Classical comets are icy visitors from the outer solar system that develop distinctively "fuzzy" comae and tails when they approach the Sun (inside the orbit of Jupiter) and their ices become warm enough to sublimate, causing the release of gas and dust. In contrast, main-belt asteroids (which have never strayed outside the orbit of Jupiter since being formed) are classically thought of as inert, rocky bodies which may have contained ice in the past but have long since been boiled dry by 4.6 billion years of close solar heating.

This simple picture, however, has recently been complicated significantly with the discovery of the class of main-belt comets.

There are currently three known main-belt comets (MBCs): 133P/(7968) Elst-Pizarro (EP), P/2005 U1 (Read) (P/Read), and asteroid 118401 (1999 RE₇₀). Orbiting completely within the main asteroid belt, the MBCs present a distinct contrast with other periodic comets (the Jupiter-family and Halley-family comets, or JFCs and HFCs) which originate in the cold outer solar system in the Kuiper Belt or Oort Cloud and are later perturbed into highly eccentric orbits passing through the inner solar system where we observe them. Unlike the JFCs and HFCs, the MBCs appear to have formed in the much warmer inner solar system, where we see them today, and so likely contain ice that is quite different in chemical and isotopic composition from that in other comets.

MBC orbits; from http://neo.jpl.nasa.gov/orbits/

133P/Elst-Pizarro: Also asteroid (7968) Elst-Pizarro. Discovered as an inactive asteroid (1979 OW₇) on July 24, 1979. Discovered to be cometary on Aug 7, 1996 by Eric Elst and Guido Pizarro using the European Southern Observatory 1.0-meter telescope in La Silla, Chile. EP's unusual orbit in the main asteroid belt was recognized immediately, and due to the unexpectedness of cometary activity on main-belt asteroids, the cause of EP's activity was uncertain, with some believing EP's activity to be the result of dust kicked up by a series of impacts. After no reports of recurring activity between 1997 and 2001, impact-driven hypotheses were eliminated by the return of activity in 2002, discovered by me and David Jewitt on August 19, 2002, using the University of Hawaii 2.2-meter telescope on Mauna Kea. Still barely active on December 28, 2002, EP was inactive by September 22, 2003 when observed by the Keck 10-meter telescope, and has thus far remained inactive up until the present-day. [ more... ]

P/2005 U1 (Read): Discovered as an active comet on October 24, 2005, by Michael Read using the Spacewatch 0.9-meter telescope on Kitt Peak in Arizona and confirmed by the Spacewatch 1.8-meter telescope on Oct 25, 26, and 27, 2005. Followup observations also made by us on the UH 2.2-meter telescope on November 10 and 19-22, 2005, and on the Gemini 8-meter telescope on Mauna Kea on November 26, 2005.

118401 (1999 RE₇₀): Discovered as an inactive asteroid on September 7, 1999, by the LINEAR 1-meter telescopes in Socorro, New Mexico. Discovered to be cometary on November 26, 2005, by me and David Jewitt as part of the Hawaii Trails project using the Gemini 8-meter telescope on Mauna Kea and confirmed by the UH 2.2-meter telescope on December 24-27, 2005, and Gemini on December 29, 2005.

• MBC activity is unambiguously "cometary" in nature, that is, driven by the sublimation of volatile material (ice). Cometary activity for all three MBCs is observed to persist over at least a month or more, much longer than would be expected for dust thrown off by an impact into an inert surface (i.e. that of a non-icy asteroid). Also, from what we know about collision rates in the main asteroid belt, the probability of serendipitously observing the immediate aftermaths of collisions on three different objects (and two collisions on a single object) in the main asteroid belt in the span of only 9 years is implausibly miniscule. We thus conclude that MBC activity is not impact-driven. Impact-triggered maybe (see below). But sublimation-driven. Which means ice.
  
  
• The MBCs constitute a fundamentally new and distinct class of comets. According to currently available orbit evolution simulations, unlike all other known comets, the MBCs show no indication of having originated in the outer solar system in the Kuiper Belt or Oort Cloud. They instead formed where we see them, in the main asteroid belt, and are the only comets known to originate in the inner solar system and not simply be transferred here from the outer solar system. Objects that formed in different parts of the solar system (i.e. at different distances from the Sun) typically differ in composition, and this is likely the case for the MBCs as compared to other comets.
  
  
• The MBCs demonstrate that ice exists in small main-belt asteroids and may be quite common. Other than the MBCs, ice has only been reported on Ceres, the largest main-belt asteroid. Ceres's large size (~950 km across), however, means that deeply buried ice has a much better chance of surviving 4.6 billion years of close solar heating only ~3 AU from the Sun than ice on asteroids only 5 km across or less. The fact that three MBCs have now been discovered from relatively limited observational data indicates that there are probably many more waiting to be discovered.
  
  
• MBCs must have only "turned on" recently. Once they enter the inner solar system and begin outgassing for the first time, comets from the outer solar system typically only remain active for about 10,000 years before having most of their ice sublimated away and going dormant, or losing so much mass that they simply disintegrate. All indications are that the MBCs have occupied their current orbits in the inner solar system since the solar system's formation, 4.6 billion years ago. Had they been outgassing since then, they would have exhausted their ice supplies long ago and could not possibly be active now. Instead, they must have been dormant until very recently.
  
  
• MBC activity may be triggered by impacts (though subsequently sustained by the sublimation of volatile ice). Because comets typically become active when they experience dramatically increased solar heating (the result of travelling in orbits that pass from the outer solar system to the inner solar system and back again), and MBC orbits are comparatively much more circular (making solar heating effectively constant), another mechanism must be responsible for the activation of MBCs. We hypothesize this triggering mechanism to be excavating impacts from other asteroids that "dig up" buried (and thus preserved) ice, exposing it to the heat of the Sun.
  
  
• MBCs may have been a dominant source of terrestrial water and therefore a key to the rise of life on Earth. There are indications that the Earth's water was delivered by long-ago impacts from icy objects from the main asteroid belt or from the outer solar system. The isotopic composition of comet water does not appear to match that of ocean water, however, implying that the bulk of our water came from the main asteroid belt, i.e. from the long-lost siblings of today's MBCs. Further study of MBC ice, perhaps by visiting spacecraft, particularly focusing on isotopic analysis, would certainly help test this and give us further insight into how life on Earth began.

Images of known MBCs from UH 2.2-meter telescope data. All MBCs are clearly cometary, at least in the observational and physical sense (i.e., they appear "fuzzy" and all indications are that this fuzziness is caused by the ejection of dust by the sublimation of volatile material, most likely water ice).		Plot of semimajor axis versus eccentricity for all numbered asteroids (small black dots) and comets (large blue dots) tabulated by JPL as of December 14, 2005, and main-belt comets 133P/Elst-Pizarro, P/2005 U1 (Read), and 118401 (1999 RE70) (large red dots). Dynamically speaking, the orbits of MBCs clearly have much more in common with those of ordinary main-belt asteroids (the dense concentrations of black dots with semimajor axes of about 2.1 AU to 3.3 AU on this plot) than those of any other known comets.