Glimpse-C02: Metal-Rich Globular Cluster

• Glimpse-C02 is an ultra-obscured, metal-rich globular cluster in the Milky Way, defined by a detailed color–magnitude diagram that reveals an old age (~11.9 Gyr), precise distance, and metallicity.
• Deep HST/WFC3/IR imaging combined with adaptive-optics observations enabled the first clear identification of the main sequence turnoff and refined key parameters through MCMC isochrone fitting.
• Advanced differential reddening corrections and structural analyses indicate a highly concentrated, dynamically evolved cluster with significant extinction variations across its field.

Searching arXiv for papers specifically about “Glimpse-C02” and its observational characterization. Glimpse-C02, also written GLIMPSE-C02, is an ultra-obscured Milky Way globular cluster discovered in the GLIMPSE mid-IR survey near the Galactic plane at Galactic coordinates $l=14.129^\circ$, $b=-0.644^\circ$. It is now characterized as an old, relatively metal-rich globular cluster affected by extreme and strongly spatially variable extinction. Deep HST/WFC3/IR imaging has produced the first color–magnitude diagram reaching the main sequence turnoff, enabling revised determinations of its reddening, distance, metallicity, age, structural parameters, and dynamical state (Loriga et al., 31 Mar 2026).

1. Identification and observational history

Early near-infrared work identified GLIMPSE-C02 as a heavily reddened, metal-rich system at $D\approx 4.6\pm0.7$ kpc, with $A_V\approx24.8\pm3$ and $[\mathrm{Fe/H}]=-0.33\pm0.14$, but only a portion of the red giant branch within $60$ arcsec was detected. That limited depth left its age and detailed structure uncertain (Loriga et al., 31 Mar 2026).

Subsequent adaptive-optics imaging with Subaru/IRCS using the RAVEN multi-object adaptive optics demonstrator improved the view of the crowded central region. In those observations, GC02 was imaged in $J$, $H$, and $K^\prime$, together with a narrow-band filter centered near $3.05\,\mu\mathrm{m}$. The achieved full width at half maximum was $b=-0.644^\circ$0 arcsec in $b=-0.644^\circ$1, $b=-0.644^\circ$2, and $b=-0.644^\circ$3, and $b=-0.644^\circ$4 arcsec in the $b=-0.644^\circ$5 narrow band. That study derived $b=-0.644^\circ$6 mag and $b=-0.644^\circ$7 kpc from the red clump, but argued that the absence of a sub-giant-branch signature in the $b=-0.644^\circ$8-band luminosity function was difficult to reconcile with an old globular cluster at that distance (Davidge et al., 2016).

The decisive revision came from HST/WFC3/IR imaging. Those data extend roughly $b=-0.644^\circ$9 magnitudes in the near-infrared color–magnitude diagram and, for the first time, identify the cluster main sequence turnoff. This deeper view changed the evidentiary basis for classification by allowing direct age fitting rather than inference from the upper giant branch and luminosity-function morphology alone (Loriga et al., 31 Mar 2026).

2. Imaging, photometric reduction, and the deep color–magnitude diagram

The HST analysis used WFC3/IR observations from program GO 17918, with four F110W images and four F160W images. The WFC3/IR pixel scale is about $D\approx 4.6\pm0.7$0 arcsec pixel$D\approx 4.6\pm0.7$1 and the nominal field of view is about $D\approx 4.6\pm0.7$2 arcsec$D\approx 4.6\pm0.7$3. To extend the radial analysis beyond the HST footprint, the study also incorporated VVVX survey images from VISTA/VIRCAM in $D\approx 4.6\pm0.7$4 and $D\approx 4.6\pm0.7$5 (Loriga et al., 31 Mar 2026).

The HST photometry was reduced with DOLPHOT, using an F160W drz frame as reference and applying PAM and CTE corrections. A $D\approx 4.6\pm0.7$6 detection threshold was adopted, photometry was placed on the VEGAMAG system with aperture corrections, and catalog cleaning used PSF-fit quality cuts with a $D\approx 4.6\pm0.7$7 clipping. VIRCAM data were reduced with DAOPHOT/ALLFRAME, calibrated to 2MASS, and astrometrized with Gaia DR3 (Loriga et al., 31 Mar 2026).

The resulting HST color–magnitude diagram in $D\approx 4.6\pm0.7$8 spans about $D\approx 4.6\pm0.7$9 magnitudes. The main sequence turnoff is identified at $A_V\approx24.8\pm3$0 and $A_V\approx24.8\pm3$1, and the photometry reaches about $A_V\approx24.8\pm3$2 magnitudes below the turnoff. The diagram shows a well-populated extended red giant branch characteristic of old populations and a low-significance red clump at $A_V\approx24.8\pm3$3, $A_V\approx24.8\pm3$4; no blue horizontal branch is apparent, which is consistent with a metal-rich globular cluster (Loriga et al., 31 Mar 2026).

3. Extinction and differential reddening

The dominant observational complication in GLIMPSE-C02 is extinction. The HST analysis adopted the Cardelli–Clayton–Mathis reddening law with $A_V\approx24.8\pm3$5, together with extinction coefficients $A_V\approx24.8\pm3$6 and $A_V\approx24.8\pm3$7, using the standard relation

$A_V\approx24.8\pm3$8

Because the cluster field is strongly and non-uniformly reddened, a star-by-star differential reddening correction was required (Loriga et al., 31 Mar 2026).

That correction followed the method of Pallanca et al. (2019). A mean ridge line was traced along the RGB, SGB, and upper main sequence using well-measured likely members. For each star, a local CMD was built from $A_V\approx24.8\pm3$9 neighbors within a maximum radius of $[\mathrm{Fe/H}]=-0.33\pm0.14$0 arcsec, and the ridge line was shifted along the reddening vector in steps of $[\mathrm{Fe/H}]=-0.33\pm0.14$1 until it matched the local CMD. The resulting map across the WFC3/IR field shows patchy structure and variations up to $[\mathrm{Fe/H}]=-0.33\pm0.14$2 mag. Applying this map substantially sharpens both the turnoff and the giant branch, although residual broadening remains (Loriga et al., 31 Mar 2026).

The revised mean reddening is

$[\mathrm{Fe/H}]=-0.33\pm0.14$3

which is substantially lower than the $[\mathrm{Fe/H}]=-0.33\pm0.14$4 quoted in the Harris catalog and in Kurtev et al. (2008) (Loriga et al., 31 Mar 2026). Earlier adaptive-optics work had derived a near-infrared extinction estimate of $[\mathrm{Fe/H}]=-0.33\pm0.14$5 under the Nishiyama et al. (2009) extinction law, and did not find strong evidence for substantial differential reddening near the cluster center (Davidge et al., 2016). The contrast between these treatments reflects both the change of extinction law and the much greater spatial leverage of the HST reddening map.

4. Distance, metallicity, and absolute age

A first-guess comparison with the bulge globular cluster NGC 6440 placed GLIMPSE-C02 near $[\mathrm{Fe/H}]=-0.33\pm0.14$6 and $[\mathrm{Fe/H}]=-0.33\pm0.14$7. The final parameter determination came from an MCMC isochrone fit to the differential-reddening-corrected CMD for stars within $[\mathrm{Fe/H}]=-0.33\pm0.14$8 arcsec, using BaSTI isochrones with $[\mathrm{Fe/H}]=-0.33\pm0.14$9, standard helium $60$0, and models including overshooting and RGB mass-loss (Loriga et al., 31 Mar 2026).

The best-fit solution is:

• $60$1
• $60$2
• $60$3 kpc
• $60$4 kpc
• $60$5 Gyr
• $60$6 (Loriga et al., 31 Mar 2026)

The heliocentric distance follows from the standard relation

$60$7

Parameter	Value
Mean reddening	$60$8
True distance modulus	$60$9
Distance from the Sun	$J$0 kpc
Galactocentric distance	$J$1 kpc
Metallicity	$J$2
Absolute age	$J$3 Gyr

These values confirm GLIMPSE-C02 as an old, relatively metal-rich globular cluster. The smaller Galactocentric distance, relative to earlier estimates, suggests that the system lies closer to the Galactic center than previously thought and strengthens its classification among bulge globular clusters (Loriga et al., 31 Mar 2026).

5. Structural parameters, luminosity, and dynamical state

The photometric center was measured from iterative barycenters of color–magnitude-selected stars and converged to

$J$4

with uncertainties of $J$5 in right ascension and $J$6 in declination. This position is offset by $J$7 from the Harris catalog value (Loriga et al., 31 Mar 2026).

The resolved star-count density profile combines HST data in the inner $J$8 arcsec and VIRCAM data from $J$9 to $H$0 arcsec. After subtraction of the field background, an MCMC fit with King models gave

$H$1

where the concentration parameter is defined as

$H$2

A surface-brightness fit to the F160W image yielded consistent values, including

$H$3

The high concentration is indicative of an advanced dynamical evolution stage (Loriga et al., 31 Mar 2026).

From the integrated F160W surface-brightness profile, the cluster has

$H$4

using $H$5 and $H$6 (Loriga et al., 31 Mar 2026). The half-mass relaxation analysis gives

$H$7

again consistent with an evolved dynamical state (Loriga et al., 31 Mar 2026).

6. Evolution of the cluster’s interpretation

The classification of GLIMPSE-C02 changed as deeper data became available. The adaptive-optics study based on Subaru/IRCS and archival Spitzer photometry found that the $H$8-band luminosity function was flat for $H$9 and that the expected sub-giant-branch rise near $K^\prime$0–$K^\prime$1 was absent if the cluster lay at about $K^\prime$2 kpc. On that basis, the study concluded that GC02 was more consistent with an intermediate age of about $K^\prime$3–$K^\prime$4 Gyr than with an old $K^\prime$5 globular (Davidge et al., 2016).

The later HST/WFC3/IR analysis reached the main sequence turnoff directly, corrected the field with a high-resolution differential reddening map, and obtained the first absolute age determination, $K^\prime$6 Gyr (Loriga et al., 31 Mar 2026). This shift from luminosity-function inference to turnoff-based dating largely resolves the earlier ambiguity. A plausible implication is that the previous intermediate-age interpretation was driven by the joint effects of limited depth, severe extinction, and crowding rather than by intrinsic youth.

In its current observational synthesis, GLIMPSE-C02 is an old and metal-rich globular cluster, one of the most extincted such systems in the Milky Way, likely located in the inner bulge region and in an advanced stage of dynamical evolution (Loriga et al., 31 Mar 2026).