Calcium-rich Gap Transients Overview

• Calcium-rich gap transients are faint, fast-evolving stellar explosions with intermediate luminosities, rapid nebular phase evolution, and spectra dominated by calcium emission.
• They are often found in old, quiescent galaxies at large galactocentric offsets, suggesting progenitor systems with significant kick velocities or delayed mergers.
• Their extreme calcium yields relative to ejecta mass provide key insights into nucleosynthesis and the enrichment of the intracluster medium, highlighting multiple progenitor channels.

Calcium-rich gap transients are a class of faint, fast-evolving stellar explosions defined observationally by luminosities intermediate between novae and supernovae, rapid evolution to the nebular phase, and nebular spectra dominated by calcium emission. In the empirical formulation introduced from Palomar Transient Factory discoveries, the class is characterized by peak luminosities in the “gap” between novae and supernovae, rapid photometric evolution, large photospheric velocities, early spectroscopic evolution into the nebular phase, and peculiar nebular spectra dominated by Calcium (Kasliwal et al., 2011). Subsequent work established that these events are also unusual in their environments: they are frequently found at large galactocentric offsets, often in old stellar populations, and often without any detectable underlying quiescent source at the explosion site (Lyman et al., 2014). The combination of Ca-dominated nucleosynthetic signatures, small ejecta masses, and extreme locations has made Ca-rich gap transients a central case study in the broader problem of mapping fast optical transients onto progenitor channels.

1. Defining observational properties

The canonical Ca-rich gap transient is faint relative to ordinary supernovae but brighter than classical novae. Early sample-defining work reported peak absolute magnitudes of $M_R = -15.5$ to $-16.5$, rise times of $\sim 12$–15 days, photospheric velocities of $\sim 6000$ to $11000\ {\rm km\ s^{-1}}$, and an early transition to the nebular phase within $\sim 1$ to 3 months (Kasliwal et al., 2011). Light-curve modeling for the original sample yielded ejecta masses in the range $0.4$–$0.7\ M_\odot$, reinforcing the inference that these explosions occupy a distinct low-ejecta-mass regime rather than the low-luminosity tail of standard SN Ia or core-collapse populations (Kasliwal et al., 2011).

Their most widely used nebular discriminator is the calcium-to-oxygen line ratio,

$$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$

In large spectroscopic compilations, Ca-rich transients occupy the highest region of $R_{\rm [Ca\,II]/[O\,I]}$, with values $-16.5$0–$-16.5$1, whereas stripped-envelope SNe typically lie at $-16.5$2–$-16.5$3 and Type II SNe at $-16.5$4–$-16.5$5 (Prentice et al., 2022). In the ZTF Census of the Local Universe search, a practical classification threshold was that nebular spectra show $-16.5$6 at any observed nebular phase (De et al., 2020). Line-width measurements further show that the mean characteristic width of the [Ca II] emission line is less than the [O I] emission line for all SN types, implying that [Ca II] typically originates from deeper ejecta layers; for Ca-rich transients the median [Ca II] FWHM was reported as $-16.5$7 (Prentice et al., 2022).

Although the class definition was built around a canonical faint-and-fast phenotype, later discoveries broadened the observed parameter space. ATLAS19dqr/SN 2019bkc rose in $-16.5$8–$-16.5$9 d and declined by $\sim 12$0 mag, making it one of the most rapidly evolving light curves known for a stellar explosion (Prentice et al., 2019). SN 2022oqm reached $\sim 12$1 mag and required three distinct power sources to explain its multi-peaked light curve, making it unusually bright for the class (Yadavalli et al., 2023). This observed spread suggests that “Ca-rich gap transient” is primarily a phenomenological classification anchored in nebular chemistry and rapid evolution, not a guarantee of a single explosion mechanism.

2. Spatial distribution and host-galaxy environments

A defining environmental result is that many Ca-rich gap transients do not trace the stellar light of their hosts. Deep VLT and HST imaging of nearby examples found no convincing underlying quiescent sources at the transient positions, allowing the exclusion of several in-situ progenitor possibilities (Lyman et al., 2014). For SN 2005E and SN 2012hn, the reported limits reached $\sim 12$2 and $\sim 12$3, respectively, with no underlying massive stars, clusters, globular clusters, or dwarf galaxies detected at or very near the transient positions (Lyman et al., 2014). These non-detections independently reject a massive-star origin for those events and strongly disfavor formation in globular clusters or known classes of dwarf satellites.

The offset distribution is correspondingly extreme. Reported projected galactocentric distances include 23 kpc for SN 2005E, 43 kpc for PTF 09dav, and 34 kpc for PTF 11bij, with $\sim 12$4 events at $\sim 12$5 kpc in projection (Lyman et al., 2014). Two later PTF events extended this picture: PTF12bho was identified as an intra-cluster transient in the Coma Cluster, while PTF11kmb lay at a physical offset of $\sim 12$6 kpc from the most likely host galaxy (Lunnan et al., 2016). In that sample, the offset distribution was found to be significantly more extreme than that seen for Type Ia supernovae or even short-hard gamma-ray bursts, implying that if kicks are responsible, they require larger kick velocities and/or longer merger times than sGRBs (Lunnan et al., 2016). A dedicated rate-and-efficiency study showed that the preference for large offsets is intrinsic rather than a survey artifact: under the hypothesis that Ca-rich events follow the stellar mass profile, the probability of all observed events being as far from their host light as the PTF sample was reported as $\sim 12$7 (Frohmaier et al., 2018).

Host-galaxy demographics reinforce the inference of an old progenitor population for a substantial fraction of the class. About half of Ca-rich transients occur in early-type galaxies with no ongoing star formation, with $\sim 12$8 in E/S0 hosts in one compilation (Lyman et al., 2014). Broad-band host-galaxy modeling across all known and suspected hosts found stellar masses spanning $\sim 12$9 to $\sim 6000$0, metallicities from $\sim 6000$1 to $\sim 6000$2, and very low current star formation rates for most E-type hosts; $\sim 6000$3 of all hosts were quiescent with $\sim 6000$4 (Dong et al., 2022). A later CLU-based comparison found that hydrogen-poor Ca-rich gap transient hosts and 91bg-like SN hosts occupy a similar parameter space of mass and sSFR and peak at $\sim 6000$5 Myr in the inferred delay-time distribution, much longer than normal SN Ia ($\sim 6000$6 Myr) or Type II ($\sim 6000$7 Myr) populations (Scherbak et al., 29 Sep 2025). At the same time, a minority of well-observed events clearly occur in star-forming environments, indicating that host demographics alone do not support a single-channel interpretation.

3. Rates and population demographics

The absolute rate of Ca-rich gap transients remains uncertain because it depends strongly on survey completeness, luminosity-function assumptions, and whether the most offset events are included. A Monte Carlo analysis of the Palomar Transient Factory survey, based on three events and explicit modeling of detection efficiencies, measured a volumetric rate

$\sim 6000$8

equivalent to $\sim 6000$9–$11000\ {\rm km\ s^{-1}}$0 of the local volumetric Type Ia SN rate (Frohmaier et al., 2018). A systematic ZTF-CLU search, correcting for CLU catalog incompleteness and missing $11000\ {\rm km\ s^{-1}}$1 offsets, derived a lower value,

$11000\ {\rm km\ s^{-1}}$2

or approximately $11000\ {\rm km\ s^{-1}}$3 of the SN Ia rate, with the abstract summarizing the result as at least $11000\ {\rm km\ s^{-1}}$4 of the SN Ia rate (De et al., 2020). The coexistence of these estimates indicates that the rate is not yet a settled quantity; the discrepancy is methodological rather than merely statistical.

Host-population inference points toward a delayed channel for a substantial portion of the sample but does not exclude prompt channels. Using nonparametric star-formation histories for nine well-sampled hosts, one study convolved the recovered SFHs with core-collapse and Type Ia delay-time distributions and found that in quiescent hosts the Type Ia-like SN rate vastly exceeds the core-collapse rate, with $11000\ {\rm km\ s^{-1}}$5, whereas some star-forming hosts reach $11000\ {\rm km\ s^{-1}}$6 core-collapse probability (Dong et al., 2022). The same analysis concluded that at least $11000\ {\rm km\ s^{-1}}$7 of Ca-rich gap transients must arise from delayed, Type Ia-like channels rather than exclusively from young massive stars (Dong et al., 2022). This does not demonstrate that the entire class is thermonuclear; instead, it statistically excludes an exclusively core-collapse origin.

Population structure within the class also appears nontrivial. In the ZTF-CLU sample, eight Ca-rich gap transients and ten literature events defined a likely continuum of spectroscopic properties ranging from SN Ia-like “Ca-Ia” objects to SN Ib/c-like “Ca-Ib/c” objects at peak light (De et al., 2020). Within the Ca-Ib/c group, the same analysis identified red and green populations distinguished by $11000\ {\rm km\ s^{-1}}$8 mag versus $11000\ {\rm km\ s^{-1}}$9 mag at $\sim 1$0-band peak, with the redder objects showing stronger line blanketing, slower light curves, weaker He lines, and lower nebular [Ca II]/O I. This phenomenology suggests that rate estimates may conflate multiple physical channels under a common nebular definition.

4. Thermonuclear models and kicked compact binaries

Several thermonuclear pathways have been advanced to explain the old environments, low ejecta masses, and calcium-dominated nebular spectra. The environmental analysis that ruled out in-situ formation concluded that the preferred scenario is a high-velocity, kicked system that explodes far from its natal site after a long delay, with neutron star–white dwarf mergers identified as a promising progenitor system (Lyman et al., 2014). In this framework, the same ingredients used to explain large-offset short gamma-ray bursts—natal kicks and merger delay times—are repurposed for a lower-energy, calcium-dominated transient, but with even more extreme offset requirements.

A different thermonuclear line of interpretation centers on helium-shell explosions on low-mass white dwarfs. The ZTF-CLU spectroscopic continuum, volumetric rates, and old environments were argued to be consistent with the explosive burning of He shells on low-mass white dwarfs (De et al., 2020). In that picture, Ca-Ia and red Ca-Ib/c objects are consistent with double detonations of He shells with high He burning efficiency, while green Ca-Ib/c objects could arise from less efficient He burning scenarios such as detonations in low-density He shells or He shell deflagrations (De et al., 2020). A concrete event-level realization was presented for SN 2016hnk, whose spectra and light curves were modeled by the detonation of a $\sim 1$1 helium shell on the surface of a $\sim 1$2 C/O white dwarf; the event produced $\sim 1$3 of $\sim 1$4 and $\sim 1$5 of ejecta (Jacobson-Galán et al., 2019).

Remnant chemistry provides an independent thermonuclear constraint. In G306.3$\sim 1$60.9, spatially resolved X-ray spectroscopy yielded mass-weighted abundances relative to solar of Si $\sim 1$7, S $\sim 1$8, Ar $\sim 1$9, Ca $0.4$0, and Fe $0.4$1, with abundance ratios $0.4$2 and $0.4$3 (Weng et al., 2021). Neither Type Ia nor core-collapse models reproduced the observed pattern, whereas He-shell detonation on an accreting white dwarf matched the required ordering

$0.4$4

once ejecta were mixed with $0.4$5–$0.4$6 of interstellar medium (Weng et al., 2021). That study preferred He-shell detonation over WD+NS merger models because the latter required too little ISM mass and underproduced S and Fe (Weng et al., 2021).

Other thermonuclear scenarios remain viable but less secure. Tidal detonation of a low-mass white dwarf by a black hole or possibly a neutron star was proposed as a mechanism capable of reproducing faint, rapidly evolving optical transients with strong calcium and little iron-peak production, but the model faces environmental difficulties because deep imaging rules out most globular clusters and dwarf galaxies at explosion sites, and a late Chandra observation of SN 2012hn provided only weak constraints on detonator mass (Sell et al., 2015). Late-time HST photometry of SN 2019ehk found a pure radioactive decay power source with $0.4$7, $0.4$8 d, and no statistical evidence for incomplete positron trapping; that analysis favored the tidal disruption of a low-mass white dwarf by a hybrid white dwarf over core-collapse or NS+WD scenarios for that event (Jacobson-Galán et al., 2020). Taken together, the thermonuclear literature supports several distinct channels rather than a single settled model.

5. Core-collapse channels and evidence for heterogeneity

Despite the strong association of many Ca-rich gap transients with old, quiescent environments, several well-observed events support stripped-envelope core-collapse origins. The most direct case is the discovery of a solar-type star in a close, eccentric binary with a neutron star within the young Galactic remnant RCW 86; the star is strongly polluted with calcium and other heavy elements, with calcium enhanced by $0.4$9 solar (Gvaramadze et al., 2017). Because the surviving compact object is a neutron star, the explosion was unambiguously core collapse, and the authors concluded that some calcium-rich supernovae can arise from core-collapse explosions of stripped massive stars (Gvaramadze et al., 2017).

iPTF15eqv provided a second strong core-collapse argument. It had among the highest [Ca II]/[O I] emission-line ratios observed to date, yet was more luminous and decayed more slowly than other Ca-rich transients; optical and near-infrared data were interpreted as the explosion of a $0.7\ M_\odot$0 solar mass star stripped of its H-rich envelope via binary interaction, with distinct chemical abundances and ejecta kinematics suggesting electron-capture processes (Milisavljevic et al., 2017). Deep radio limits implied a clean environment with $0.7\ M_\odot$1 within a radius of $0.7\ M_\odot$2 cm, and Chandra limits ruled out tidal disruption of a white dwarf by a black hole for masses $0.7\ M_\odot$3 solar masses (Milisavljevic et al., 2017). This event explicitly challenged the view that spectroscopically classified Ca-rich transients only originate from white-dwarf progenitor systems.

Double-peaked and star-forming-environment events further complicate the class. iPTF 16hgs occurred in the outskirts of a low-metallicity ($0.7\ M_\odot$4), star-forming dwarf spiral galaxy and showed a unique first blue peak followed by a main Ca-rich-like peak with $0.7\ M_\odot$5, $0.7\ M_\odot$6 erg, and a nebular [Ca II]/[O I] ratio of $0.7\ M_\odot$7 (De et al., 2018). The first peak could be explained either by outwardly mixed $0.7\ M_\odot$8 of $0.7\ M_\odot$9, reminiscent of some He-shell detonation models on white dwarfs, or by shock cooling of an extended envelope with $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$0 and $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$1, consistent with a highly stripped core-collapse progenitor (De et al., 2018). SN 2019ehk, initially classified as a Type Ib SN, evolved into a Ca-rich transient with an early first peak, a star-forming environment, $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$2, and $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$3 erg; it was proposed as another ultra-stripped envelope SN candidate, with the first optical peak identified as the key diagnostic of a young subpopulation among Ca-rich transients (Nakaoka et al., 2020).

Very recent multiwavelength work strengthens the case that some Ca-strong transients are He-rich stripped-envelope explosions interacting with compact circumstellar material. SN 2025coe showed Type Ib-like near-infrared spectra, early Swift X-ray detections, and radio non-detections that implied interaction with $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$4 of circumstellar material extending to $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$5 cm; the paper noted that similar nearby high-density CSM is seen in the other two X-ray detected CaSTs and suggested that either intensive mass-loss or some polluting mechanism may be a common feature of this subclass (Kumar et al., 26 Jan 2026). SN 2022oqm, by contrast, remained ambiguous: its multi-peaked light curve, $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$6, and $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$7 could be explained either by helium detonation on an unexpectedly massive white dwarf or by a peculiar stripped-envelope explosion inside dense, H- and He-poor circumstellar material (Yadavalli et al., 2023). The current evidence therefore supports a mixed population of Ca-rich transients rather than a single progenitor taxonomy.

6. Nucleosynthesis, remnants, and astrophysical significance

The astrophysical importance of Ca-rich gap transients derives from their unusually large calcium yields relative to their ejecta masses and rates. For SN 2005E, modeling suggested an ejecta mass of $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$8, nearly half of which was Calcium, with $$R_{\rm [Ca\,II]/[O\,I]}= \frac{F([\mathrm{Ca\,II}]\ \lambda\lambda 7291,7323)} {F([\mathrm{O\,I}]\ \lambda\lambda 6300,6364)}.$$9 of Calcium produced—$R_{\rm [Ca\,II]/[O\,I]}$0–$R_{\rm [Ca\,II]/[O\,I]}$1 times more than typical Type Ia or core-collapse supernovae (Mulchaey et al., 2014). This extreme Ca productivity motivated chemical-evolution studies of the intracluster medium. In one abundance-fitting analysis, adding a Ca-rich gap transient component to the usual core-collapse plus Type Ia mixture improved the $R_{\rm [Ca\,II]/[O\,I]}$2 goodness of fit from 84 to 2, with best-fit Ca-rich rates of about $R_{\rm [Ca\,II]/[O\,I]}$3–$R_{\rm [Ca\,II]/[O\,I]}$4 that of SN Ia, sufficient to reproduce nearby rich-cluster X-ray abundances (Mulchaey et al., 2014). Their preference for galaxy outskirts and intracluster or intragroup locations makes these explosions efficient polluters of diffuse gas reservoirs because their nucleosynthetic products are more easily incorporated into the intracluster medium (Mulchaey et al., 2014).

Nebular and remnant spectroscopy have become the main tools for connecting transient phenomenology to nucleosynthesis. Large-sample line-ratio work showed that Ca-rich transients are completely separated from other H-poor transients in $R_{\rm [Ca\,II]/[O\,I]}$5, with weak or absent [O I] and steeply evolving calcium-line ratios (Prentice et al., 2022). In G306.3$R_{\rm [Ca\,II]/[O\,I]}$60.9, the high Ca/Si and Ar/Si ratios and the shape of the abundance curve led to the proposal that the remnant is likely the first identified Galactic Ca-rich transient remnant, though the conclusion was explicitly tied to the measured abundance ratios and existing nucleosynthesis models (Weng et al., 2021). RCW 86 adds a different kind of remnant signature: a solar-type companion polluted by Ca-rich ejecta inside a young SNR provides direct evidence that at least some Ca-rich-like explosions are compatible with stripped, binary-mediated core collapse (Gvaramadze et al., 2017).

The class also matters for the interpretation of transient taxonomy itself. Because some objects resemble 91bg-like SN Ia at peak, others resemble Type Ib/c SNe, and still others exhibit double-peaked light curves, nearby dense CSM, or unusual X-ray behavior, a late-time calcium-dominated nebular spectrum is best understood as a chemically defined endpoint rather than a unique physical mechanism (De et al., 2020). A plausible implication is that Ca-rich gap transients occupy an intersection of low ejecta mass, high calcium production efficiency, and rapid transparency that can be reached by multiple channels—He-shell detonations on sub-Chandrasekhar white dwarfs, kicked compact binaries with long delay times, and some stripped low-mass core-collapse explosions. The central unresolved problem is therefore not whether the class is real, but how many physically distinct routes can populate it.