Dust, not rotation: LAMOST cracks the broad main sequence of NGC 1647

For decades, NGC 1647 has been one of those open clusters that astronomers walk past on their way to flashier targets. It sits some 1,800 light-years away in Taurus, easy to spot in binoculars near Aldebaran, and was first catalogued by William Herschel in 1784. Around ninety confirmed members, all but a handful on the main sequence, no spectacular giants demanding attention. A textbook young cluster — except that its main sequence has always looked oddly fuzzy. Stars that should line up neatly in a colour-magnitude diagram instead spread into a thick, smeared band, as if the cluster could not make up its mind about how old its members are.

A new spectroscopic study led by Antonio Frasca at INAF’s Catania Astrophysical Observatory, in collaboration with researchers at Beijing Normal University, the University of Tübingen, and INAF Bologna, finally settles the question. The answer is not exotic. It is dust. But the way the team got to that answer is what makes the work matter.

The largest spectroscopic survey ever attempted on this cluster

Frasca and colleagues used the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) in China, complemented with astrometric and photometric data from ESA’s Gaia satellite, to obtain medium-resolution spectra of 347 stars that are confirmed or candidate members of NGC 1647. That is, by a wide margin, the largest spectroscopic sample ever assembled for this cluster. Until now, NGC 1647 had been studied with only a handful of stars and a few decades-old photometric surveys. The Frasca team’s work appeared as a preprint on arXiv on April 13, 2026.

Using the ROTFIT code, the team derived atmospheric parameters — effective temperature, surface gravity, and metallicity — together with radial velocity and projected rotational velocity, for 158 unique stars. They additionally identified four new double-lined spectroscopic binaries (SB2s) and reported the radial velocities of both components in each system. Rotation periods for 160 stars were extracted from TESS photometry.

The radial-velocity distribution turned out to be remarkably clean: a nearly symmetric peak centred at −5.32 km/s with a dispersion of only 1.57 km/s. That is the kinematic fingerprint of a real, gravitationally coherent stellar group, not a chance projection. For cooler members with effective temperatures below 7,000 K, the team derived a mean metallicity of −0.08 dex, slightly subsolar and consistent with the few previous estimates that had relied only on the cluster’s two giants. And the cluster age came out at approximately 203 million years, in agreement with the long-standing 150–200 Myr range from isochrone fitting.

So far, so unsurprising. The interesting part is what the spectra revealed about the smearing of the main sequence.

The eMSTO puzzle

Many young and intermediate-age open clusters display what stellar astrophysicists call an extended main-sequence turn-off, or eMSTO — a broadening of the upper main sequence that should not exist if all the stars formed at the same time from the same molecular cloud. The phenomenon has triggered an entire subfield of debate. Two physical explanations have dominated the discussion for the past fifteen years.

The first is rotation. Fast-rotating stars are deformed by centrifugal force, dimmer at their poles, and behave thermally as if they were slightly cooler and slightly older than non-rotating siblings. A cluster with a wide spread in rotational velocities should show a broadened turn-off as a natural consequence.

The second is age spread. If star formation in the cluster’s progenitor cloud was not instantaneous — if it lasted tens of millions of years — then the cluster genuinely contains stars of subtly different ages, and the main sequence will reflect that.

A third candidate, less glamorous but stubbornly persistent in the literature, is differential reddening: dust within or in front of the cluster does not absorb starlight uniformly. Some lines of sight are dustier than others. Stars seen through thicker patches appear redder and dimmer; stars seen through thinner patches appear bluer and brighter. The result, on a colour-magnitude diagram, mimics an age spread without there being one.

Where the new data point

Frasca’s team analysed the spectral energy distributions of 160 likely members of NGC 1647 and found that the cluster does indeed sit behind significant interstellar material. The average extinction across the field reaches 1.1 magnitudes in the visual band — far from negligible — and, crucially, it is not uniform. Different members suffer different amounts of absorption depending on where they fall within the cluster’s footprint. NGC 1647 is well known to lie behind the Taurus dark cloud complex, only about 160 parsecs in front of it, and the new analysis shows that complex casts a patchy, structured shadow across the cluster.

The decisive test came when the team correlated the reddening of each star with two physical quantities: the star’s offset from the lower boundary of the main sequence in the colour-magnitude diagram, and the star’s rotational velocity.

The first correlation is strong. The second is essentially zero.

That is the whole story in one sentence. Stars that appear shifted within the broadened main sequence are shifted in exactly the way you would expect if differential reddening — and not their spin — were responsible. The conclusion the authors draw is direct: differential reddening is mainly responsible for the extended main-sequence turn-off observed in NGC 1647.

Why this matters beyond one cluster

It would be a mistake to read this result as «case closed for eMSTOs in general.» It isn’t. In other clusters — particularly older ones, and clusters where reddening has already been carefully characterised — rotation almost certainly does drive a real broadening of the main sequence. There is genuine physics there. What the NGC 1647 result does is force a methodological discipline onto the field: before invoking exotic stellar physics to explain a fuzzy turn-off, you must rule out the dust. And ruling out the dust requires the kind of dense, star-by-star spectroscopic mapping that LAMOST is uniquely positioned to provide.

This is also a quiet vindication of the LAMOST programme as a whole. A 4-metre survey telescope optimised for fibre-fed spectroscopy was never going to make headlines for individual discoveries. What it does is dismantle ambiguities at scale, one cluster at a time. NGC 1647 is now better characterised than at any point in the 240 years since Herschel saw it.

For those of us who teach stellar astrophysics, there is a useful pedagogical takeaway as well. Open clusters are routinely sold to students as the cleanest laboratory in stellar physics — same age, same composition, same distance. NGC 1647 is a reminder that «same distance» can quietly mean «same distance through wildly different columns of dust,» and that the cleanest laboratory in the sky still has a layer of grime on its windows.

The study is titled NGC 1647: A young open cluster with a broad main sequence observed with LAMOST, by A. Frasca, M. Qin, J. Alonso-Santiago, G. Catanzaro, J. N. Fu, J. Y. Zhang and A. Bragaglia, posted as arXiv:2604.11765 with DOI 10.48550/arxiv.2604.11765.

© 2026 SKYCR.ORG | Homer Dávila Gutiérrez, FRAS. All rights reserved. Total or partial reproduction prohibited without express authorization. Original source: A. Frasca et al., NGC 1647: A young open cluster with a broad main sequence observed with LAMOST, arXiv:2604.11765 (2026).