NGC 5466's Stream Gap May Not Require a Dark Matter Subhalo

Globular cluster NGC 5466 is an ancient, faint ball of stars in the Milky Way's halo, roughly 52,000 light-years away. For nearly two decades, astronomers have known that it trails a long ribbon of stars—a tidal stream pulled from the cluster by the Galaxy's gravity. New deep imaging with the Subaru Telescope now traces that stream using some of its faintest, lowest-mass stars, revealing a gap, a perpendicular component, and the slow signature of a cluster losing its lightest members.

A filter tuned to calcium and metal-poor stars

The observations use the Hyper Suprime-Cam (HSC) on the Subaru Telescope and a narrowband filter called NB395. A narrowband filter transmits only a narrow slice of starlight. NB395 is centered near 3954 angstroms, in the violet-blue part of the spectrum, and covers the calcium H and K absorption lines, which make the filter sensitive to stellar metallicity. By combining NB395 with broader g- and i2-band images, the team could separate NGC 5466's metal-poor stars from foreground field stars and background galaxies.

The paper, posted to arXiv as a preprint, reports observations taken in 2022 and 2023. The researchers selected likely cluster members with a k-nearest-neighbor algorithm applied to a color-color-magnitude diagram—a sorting method that groups stars by brightness and color. They reached stars as faint as i2 approximately 23.5, deep enough to include low-mass main-sequence stars. They also cross-checked their photometry against Gaia XP spectra and corrected a systematic offset in the crowded center.

Metallicities agree in the core, but the outskirts are messier

Metallicity is the abundance of elements heavier than helium, often measured as [Fe/H], the iron-to-hydrogen ratio relative to the Sun. NGC 5466 is metal-poor, with a known average [Fe/H] near -2.0. The NB395-based metallicities broadly agree with previous Pristine and SEGUE measurements, and the high-probability members near the cluster center match that known value.

Farther out, however, the inferred metallicities shift slightly toward more metal-rich values. The authors do not claim this is a true iron gradient. The trend could come from leftover contamination, or from multiple stellar populations within the cluster. Such populations can differ in light elements such as carbon, nitrogen, and oxygen, changing the blue and ultraviolet light that NB395 samples without changing iron. Spectroscopy will be needed to tell the difference.

A gap that may not need a dark matter culprit

The member stars clearly outline NGC 5466's tidal stream. Beyond the cluster's tidal radius—16.1 arcminutes on the sky, about 240 light-years at the cluster's distance—the azimuthally averaged radial surface density profile follows a steep power-law decline, with a slope of -4.53. That is consistent with earlier work. The team also detects a power-law component perpendicular to the main stream, a hint that stripped stars have spread out during multiple passages through apogalacticon—the far point of the cluster's Galactic orbit.

Most striking is a density gap about 200 parsecs, or roughly 650 light-years, in projected distance from the cluster center. Gaps in tidal streams are often attributed to encounters with invisible dark matter subhalos, which can perturb the stream like a rock in a current. But the researchers compared their map with N-body simulations from the e-TidalGCs project—computer models that follow the gravity of many particles—which model the cluster's disruption without dark matter subhalo encounters. The simulated gap appears close to the observed one, within about 0.2 degrees, though not exactly at the same place.

If the match holds, the gap could arise from NGC 5466's own orbital history: a recent pericentric passage—a close approach to the Milky Way's center—or a crossing of the Galactic disk could have enhanced mass loss and carved a density dip. That is a suggestive result, not a closed case. Radial velocities of stream stars, which future Subaru spectroscopy could provide, would show whether the gap is accompanied by a localized velocity kick expected from a subhalo encounter.

Where the lightest stars go

The study also measures the mass function of main-sequence stars between about 0.55 and 0.75 solar masses. The mass function describes how many stars exist at each mass. Inside the tidal radius, its slope changes strongly with radius: the team finds a negative radial gradient of -0.91 plus or minus 0.06. This means the cluster is mass-segregated—heavier stars have sunk toward the center while lighter stars have drifted outward.

Beyond the tidal radius, the mass function along the stream is flatter than in off-stream regions. The authors interpret this as preferential stripping of low-mass stars: after mass segregation pushed them outward, the Milky Way's tidal field removed them more easily. The paper also sees quasi-periodic fluctuations in the mass-function slope farther out, but cautions that they may be partly artificial, caused by completeness corrections or lingering background galaxies.

What the stream can and cannot tell us

The work is a preprint, and several uncertainties remain. Residual contamination and possible offsets in the inferred metallicities beyond 10 arcminutes could affect the outer metallicity trend. The NB395 filter is also sensitive to blue and ultraviolet light affected by light-element changes, so the apparent metallicity shift may not be a true iron abundance gradient. The gap match to simulations is broadly consistent but not exact. And the perpendicular component extends beyond the observed footprint, so its nature is still open.

Still, the study shows how a metal-sensitive filter on a wide-field telescope can turn faint stars into a map of a cluster's destruction. NGC 5466 is not being destroyed in a single event; it is gradually losing stars through tides. Its stream preserves a record of internal dynamics and Galactic tides, and the new map suggests that some of the most important clues are written in the faintest, lightest stars.