involve a transition from solid iron to liquid. A few high-density meteoroids were also found in a small survey of very faint (down to þ7 magnitude) meteors in Kikwaya et al. (2009). Six meteors with complete light curves from very sensitive, small field intensified cameras were carefully modelled, and three of them had densities over 4000 kg m 3. The three meteors had speeds of 30 km s 1 or less, ablated at heights under 100 km, and had light curves which peaked early in the trajectory; they were also among the faintest of the sample.
http://dx.doi.org/10.1016/j.pss.2015.03.022 0032-0633/& 2015 Elsevier Ltd. All rights reserved.
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
2. Observations The Canadian Automated Meteor Observatory (CAMO) consists of two stations separated by approximately 45 km; Elginfield, at 43.193°N, 81.316°W, and Tavistock, at 43.265°N, 80.772°W. Each site houses three intensified video cameras; one pair of wide and narrow field cameras which incorporate computer-controlled mirrors to track meteors in the sky (the tracking system), and one camera referred to as the influx system. The influx system uses a Gen III ITT NiteCam image intensifier (18 mm until November 2011 (Elginfield) and July 2012 (Tavistock), and 25 mm thereafter), coupled to a Cooke pco.1200 camera, with a 50 mm f/0.95 objective lens. The camera records video at 20 frames per second, 14 bit optical depth; the progressively scanned images were 1K 1K before the intensifier change, and 1600 1200 pixels after. It has a 20° field of view, and an effective limiting meteor magnitude of þ 6.5M. The main purpose of the system is monitoring meteoroid flux. The tracking system has two 18 mm Gen III NiteCam image intensifiers coupled to ImperX VGA-120 cameras. The wide field system has a 25 mm f/0.85 lens and a field of view of 28°; the narrow field system has a 545 mm f/11 telescope as an objective and a field of view of 1.5°. It is not used in this study. The wide field system runs at 80 progressive frames per second, 640 480 pixels, with an optical depth of 12 bits. It has an effective limiting magnitude of þ4M, and is used to automatically detect meteors for tracking with the narrow field system. The system is described in detail in Weryk et al. (2013). CAMO has been recording data since June of 2009. The influx system, which was the first system to become operational, has recorded 5092 two-station meteors until 2012, while the tracking system recorded just over 4000 two station meteors between June 2010 and October 2014. Meteors recorded in the wide field tracking camera are detected and reduced in realtime using the ASGARD software package (Weryk et al., 2008). The threshold for detection is set relatively high, since only brighter meteors can reliably be tracked. The influx system uses MeteorScan (Gural, 1997) to detect meteors after night's data have been recorded; the process has the thresholds set relatively low, and false alarms are mostly removed
by comparing the two systems to find common meteors; only two station meteors are saved. After detection, meteors from the influx system are reduced manually using the METAL software package (Weryk and Brown, 2012). Since we are analysing for this work only those meteors which begin and end in the field of view, the relative area of overlap of each system as a function of height is of interest. The tracking system has a peak overlap area at 90 km; the overlap area is half as large at 65 km and 143 km, and there is no overlap above 165 km. The influx system has a peak overlap area at 120 km, and falls to half that at 82 km, and a quarter at 72 km. The relative overlap area is still 60% of the maximum at 180 km, the highest height for which it has been computed.
3. Height and speed distributions Rather than plotting histograms of observed speeds and beginning heights, it is more useful to plot meteors on a speed/ begin height graph, since this makes the dependence of begin height on speed clear. Ideally, the zenith angle would also be included, but the effect of the zenith angle on the initial height is much less than the effect of the speed. Of the 5092 meteors observed by the influx system, we removed those which did not both begin and end in the field of view, those with a convergence angle (angle between the planes formed by the meteor's path and each of the cameras) less than 5°, meteors for which orbits could not be determined, and meteors with very few points. This left 3491 high quality meteors. For the mirror system, of the 4052 two-station meteors, 3521 passed the same filters. The higher fraction of useful meteors in the mirror system reflects the lack of detection of faint meteors for this system, meaning those meteors which are recorded have longer trails and better signal-to-noise. Fig. 1 shows density plots of begin heights versus speed for the influx and tracking systems. Both show the same trend of increasing begin height with speed, and a rough division into two populations: a (presumably weaker or more volatile) higher population and a (presumably stronger or more refractory) lower population. These two populations cover the full range of speeds.
Fig. 1. Begin height and speed distributions for the influx system (left) and tracking system (right).
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
The higher and lower populations in the begin height–speed plots represent two peaks in the kB parameter distribution. Zenith angle has only a small effect on the begin height, so the kB parameter increases from top (most volatile) to bottom (most refractory) in the plot. The higher population corresponds to Ceplecha's C group of cometary material, and the lower to Ceplecha's A group, which he associated with ordinary chondritic material (Ceplecha, 1988). The two CAMO systems observed meteors at similar times, but the distribution of meteors is very different for the two systems. The influx system observes meteors mostly below 30 km s 1, with a strong peak in the higher population around 30 and a second on the lower population around 12 km s 1. The tracking system has its strongest peak in the higher population just under 70 km s 1, and a second peak just over 30 km s 1 and just below the higher population. Note that height biases do not explain the distribution of data: the tracking system has relatively more collecting area at low heights, and the influx system has relatively more collecting area at higher heights. If the collecting area were uniform with height for both systems, the differences between the two datasets would be even more pronounced. The main difference between the influx and tracking systems is their limiting sensitivity. Fig. 2 shows the distribution of peak magnitudes in the two datasets: there is a small amount of overlap, but most of the meteors observed by the tracking system are between 0 and þ3 magnitude, while the influx system meteors are mainly between þ3 and þ6. Plotting only the small number of influx meteors brighter than 3rd magnitude produces a plot very similar to the tracking data, as shown in Fig. 3. Most of the slow meteors are also very faint. The influx data includes 384 shower meteors out of 3491, which is both because sporadic meteors begin to dominate at lower masses, and because the data used in this study does not include any nights of some major showers, including the Geminids (there are only two Geminid meteors in this set of influx data). The dominant showers are the North and South Taurids (81 and 85, respectively), the Perseids (45) and the Orionids (31). The tracking system has 705 shower meteors out of 3521, in which the Orionids (114), North and South Taurids (60 and 105), Geminids (89), and Perseids (77) are prominent. These showers can be seen in the begin height–speed plots of Fig 1: the Taurids are the cluster just under 30 km/s in the higher population, the Geminids are the cluster around 35 km/s in between the high and low populations in the tracking system, and the Perseids and Orionids are in the higher population at around 60 and 65 km/s, respectively.
Fig. 2. Distribution of peak magnitudes for the two CAMO camera systems.
3
Similar plots of end heights against speed were also plotted, but they did not reveal anything unexpected. In general, meteors with the lowest begin heights also had the lowest end heights, and vice versa. Fig. 4 shows the radiants of the meteors from the influx and tracking systems in Sun-centred ecliptic coordinates. In these
Fig. 3. Begin height and speed distribution for the influx systems, for only those meteors with peak magnitudes brighter than þ3.
Fig. 4. Radiant distributions in Sun-centred ecliptic coordinates, for the influx and mirror systems. In these plots, the apex of the Earth's way is in the centre, and the horizontal axis is the ecliptic plane. The Sun is to the left.
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
plots, the centre is at the apex, the direction of the velocity vector of the Earth; the Sun is to the left (thus the lack of radiants in this area), and the antisun point to the right. Most of the tracking system meteors are in the north and south apex sporadic sources and the antihelion source; the influx system shows meteors in those sources but also a large number of meteors outside the sporadic sources.
4. The low, slow population Of particular interest is the faint population of slow meteors ablating very low in the atmosphere. For simplicity, this population was defined as any meteor with a begin height of less than 86 km. This population represents only 3% of meteors brighter than þ 3, but comprises more than 6% of meteors between þ6 and þ3. A total of 263 of the influx system meteors have begin heights lower than 86 km; only 108 of the less sensitive tracking system begin under 86 km. For this analysis, only the influx data are used. Since the tracking system meteors are reduced automatically, the uncertainties in their speeds and heights are higher than the manually reduced influx meteors. The distributions of orbital parameters of all meteors observed with the influx system, and of the low, slow population, are shown in Fig. 5. The low, slow meteors have Tisserand parameters with respect to Jupiter which are generally greater than three, indicating a probable asteroidal origin. The inclinations of these particles are lower than 30°, mostly between 5° and 20°, which supports an asteroidal origin. The eccentricities, on the other hand, are higher than one would expect for orbits evolving under Poynting–Robertson drag, with a majority between 0.3 and 0.7. The semimajor axes are between 1 and 2 AU,
which is not very different from the general population. It is worth noting that radiation forces like Poynting–Robertson and close gravitational encounters can alter the Tisserand parameter, so it is possible that some of this population comes from Jupiter family comets instead of asteroids. Fig. 6 shows the radiants of all the meteors in the low, slow population. In spite of their low inclinations, the radiants show a wide scatter about the ecliptic. A few are in the outer antihelion sporadic source, but most are not close to any of the sporadic sources. The meteors, because of their low speeds, can strike the Earth from nearly any direction except the apex, including near the poles (in spite of their low inclinations) and the antapex point. Because these meteors ablate low in the atmosphere, they are presumably more refractory than meteors at the same speeds ablating higher in the atmosphere. Fig. 7 shows the kB parameters
Fig. 6. Radiants of the low, slow population in Sun-centred ecliptic coordinates.
Fig. 5. Distributions of orbital elements for all influx meteors and for the low, slow population. The Tisserand parameter in the last panel is with respect to Jupiter.
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
of all influx meteors and the low, slow population. The low, slow meteors, unsurprisingly, are all part of the secondary peak in the distribution, indicating a stronger, more refractory composition. They are clearly in Ceplecha's A group, which he links to stony material. The shape of meteor light curves is often used to determine the properties of the meteoroid. Classical meteor ablation theory (e.g. Ceplecha et al., 1998) predicts that solid, monolithic bodies will have light curves which peak toward the end of the luminous trail; bodies which fragment into many pieces of various sizes should have symmetric light curves (Hawkes and Jones, 1975). The F parameter is often used to characterize light curves: it is the ratio of the difference between the begin height of the meteor and the height of maximum light, and the difference between the begin and the end height. A small F parameter indicates that a meteor peaks early; one close to 0.5 is symmetric, and an F value close to 1 indicates that the meteor is late peaked. Fig. 8 shows the F parameters of all the meteors, and of the low, slow population. The latter are much more likely to have earlypeaked light curves than other meteors. This is odd, since we expect this low population to be stronger than average meteoroid material, and thus might expect classical light curve shapes.
Fig. 7. Distribution of kB parameters for all influx meteors and for the low, slow population.
Fig. 8. F parameter distributions for all influx meteors and for the low, slow population. The F parameters indicates where the peak light occurs; 0.5 indicates that the light curve is symmetric, an F close to 1 indicates the peak near the end of the curve, and an F close to 0 indicates an early peak.
5
5. Discussion Height and speed plots can provide information on the populations sampled by an observing system and the characteristics of the meteoroids in those populations, particularly their physical and thermal properties. It is particularly interesting that both fast meteors in retrograde orbits and slow meteors in prograde orbits show both a high and a low population in the height speed plot, implying that both strong and weak materials are present in both long period (retrograde) comets and prograde comets. It is unlikely that this represents inhomogeneities in the parent body (for example, matrix and refractory inclusions), since showers tend to cluster tightly in the height/speed plots. Future work comparing the orbits of strong and weak materials in both populations may clarify this issue. This study emphasizes that the characteristics of the observed meteor population vary significantly with limiting magnitude, even for systems which are otherwise very similar. The radiants and speed distributions of the populations observed with video systems whose sensitivity varies by just two and a half magnitudes show very different characteristics in speeds and strengths. At smaller masses, there is an important population of slow meteors which ablate low in the atmosphere. This population is not seen in radar studies at similar masses (see, for example Campbell-Brown, 2008) in any significant numbers, likely because at slow speeds very little ionization is produced. Radar studies tend to show meteoroids grouped tightly in the sporadic sources, missing many of the slower meteors which occur outside the sources. The low, slow population of meteors have orbits with Tisserand parameters with respect to Jupiter greater than 3, which points to an asteroidal origin. The low inclinations of the meteoroids support this sort of origin. It is possible, however, that the Tisserand parameters have been changed since the particles were released from their parent bodies; the Tisserand parameter is only conserved in the presence of secular perturbations from Jupiter, and can be altered by close gravitational encounters with any of the planets or (of particular importance for objects in this size range) by radiation forces. For millimeter sized particles, the Poynting– Robertson effect may be particularly effective at changing the Tisserand as orbits are circularized and reduced in size by the radiation drag force. Particles in Jupiter family comet orbits, with Tisserands initially between 2 and 3, may be altered into orbits with Tj greater than 3 by this effect. It must be noted, however, that the eccentricities of the low, slow particles are relatively high, which argues against Poynting–Robertson being a strong effect in their orbital evolution. Neither the asteroidal nor Jupiter family comet origins are entirely convincing at this point; future work involving the modelling of possible orbital evolution of these particles may clarify the origins of these particles. One possible scenario for the formation of these particles is evolution from the asteroid belt: the fact that they are particularly strong (and similar to iron particles seen by Borovička et al., 2005) may indicate that they are the longest-lived members of a population of asteroidal material evolving toward the sun, with weaker particles having been destroyed in collisions. Although these particles are among the strongest in the sample, they have kb parameters less than 8, which is the cutoff used by Ceplecha (1967) to distinguish between stony particles and true asteroidal material. He calculated the kb parameter of two fireballs which had only iron lines, and an artificial iron meteoroid, and found in all three cases that the parameter was greater than 8. This may indicate that the particles observed here are not solid iron. Since Borovička et al. (2005) observed similar meteors with only iron in their spectra, these objects could be grains of iron sulfide, which have a lower melting point and density than pure iron or iron–nickel
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
alloys. Sulfur does not produce optical light in meteors, which means that iron and iron sulfides cannot be distinguished optically. Another interesting property of this population is the earlypeaked light curves. We expect strong meteoroids (which ablate lower in the atmosphere) to have classical, late-skewed light curves. The data collected here show a broader trend (which will be explored in a future paper) with meteors with high kb parameters (lower begin heights relative to other meteors with the same speeds) having earlier peaks to their light curves. It is not at all clear why this should be: Borovička et al. (2005) suggest that, in their sample of 14 iron meteoroids, the sudden onsets were due to melting and spraying of the iron particles. Much work remains to understand this population of meteoroids, including detailed models of meteoroid ablation to explain their unexpected light curve shapes, dynamical modelling to understand their origin, and further observations to understand their distribution and nature.
Acknowledgements
and Engineering Research Council of Canada and the Canadian Foundation for Innovation program. The author also thanks the reviewers for their helpful comments.
References Borovička, J., Koten, P., Spurný, P., Boček, J., Štork, R., 2005. Icarus 174, 15. Campbell-Brown, M., 2008. Icarus 196, 144. Ceplecha, Z., 1967. Smithsonian Contributions to Astrophysics, vol. 11, p. 35. Ceplecha, Z., 1988. Bull. Astron. Inst. Czechoslov. 39, 4. Ceplecha, Z., Borovička, J., Elford, W.G., ReVelle, D.O., Hawkes, R.L., Porubčan, V., Šimek, M., 1998. Space Sci. Rev. 84, 327. Gural, P., 1997. WGN 25, 136. Hawkes, R.L., Jones, J., 1975. Mon. Not. R. Astron. Soc. 173, 339. Kikwaya, J., Campbell-Brown, M., Brown, P., Hawkes, R., Weryk, R., 2009. Astron. Astrophys. 497, 851. Kikwaya, J., Campbell-Brown, M., Brown, P., 2011. Astron. Astrophys. 530, A113. Štork, R., Koten, P., Borovička, J., Spurný, P., 2002. In: Warmbein, Barbara (Ed). Proceedings of Asteroids, Comets, Metors 2002, ESA SP-500, Noordwijk, Netherlands, p. 189. Weryk, R., Brown, P., Domokos, A., Edwards, W., Krzeminski, Z., Nudds, S., Welch, D., 2008. Earth Moon Planets 102, 241. Weryk, R., Brown, P., 2012. Planet. Space Sci. 62, 132. Weryk, R., Campbell-Brown, M., Wiegert, P., Brown, P., Krzeminski, Z., Musci, R., 2013. Icarus 225, 614.
The author acknowledges funding support for CAMO from NASA co-operative agreement NNX11AB76A, the Natural Sciences
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
Contents lists available at ScienceDirect
Planetary and Space Science journal homepage: www.elsevier.com/locate/pss
A population of small refractory meteoroids in asteroidal orbits M. Campbell-Brown n University of Western Ontario, Department of Physics and Astronomy, London, ON, Canada N6A 3K7
art ic l e i nf o
a b s t r a c t
Article history: Received 8 January 2015 Received in revised form 2 March 2015 Accepted 25 March 2015
More than 7000 two-station meteors observed with two different video systems, both parts of the Canadian Automated Meteor Observatory, have been analysed. The more sensitive (limiting magnitude þ6.5) influx system shows a significant population of slow meteors with begin heights under 86 km, while the less sensitive (limiting magnitude þ 4) tracking system shows many more fast meteors ablating at high altitudes. The low, slow population has asteroidal orbits with low inclinations and moderate eccentricities, and radiants which are not, in general, associated with the sporadic sources. In spite of their low begin heights, which imply that they are strong and refractory, the meteors have early peaked light curves which are not predicted by classical ablation theory for non-fragmenting objects. & 2015 Elsevier Ltd. All rights reserved.
Keywords: Meteor Interplanetary dust
1. Introduction The ablation height and speed distributions of meteors in the Earth's atmosphere are of great interest because of what they reveal about the origin and composition of asteroidal and cometary material striking the Earth. The height at which a meteoroid begins to ablate depends on its speed, the angle at which it collides with the atmosphere, and its physical strength; in general, bodies moving faster, entering at shallower angles and with weaker or more volatile composition ablate higher. The effect of entry angle is the weakest effect: Štork et al. (2002) show the effect of speed and entry angle on begin and end heights; the speed has a strong effect, but the effect of zenith angle is only apparent in meteors with radiants less than 10° from the horizon. Begin heights are not strongly affected by the photometric mass, but there is a significant association between end height and mass (with more massive meteoroids having lower end heights). Ceplecha (1967) introduced the kB parameter to characterize the point at which the meteoroid had received enough energy to begin intensive ablation. He defined it as
kB = log10ρB + 2.5 log10 (v∞ ) − 0.5 log10 cos(zR )
(1)
where ρB is the density of the atmosphere at the height at which the meteoroid becomes visible, in g cm 3, v∞ is the decelerationcorrected speed of the meteoroid, in cm s 1, and zR is the zenith distance of the radiant. kB should depend only on the thermal properties of the meteoroid, notably the thermal conductivity, the specific heat, and the density. Later Ceplecha (1988) used the kB n
Tel.: þ 1 5196612111. E-mail address: [email protected]
parameter and orbital parameters to divide meteoroids into five strength categories, corresponding to asteroidal, ordinary chondrites, carbonaceous chondritic material, and strong and weak cometary material. The boundaries between the categories must be adjusted with detector sensitivity, since the beginning height of a meteor is higher in more sensitive systems (Kikwaya et al., 2011). Among the populations of meteors which can be identified based on begin height and speed, there is an interesting population with low speeds and begin heights which are low even among slow meteors. Borovička et al. (2005) examined spectra and orbits of 97 meteors, mostly between þ 3 and 1 magnitude. Of these, 14 were identified as iron-rich, with spectra almost exclusively composed of iron lines. This group had low speeds and low begin heights, and all of them had asteroidal orbits, with Tisserand parameters with respect to Jupiter of more than 3; all but two had inclinations of 10° or less. Most were very faint: all but one were fainter than þ2.5 and most peaked between þ 3 and þ 4. All of the meteors had light curves which peaked in the early part of the trajectory, contrary to what would be expected from classical single body theory, and most had sudden onsets, which started with the brightest part of the trail. The authors speculated that this might involve a transition from solid iron to liquid. A few high-density meteoroids were also found in a small survey of very faint (down to þ7 magnitude) meteors in Kikwaya et al. (2009). Six meteors with complete light curves from very sensitive, small field intensified cameras were carefully modelled, and three of them had densities over 4000 kg m 3. The three meteors had speeds of 30 km s 1 or less, ablated at heights under 100 km, and had light curves which peaked early in the trajectory; they were also among the faintest of the sample.
http://dx.doi.org/10.1016/j.pss.2015.03.022 0032-0633/& 2015 Elsevier Ltd. All rights reserved.
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
2. Observations The Canadian Automated Meteor Observatory (CAMO) consists of two stations separated by approximately 45 km; Elginfield, at 43.193°N, 81.316°W, and Tavistock, at 43.265°N, 80.772°W. Each site houses three intensified video cameras; one pair of wide and narrow field cameras which incorporate computer-controlled mirrors to track meteors in the sky (the tracking system), and one camera referred to as the influx system. The influx system uses a Gen III ITT NiteCam image intensifier (18 mm until November 2011 (Elginfield) and July 2012 (Tavistock), and 25 mm thereafter), coupled to a Cooke pco.1200 camera, with a 50 mm f/0.95 objective lens. The camera records video at 20 frames per second, 14 bit optical depth; the progressively scanned images were 1K 1K before the intensifier change, and 1600 1200 pixels after. It has a 20° field of view, and an effective limiting meteor magnitude of þ 6.5M. The main purpose of the system is monitoring meteoroid flux. The tracking system has two 18 mm Gen III NiteCam image intensifiers coupled to ImperX VGA-120 cameras. The wide field system has a 25 mm f/0.85 lens and a field of view of 28°; the narrow field system has a 545 mm f/11 telescope as an objective and a field of view of 1.5°. It is not used in this study. The wide field system runs at 80 progressive frames per second, 640 480 pixels, with an optical depth of 12 bits. It has an effective limiting magnitude of þ4M, and is used to automatically detect meteors for tracking with the narrow field system. The system is described in detail in Weryk et al. (2013). CAMO has been recording data since June of 2009. The influx system, which was the first system to become operational, has recorded 5092 two-station meteors until 2012, while the tracking system recorded just over 4000 two station meteors between June 2010 and October 2014. Meteors recorded in the wide field tracking camera are detected and reduced in realtime using the ASGARD software package (Weryk et al., 2008). The threshold for detection is set relatively high, since only brighter meteors can reliably be tracked. The influx system uses MeteorScan (Gural, 1997) to detect meteors after night's data have been recorded; the process has the thresholds set relatively low, and false alarms are mostly removed
by comparing the two systems to find common meteors; only two station meteors are saved. After detection, meteors from the influx system are reduced manually using the METAL software package (Weryk and Brown, 2012). Since we are analysing for this work only those meteors which begin and end in the field of view, the relative area of overlap of each system as a function of height is of interest. The tracking system has a peak overlap area at 90 km; the overlap area is half as large at 65 km and 143 km, and there is no overlap above 165 km. The influx system has a peak overlap area at 120 km, and falls to half that at 82 km, and a quarter at 72 km. The relative overlap area is still 60% of the maximum at 180 km, the highest height for which it has been computed.
3. Height and speed distributions Rather than plotting histograms of observed speeds and beginning heights, it is more useful to plot meteors on a speed/ begin height graph, since this makes the dependence of begin height on speed clear. Ideally, the zenith angle would also be included, but the effect of the zenith angle on the initial height is much less than the effect of the speed. Of the 5092 meteors observed by the influx system, we removed those which did not both begin and end in the field of view, those with a convergence angle (angle between the planes formed by the meteor's path and each of the cameras) less than 5°, meteors for which orbits could not be determined, and meteors with very few points. This left 3491 high quality meteors. For the mirror system, of the 4052 two-station meteors, 3521 passed the same filters. The higher fraction of useful meteors in the mirror system reflects the lack of detection of faint meteors for this system, meaning those meteors which are recorded have longer trails and better signal-to-noise. Fig. 1 shows density plots of begin heights versus speed for the influx and tracking systems. Both show the same trend of increasing begin height with speed, and a rough division into two populations: a (presumably weaker or more volatile) higher population and a (presumably stronger or more refractory) lower population. These two populations cover the full range of speeds.
Fig. 1. Begin height and speed distributions for the influx system (left) and tracking system (right).
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
The higher and lower populations in the begin height–speed plots represent two peaks in the kB parameter distribution. Zenith angle has only a small effect on the begin height, so the kB parameter increases from top (most volatile) to bottom (most refractory) in the plot. The higher population corresponds to Ceplecha's C group of cometary material, and the lower to Ceplecha's A group, which he associated with ordinary chondritic material (Ceplecha, 1988). The two CAMO systems observed meteors at similar times, but the distribution of meteors is very different for the two systems. The influx system observes meteors mostly below 30 km s 1, with a strong peak in the higher population around 30 and a second on the lower population around 12 km s 1. The tracking system has its strongest peak in the higher population just under 70 km s 1, and a second peak just over 30 km s 1 and just below the higher population. Note that height biases do not explain the distribution of data: the tracking system has relatively more collecting area at low heights, and the influx system has relatively more collecting area at higher heights. If the collecting area were uniform with height for both systems, the differences between the two datasets would be even more pronounced. The main difference between the influx and tracking systems is their limiting sensitivity. Fig. 2 shows the distribution of peak magnitudes in the two datasets: there is a small amount of overlap, but most of the meteors observed by the tracking system are between 0 and þ3 magnitude, while the influx system meteors are mainly between þ3 and þ6. Plotting only the small number of influx meteors brighter than 3rd magnitude produces a plot very similar to the tracking data, as shown in Fig. 3. Most of the slow meteors are also very faint. The influx data includes 384 shower meteors out of 3491, which is both because sporadic meteors begin to dominate at lower masses, and because the data used in this study does not include any nights of some major showers, including the Geminids (there are only two Geminid meteors in this set of influx data). The dominant showers are the North and South Taurids (81 and 85, respectively), the Perseids (45) and the Orionids (31). The tracking system has 705 shower meteors out of 3521, in which the Orionids (114), North and South Taurids (60 and 105), Geminids (89), and Perseids (77) are prominent. These showers can be seen in the begin height–speed plots of Fig 1: the Taurids are the cluster just under 30 km/s in the higher population, the Geminids are the cluster around 35 km/s in between the high and low populations in the tracking system, and the Perseids and Orionids are in the higher population at around 60 and 65 km/s, respectively.
Fig. 2. Distribution of peak magnitudes for the two CAMO camera systems.
3
Similar plots of end heights against speed were also plotted, but they did not reveal anything unexpected. In general, meteors with the lowest begin heights also had the lowest end heights, and vice versa. Fig. 4 shows the radiants of the meteors from the influx and tracking systems in Sun-centred ecliptic coordinates. In these
Fig. 3. Begin height and speed distribution for the influx systems, for only those meteors with peak magnitudes brighter than þ3.
Fig. 4. Radiant distributions in Sun-centred ecliptic coordinates, for the influx and mirror systems. In these plots, the apex of the Earth's way is in the centre, and the horizontal axis is the ecliptic plane. The Sun is to the left.
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
plots, the centre is at the apex, the direction of the velocity vector of the Earth; the Sun is to the left (thus the lack of radiants in this area), and the antisun point to the right. Most of the tracking system meteors are in the north and south apex sporadic sources and the antihelion source; the influx system shows meteors in those sources but also a large number of meteors outside the sporadic sources.
4. The low, slow population Of particular interest is the faint population of slow meteors ablating very low in the atmosphere. For simplicity, this population was defined as any meteor with a begin height of less than 86 km. This population represents only 3% of meteors brighter than þ 3, but comprises more than 6% of meteors between þ6 and þ3. A total of 263 of the influx system meteors have begin heights lower than 86 km; only 108 of the less sensitive tracking system begin under 86 km. For this analysis, only the influx data are used. Since the tracking system meteors are reduced automatically, the uncertainties in their speeds and heights are higher than the manually reduced influx meteors. The distributions of orbital parameters of all meteors observed with the influx system, and of the low, slow population, are shown in Fig. 5. The low, slow meteors have Tisserand parameters with respect to Jupiter which are generally greater than three, indicating a probable asteroidal origin. The inclinations of these particles are lower than 30°, mostly between 5° and 20°, which supports an asteroidal origin. The eccentricities, on the other hand, are higher than one would expect for orbits evolving under Poynting–Robertson drag, with a majority between 0.3 and 0.7. The semimajor axes are between 1 and 2 AU,
which is not very different from the general population. It is worth noting that radiation forces like Poynting–Robertson and close gravitational encounters can alter the Tisserand parameter, so it is possible that some of this population comes from Jupiter family comets instead of asteroids. Fig. 6 shows the radiants of all the meteors in the low, slow population. In spite of their low inclinations, the radiants show a wide scatter about the ecliptic. A few are in the outer antihelion sporadic source, but most are not close to any of the sporadic sources. The meteors, because of their low speeds, can strike the Earth from nearly any direction except the apex, including near the poles (in spite of their low inclinations) and the antapex point. Because these meteors ablate low in the atmosphere, they are presumably more refractory than meteors at the same speeds ablating higher in the atmosphere. Fig. 7 shows the kB parameters
Fig. 6. Radiants of the low, slow population in Sun-centred ecliptic coordinates.
Fig. 5. Distributions of orbital elements for all influx meteors and for the low, slow population. The Tisserand parameter in the last panel is with respect to Jupiter.
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
of all influx meteors and the low, slow population. The low, slow meteors, unsurprisingly, are all part of the secondary peak in the distribution, indicating a stronger, more refractory composition. They are clearly in Ceplecha's A group, which he links to stony material. The shape of meteor light curves is often used to determine the properties of the meteoroid. Classical meteor ablation theory (e.g. Ceplecha et al., 1998) predicts that solid, monolithic bodies will have light curves which peak toward the end of the luminous trail; bodies which fragment into many pieces of various sizes should have symmetric light curves (Hawkes and Jones, 1975). The F parameter is often used to characterize light curves: it is the ratio of the difference between the begin height of the meteor and the height of maximum light, and the difference between the begin and the end height. A small F parameter indicates that a meteor peaks early; one close to 0.5 is symmetric, and an F value close to 1 indicates that the meteor is late peaked. Fig. 8 shows the F parameters of all the meteors, and of the low, slow population. The latter are much more likely to have earlypeaked light curves than other meteors. This is odd, since we expect this low population to be stronger than average meteoroid material, and thus might expect classical light curve shapes.
Fig. 7. Distribution of kB parameters for all influx meteors and for the low, slow population.
Fig. 8. F parameter distributions for all influx meteors and for the low, slow population. The F parameters indicates where the peak light occurs; 0.5 indicates that the light curve is symmetric, an F close to 1 indicates the peak near the end of the curve, and an F close to 0 indicates an early peak.
5
5. Discussion Height and speed plots can provide information on the populations sampled by an observing system and the characteristics of the meteoroids in those populations, particularly their physical and thermal properties. It is particularly interesting that both fast meteors in retrograde orbits and slow meteors in prograde orbits show both a high and a low population in the height speed plot, implying that both strong and weak materials are present in both long period (retrograde) comets and prograde comets. It is unlikely that this represents inhomogeneities in the parent body (for example, matrix and refractory inclusions), since showers tend to cluster tightly in the height/speed plots. Future work comparing the orbits of strong and weak materials in both populations may clarify this issue. This study emphasizes that the characteristics of the observed meteor population vary significantly with limiting magnitude, even for systems which are otherwise very similar. The radiants and speed distributions of the populations observed with video systems whose sensitivity varies by just two and a half magnitudes show very different characteristics in speeds and strengths. At smaller masses, there is an important population of slow meteors which ablate low in the atmosphere. This population is not seen in radar studies at similar masses (see, for example Campbell-Brown, 2008) in any significant numbers, likely because at slow speeds very little ionization is produced. Radar studies tend to show meteoroids grouped tightly in the sporadic sources, missing many of the slower meteors which occur outside the sources. The low, slow population of meteors have orbits with Tisserand parameters with respect to Jupiter greater than 3, which points to an asteroidal origin. The low inclinations of the meteoroids support this sort of origin. It is possible, however, that the Tisserand parameters have been changed since the particles were released from their parent bodies; the Tisserand parameter is only conserved in the presence of secular perturbations from Jupiter, and can be altered by close gravitational encounters with any of the planets or (of particular importance for objects in this size range) by radiation forces. For millimeter sized particles, the Poynting– Robertson effect may be particularly effective at changing the Tisserand as orbits are circularized and reduced in size by the radiation drag force. Particles in Jupiter family comet orbits, with Tisserands initially between 2 and 3, may be altered into orbits with Tj greater than 3 by this effect. It must be noted, however, that the eccentricities of the low, slow particles are relatively high, which argues against Poynting–Robertson being a strong effect in their orbital evolution. Neither the asteroidal nor Jupiter family comet origins are entirely convincing at this point; future work involving the modelling of possible orbital evolution of these particles may clarify the origins of these particles. One possible scenario for the formation of these particles is evolution from the asteroid belt: the fact that they are particularly strong (and similar to iron particles seen by Borovička et al., 2005) may indicate that they are the longest-lived members of a population of asteroidal material evolving toward the sun, with weaker particles having been destroyed in collisions. Although these particles are among the strongest in the sample, they have kb parameters less than 8, which is the cutoff used by Ceplecha (1967) to distinguish between stony particles and true asteroidal material. He calculated the kb parameter of two fireballs which had only iron lines, and an artificial iron meteoroid, and found in all three cases that the parameter was greater than 8. This may indicate that the particles observed here are not solid iron. Since Borovička et al. (2005) observed similar meteors with only iron in their spectra, these objects could be grains of iron sulfide, which have a lower melting point and density than pure iron or iron–nickel
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i
M. Campbell-Brown / Planetary and Space Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
alloys. Sulfur does not produce optical light in meteors, which means that iron and iron sulfides cannot be distinguished optically. Another interesting property of this population is the earlypeaked light curves. We expect strong meteoroids (which ablate lower in the atmosphere) to have classical, late-skewed light curves. The data collected here show a broader trend (which will be explored in a future paper) with meteors with high kb parameters (lower begin heights relative to other meteors with the same speeds) having earlier peaks to their light curves. It is not at all clear why this should be: Borovička et al. (2005) suggest that, in their sample of 14 iron meteoroids, the sudden onsets were due to melting and spraying of the iron particles. Much work remains to understand this population of meteoroids, including detailed models of meteoroid ablation to explain their unexpected light curve shapes, dynamical modelling to understand their origin, and further observations to understand their distribution and nature.
Acknowledgements
and Engineering Research Council of Canada and the Canadian Foundation for Innovation program. The author also thanks the reviewers for their helpful comments.
References Borovička, J., Koten, P., Spurný, P., Boček, J., Štork, R., 2005. Icarus 174, 15. Campbell-Brown, M., 2008. Icarus 196, 144. Ceplecha, Z., 1967. Smithsonian Contributions to Astrophysics, vol. 11, p. 35. Ceplecha, Z., 1988. Bull. Astron. Inst. Czechoslov. 39, 4. Ceplecha, Z., Borovička, J., Elford, W.G., ReVelle, D.O., Hawkes, R.L., Porubčan, V., Šimek, M., 1998. Space Sci. Rev. 84, 327. Gural, P., 1997. WGN 25, 136. Hawkes, R.L., Jones, J., 1975. Mon. Not. R. Astron. Soc. 173, 339. Kikwaya, J., Campbell-Brown, M., Brown, P., Hawkes, R., Weryk, R., 2009. Astron. Astrophys. 497, 851. Kikwaya, J., Campbell-Brown, M., Brown, P., 2011. Astron. Astrophys. 530, A113. Štork, R., Koten, P., Borovička, J., Spurný, P., 2002. In: Warmbein, Barbara (Ed). Proceedings of Asteroids, Comets, Metors 2002, ESA SP-500, Noordwijk, Netherlands, p. 189. Weryk, R., Brown, P., Domokos, A., Edwards, W., Krzeminski, Z., Nudds, S., Welch, D., 2008. Earth Moon Planets 102, 241. Weryk, R., Brown, P., 2012. Planet. Space Sci. 62, 132. Weryk, R., Campbell-Brown, M., Wiegert, P., Brown, P., Krzeminski, Z., Musci, R., 2013. Icarus 225, 614.
The author acknowledges funding support for CAMO from NASA co-operative agreement NNX11AB76A, the Natural Sciences
Please cite this article as: Campbell-Brown, M., A population of small refractory meteoroids in asteroidal orbits. Planetary and Space Science (2015), http://dx.doi.org/10.1016/j.pss.2015.03.022i