1 Introduction

The topics of this review are the minor planets in short orbital periods (<20 years) that are potential parent bodies of our ecliptic (and toroidal) meteor showers. They derive from two source regions: the asteroid belt between Mars and Jupiter and the Kuiper Belt beyond Neptune, specifically its Scattered Disc. Some of the planetesimals that were formed in the region between Jupiter and Neptune can now be found in the Oort cloud and are the source of our Long-Period and Halley-type comets.

The International Astronomical Union has not yet adopted clear definitions on what the terms “asteroid” or “comet”, or even “minor planet” refer to. All these objects fall under the category “small solar system bodies”, including meteoroids.

The name “asteroid” is used by some to imply all minor planets that appear star-like. Others restrict the name to planetesimals that were formed in the region between Mars and Jupiter. I will adopt an even stricter definition: those that were formed in the region between Mars and Jupiter and also have lost most of their unbound water. Most remaining water in asteroids was incorporated into mineral structures, mostly forming clays, resulting in strong rocky materials. Asteroids are the source of our meteorites, material strong enough to survive the impact in Earth’s atmosphere. Most are suspected to originate from the inner parts of the asteroid belt, where S-type asteroids are common.

The name “comet” is used by some to imply all minor planets that have a fuzzy halo or tail. Others restrict the name to planetesimals formed from dust grains coated with a layer of water ice. Such grains did not exist in the inner solar system, where small rocky planets were formed. Water ice was present in the region where rapid dust accumulation resulted in the growth of Jupiter, our most massive planet. All planetesimals formed in the neighbourhood of Jupiter and outwards are comets. Because of the presence of volatile ices (and abundant organic molecules), the dust of comets tends to be very fragile as soon as the ice evaporates. A loose agglomerate of dust grains remains, such as collected from comet 81P/Wild 2 by NASA’s Stardust mission (Brownlee et al. 2006; Zolensky et al. 2006).

It was recently realised that the border between “asteroids” and “comets” may well be diffuse and is somewhere in the asteroid belt. Some very primitive asteroids could still contain water ice and result in comet-like activity following a collision or when perturbed inwards. Several main belt asteroids have been discovered that showed brief cometary activity (Hsieh and Jewitt 2006). These objects are strictly comets, in my definition and that of Hsieh and Jewitt, but they are comets from a third source region: just inside the orbit of Jupiter.

Near-Earth Objects (NEO) are asteroids and comets whose orbits have a perihelion distance q < 1.3 AU, which can bring them close to Earth’s orbit. Asteroids are perturbed into near-Earth orbits through the action of the ν6 secular resonance (line of apsides of the asteroids move at the same rate as that of Saturn) on the inside of the asteroid belt, and numerous mean motion resonances with Jupiter throughout the asteroid belt, notably the 3:1 mean motion resonance. Once a large or small asteroid attains an orbit that resonates with that of Jupiter, it will quickly change eccentricity and the perihelion distance will decline. The aphelion of the orbit stays in the asteroid belt (2.5–4 AU). Asteroid belt comets would also be perturbed by mean motion resonances, perhaps most notably by the 2:1 resonance.

Comets are perturbed into near-Earth orbits through the action of resonances with Neptune in the scattered disc of the Kuiper Belt. They gradually evolve into orbits with a perihelion inside that of Neptune (when they are called “Centaurs”) and can then be captured by Uranus, Saturn, and finally Jupiter. When they loose momentum in a close encounter, they end up having an aphelion close to the orbit of the planet and a perihelion much further inwards. Jupiter Family Comets have an aphelion near the orbit of Jupiter, until they are decoupled from Jupiter in a close encounter with one of the terrestrial planets, which can lower the aphelion distance (Everhart 1972).

2 A Brief History of Associating Asteroids to Meteor Showers

The first fifteen asteroid-looking NEO were discovered in 1898 (433 Eros), 1911 (719 Albert), 1918 (887 Alinda), 1929 (1627 Ivar), 1924 (1036 Ganymed), 1932 (1862 Apollo, the first Earth-crossing asteroid) and (1221 Amor), 1936 (2101 Adonis), 1937 (69230 Hermes), 1947 (2201 Oljato), 1948 (1685 Toro) and (1863 Antinous), 1949 (1566 Icarus), 1950 (29075), and 1951 (1620 Geographos).

When Whipple (1938) calculated the first meteoroid orbits from his multi-station photographic project, he pointed out that several of his ecliptic short period orbits were similar to those of Apollo type asteroids Apollo, Adonis (which he called Anteros), and Hermes. They had a similar low inclination (1.9–6°) and one had a semi-major axis of only 1.91 AU, shorter than that of Jupiter Family Comets. Whipple’s 1936–1937 results for the orbit of the Geminids, with \( \langle {\text{a}}\rangle \) = 1.396 and \( \langle {\text{e}}\rangle \) = 0.900 (Lovell 1954), implied a stream that was unique compared to orbits of comets or asteroids known at that time, but not unlike the orbit of Icarus discovered shortly thereafter.

Cuno Hoffmeister (1948) first recognized the complex of ecliptic showers and pointed out that this could well be part of the system of minor planets. He noticed a resemblance between the orbit of Adonis and that of his Scorpiid-Sagittariid Complex, but he found a discouraging difference of some 25° between the longitude of perihelion of the asteroid and that of the middle of the meteor complex. He also noticed a similarity between his Piscids Complex and asteroid Hermes, and his Virginids Complex and Apollo. Plavec (1953, 1954) investigated the evolution of these proposed shower and asteroid associations, but did not confirm their generic relationship.

Later, Sekanina (1976) reinforced the orbital similarity of Adonis with six streams detected in the Harvard Meteor Project radar data, radiating from Sagittarius, Aquarius, and Capricorn, but he could not prove the evolutionary relationship more quantitatively. Besides Adonis, he pointed to the possible existence of associations of meteor streams with minor planets 433 Eros (Sekanina’s xi Cygnids), 1627 Ivar (his August mu Draconids), 1566 Icarus (#171 Daytime Arietids and his Taurids-Arietids), 1862 Apollo (#66 Northern omega Scorpiids), 69230 Hermes (#156 N. Daytime May Arietids and #234 October epsilon Piscids), 1620 Geographos (#39 N. alpha-Leonids), 1685 Toro (his January Aquariids), 1950 DA (#39 N. alpha-Leonids and #133 April Psi Ursae-Majorids), 1959 LM (6 toroidal streams in June), 4788 P-L (his Canes Venaticids), 1973 NA (#187 psi Cassiopeiids), and 1973 EC (his kappa Geminids and his lambda Aurigids).

None of these proposed associations were particularly convincing. The tool used to make such associations was the Dissimilarity criterion (D criterion) introduced by Southworth and Hawkins (1963) and several varieties since. It makes a comparison between two sets of orbits and quantifies how much they differ. With many ecliptic streams much dispersed and very badly described by observations, it was not too difficult to find potential parent bodies.

Sekanina also identified as associated with meteor showers a series of deeply penetrating fireballs of the Prairy Network and European Fireball Network, the type of fireballs that are suspected meteorite droppers. This would suggest that asteroids, like comets, travel with a stream of debris. These early results suggested that all near-Earth objects with a node near Earth orbit had an associated meteoroid streams, irrespective of taxonomy.

The problem with the proposed meteoroid streams from an asteroidal origin are the high cosmic ray exposure ages of meteorites. They measure the time since the meteoroid was away from a larger parent body that generated neutrons by cosmic ray impacts, either sitting on the surface or being part of the parent body. Those timescales range typically from 1 to 30 million years. Over such long timescales, the streams will loose cohesion, disperse widely, and can get separated from the parent asteroid (Levin 1956). An asteroidal meteoroid stream is possible only if the meteoroid subsequently breaks in a collision, or otherwise, not long before some of the fragments hit Earth (Pauls and Gladmann 2005). The search for asteroidal streams was pursued by A. K. Terentjeva (1968, 1989), who has published many possible associations from fireball orbit surveys. Table 9 in Jenniskens (2006) gives a list of the more likely potential asteroidal meteoroid streams, after separating out the deeply penetrating fireballs from others, most of which are only pairs or triplets of similar orbits. All of those proposed streams need confirmation, before it can be certain that these are meteoroids from the same parent bodies. As it stands, no asteroidal meteoroid streams are established.

3 A Brief History of Association with Dormant Comets

For years, the relationship between asteroids, Jupiter family comets, and meteoroids was widely debated, but little was known about the dynamical processes that determined their interrelationships. Following Whipple’s (1950) formulation of a comet model, where the sublimation of water vapor caused the comet to accelerate and loose mass by ejecting ice and dust in space, it was realised that comets can get “defunct” or dormant after having exhausted their gas reserves (Samoilova-Yakhontova 1950; Öpik 1963). There were in fact examples, such as comet 28P/Neujmin 1, which was stellar in appearance only two weeks after its perihelion passage during discovery in 1913. Only shortly afterwards a faint coma and tail were detected. With an orbital period of 18 yrs, this was not likely an asteroid. Later, Öpik (1968) argued from the lack of a known mechanism at the time to turn circular asteroidal orbits into eccentric orbits, that many of the Apollo asteroids had to be dormant comet nuclei, and argued that these comets were the source of meteorites. Kresák (1987) pointed to evidence of dormant phases in the aging of periodic comets from missed comet apparitions. In the mid 1980s, the list of asteroid-looking objects moving on cometary orbits increased significantly (Kresák and Stohl 1989).

Since that time, the orbital dynamics is better understood and we now guess that only 5–9% of NEO are dormant comets. Weissman et al. (2002) and Binzel and Lupisho (2006) have given reviews of the physical characteristics of such objects and the dynamical studies that estimate their abundance. These objects are now typically recognized by their Jupiter-Family-Comet orbits (Tisserand parameter between 2 and 3) and dark albedo (A’Hearn 1985). Another way of recognizing dormant comets among the population of Near Earth Objects (NEOs) is the presence of a meteoroid stream from past cometary activity.

3.1 The Association with Meteor Showers

In 1983, a fast moving object was discovered in IRAS observations of the sky at mid-Infrared wavelengths and Whipple (1983) realised that this asteroidal-looking object 3200 Phaethon moved among the Geminids. Phaethon shows no cometary activity (McFadden et al. 1985; Hsieh and Jewitt 2005) and has such a high Tisserand invariant that it was suspected to be an asteroid, perhaps just an interloper, or an object that generated meteoroids from a collision with a small main belt asteroid near aphelion (Hunt et al. 1985). Gustafson (1989), however, demonstrated that the Geminids were generated at perihelion, not at aphelion. He suspected activity over an extended period of time, from a now dormant comet. Phaethon was thought to be the rocky core of a de-volatilized comet, the missing link showing that comets can evolve into Apollo-type asteroids. In the words of Hughes (1985): “The discovery of the asteroid-like object 1983 TB in the Geminid stream has strengthened the possibility that some comets can either choke themselves to death by forming a thick crust, or have a core of volatile-free material that remains after the majority of the gas and dust has escaped.”

Earlier, Whipple and Hamid (1952) had discussed the Taurid stream in connection with 2P/Encke and concluded that other objects than this comet had to contribute to the formation of the stream. Napier and Clube (1979) and Napier (1983) proposed that minor planets 2201 (Oljato), 2212 (Hephaistos), 5025 P-L, 1979 XB, 1982 TA, 1984 KB, 1987 SB, 1991 TB and others were such comet fragments, now dormant, together forming a massive Taurid Complex (Clube and Napier 1984; Bailey et al. 1986; Clube 1986, 1987; Olsson-Steel 1987, 1988; Steel et al. 1991; Asher et al. 1992; Porubçan and Kornos 2002). The main premise of a progressively disintegrating comet has held up, but the original comet was not quite as big as needed to justify their hypothesis of frequent past terrestrial catastrophic events. Nearly all proposed parent bodies have since been dismissed as asteroids, based on taxonomy (Jenniskens 2006). Most are O or S-type asteroids that became NEO through the ν6 secular resonance mechanism that is also responsible for some of our meteorites.

The discovery of Phaethon and the possible existence of a Taurid Complex resulted in a new search for associations of asteroid-looking objects and meteor showers. Olsson-Steel (1987, 1988) linked 1566 (Icarus) to the Daytime Arietids, now thought to be associated with the Marsden Sungrazers instead. He, too, pursued the idea that some of the NEO could be dormant comets and therefore associated with meteoroid streams from past activity.

Following on this work, Hasegawa et al. (1992) published a series of theoretical radiants and considered orbits of NEO up to the end of 1989. Drummond (1982, 1991) has compared the orbital elements of 139 NEO to meteoroid streams up to 1990 KA. He also compared the orbits of meteorite falls to those of minor planets, and like Halliday et al. (1990), identified four possible streams among meteoroid dropping fireballs. Kostolansky (1998) searched for asteroid parent bodies for 4409 photographed meteor orbits. Babadzhanov (1998) investigated the orbital evolution of candidate Taurid complex bodies over long enough periods of time to complete a nutation cycle and identified observed meteoroid streams at the four possible nodes for all objects. None of the proposed associations have been confirmed (but see Beech 2006).

4 The New Era: Comet Disintegration as the Major Source of Dust

The massive disruption of comets was recognized as a possible source of meteoroids, but such disruptions were deemed too rare among active Jupiter Family Comets to be a significant source of our meteor showers (Hughes 1985). The state of affairs before 2003 was best expressed by Hughes saying: “There is no reason why the parent comet should undergo perturbations of a similar magnitude so even though they started in the same place, the stream and comet can quickly separate as time passes. This is probably the only satisfactory explanation as to why two out of three of the streams in Cook’s (1973) list do not have recognizable parents.” The parent bodies were somewhere, but they had now evolved beyond their streams. “It seems”, according to Hughes, “that in the large majority of cases comets decay gently.”

In my 2006 book “Meteor Showers and their Parent Comets”, I have argued that all of that changed in 2003, when it was discovered that 2003 EH1 moves among the highly inclined Quadrantid meteoroid stream, with only a 1 in 2 million chance of being a coincidental interloper (Jenniskens 2003, 2004). The association has since been studied by Williams et al. (2004), who confirmed that a comet observed in 1491 (C/1490 Y1) could well be the moment of breakup that generated the Quadrantid stream. Alternatively, Wiegert and Brown (2005a) have calculated backward in time the orbits of photographed Quadrantids, to conclude that the stream may be as young as two hundred years. Note, however, that the dispersion in the backward integrated orbits rapidly increases only when integrated to before 1490.

Over such a short timescale, the dispersion of dust can be simulated in numerical modeling, and from the distribution of nodes and the activity of the shower in Earth’s path, a mass can be calculated for the whole stream. That mass (Table 1) is of order 1 × 1013 kg, needing a thousand years to generate during normal comet activity.

Table 1 Mass estimates of remaining comet fragments and their meteoroid stream, in units of 1 billion kg, after Jenniskens (2006)
Full size table

There is mounting evidence, in my opinion, that in fact most of our streams originate from discrete breakup events, rather than long episodes of gradual water vapor outgassing. Jenniskens and Lyytinen (2005) demonstrated that 2003 WY25 can be a fragments of an 1819 (or shortly before) breakup of D/1819 W1 (Blanpain) and that the dust of such a breakup would have evolved into Earth’s path to create the 1956 Phoenicids. 2003 WY25 has since been found to be weakly active at perihelion (Jewitt 2006). Watanabe et al. (2006), too, recognized that the 1956 Phoenicids could have been the product of a breakup in 1819.

After this, it was found that 2002 EX12 moves among the alpha-Capricornids (Wiegert and Brown 2005b; Jenniskens 2006). This minor planet is now better known by the name 169P/NEAT, after it was found that the object was weakly active at perihelion (Jäger and Hale 2005), after (!) we associated the object with the alpha-Capricornids. Again, the comet was only weakly active at perihelion, not active enough to account for the massive stream. The proposed shower formation age (∼AD 10) is still very uncertain.

The best documented case of comet fragmentation is that of the Sungrazers. Among the various sungrazer comet groups are the Marsden and Kracht Sungrazers which move in prograde orbits. These are small comet fragments that are detected only because they pass close to the Sun during perihelion, at which time they brighten from backscattered sunlight and pass the field of view of spaceborne Sun observatories. They are observed too briefly for a good orbit determination. Seargent (2002) first recognized the similarity in orbital elements between the Daytime Arietids and Marsden sungrazers and when it was recognized that some sungrazer comets return on a short-period comet orbit, the orbit could be greatly improved and the association was established. Sekanina and Chodas (2005) have argued that the Daytime Arietids and delta-Aquariids were created after 1059 AD. Shortly before that time, the Machholz family progenitor broke and a train of comet fragments had a close encounter with Jupiter in 1059 AD, which accelerated its evolution along the nutation cycle. The meteoroid streams were created by subsequent disintegration of some of these fragments.

There is, however, an interesting discrepancy in orbital period. Most Machholz family objects (9P/Machholz 1, Marsden and Kracht Sungrazers, Delta Aquariids, even 2003 EH1 and the Quadrantids) have a semi-major axis of about 3.1 AU. The Daytime Arietids have a semi-major axis of only 1.5 AU, half this value (Campbell-Brown 2005). The reason for this discrepancy may hold clues to understanding why some of our meteor streams have a relatively short semi-major axis. Perhaps the progenitor had a close encounter with Earth before (or during?) breakup.

Further evidence of frequent comet disintegrations has come from the confirmation that, in addition to 2P/Encke, there are other comet fragments among the Taurid showers (Table 2), several objects now being discovered that are a much better match to the Taurid showers than any of the objects proposed before (Porubçan et al. 2005; Jenniskens 2006).

Table 2 (Mostly) dormant Jupiter family comets and their established meteor showers
Full size table

Finally, Ohtsuka et al. (2005) recognized that 2005 UD moves among the Daytime Sextantids. Indeed, 2005 UD and 3200 Phaethon appear to have originated from a common ancestor, with 2005 UD over time evolving into an orbit not unlike that of Phaethon today. Recently, Jewitt and Hsieh (2006) found that 2005 UD is smaller than Phaethon (1.3 ± 0.1 km), but has the same bluish color, albedo = 0.11, and similar rotation period (5.249 h), consistent with both objects originating from one parent object. The Geminids are thought to have originated from Phaethon (or more precisely from a parent body that left Phaethon and the Geminids as products) at about 1030 AD (Jenniskens 2006). The whole complex of comet fragments broke at an earlier time.

The type of disintegration is not unlike that of the recent 1995 breakup of 73P/Schwassmann-Wachmann 3, which will cause a shower of tau-Herculids in 2022 (Lüthen et al. 2001). About as much mass is released in the form of dust and small fragments than the remaining mass (Table 1). That said, the mechanism of fragmentation may well be very different.

Few of these associations have been studied in detail thus far, but those that have point at a recent formation history of our meteor showers. All strong showers identified containing now dormant or weakly active comets have fragmented in the last 2,000 years (Table 1).

5 Streams from (Weakly) Active Jupiter Family Comets: Also From Disintegration?

I would like to add here that comet disintegration may even play a role in creating meteoroid streams from active Jupiter Family Comets, perhaps dominating the mass loss from the normal water vapor outgassing as envisioned by Whipple (1951). From the streams listed in Table 3 (in that sequence), the situation is as follows:

Table 3 (Weakly active) Jupiter family comets and their meteor showers
Full size table

Comet 26P/Grigg-Skjellerup was visited by Giotto after its Halley flyby and scattered light was observed from a cloud of particles, which suggested to McBride et al. (1997) that a larger fragment had come off.

Comet 21P/Giaconini-Zinner is an active comet and the Draconids are thought to have originated from normal comet outgassing. However, the meteor magnitude distribution in the stream is high, indicative of agressive disintegration of the dust after ejection. This implies evaporation from a water-rich layer, possibly freshly exposed in a comet breakup.

The Andromedids of comet 3D/Biela were created in a breakup in 1842/43 AD. In a recent paper, Jenniskens and Vaubaillon (2007) investigated the cause of the 1872 and 1885 Andromedid storms and concluded that the dust encountered was that generated during the continued fragmentation in the 1846 and 1852 returns. The mass generated in the 1842/43 breakup did not meet with Earth orbit. Normal activity from prior years did not result in meteor showers, but there were no favorable dust trail crossings.

Sykes and Walker (1992) found that the IRAS dust trail of comet 7P/Pons-Winnecke could have been created in about 1 orbit in normal comet activity and it is not clear if this activity is normal. The comet is known for periodic outbursts of June Bootids from orphan trails, now much different from that of the comet orbit.

D/1978 R1 (Haneda-Campos) is now lost and probably also a dormant comet. It was responsible for a brief, but strong, shower in October, the #233 October Capricornids, with considerable mass (Wood 1988). The comet itself was only seen in 1978 and has not yet reappeared as a dormant comet.

6 Not So Well Established Associations

Given that all strong streams seem to have such remnant fragments, it is likely that many more associations will be recognized. Until now, however, a very small number have been associated with meteoroid streams, which is somewhat surprising given the number of known NEO. As of January 1, 2007, 701 Near-Earth Asteroids greater than 1 km in size have been identified as well as 64 Near Earth Comets (JPL website). In total, about 4407 NEOs have been discovered, 822 of which are potentially hazardous. This number is expected to increase in the near future. The estimated population larger than 1 km is about 1,100, while the population greater than 140 m in size is about 100,000 objects.

Reason for that low number may be the difficulty of establishing an association with a NEO in a low-inclination orbit of moderate eccentricity. The likelihood of chance associations increases dramatically with the increase in the population density of NEO in a, e, i space. The problem with lower inclination streams is that many more potential candidate parent bodies exist, as demonstrated by many proposed identifications before 2003 that did not pan out.

Ways to decrease the likelihood of chance association are: 1) to better describe the meteoroid stream so that dynamical studies are possible that trace the stream back to its point of origin; and 2) to discover that the proposed parent body has features expected for a (mostly) dormant comet nucleus, such as weak activity at perihelion, a dark nucleus (e.g., A’Hearn 1985; Dandy et al. 2003; Binzel and Lupishko 2006), or be dynamically related to Jupiter Family Comets (e.g., Bottke et al. 2002).

Fortunately, we can now make a strong argument that the association of parent bodies and their streams ought to be relatively tight. In search of other such associations, we are looking for (remnants of) parent bodies that can have created meteoroid streams within the past nutation cycle of the secular orbital evolution (one rotation of the nodal line relative to the line of apsides, which takes typically less than 4000 years). Some of these objects can now pass more than 0.2 AU from Earth’s orbit. 2003 EH1, for example, passes at 0.21 AU due to periodic perturbations by Jupiter at aphelion. In the same way, 2002 EX12 does not pass close to Earth’s orbit, but is found along the evolving orbit of the alpha-Capricornid shower, just slightly further along the nutation cycle than the meteoroids we recognize at Earth.

Table 4 lists associations made in this manner by searching for theoretical radiants close to those of observed meteor showers, and then testing how dissimilar the orbits are relative to those expected from the nutation cycle. The table gives the Tisserand parameter with respect to Jupiter and a dissimilarity criterion derived from invariants of secular perturbation (DB).

Table 4 Less certain associations with not so well established showers
Full size table

Obrubov (1991) and Babadzhanov (1989) have first used invariants in secular perturbation theory derived by Lidov (1961, 1962) to search for asteroid–comet associations along the nutation cycle. Their first invariant is derived from the constant energy and momentum (related to the Tisserand parameter with respect to Jupiter). Their second invariant is a consequence of a perturbing function in the elliptical twice-averaged three-body problem being constant, as derived by Lidov (1961, 1962). Their third invariant is that of the longitude of perihelion, which moves much slower than the nodal line. In a recent paper, Jenniskens (2007) defined a dissimilarity criterion (DB) based on these invariants. Table 4 (update from those given in Jenniskens 2006, until January 1, 2007) is ordered according to this dissimilarity criterion. Associations with DB < 1.0 are thought to be siblings. Associations with DB = 1.0–1.5 are thought to be aunts and uncles, like the Machholz family comet showers. Hence, the most likely associations are those with DB < 1.0. The listed associations are less likely going down the list. Jenniskens (2007) also gives other criteria to evaluate the likelihood of association, including one proportional to the population density of NEO in a, e, i space.

7 Rapid Evolution of Jupiter Family Comets

Decoupling occurs due to a series of encounters with Earth and Venus. According to Wetherill (1991), shortly after decoupling, Jupiter Family comets have orbits with a = 2.1–2.5 AU, Q < 4.35 AU, and q mostly just outside Earth’s orbit. Those with q < 1 AU are distributed mostly between 0.5 and 0.9 AU. Many of our potential parent bodies are objects with semi-major axis above a = 2.5 AU. These are Jupiter Family Comets that are not yet fully decoupled from Jupiter.

The typical lifetime for decoupling is 100,000 to 1,000,000 years (Wetherill 1991). This is much longer than the lifetime of an active comet (∼12,000 years according to Levison and Duncan 1997) and the typical nutation cycle (∼4,000 years). Hence, most not-yet decoupled Jupiter-family comets are expected to be dormant. This is consistent with finding many dormant objects in this transition regime.

Our lists include 24 candidate dormant (or weakly active) comets that appear to be Jupiter Family comets (TJ = 2–3). This is a significant fraction of all such objects, estimated at 123 ± 41 by Binzel and Lupishko (2006). If all these objects are confirmed parent bodies, then this would imply that these objects break on a time scale equal or less that of the rate of meteoroid streams evolving into Earth orbit (<2,000 years).

In each fragmentation, about half of the mass of the comet is lost in the form of meteoroids (Table 1). A typical comet would evolve from a diameter of D = 3 km to D < 0.5 km in only eight disruptions. For the lifetime of Jupiter family comets of 12,000 years, this would imply a period between disruptions of about 1500 years, in good agreement. If the comet would stay at its most active over that period of time, the same amount of mass would be lost, but that is clearly not the case. Hence, fragmentation is the main mass loss mechanism for dormant comets in the inner solar system.

Already 20 years ago, Kresák and Kresákova (1987) concluded that the visible release of dust from Jupiter Family Comets was insufficient to maintain the zodiacal cloud in equilibrium. They were first to suggest that “the progressive decay of the dark matter, including extinct cometary nuclei, their fragments, and products of asteroidal collisions, represents the dominant source of replenishment of the interplanetary dust complex.”

8 Origin of Phaethon

This young dynamical lifetime of Jupiter Family Comets creates a problem in explaining the origin of 3200 Phaethon and 2005 UD, which currently move in an asteroid-like orbit (TJ >> 3). Close encounters with the terrestial planets are needed, but those are infrequent. It is possible that their predecessor originated from among the most primitive asteroids in the (outer) asteroid belt.

As noticed in the past, quite a number of our meteor showers have a small semi-major axis, and may be related to such outer belt comet-asteroid transition objects (Table 4). However, many of these streams are in doubt and need to be established first.

9 Further Work

Before any of the associations listed in Table 4 can be established, we need to be certain that proposed streams exist, as in being streams of meteoroids from the same parent body. Observational programs are encouraged that can help confirm the existence of the streams.

To help confirm the association of a NEO with a given meteoroid stream, dynamical studies are needed that trace the meteoroid stream and the proposed parent body to the time of fragmentation. The dispersion of dust reveals the age of a stream. Orbits more precise than those derived by radar are needed for such studies. This calls for a significant push to better characterize the known meteoroid streams by photographic, digital CCD, and intensified video techniques.

The likelihood of an association can also be increased from studies of the proposed parent body. Table 4 can serve as a list of priority for taxonomic studies of NEOs, to address some of the other criteria that could help identify dormant comets (Binzel and Lupisho 2006): they should have low geometric albedo (<0.075), taxonomic classes D, P, or C (possibly F or B), and rotation rates lower on average than the mean rate of asteroidal NEOs.

As a more general course of action, the fragmentation mechanisms needs to be better understood (Hughes 1990; Gronkowski 2007). There may be more than one. The streams themselves may give information about their cause. In the case of the Andromedids, for example, most mass in the resulting meteoroid stream is in the form of small particles, presumably because rapid evaporation of residual ices in the comet boulders broke the larger meteoroids. This could also be why few meteoroid streams are known for the prevalence of boulders. In the case of comet 2P/Encke, on the other hand, the gas drag limit would predict no larger than ∼kilogram sized meteoroids, but the largest observed Taurids are at least two orders of magnitude in mass bigger. These meteoroids could originate from parts of the comet that had already lost much of their volatiles. Other such cases might be found in a search for meteoroid stream association in the population of tens to hundreds of kilogram objects.