Overview of Water-Ice in Asteroids—Targets of a Revolution by LSST and JWST

Abstract

Water-ice occurs inside many minor bodies almost throughout the Solar System. To have an overview of the inventory of water-ice in asteroids, beside the general characteristics of their activity, examples are presented with details, including the Hilda zone and among the Trojans. There might be several extinct comets among the asteroids with only internal ice content, demonstrating the complex evolution of such bodies. To evaluate the formation of ice-hosting small objects, their migration and retention capacity by a surface covering dust layer are also overviewed to provide a complex picture of volatile occurrences. This review aims to support further work and search for sublimation-induced activity of asteroids by future missions and telescopic surveys. Based on the observed and hypothesized occurrence and characteristics of icy asteroids, future observation-related estimations were made regarding the low limiting magnitude future survey of LSST/Vera Rubin and also the infrared ice identification by the James Webb space telescope. According to these estimations, there is a high probability of mapping the distribution of ice in the asteroid belt over the next decade.

1. Introduction

In recent years, it has become evident that asteroids cancontain water-ice (see, e.g., [1,2,3,4]). This topic is increasingly important as new study opportunities have opened up with the powerful new James Webb Space Telescope (JWST) and the upcoming Vera Rubin C. Observatory (formerly known as the Large Synoptic Survey Telescope—LSST). We review the occurrence of water-ice in small bodies located in a broad region of the inner Solar System that encompasses the near-Earth environment, the classical Asteroid Belt, and the orbital region of Jupiter. Then, we discuss observation ideas using the capabilities of the JWST and LSST/Vera Rubin telescopes.

The theoretical borderline between asteroids and cometary nuclei has become more ill-defined recently (see Figure 1). This work focuses on “classical” asteroids according to the Tisserand criteria with the exception of Ceres. While Ceres might be unique in several respects among the nearby asteroids, including its origin, it has been discussed in detail in several recent works [5,6]. In theory, the occurrence of ${\mathrm{H}}_2$O increases with the solar distance due to condensation during planet formation [7]. However, some observations indicate that the situation is more complex, and evidence from meteorites also indicates liquid water-related chemical alteration in asteroid interiors from co-accreted primordial ice [8]. On the other hand, there is no strong consensus on the observable signatures of former ice in meteorite parent bodies, and in particular, the possible existence of current ice based on meteorites. This review also provides information on these research topics.

Small icy objects hold clues about the formation of the Solar System and the delivery of water to the Earth. As the origin of life is closely related to water, the analysis of icy objects provides relevant information for astrobiology research. In the last decade, several such objects have been observed. Although originally classified as asteroids, they were later found to exhibit cometary activity or to host water-ice. Therefore, while they should remain categorized as asteroids, their activity and its underlying causes warrant further investigation. This review provides an inventory of such objects with specific information on their water-ice-related aspects in order to summarize for further research water-ice occurrence around and inward of the orbit of Jupiter. The motivation of this work is to collect the related observations and arguments in order to provide a wide overview of this topic, especially estimations of future observations by LSST/Vera Rubin and JWST with the expected discovery of a large number of active objects, along with their spectral properties.

1.1. Active Ice-Bearing Small Bodies

Direct observation of ice signatures on the surface of a small body is possible both with in situ space missions, as recently illustrated by the Rosetta rendezvous mission with ${\mathrm{H}}_2\mathrm{O}$ and ${\mathrm{CO}}_2$ ice found on the nucleus of comet 67P/Churyumov–Gerasimenko [10,11,12], and by remote sensing via spectroscopic detection of the water-ice signature, as in the case of Themis (see [13,14]) and Cybele [15]. The (water-ice-driven) activity of small bodies is only possible when the bodies are close enough to the Sun to drive water-ice sublimation (∼3 au), and the ice is close enough to the surface to be within the diurnal or orbital thermal wavefront.

The discoveries in the last two to three decades have broadened the classification of small bodies. The formerly used definitions of comets were provided by [16], who discussed the processes that transform comets into asteroids. Later, the discovery of asteroids with a comet-like appearance initiated a dispute about this definition [17,18,19]. The essential principle is that volatiles are needed to produce classical cometary activity on asteroids ([19,20]; see also a review by [21]) presenting repeated volatile-driven activity.

Some asteroids exhibiting transient comet-like comae and/or tails might be considered active asteroids or dual-class objects, i.e., comets/asteroids (see also [9,19,21,22,23,24,25]). In addition tocometary activity by sublimation, there are numerous possible further mechanisms for driving mass loss in microgravity [9]. For instance, P/2010 A2 has been observed to display dust activity (Figure 2), where the slow evolution of the debris without observable gas content is best explained by a collision with ejecta subsequently influenced by radiation pressure, resulting in a complex structure compatible with a collisional origin.

Solar heating of a cometary surface supplies the energy required to sustain gaseous activity, which, in turn, drives dust removal. In this dynamic environment, both the coma and the nucleus undergo physical and compositional changes throughout the orbit. Ref. [26] reported observations of two opposite seasonal color cycles in the coma and on the surface of comet 67P/Churyumov–Gerasimenko during its perihelion passage, attributing these variations to seasonal effects on sublimation and dust release activity.

The sublimation cycle represents the fraction of time during which the body remains active [27]. Sublimation plays a fundamental role, as the release of gas and dust is driven by the sublimation of ice. The duration of this active phase (that is, the duty cycle or sublimation cycle) depends on the lifetime of the asteroid (≤10⁻²) and/or the age of its collisional asteroid family. For example, the Lixiaohua family—which includes 313P/Gibbs—is a member (≤10⁻³).

Observations of comet 67P/Churyumov–Gerasimenko have revealed that its activity exhibits diurnal variations in intensity, driven by changing insolation conditions. Ref. [28] reported surface water-ice detection that appears and disappears in a cyclic pattern, closely following local illumination conditions, providing a source of localized activity. The resulting water cycle appears to play an important role in the evolution of the comet, leading to cyclical modifications in the relative abundance of surface water-ice.

1.2. Currently Used Groups for Ice-Rich Bodies in the Solar System

Based on various evidence, including the observations outlined in this work, the following minor body groups are frequently used in the literature. These are not well-established separate groups as there are overlaps, problems, and sometimes even contradictions:

• Active asteroids or activated asteroids are small bodies that have asteroidal orbits but a comet-like visual appearance [9,25]. Mass-loss activity can be driven by ice sublimation or external mechanisms, for example, activation by impact (exposing buried ice for sublimation [17,18,20,21,25,29]), by material shedding due to rapid rotation [9,25], or even disintegration (e.g., [9,25]), but not as a repeated activity, although thermal fracture can also cause mass shedding somewhat periodically, as proposed for Bennu [30,31].
• Main-belt comets (MBCs, [17,18]) are active minor bodies on asteroidal orbits within the main belt, with repeated ice-sublimation-driven activity at heliocentric distances that are consistent with water-ice sublimation. There are also active or activated asteroids in the near-Earth space (NEOs); hence, “active asteroid” might be a more general term than the previously used “main-belt comet (MBC)”. In certain cases, episodic impacts can produce ejecta and related coma-like features, but such a dust coma or tail is short-lived—thus, they should be separated from MBCs. Such collisions are implicated in the asteroids (596) Scheila and, perhaps, P/2010 A2 (LINEAR), and indirectly as a trigger for activity (through exposing buried ice), but not as repeatedly as in the cases of 133P and 238P.
• Icy objects have a dominant mass fraction of ice, which produces strong activity closer to the Sun. This group includes Jupiter family comets and other short-period comets, long-period comets, Trojan objects, Centaurs, and Trans-Neptunian Objects (TNOs), which are usually outside the orbit of Jupiter. These objects might show activity around the perihelion.

“Classical” comets among NEOs are activated by the sublimation of water-ice. This produces a persistent coma of gas and dust surrounding the nucleus, along with the development of typical gas and dust tails [16]. The real comet classification can be confirmed by tracing its orbital evolution from its original source region (e.g., ecliptic comets or Jupiter-family comets + 2P/Encke or Oort cloud) to the near-Earth region. In contrast, main-belt comets are not considered classical comets due to differences in their structure and material composition. If an MBC’s orbit evolved into the near-Earth region, it cannot be regarded as a classical, primordial comet from the outer Solar System. Nevertheless, a native MBC can exhibit a permanent coma similar to that of a classical comet, which complicates efforts to distinguish between native asteroid-comets and classical comets. Classical comets are presumed to be primordial bodies that formed farther from the Sun than native MBCs, although they could have undergone certain alteration effects. Though only minimal alteration can occur at large solar distances, rare and low velocity collisions, especially in the early periods of the Solar System, might have some effects [32]. Limited early radiogenic [33] or serpentinization-generated heat [34] could also influence interiors, while long-term exposure of galactic cosmic rays could exert a slight influence on the surface of these objects [35].

An example of an activated NEA is (3200) Phaethon, often referred to as a “rock comet” due to its very small perihelion distance of 0.13 AU. The Sun’s heat causes fractures similar to mudcracks in a dry lake bed [36]. In very rare cases, recent collisions among NEAs can also produce irregular dust and debris clouds. Such events—like those observed for P/2010 A2 and (596) Schelia in the main belt—serve as analogs for this type of activation mechanism and suggest that similar processes may occasionally occur among NEAs.

This work is structured as follows: The specific aspects of interesting example objects (Section 2.1), including Trojan objects (Section 2.3), and in the near-Earth Zone (Section 2.4), are presented in detail. We discuss the origin of water content in our celestial bodies of interest in Section 3, and finally, outline the next steps in their exploration in Section 4, to reach conclusions on specific suggestions and expectations from the LSST/Vera Rubin (Section 4.1) and JWST (Section 4.3). Although the aim is a comparative presentation and evaluation of various ice-hosting asteroids, a perfectly homogeneous overview was not possible as the observations—and occasionally the model calculations—are relevant only for specific parts of the orbit, depending on when and where the object was located during the observed or modeled period. However, these data still contribute to forming a complete picture of the activity of these bodies. While the signatures of past ice on or within these bodies are also important, we focus here on currently active objects and the present occurrence and sublimation of ice.

2. Icy Objects in the Main Belt

Below, we address the detailed observations related to possible ice occurrence and activation mechanisms producing mass loss in MBCs to provide specific insight into their behavior. We chose to focus on MBCs (see the specific objects listed in Table 1) because their periodic behavior is the strongest evidence for water-ice presence in them, while maintaining that other active asteroids may not necessarily contain ice and are shedding mass through processes that do not require ice. The presence of water-ice was identified using the broad absorption bands at 1.5 and $2.05 \mu$m [37] and also by the $3 \mu$m band absorption features [38], while the rarity of the $1.65 \mu$m band indicates mainly amorphous and not crystalline ice. High-resolution spectra of the comet 8P/Tuttle obtained by the United Kingdom Infrared Telescope in the range $0.34 \mu$m [39] showed several ${\mathrm{H}}_2\mathrm{O}$ features previously unseen in cometary spectra. Water emission at $2.7 \mu$m was also identified from the 238P/Read main-belt comet [40] and from the 358P/Panstarrs (P/2012 T1) active asteroid [41]. In certain cases, sublimation may only be one possible explanation and there are other mechanisms that could also contribute. Based on Pan-STARRS1 discovery statistics, of the order of ∼50–150 currently active MBCs comparable in brightness to the known MBCs are expected to exist [42]. Their activity strength depends on their orbital position at the time of observation (Figure 3).

There are several natural processes that can cause asteroids to lose mass. For example, the sublimation of volatile materials is one such mechanism. When ice-rich asteroids heat up, the sublimation of ice in the near-surface leads to comet-like activity, ejecting gas and dust into space [43]. A different path can be through tidal interactions. During close encounters with a planet or with the Sun, the intense gravitational pull can strip away some material or even fully disrupt the asteroid [44]. Rotational spin-up due to solar radiation (YORP effect) can, on longer timescales, increase the spin of the asteroid to critical speeds that can lead to centrifugal mass shedding or even a complete breakup [45]. Furthermore, collisions at high velocities are also a primary mechanism through which asteroids can lose mass. An energetic impact can even catastrophically fragment an asteroid, shattering the parent body and dispersing debris that may eventually re-accumulate into smaller bodies [46].

Table 1. Summary of the discussed target observations in the inner Solar System.

Body or Group	Location/Family	Ice Presence	Additional Notes	References
133P/Elst-Pizarro	Main-Belt comet	Inferred from recurrent activity	-	[20]
238P/Read	Main-Belt comet	Inferred from recurrent activity	Exhibits the strongest water ice sublimation among the repeatedly active objects in the main belt	[43,47,48]
313P/Gibbs	Main-Belt comet	Inferred from recurrent activity	-	[27]
Vesta	Main-Belt asteroid	Inferred from radar analysis and high hydrogen content	Visited by NASA Dawn mission	[49,50]
Themis	Outer Main Belt	Inferred from ground-based spectroscopy	-	[9,13,14]
Antiope	Outer Main Belt (Themis family)	Inferred from ground-based spectroscopy	-	[51]
Cybele	Main Belt	Inferred from ground-based spectroscopy	Representative of the Cybeles asteroid family, which exhibits similar properties	[9,52]
Quasi-Hilda comets	3:2 Orbital resonance zone with Jupiter	Inferred from observable activity including outbursts	-	[53]
Jovian Trojans	Co-orbiting resonance with Jupiter	Undetected or unconfirmed	-	-
Near-Earth Objects	Near-Earth environment within 1.3 AU at perihelion	Inferred from ground-based spectroscopy, dedicated missions and returned samples	Notable bodies explored include 107P/Wilson–Harrington, Don Quixote, Phaethon, Bennu and Ryugu	[54,55,56]

Table 1. Summary of the discussed target observations in the inner Solar System.

Body or Group	Location/Family	Ice Presence	Additional Notes	References
133P/Elst-Pizarro	Main-Belt comet	Inferred from recurrent activity	-	[20]
238P/Read	Main-Belt comet	Inferred from recurrent activity	Exhibits the strongest water ice sublimation among the repeatedly active objects in the main belt	[43,47,48]
313P/Gibbs	Main-Belt comet	Inferred from recurrent activity	-	[27]
Vesta	Main-Belt asteroid	Inferred from radar analysis and high hydrogen content	Visited by NASA Dawn mission	[49,50]
Themis	Outer Main Belt	Inferred from ground-based spectroscopy	-	[9,13,14]
Antiope	Outer Main Belt (Themis family)	Inferred from ground-based spectroscopy	-	[51]
Cybele	Main Belt	Inferred from ground-based spectroscopy	Representative of the Cybeles asteroid family, which exhibits similar properties	[9,52]
Quasi-Hilda comets	3:2 Orbital resonance zone with Jupiter	Inferred from observable activity including outbursts	-	[53]
Jovian Trojans	Co-orbiting resonance with Jupiter	Undetected or unconfirmed	-	-
Near-Earth Objects	Near-Earth environment within 1.3 AU at perihelion	Inferred from ground-based spectroscopy, dedicated missions and returned samples	Notable bodies explored include 107P/Wilson–Harrington, Don Quixote, Phaethon, Bennu and Ryugu	[54,55,56]

2.1. Population Statistics of Main-Belt Comets

The inventory of the MBC population has been steadily increasing and as they are only briefly active (few months of a 5–6 year long orbit), there are many more bodies yet to be discovered [21]. At the time of writing there are 18 active asteroids: 14 MBCs, one active dwarf planet (Ceres—not discussed here), and three active NEOs. There are eight MBCs that exhibit ice sublimation-related outgassing and mass-loss activity (see Table 3 of [25]). Below, we discuss in more detail a few of the prominent examples.

2.1.1. 133P/Elst-Pizarro

The main-belt asteroid (7968) Elst-Pizarro or main-belt comet 133P/Elst-Pizarro (133P) resides in the Themis dynamical family of the outer regions of the main belt; in 1996, a cometary tail clearly showed around the perihelion (Figure 4).

Ref. [20] estimated from the recurring activity that a trail production rate of about ${10}^{-3}$ to ${10}^{-2}$ kg ${\mathrm{s}}^{-1}$ would require areas of exposed ice ranging from a few tens to a few hundred square meters (Figure 5). The slow ejection of the particles suggests a weak gas flow, as would be expected from water-ice sublimating from beneath a thermally insulating mantle or from an active area exposed in a tilted orientation to the Sun. Ref. [20] also estimated that the ice sublimation rate from the surface at 160 K is ${10}^{-7}$ kg ${\mathrm{m}}^{-2}{\mathrm{s}}^{-1}$ and the corresponding recession rate of a sublimating ice surface is ${10}^{-10}$ m ${\mathrm{s}}^{-1}$, assuming 1000 kg ${\mathrm{m}}^{-3}$ as the bulk density. At this recession rate, a dark sphere of ice with a radius of 2 km could survive for only ∼${10}^6$ yr in the orbit of 133P.

Ref. [20] proposed two scenarios for the origin of 133P:

• EP could be a Jupiter-family comet (JFC) that has evolved into an asteroid-like orbit via planetary gravitational scattering and non-gravitational forces, meaning its membership in the Themis family would necessarily be simply coincidental. However, the dynamical path from a cometary orbit to an asteroidal orbit is not well established, due to the general lack of on non-gravitational forces in most current dynamical models.
• Alternatively, 133P could be a true Themis family member whose subsurface ice has been recently exposed.

2.1.2. 238P/Read

Object 238P could be considered as an Encke-type comet, and it also exhibits the strongest water-ice sublimation among the repeatedly active objects in the main belt. It shows a photometric excess implying the presence of a coma [58] and exhibited repeated activity near its perihelion in 2005 and 2010 [47]. A coma dust mass of the order of ${10}^5$ kg and production rates estimated near ∼0.1 kg ${\mathrm{s}}^{-1}$ are suggested [25]. The authors of [40] calculated a lower production rate of 0.30 ± 0.03 kg ${\mathrm{s}}^{-1}$ from JWST measurements acquired in 2022, near its expected peak brightness. The total dust mass of 238P/Read is around the same magnitude as that of the majority of known MBCs [48]. This C-type object is estimated to have a 0.4 km equivalent diameter, assuming an albedo of 0.05 [59]. According to its dynamical lifetime, 238P is dynamically unstable with a survival time of the order of 20 Myr, and is hypothesized to have diffused in eccentricity from its original location within the Themis family due to its proximity to the 2:1 mean-motion resonance with Jupiter [60].

Ref. [40] presented James Webb Space Telescope observations revealing that MBC 238P/Read exhibits a coma of water vapor, but lacks a significant ${\mathrm{CO}}_2$ gas coma. Water emission at $2.7 \mu$m was also identified [40]. Their findings demonstrate that the activity of comet Read is driven by water-ice sublimation and implies that MBCs are fundamentally different from the general cometary population. Regardless of whether comet 238P/Read formed under different conditions or experienced a unique evolutionary path, it is unlikely to be a recent interloper from the outer Solar System into the asteroid belt.

2.1.3. 288P/2006 VW139

Object 288P/2006 VW139 is a C-type [61] binary asteroid with cometary-like activity identified by Pan-STARRS images in November 2011, which led to its cometary designation (Figure 6). Considering it as a comet, it is the first cometary object of a binary system, possibly formed by the YORP effect [62]. In December of 2011, the Hubble Space Telescope also confirmed its activity around perihelion, with models indicating dust ejection speeds of 0.06–0.3 m ${\mathrm{s}}^{-1}$, and particle sizes around 10–300 μm [61]. The dust production might be concentrated around the B component [63] and could be produced both by ice sublimation and centrifugal forces.

2.1.4. 313P/Gibbs

Discovered in 2014, this MBC is located near at least eight other active asteroids in the outer belt in terms of orbital elements. It displayed a fan-shaped dust tail in both 2003 and 2014 that is well approximated by syndyne dust emission models (assuming the motion of cometary dust particles is controlled only by solar gravity and solar radiation pressure) [27,64], indicating that the dust was ejected over a period of at least 3 months. From HST observations, ref. [27] found that the mass loss peaked near 1 kg ${\mathrm{s}}^{-1}$ in October 2014 and then declined over the subsequent months by about a factor of five, at a nearly constant heliocentric distance (Figure 7). Ref. [27] also showed that dust was ejected continuously and not impulsively, and that the effective particle size was large, ∼$50 \mu$m, with an ejection speed of ∼2.5 m ${\mathrm{s}}^{-1}$.

The total dust mass ejected was $\sim {10}^7$ kg, corresponding to $\sim {10}^{-5}$ of the nucleus mass. The observations are consistent with sublimation occurring from a partially restricted area on the surface of a $\sim {10}^4 {\mathrm{m}}^2$ ice-covered area, corresponding to ∼0.2% of the nucleus surface. For ice to survive in 313P on billion-year timescales, it is required that the cycle for sublimation be ≤10⁻³. The sublimation cycle is the duty cycle, i.e., the fraction of time for which the body is active [27]. The role of the sublimation is fundamental because the gas and dust release are driven by sublimation of the ice. The fraction of the time of activity (duty cycle or sublimation cycle) depends on the lifetime of the asteroid (≤10⁻²) and/or the age of the collisional asteroid family, e.g., the Lixiaohua family, of which 313P/Gibbs is a member (≤10⁻³).

Ref. [27] estimated the nucleus equivalent radius to be 700 ± 100 m (assuming a geometric albedo of 0.05). Both 313P and MBC P/2012 T1 are associated with the ∼150 Myr-old Lixiaohua asteroid family—presumably their surfaces were exposed after collision and/or breakup during the formation of this family [65]. Their young age of ∼150 Myr with respect to the ice survival time means that these asteroids may contain enough water ice to support their sublimation activity.

2.2. Non-MBCs with Inferred Ice Content

2.2.1. Possible Ice on (4) Vesta

Object (4) Vesta is one of the largest objects in the asteroid belt. From 2011 to 2012 it was explored by NASA’s Dawn orbital mission, and was expected to be a very dry asteroid. When the probe moved out of the line of sight of our planet, its radio waves bounced off the asteroid surface (grazing incidence) before being received on Earth and the analysis of the scattering behavior by [50] led to mapping of the surface at the scale of a few centimeters to tens of centimeters. The results suggest that unlike the Moon, Vesta’s surface roughness variations cannot be explained by cratering processes only. In particular, the occurrence of increased hydrogen concentrations within large smoother terrains (over hundreds of square kilometers) suggests that potential ground-ice presence may have contributed to the formation of Vesta’s current surface texture (Figure 8).

Ref. [50] observations are also consistent with geomorphological evidence of transient water flow of liquid water after buried deposits of ice were melted by the heat of the impacts [66]. However, dry mass movements cannot be excluded [67]. If ice exists on Vesta, it is probably accumulated and retained in the polar craters, possibly in a manner similar to trapped ${\mathrm{H}}_2$O ice in permanently shadowed regions (PSRs) on Mercury and the Moon. However, it is also possible that the reason for the observation is not ice but material from the infall of carbonaceous meteorites [68]. Refs. [69,70] presented the detection of hydration material on Vesta in the Dawn mission data. Alternative mechanisms to explain the water on Vesta propose delivery via secular impact processes [71,72]. Supporting this hypothesis, refs. [73,74] have shown the presence of trace volatile material in the meteorites of Vesta, which have been modeled as the result of cometary bombardment during the formation of Jupiter [75,76].

Figure 8. Candidate locations of hydrated surface units on Vesta. Dashed yellow lines indicate the two main hydrated areas. The map was obtained by an infrared mapping spectrometer of the DAWN mission in August 2011, from an altitude of 1700 miles (2700 km) (Image source: NASA/JPL-Caltech/UCLA/INAF, also see [70]).

Figure 8. Candidate locations of hydrated surface units on Vesta. Dashed yellow lines indicate the two main hydrated areas. The map was obtained by an infrared mapping spectrometer of the DAWN mission in August 2011, from an altitude of 1700 miles (2700 km) (Image source: NASA/JPL-Caltech/UCLA/INAF, also see [70]).

2.2.2. (24) Themis

Object (24) Themis is a 100 km diameter asteroid, and the parent body of the Themis collisional asteroid family in the outer region of the main belt. Its infrared spectra displayed spectroscopic absorption features consistent with the presence of water-ice at $3.1 \mu$m and organics at $3.3 \mu$m and $3.6 \mu$m [13,14]. The upper limit of the water production rate was estimated to be 4.5 × ${10}^{27}$ mol ${\mathrm{s}}^{-1}$ from the observations [77]. Ref. [14] considered possible sources of surface ice on (24) Themis, and found stable ice may exist in the subsurface, possibly exposed by impact gardening, sublimation or both. The sublimation activity might deposit a thin layer of frost on the surface.

Daily or seasonal thermal pulses reaching subsurface ice could produce sublimation, and as water vapor encounters a cold surface at night around aphelion, frost would form. Current sublimation activity may be a non-equilibrium process because any ice that might have originally formed within reach of the daily or seasonal thermal pulse skin depth should have already sublimated [78]. Regarding the extent of the ice-covered surface area on (24) Themis, ref. [9] found that if ice is present at its surface, it should be relatively clean and confined to a limited spatial area, as confirmed by [77].

A number of members of the Themis family show evidence for hydration. The binary asteroid (90) Antiope might have some surface water-ice [51] based on the 3 micron absorption. The bowl-shaped absorption features are consistent with those from water ice, as in the spectrum of (24) Themis. Observations of the spectra of Themis asteroids are consistent with the carbonaceous chondritic meteorites with different degrees of aqueous alteration (see [21] and references therein), indicating that the large parent body could be made up of significant amounts of water-ice that differentiated or maintained layers of pristine/unheated material [79].

Recent systematic observational surveys [80] and dynamical analyses [81] suggest that some JFCs may have originated in the main asteroid belt, rather than in the outer Solar System. This possibility is especially compelling given the presence of icy main-belt objects in the Themis asteroid family. However, as noted in the earlier discussion of NEO comets, Themis-family MBCs are not classical, primordial comets from the Kuiper belt or the Oort cloud.

2.2.3. (65) Cybele

The asteroid (65) Cybele is the largest member of the Cybele asteroid group (its members are also known as the “Cybeles”). The infrared spectrum of (65) Cybele shows that it is a primitive asteroid, and has an absorption feature that is similar to those of (24) Themis that are explained by the presence of water-ice. The asteroid may be covered by a layer of fine silicate dust mixed with small amounts of water-ice and organic solids [52].

Upper limits on water production rates were derived from optical spectra of (24) Themis and (65) Cybele [82]. No emission lines—expected from resonance fluorescence in gas sublimated from the ice—were detected. For both asteroids, the derived upper limits for water production are ≤400 kg ${\mathrm{s}}^{-1}$ (5$\sigma$) assuming a cometary ${\mathrm{H}}_2$O/CN ratio to calculate water production rate from the spectra [52]. Ref. [82] ruled out models in which a large fraction of the surface is covered by high-albedo (“fresh”) water-ice, as the measured albedos of both bodies are low (∼0.05–0.07). Models assuming widespread low-albedo (“dirty”) ice were also excluded, as such water-ice would be warm enough to sublimate strongly, producing gaseous products that should have been detectable in the spectra. Therefore, if ice exists on these bodies, it must be relatively clean (albedo 0.3) and confined to a fraction of the Earth-facing surface ≤ 10%.

2.2.4. Quasi-Hilda Comets

The Hilda asteroids are a dynamical population, orbiting the Sun at an average distance of ∼4 AU in a 3:2 mean-motion resonance with Jupiter [83,84]. Due to their semi-major axes just beyond the traditional boundary of the classical main asteroid belt, they are often grouped with the outer main belt (e.g., [85]). Some objects in the Hilda group also show comet-like activity, suggesting a link to the trans-Neptunian region or outer Solar System material (e.g., [86,87]). A quasi-Hilda comet (QHC) is a Jupiter-family comet that interacts strongly with Jupiter and can undergo extended temporary capture by the planet. These comets are associated with the Hilda asteroid zone [53,88]. Asteroids in this zone typically have semi-major axes between 3.70 and 4.20 AU, eccentricities below 0.30, and inclinations of maximum ${20}^{\circ }$ [53]. Comets can be temporarily perturbed into this group and then later perturbed back out again [89]. Notably, 8% of the comets that escape the 3:2 resonance ultimately collide with Jupiter [53].

The quasi-Hilda comets (QHC) could be Hilda asteroids that escaped. Although the primary source region of ecliptic comets is the Kuiper belt, a small fraction of them are icy Hilda asteroids from the Hilda zone. To date, there is no spectroscopic evidence of surface ice on either Hilda asteroids or QHCs, but ice sublimation activity and observable outbursts in several such bodies suggest the presence of ice on them. For example, the presence of ice in P/2010 H2 (Vales) was reported by [90], suggesting that a higher surface albedo may be appropriate for this object. Object P/2010 H2 (Vales) is a quasi-Hilda object that exhibited a dramatic outburst in April, 2010 brightening by over 7.5 magnitudes [91]. While its orbital elements put it into the space of the Hilda group, its smaller semi-major axis of 3.85 AU, high inclination of 14.3°, and its past close encounters with Jupiter mark it to be more consistent with temporarily captured Jupiter-family comets [53,92]. Ref. [93] concluded that the observed high particle ejection speeds (reaching up to 210 m ${\mathrm{s}}^{-1}$) are consistent with gas-drag acceleration from sublimating ice. In contrast, they are incompatible with other proposed driving mechanisms, such as rotational instability, thermal or desiccation stress fracture, and electrostatic repulsion, all of which were consequently ruled out. Although asteroid impacts can produce high-speed ejecta, this explanation is considered unlikely for quasi-Hilda comets due to their small population size and short dynamical lifetime. An updated inventory of QHCs is available in [53].

Active asteroids can be driven by several low-energy processes beyond sublimation gas drag and impacts [93]. These include rotational disruption of the parent body, cracking due to thermal expansion and/or desiccation stresses, and electrostatic ejection of fine dust. However, none of these processes can adequately explain the high speeds attained by the ejecta observed in P/Vales. For instance, in the case of rotational disruption, the released material typically escapes with a velocity comparable to the body’s equatorial rotation speed, which is generally ≤1 m ${\mathrm{s}}^{-1}$ at breakup for a kilometer-sized object. The nucleus of P/Vales is likely a small kilometer or sub-kilometer-sized body, as suggested by its faintness. The typical size and rotation periods of small quasi-Hilda asteroids or comets were determined by [94], with diameters ranging from 0.73 km to 10.43 km, and rotational periods from 1.633 h to 11.163 h. The small diameter and slow rotation are not able to assure the activity and ejected material speed.

The possible sources of activity of such icy objects between 2 and 5 AU of heliocentric distances were analyzed by [95]. These authors concluded that their observed dust activity can be connected to water-ice sublimation (MBCs) or crystallization of amorphous water ice (QHCs) as well as external causes, such as impacts or orbital temperature changes (see [9,21,30]). Ref. [96] concluded that systematic differences exist in the physical parameters of MBCs and QHCs: MBCs show smaller values of Af$\varrho$ (usually less than 10 cm). In addition, these measurements seem to confirm that the sizes of MBC nuclei are smaller than those of QHCs and JFCs.

2.3. Potential Ice in the Jovian Trojans

Jupiter Trojans are dark, spectroscopically reddish bodies, usually with featureless spectra [97]. Their densities range between 0.8 and 2.5 g ${\mathrm{cm}}^{-3}$, as seen in, for example, Paroclus [98,99], (65489) Ceto/Phorcys [100], and (3548) Eurybates [101], with the lower values possibly indicating the presence of water-ice inside. No ice has been spectroscopically detected on the Trojans [22,102,103], which is not surprising given that the surface temperatures (∼150 K) cause moderately rapid loss of exposed ice by sublimation [104]. However, compared to the Jupiter Trojans, (24) Themis and (65) Cybele have even higher surface temperatures, indicating that 150 K does not necessarily exclude ice at this range of solar distances. Jovian Trojans are a large asteroid group co-orbiting the Sun with Jupiter locked in a 1:1 mean-motion resonance, librating around the Sun–Jupiter L4 and L5 Lagrange points in long-term stable orbits. They are believed to have been populated in the early Solar System, for example, via capture during the formation and migration of Jupiter [104,105,106].

Thermal modeling by [107] demonstrated that water-ice can be stable for the lifetime of the solar system if covered by ∼10 m of dust at the equator, or ∼10 cm near the poles. In addition, ref. [107] also concluded that Trojans might have volatile-rich interiors and volatile-poor surfaces. Model-fitting of the near-infrared spectroscopic observations of Trojans has been used to put an upper limit of water-ice on their surfaces (e.g., [108,109] and references therein). Ref. [110] used JWST/NIRSpec reflectance spectra to analyze the NIR reflectance of five Jupiter Trojans and found a broad OH absorption feature centered around $3 \mu$m, that is more comparable to the characteristics of water-ice-rich Centaurs, KBOs, and the asteroid 238P. This feature might be due to small patches of water-ice. Although not considered as classical asteroids, it is worth mentioning that surveying the color distribution of Neptune’s Trojans (much farther than Jupiter’s Trojans), there are some that resemble ultra-red objects, as in the Kuiper-belt. Ref. [111] found that based on JWST NIRSpec measurements, their targets were consistent with a thin ${\mathrm{H}}_2$O ice and ${\mathrm{CO}}_2$ ice layer coverage.

Centaurs are small Solar System bodies orbiting between Jupiter and Neptune, typically with perihelions just beyond Jupiter (q > 5.2 AU) and a semi-major axis around 30 AU [112]. They are believed to be a population likely originating from the Kuiper Belt, transitioning from trans-Neptunian reservoirs into the inner parts of the Solar System, with most of them eventually evolving into short-period comets [113]. KBOs (Kuiper-Belt Objects) are small, icy bodies with orbits beyond Neptune, roughly between 30–50 AU from the Sun. They are thought to be largely pristine remnants from the accretion disk of the Sun, preserving evidence of the processes that took place during the formation of the early Solar System [114,115].

2.4. Potential Ice in Near-Earth Asteroids

The first NEO that exhibited cometary activity was 107P/Wilson–Harrington (hereafter W-H) initially discovered in 1949 as a comet with a coma and a tail (Figure 9). In 1979, it was rediscovered as an Apollo orbit-class Mars-crosser asteroid 1979 VA. Multiple observations over the years suggest that this object is a largely inactive comet that underwent an outburst in 1949. Ref. [54] calculated that the plausible maximum volatile release during outburst ${\mathrm{H}}_2$O and CO production rates are nearly the same: 5 × ${10}^{27}$ molecules ${\mathrm{s}}^{-1}$. NEO (4015) 1979 VA, originally discovered by [116] and later rediscovered in 1992 by [117] on a prediscovery Palomar Observatory Sky Survey (POSS) plate, was identified as the same object as the active comet 107P/1949 W1 (Wilson–Harrington). The history of both successful and unsuccessful attempts to detect activity in 107P is detailed in [54].

NEO (3552) Don Quixote, discovered in 1983, has an exceptionally eccentric orbit and is classified as an object of the Amor group, a Mars-crosser, and a Jupiter-crosser, as well as a centaur, and an extinct comet. Thermal modeling of combined photometric data on the NEO (3552) Don Quixote reveals a diameter of ${18.4}_{-0.4}^{+0.3}$ km and an albedo of $0.{03}_{-0.01}^{+0.02}$, confirming Don Quixote as the third-largest known NEO [55]. Based on thermal-infrared observations made with the Spitzer Space Telescope at the 3.6 and $4.5 \mu$m bands, ref. [118] discovered cometary activity in Don Quixote with a coma and a tail in the $4.5 \mu$m but not in the $3.6 \mu$m band. This is consistent with molecular band emission from ${\mathrm{CO}}_2$. [118] derived an upper limit on the dust production rate of 1.9 kg ${\mathrm{s}}^{-1}$ and a ${\mathrm{CO}}_2$ gas production rate of (1.1 ± 0.1) × ${10}^{26}$ molecules ${\mathrm{s}}^{-1}$. Ref. [55] also derived an upper limit on the dust production rate of 1.9 kg ${\mathrm{s}}^{-1}$ and reported a ${\mathrm{CO}}_2$ gas production rate of (1.1 ± 0.1) ×${10}^{26}$ molecules ${\mathrm{s}}^{-1}$. Spectroscopic observations with the Spitzer Infrared Spectrograph indicate the presence of fine-grained silicates—possibly pyroxene-rich—on the surface of Don Quixote. These findings suggest that ${\mathrm{CO}}_2$ can be present in near-Earth space over long timescales and may explain why Don Quixote’s cometary nature remained undetected for nearly three decades.

Apollo asteroid (3200) Phaethon is an active asteroid in an orbit that brings it closer to the Sun at a perihelion distance of 0.14 AU. With its size of 5.8 km (geometric albedo 0.10) and volume equivalent diameter of 5.4 km ([25,119] and JPL Small-Body Database), it is among the largest known NEOs. It is considered as the parent body of the Geminids meteor shower by cometary activity, although other scenarios have also been proposed to explain the formation of the meteor stream: around perihelion extreme subsolar surface temperatures (∼1000 K) could be reached, triggering thermal fracture and dehydration of water-bearing minerals (see [9], and references therein).

Asteroid (101955) Bennu was analyzed in detail by the OSIRIS-REx NASA mission, which observed dust grains being ejected from the surface [31]; however, such activity cannot produce a comet-like coma. For the activity thermal fatigue, thermal cracking or desiccation cracking are realistic reasons for the ejection process, but still ice sublimation could not be ruled out (see e.g., [30,31,120]). Thermal modeling suggests that based on temperature analysis, the occurrence of ice in Bennu’s polar region is possible if buried in the top meter(s) [121]. However, since grain ejection occurs mainly from low latitude locations, it may not come from the sublimation of ice.

2.5. Dark Comets

Dark comets are small bodies that show no visible coma but exhibit significant non-gravitational accelerations consistent with outgassing of volatiles. These objects represent a potentially widespread class of small bodies that further populate the continuum between asteroids and comets, with their active nature inferred from their orbital motion [122]. They are analogous to the first interstellar object, 1I/’Oumuamua, which was discovered in 2017 as it passed through the inner Solar System, and was the first recognized member of this class. Its speed and path around the Sun do not match those of typical asteroids, yet it lacks a bright tail or nucleus (icy core) typically associated with comets. However, ’Oumuamua exhibited erratic motions that are consistent with gas escaping from its surface.

The inventory of the dark comet object class began to grow when ref. [123] reported statistically significant detections of non-radial, non-gravitational accelerations based on astrometric data for the photometrically inactive NEOs 1998 KY26, 2005 VL1, 2016 NJ33, 2010 VL65, 2016 RH120, and 2010 RF12. Following these discoveries, about a dozen additional dark comets were identified. The magnitudes of the non-gravitational accelerations are greater than those typically caused by the Yarkovsky effect, and there are no radiation-based, non-radial effects that can account for such large values. The Yarkovsky effect is the force acting on rotating asteroids produced by the anisotropic momentum transfer produced by the thermal emission of photons. While the solar irradiation is mainly received around local noon, the thermal re-radiation mainly happens during local afternoon–evening. The effect influences the orbital characteristics mainly of asteroids smaller than about 10 km diameter in the long term. Therefore, we hypothesize that these accelerations are driven by outgassing, and calculate implied ${\mathrm{H}}_2$O production rates for each object. The apparent lack of visible comae or photometric activity is attributed to the absence of surface dust and the low levels of gas production.

3. Origin of Ice in the Inner Solar System Small Bodies

Ice in small bodies can either be primordially accreted (depending on the conditions during accretion), subsequently deposited, e.g., delivered by impact events on icy bodies, stored later in shadowed locations similar to ice in the polar regions of the Mercury and Moon (see, for example, [124,125,126,127]), or originated from a water cycle on the object (similar to what is seen on 67P [26,28]). The surface temperature and interconnected porosity are the main factors affecting long-term ice stability, along with the internal temperature, which is influenced by orbital evolution, radioactive decay, and exothermic reactions such as serpentinization [34]. Based on the modeling of [128], the orbital configuration, thermal inertia, dynamical age, and rotational axis tilt [129] also influence ice retention in the shallow subsurface of asteroids.

The surface temperatures experienced today for most main-belt asteroids are generally too high to allow ice to remain stable across their surfaces. The majority of ice loss occurs long before an asteroid reaches an NEA- or JFC-like orbit. Ref. [78] modeled the temperature distribution for asteroids as a function of rotation rate, obliquity, semi-major axis and thermal inertia, finding that beyond roughly 3.1–3.2 AU, ice could remain stable in the shallow subsurface at low latitudes, where it could plausibly be liberated by impacts. This modeling is consistent with the observations of activity by MBCs driven by sublimation of newly excavated ice. Ref. [128] showed that the lowest amount of ice loss occurs in the polar regions of bodies with small axis tilt, which remain so cold that ice does not retreat beyond the influence of the seasonal thermal wave (often less than 5 m) by the time the objects have reached NEA- or JFC-like orbits. As sublimation continues, an ice-free lag deposit forms, further damping the sublimation process. It is not known whether outgassing is occurring today on objects other than the MBCs. However, observations of Ceres by [130] indicate OH emission from its sunlit pole, but as previously noted, Ceres is not the focus of this overview.

Migration of various icy bodies substantially influenced their distribution in the Solar System and is probably a fundamental process of ice enrichment in the inner Solar System. Icy bodies are thought to have been migrated or been implanted from the trans-Neptunian region to the inner Solar System [131]. The unstable 2:1 mean-motion resonance location just around the outskirts of the Themis zone can dynamically alter the orbits of main-belt asteroids to JFC ones [81,132]. Refs. [81,132] suggested that some members of the Jupiter-family comets may originate from the main belt, escaping it within 100 Myr. In the “Nice model” [133,134], Trans-Neptunian Objects (TNOs) are injected into the outer regions of the main belt [135]. Most of them are dynamically lost or collisionally destroyed and their remnants may be D-type asteroids. Similarly, in the “Grand Tack” model [136], both rocky and icy small bodies are scattered throughout the Solar System. There are other complementary and even contradictory alternatives for the water and volatile transport in the Solar System. Observational data do not favor either model as of present; for example, the lunar cratering suggests a steady decline in impacts rather than a late heavy bombardment according to some studies [137], and some evidence derived from Vesta observations [138]. Several studies offer alternative scenarios proposing different mechanisms for giant planet formation and their role in transporting volatiles inward, e.g., [75,139,140,141,142]. Additionally, ref. [143] argues that the Grand Tack may more likely disrupt the Solar System than result in its current form. The early Solar System’s architecture can be explained without the Nice or Grand Tack models by classical dynamical mechanisms. A massive primordial asteroid belt was likely cleared by Mars-sized embryos and resonance effects [144,145,146], simultaneously scattering water-rich asteroids inward to hydrate Earth, with comets contributing little [147,148]. Later bombardments, including the LHB, can arise from secular resonance sweeping during nebula dispersal or slow giant planet migration [149,150], destabilization of an inner E-belt [151], or the chaos of a lost fifth terrestrial planet [152]. Finally, Mars’s small mass and the belt’s low density can be explained by a truncated inner disk [153], showing that even without major planetary migration, gravitational interactions and early disk structure can reproduce the Solar System’s key features. The remixing of the asteroids following the migration of giant planets seems to happen naturally, not just in the case of our Solar System. For example, ref. [154] showed that the formation of giant planets in HD 163296 excited planetesimals (most from beyond the CO snowline), scattering them both inward, where they enrich forming terrestrial and giant planets with volatiles, and outward into interstellar space.

Before Jupiter’s migrations, the asteroids had varied composition with their distance from the Sun. Rocky asteroids dominated the inner region, while more primitive and icy asteroids dominated the outer region. Ref. [135] made simulations of dynamical evolution of trans-Neptunian primordial small bodies in order to determine the contamination of the main asteroid belt. They have shown that the violent dynamical evolution of the giant-planet orbits required by the Nice model [133,134,155] leads to the insertion of primitive Trans-Neptunian Objects into the outer belt. They found that a substantial population of Trojans (32 of the 230 captured objects, or 13%) and Hilda asteroids (8%) are produced. In addition, they found captured comets in orbits in the main asteroid belt with semi-major axes as low as 2.68 AU. Ref. [60] performed numerical simulations of the dynamics of MBCs, and showed that they were formed in situ as the remnants of the breakup of large icy asteroids. They have shown that among scattered MBCs, approximately 20% reach the region of terrestrial planets, where they might have contributed to the accumulation of water on Earth. Their simulations also shown that collisions among MBCs and small objects could have contributed to triggering cometary activity of these bodies. Ref. [136] reported their results of early Solar System simulation that show how the inward migration of Jupiter to 1.5 AU and its subsequent outward migration lead to the planetesimal disk truncated at 1 AU; the terrestrial planets then formed from this disk over the next 30–50 million years. Their simulations have shown that S-type objects dominate in the inner belt, while C-type objects dominate in the outer belt. They have also shown that C-type material entered the terrestrial region, possibly contributing to the water on Earth. Recent simulations from [156] suggested that the dynamical pathways that inserted material from beyond Jupiter into the main asteroid belt may continue to be active today, while those sending debris from the terrestrial territory into the belt may be currently inactive or significantly weaker. The asteroid belt is capable of supplying most of the observed NEOs, with particular zones providing them via powerful and diffusive resonances [157]. Although asteroids and meteoroids dominate the NEO population, comets are also expected to be important contributors to the overall NEO population.

Incorporation of ice into asteroids around their formation is possible under favorable nebular conditions, if surface shielding of deeper ice is possible [60], and these objects became transferred inside the belt by collisional fragmentation of earlier objects. The visible spectra of asteroid–comet transition objects are compatible with the idea of origin from the JFCs, as was presented by [15]. They concluded that only the activated near-Earth Object (3200) Phaeton can be related to comets but the investigated main-belt comet 133P is unlikely to have a cometary origin—however, sodium fluorescence and anti-sunward tail were recently suggested to influence perihelion brightness [158]. Ref. [43] claimed that there is a potential contamination of the main belt by Jupiter-family comets. They investigated the orbital evolution of objects that are close to the orbital boundary between asteroids and comets (their Tisserand parameter with respect to Jupiter is nearly 3). Due to gravitational interactions with terrestrial planets and temporary trapping by mean-motion resonances with Jupiter, the fraction of the Jupiter-family comets evolved into main-belt-like orbits on million-year timescales could be as large as ∼0.1–1%. However, most such main-belt captures would be transient, and long-term stable orbits with both small eccentricities and inclinations should be rare.

Ref. [159] found that non-gravitational effects on 324P/La Sagra are large, and [160] concluded that this may be because it is an active MBC. Otherwise, the non-gravitational effects of in situ-formed icy main-belt asteroids are significantly smaller than those of icy asteroids of JFC origin [14,159]. Ref. [14] concluded that the ice in MBCs might be primordial and would have been protected from solar insolation due to the formation of a poorly conducting surface layer [78]. Discoveries by [161] provided more evidence that the asteroid main belt hosts a population of bodies that were formed in the outskirts of the Solar System, namely, the main-belt asteroids (203) Pompeja and (269) Justitia originating from the TNO region. Although, the surfaces of these two asteroids do not show evidence for ice, the results by [161] further support transport of TNO ice to the inner Solar System.

Jupiter Trojans may contribute to the comet population through dynamical instabilities and collisional ejection [162]. Once removed from the vicinity of the Lagrangian L4 and L5 points, they quickly lose dynamical traces of their origin [162]. A further source might be Neptune Trojans (if they exist), as these are unstable and a dynamical route delivers them as centaurs [163] and could provide JFCs (ecliptic comets) ([164] and references therein). Ref. [164] showed that their contribution to that population could be of the order of ∼3%. Finally, ref. [165] suggests that there is the possibility of large TNOs being injected into the outer main belt.

Based on the comparison of the items listed in Table 1, in most cases (for most objects), the activity was identified as ejected material. This might be, but is not necessarily connected to water-ice sublimation, while even fewer spectral-based ice identifications have occurred. Keeping in mind the limited capability of optical telescopes, more active objects are expected to exist, and (excluding the well-monitored ones) more active periods could also occur at the known targets. For spectral identification, because of its difficulty, many more objects with potentially observable water-ice signatures are waiting for discovery.

Meteorites might also provide information on the water-ice content of their parent asteroids, despite not containing liquid water. Although water-ice will likely never be found in fallen meteorites, there are various clues indicative of previous ${\mathrm{H}}_2$O content of asteroids. The ${\mathrm{H}}_2$O content of meteorites in various forms is around a few %, though in some cases up to 15%, mainly in the form of OH of clay minerals. Important exceptions besides some Martian meteorites are the CI, CM, and CR groups with a very small amount occasionally in CV and CO meteorites [166], in the asteroid belt from accreted local water. These most-abundant carrier minerals are OH-containing phyllosilicates, phosphates, and hydroxides, which occur preferentially inside primitive chondrites, and probably formed by early aqueous alteration [127]. However, many parent bodies might have co-accreted ice without substantial melting. Among them, CI, CR and CM chondrites are the richest in such minerals, which formed by extensive early aqueous alteration [167]. CM chondrites, especially, contain hydrated minerals like serpentine or carbonates.

Another, recently identified form of ${\mathrm{H}}_2$O is nanoscale water-fluid inclusions in carbonaceous chondrites [168]. While water could be present in bound form at salt crystals also, the separation of originally bound ${\mathrm{H}}_2$O molecules from those that became bound after the fall of the meteorite is very difficult. However, evaluating the hydrogen and oxygen isotopic content differences between the given NaCl crystal and the rest of the meteorite indicates it might not have been originally accompanied with the given meteorite [169]. Cubic salt (NaCl) crystals were also found, indicating aqueous alteration, with the salt holding liquid brine pockets. The halite is radioactively dated to 4.7 ± 0.2 Gyr, and might be related to the MBCs [19,170]. Organic compounds found in the millimeter-sized halite crystals containing brine inclusions in the Zag and Monahans meteorites indicate that the asteroidal parent body where the halite precipitated potentially was asteroid 1 Ceres with complex biologically and prebiologically relevant molecules [171]. There might also be indirect evidence of formerly embedded ice grains in meteorites: in the Acfer 094 meteorite, microporosity was observed as irregularities in the very-fine-grained matrix, which might have been left behind by water-ice grains that later sublimated away. This material formed probably beyond the snow line in the protoplanetary disk as fluffy ice and dust aggregates [172]. There are a range of water content-related observations from meteorites; however, due to cosmic weathering-related spectral changes, it is difficult to link these meteorite data to source asteroids [173,174]. However, solid-state spectroscopic features of hydrated minerals have been found in 10% or more of asteroids beyond 3 AU [19].

3.1. Dynamical Stability of Orbits of Active Asteroids

Planet migrations during the early stages of Solar System evolution, as described by the Grand Tack and Nice models, could have resulted in the emplacement of icy outer Solar System objects into the main belt [134,135,136]. In addition, the non-gravitational forces owing to outgassing and the Yarkovsky effect can change the orbit of active asteroids in the long term. Ref. [159] systematically examined the non-gravitational effects of MBCs. The authors report statistically significant detections of non-gravitational effects for 313P/Gibbs and 324P/La Sagra. The non-gravitational effects—including the Yarkovsky effect—can generate migration of inactive and active asteroids in the main belt.

Studying the dynamical stability and evolution of active asteroids—including MBCs—is a key component to identify their likely source regions in the Solar System. Most MBCs have been found to be mostly stable over ${10}^8$ years or more, suggesting that they formed in situ where they can be found today (see review by [21], and references therein). These results were corroborated by [43], who studied synthetic test particles rather than real objects, concluding that objects on orbits with both low eccentricities and low inclinations are unlikely to have been recently implanted outer Solar System objects. Since the activity duration of MBCs lasts a few months in a 5–6 years orbital period (there are objects with recurrent activity), there is a chance to discover new active asteroids with intrinsic activity (not activated by current collisions). Based on the overview of [21], a serious challenge is the direct detection of escaping ${\mathrm{H}}_2\mathrm{O}$ from MBCs. Even the ice can occur in smaller (sub-kilometer)-sized bodies (e.g., 147P/Kushida–Muramatsu in the Hilda-zone, [88,175]); therefore, future sky surveys, for example, LSST/Vera Rubin, will be able to identify small active asteroids and expand their inventory.

3.2. Role of Ice in Orbital Evolution

Beside the “classical” approach of ice-related cometary activity that change the orbital characteristics by impulse moment transfer during the ejection by so-called non-gravitational forces, two further aspects should be considered. The ice on the surface influences the thermal properties through its albedo; thus, with elevated albedo, less solar irradiation is adsorbed and subsequently re-radiated, which may weaken the Yarkowski and YORP effects [176]. However, the situation is more complex and based on [177], the anisotropic reflection of the optical illumination is partly compensated by the anisotropic re-radiation in the thermal range.

Subsurface ice might also influence the emergence and rate of the Yarkowski and YORP effects mainly by two processes: depending on its porosity (and related thermal conductivity together with thermal capacity), ice below any dust layer dampens the penetration of the vertical thermal wave. In a case where bulk and low-porosity ice exists below the surface dust, the daytime warming and nighttime re-radiation is slower and the re-radiation is extended temporarily, resulting in a “less sharp” orbital impulse moment change with respect to the spatial direction. Although porous ice is expected to exist inside asteroids (and especially comet-like objects, [178]), regular temperature cycling [179,180] or other processes like crystallization of amorphous water-ice [181] or sintering [182] might decrease the porosity in theory; however, no such change in related conditions has been well observed yet. Increased thermal inertia might indicate decreased subsurface porosity and elevated thermal capacity—this twirl thermal behavior of asteroid surfaces should be surveyed and further integrated to orbital evolution.

3.3. History of Geological and Chemical Changes

In the case of small bodies, low gravity does not allow a permanent atmosphere to develop, thereby minimizing the drivers of surface change. However, under the right external triggers, small icy bodies may undergo chemical changes and surface modification [11,19,183,184]. This is most evident in the case of comets. Most comets originate in the outer Solar System, particularly from the Oort Cloud and the scattered disk of the TNO population. As they undergo gravitational interactions with planets and other Solar System bodies, their orbits evolve (see Section 4), and may occasionally drift into the inner Solar System. If their adjusted orbits bring them within the “snow line” (∼5–6 AU for the emergence of water-ice-containing comas), particularly around the perihelion phase of their elliptical orbits, cometary surfaces are subjected to enough heat from the Sun to drive volatile sublimations and the onset of activity, which is a fundamental trait of comets [16].

A smaller, but yet significant, dynamical evolution path may occur in comets as they become active due to induced torques, especially if the body is small enough and the spin axis is favorable for such changes by non-gravitational forces, e.g., gas and dust ejection. Comet 67P/Churyumov–Gerasimenko underwent changes in its spin period by around ∼20 min during the lifetime of the Rosetta mission, which could have accounted for tectonic changes in its neck region [185]. Furthermore, extensive modeling work by [186] has shown that activity-induced torques can be a fundamental factor in the splitting and reconfiguration of bi-lobed or multi-lobed small bodies.

During the mission of Rosetta to comet 67P/Churyumov–Gerasimenko, several different types of surface changes associated with activity were observed, including cliff collapses, local erosion, dust and gas ejection, transport and re-deposition of materials, aeolian-like processes, and features associated with the transient coma that develops during the activity, in addition to other surface changes (e.g., [185,187,188,189]). Comets may also experience activity driven by volatiles other than water-ice (e.g., CO and ${\mathrm{CO}}_2$ ice) at longer distances from the Sun because of the lower sublimation temperatures of these “hypervolatiles”. In the longer term, this leads to chemical alterations of the primordial volatile inventory in these icy bodies.

In summary, icy bodies may undergo extensive physical and chemical changes under the right conditions. Specific surface changes are also expected in the case of ice-containing asteroids, and based on the examples listed in this work, there is a need to explore their surface characteristics related to activity. Based on our review, the outskirts of the main belt (Hilda zone and its outer boundary) should contain a significant amount of water-ice, waiting for further in situ analysis.

4. Next Steps for Icy Asteroid Survey

Below, we list suggestions and expected outcomes for future asteroid surveys and how targeted observations could provide much more information to understand the distribution and thus the origin of ice among asteroids. Below, we summarize two topics:

• How the optical identification of comae is possible and expected to occur, primarily through LSST/Vera Rubin observations,
• How spectral water-ice identification is expected by the JWST (which already demonstrated the related capability [40])

Such observations support the planning of future Earth-based observations and space missions to small bodies, improving the understanding of ice distribution and evolution in MBCs and NEOs in the coming years (see review by [21]). The discovery rate of icy small bodies (MBCs, NEOs, and minor bodies and dwarf planets in the outer Solar System) is expected to increase substantially due to new and improved optical surveys, e.g., the Pan-STARRS telescopes [190], the Dark Energy Survey (DES) [191], the 4 m Blanco telescope [192], the NEO Surveyor [193], and the LSST/Vera Rubin.

Comparing the icy occurrence types in asteroids listed in this paper, it became clear that ice occurs at many locations where small bodies are present in the Solar System, especially including hidden shallow subsurface regions. The survey of their distribution requires observations of stochastic or recurrent activity of asteroids, which could be effectively performed using good temporal coverage by the next all-sky survey project. For example, LSST/Vera Rubin is expected to achieve approximately +22 magnitude/arcsecond limiting brightness depending on filters [194].

To obtain a relevant estimation of the ice (Figure 10) and related thermal properties of possible ice-hosting minor bodies, beside the solar distance, the albedo and thermal inertia are also needed. In Figure 10, two exponential curves show a decreasing sublimation rate with increasing solar distance toward the right. The ${\mathrm{T}}_u$ curve shows the estimated maximal interior temperature (calculated using 0.05 albedo and 95% emissivity), while the ${\mathrm{T}}_{eff}$ shows the effective surface temperature (in this case, the rotation is fast enough to inhibit temperature change with geographic longitude, e.g., local time, or has a hypothetical infinite thermal inertia). For the identification of water-ice, infrared spectral observations are needed by in situ mission or telescopic observations (see later in Section 4.3).

Using moderately low-cost CubeSats and off-the-shelf infrared detectors [196], missions could visit a few such asteroids that are active based on telescopic surveys. In these observational domains, the possibilities are worse than the LSST/Vera Rubin in the optical region as data should be recorded in the infrared—in an ideal case, the middle infrared—region, that is not realistic from the surface of the Earth. Considering that most icy asteroids orbit in the outer part of the main belt, their characteristic temperature could be best estimated in the range of 8–$10 \mu$ms approximately, according to Wien’s law [197]. However, thermal emission can be observed at longer wavelengths too, not only near the peak of the Planck function. For such observations, the currently available main instrument is the JWST, and no similar-category instrument is available for survey. Thus, to gain a better insight into the interpretation of optically identified possible ${\mathrm{H}}_2$O-related activities in the main belt, few instruments are available, and for <100 km sized objects, only large space telescope instruments will provide data with good quality—indicating an important gap to verify the existence of water-ice-induced activity regarding the thermal conditions.

4.1. Potential for LSST/Vera Rubin Observations

To obtain more relevant statistical results on the occurrence of water-ice-induced asteroid activity, a larger number of successful identifications is needed. Earlier surveys found 37 active asteroids to date [198]. After the discovery of the activity of 133P/Elst-Pizarro [20], new survey programs were started [17], including the SAMBA project, the SAFARI [199], and the Hawaii trails projects [200], leading to the discovery of 176P/LINEAR. Recently, [201,202] using the WF Camera of the 2.54 m Isaac Newton Telescope, selected 514 main-belt asteroids and dozens of them showed a coma.

The difficulty in identifying the coma by optical observations is its low surface brightness and small angular size compared to those of “regular” comets. PSF-fitting to the observed brightness distribution and searching for deviation between the observed and scaled model fit (e.g., [203]) has been used. Ref. [201] developed an automated pipeline to reduce and analyze the data using tail and coma detection methods adapted from [204]. Separation of diffuse coma from motion-produced trails was considered by [205]), to separate the flux of the nucleus and coma, while the method of [206] was applied to the nucleus fragments of D/Shoemaker–Levy 9, including, for example, for Nos. 11 and 15 (Table 1 of [206]) which were very faint and had the same 23.9 V magnitude. Using [207] with the WFPC2 of the HST, 0.0455 arcsec ${\mathrm{pixel}}^{-1}$ was achieved to “photometrically resolve” the nucleus of 147P/Kushida–Muramatsu (among others), an MBC in the Hilda-zone [88], in the presence of a surrounding coma.

While most discoveries of a coma have been determined by the 1.8 m Pan-STARRS1 and the UH 2.2 m telescope, for an even more effective survey, the 8.4 m LSST/Vera Rubin [208] should be used. The LSST/Vera Rubin is able to detect a faint coma with 14-times smaller surface brightness compared to the UH 2.2 m telescope, and 21-times fainter compared to the Pan-STARRS1 telescope. Using the planned 15 s long exposures in survey mode and the example case of 133P, the movement of the target during the exposure was approximately 0.25 arcsecond (0.016 arcsecond/second). The target’s own movement relative to the surrounding stars would not be a serious problem because of the large solar distance. In this case, aperture photometry can be useful; however, the PSF method is even better. The above argument suggests that the survey mode will be sufficient to identify an expected large number of active asteroids with LSST/Vera Rubin. In the Themis region, somewhat smaller, while in the main belt, somewhat larger smear is expected.

The first characteristic feature identification of cometary activity in MBCs was the dust tail (at a longer distance from the nucleus, it is called trail) of 133P. Ref. [20] measured the surface brightness starting from the near-nucleus dust coma along the dust tail (see Table 7, and Figures 8 and 9 of [20]), from the near-nucleus region to an 80 arcsecond distance projected to the sky. The surface brightness measurements for composite images of 133P dust trails are expressed as percentages of nuclear brightness per linear arcsecond. For example, on 19 August 2002, one surface brightness unit was equivalent to 25.1 R magnitude per linear arcsecond, and along the trail, it varied between <1 and 6% of nuclear brightness per linear arcsecond. On 19 August 2002, 133P was observed with the UH 2.2 m telescope and the effective exposure time was 2500 s. The 133P was at 2.050 AU geocentric distance; hence, at 80 arcseconds from the nucleus, the sky-projected trail length was approximately 119,000 km. For example, the total mass of the dust coma ∼1.7 × ${10}^5$ kg of the dust coma of 133P was observed with the UH 2.2 m telescope in August of 2002 [20] but the LSST/Vera Rubin was able to detect a dust coma with a total mass of 1.2 × ${10}^4$ kg. The ratio of the total flux in the dust trail to that from the 133P nucleus was ∼1.3 on 19 August 2002, which means that the total brightness of the dust trail was brighter with ∼0.3 R magnitude than the nucleus itself, i.e., the total surface brightness of the dust trail was 25.1–0.3 = 24.8 R magnitude per linear arcsecond. For LSST/Vera Rubin, the limiting magnitude is around 24, and allows a somewhat better coverage regarding space and time than PanSTARRS because of fast progress (detector readout, storing, and automatized data checking).

4.2. Upcoming Facilities for Asteroid Research

There are several approved observatories aimed to start in the next 5–10 years which will enhance asteroid science. The NEO (Near-Earth Object) Surveyor space telescope is set to launch no earlier than September, 2027, with the objective to detect comets and asteroids that may be potential hazards to Earth [209]. The ESA Flyeye 1 telescope, an automated wide-field survey to detect and track NEO asteroids and comets [210], has received first light in the summer of 2025, with plans already in motion to deploy Flyeye-2 in the southern hemisphere by 2028 [211]. The SiTian project, led by the Chinese Academy of Sciences and planned to begin operations by 2032, will monitor main-belt asteroids and NEOs and obtain their light curves [212], while the upcoming Xuntian, or CSST (Chinese Space Station Telescope), planned for launch in 2026, could also aid asteroid characterization [213,214]. The ngRADAR (Next-Generation Radar program), a planetary defense radar under development, will create an active radar system using the GBT (Green Bank Telescope) and will also have capabilities to study NEOs, comets, and asteroids. As new telescopes and continued advances in observational capabilities expand the capacity to study small solar system bodies, this field is expected to develop further, reinforcing the value of periodic reviews.

4.3. Potential for Infrared Spectral Ice Identification

For the spectral identification of water-ice on more (and fainter) objects, the JWST could be an ideal instrument, despite the moderately small observation time expected to be dedicated for such observation activity. The authors of [215] suggest that with the help of JWST exploration, the seasonal cycles or measuring temperatures at certain depths of the icy regolith should be of high interest. The most prominent OH absorption feature on asteroid surfaces is at ∼$3 \mu$m, caused by the O-H fundamental stretching mode. This wavelength regime is not available from the ground due to water vapor in the Earth’s atmosphere; however, the NIRSpec near-infrared spectrometer of the JWST is well suited for this band, and the JWST could be an ideal instrument for water-ice spectral identification. For example, the NIRSpec instrument and the CLEAR/PRISM disperser-filter or the G235M/F170LP plus G395M/F290LP grating-filter combinations could be used to search for ice features and/or hydrated minerals, depending on the brightness of the target. Furthermore, ref. [216] showed that by combining chemical modeling with laboratory IR spectra of interstellar ice analogs, the variations in ice composition and layering result in distinct signatures, providing a benchmark for future JWST observations. Ref. [217] observed five spinel-rich asteroids, revealing a distinct $2.85 \mu$m absorption feature, which is likely linked to hydrated material or space weathering. In the works of [218], JWST NIRSpec observations of Ceres, Pallas, and Hygiea revealed similar $2.72 \mu$m Mg-OH absorption features, indicative of aqueous alteration, with additional signs of ammoniated minerals on Ceres and Hygiea.

Considering an asteroid with a visible range brightness of ∼15 magnitude with an ∼100 s exposure time, we can achieve an S/N ratio of ∼130 according to JWST’s Exposure Time Calculator, which should be suitable for ice detection. Objects of this brightness are above about 100 km in diameter in the Trojan region. In the case of an ∼20 mag asteroid, a total exposure time of ∼1 h is required to reach S/N ∼50 at ∼$3 \mu$m using the same instrument. These allow the identification of ice (if exposed on the surface) to approximately the scale of one hundred cases (from the orbital distance of the Trojan asteroids).

Based on the MPCORB.DAT database file of the Minor Planet Center, currently, 12,053 Jupiter Trojans are known, of which 1299 are brighter than 20 mag, 434 than 19 mag, 179 than 18 mag, 59 than 17 mag, and 18 than 16 mag. Apparent brightness is estimated from the Hv absolute brightness of the targets, considering the heliocentric and observer distances and assuming that the observations are performed with the JWST in the allowed solar aspect angle range (85–135 degrees). The abovementioned aspects allow ice detection for an order of more objects than previously, clarifying the occurrence of water-ice in the Trojan region. Exposure times longer and/or objects brighter than the 1 h integration time for 20 mag asteroids grant S/N > 50, which is required to obtain better spectral resolution and allow the characterization of water-ice-related features in more detail—these longer exposures should target a properly selected sample of the many asteroids showing the ∼$3 \mu$m band.

Evaluating the capabilities of the above two large observatories, the expected increase in future-identified active asteroids and the more widespread spectral-based ice identification will allow a major leap to better understand the spatial distribution as well as the temporal occurrence of activities in the coming years. The distribution of ice in the belt will help to better constrain current and future formation models, and to understand the potential habitability-related occurrence and migration of ${\mathrm{H}}_2$O in the Solar System.