The final James Webb Space Telescope image of the month of the year, directed by ESA, brings a scene that looks like a Christmas card from the depths of the Milky Way: the young star cluster Westerlund 2, immersed in the glowing clouds of the Gum 29 nebula, fills the frame with thousands of dazzling stars and complex structures of gas and dust. In a single frame, aesthetics and science are combined – a visually spectacular cosmic “fireworks display” and an exceptionally rich laboratory for studying star formation, brown dwarfs, and protoplanetary disks.

Westerlund 2 is located about 20,000 light-years from Earth in the southern constellation of Carina. It is an extremely young cluster, only about two million years old, situated in a vast star-forming region known as Gum 29. In this chaotic environment, where massive stars “carve out” the surrounding nebula with their radiation and winds, Webb now shows for the first time in the infrared range almost the entire population of stars and substellar objects, from the most massive giants down to the smallest brown dwarfs whose mass is only about ten times that of Jupiter.

The new image was created by combining data from Webb’s NIRCam (Near-Infrared Camera) and MIRI (Mid-Infrared Instrument). NIRCam records young stars piercing through the dust in the near-infrared range, while MIRI reveals glowing dusty gas and layers of warm material where planets are born in the mid-infrared. Together they provide a layered, almost three-dimensional image of a stellar “cradle” in full operation.

Fireworks of stars in the heart of the Gum 29 nebula

The upper part of Webb’s image is dominated by a dense cluster of young, extremely massive stars – the very heart of Westerlund 2. Many of them rank among the hottest and brightest stars known in the Milky Way, and part of the system likely includes Wolf–Rayet type stars, which already show strong stellar winds and a rich spectrum of emission lines. Their intense ultraviolet radiation and winds literally “sandblast” the surrounding gas, pushing it into waves, cliffs, and pillars that surround the cluster’s core.

The Westerlund 2 cluster extends roughly 6 to 13 light-years in diameter and contains about several thousand stars, from giant hot O-stars to a multitude of fainter, newly formed lower-mass stars. Since the cluster is very young, it has not yet managed to gravitationally “disperse” throughout the galaxy; all stars are still relatively densely packed in the same area. It is precisely this density and youth that make Westerlund 2 an ideal natural laboratory for studying the processes by which stars and planets form under the most extreme conditions.

The Gum 29 nebula, in which the cluster is located, is a vast bubble of ionized hydrogen and dust. In Webb’s image, the walls of this bubble are depicted as wavy, broken structures in shades of orange and red, while thinner, sparser layers of material transition into softer blue and pink tones. Radiation from the stars of Westerlund 2 permeates the entire scene, illuminating the edges of the clouds like the setting Sun coloring the edges of cumulus clouds on Earth.

The same scene was recalled about ten years ago by the famous Hubble image published on the occasion of that telescope’s 25th anniversary. Hubble then showed a firework of about three thousand young stars and chaotic gas clouds in the visible and near-infrared range. However, Webb’s instruments can peer even deeper through the dust and reveal even those much fainter light sources that Hubble could not clearly resolve – especially cooler, low-mass stars and brown dwarfs hiding in the shadows of the clouds.

Six-pointed stellar sparks and glowing walls of gas

The visual impression of Webb’s image of Westerlund 2 can be deceptive at first glance: in the sky, it seems as if all the stars are part of the same cluster. But many of the brightest stars with spectacular six-pointed diffraction “spikes” are actually located much closer to Earth. These are stars from our spiral arm of the Milky Way, whose light is particularly strongly refracted by the geometry of Webb’s mirrors and struts, creating the recognizable spike pattern.

The true members of Westerlund 2 are mostly much fainter, but there are surprisingly many of them in Webb’s infrared view. In the central part of the image, we see a dense cluster of tiny, pointy dots – young stars on the main sequence and pre-main-sequence objects, which are just finishing their formation phase. Some of them are still embedded in small cocoons of gas and dust, visible as crumpled, dense “knots” in otherwise more transparent clouds.

Below and around the main cluster, large, almost sculpturally shaped walls and pillars of gas stand out. These are zones where the clouds have already been eroded by the radiation of the most massive stars; the edges of these structures are illuminated like the rims of clouds at sunset, but in the infrared, so the colors transition from dark red to bright orange. In these boundary layers, there are also dense pillars and “fingers” of material that protect the inner, even cooler nests of stars from the strongest radiation.

In the thinner parts of the nebula, between denser clouds, infrared radiation shows soft bluish and pink shades – traces of sparser gas and dust floating between the two main star-forming zones. The whole image thus functions as a map of density and temperature: warmer and denser parts glow in redder tones, while cooler and sparser material transitions into milder, more transparent colors.

How Webb “sees” through dust: NIRCam and MIRI in tandem

The key to such a detailed image lies in the combination of Webb’s two main cameras. NIRCam, the near-infrared camera, is sensitive to wavelengths from approximately 0.6 to 5 micrometers. It is Webb’s main “workhorse” for sharp images of young stars and protostars, but also the tool with which the telescope was originally calibrated and focused. Thanks to high spatial resolution and sensitivity, NIRCam can distinguish individual stars even in extremely dense clusters like Westerlund 2, where thousands of light sources are located almost “shoulder to shoulder”.

MIRI, the mid-infrared instrument, works at even longer wavelengths, roughly from 5 to 28 micrometers. In this range, warm dusty gas and dust particles heated by stellar radiation radiate, as do layers of molecular gas in the vicinity of young stars and protoplanetary disks. MIRI thus provides a map of which parts of the nebula heat is coming from and where dense, still active pockets of material are located from which stars and planets continue to form.

In the depiction of Westerlund 2, data from NIRCam and MIRI have been carefully merged into a single composite image. Specific infrared colors were chosen to emphasize the differences between hot, massive stars, cooler, newly formed low-mass stars, and clouds of gas and dust. Behind the aesthetically appealing result stands precise scientific color “coding,” which allows astronomers to read a multitude of different physical information from a single image – from temperature and density to the structure and depth of the clouds.

Hunting for brown dwarfs: a complete census of substellar neighbors

One of the most important scientific results hidden behind this image is the first nearly complete census of brown dwarfs in Westerlund 2. Brown dwarfs are objects “between” stars and planets: they are massive enough to initiate deuterium fusion in their cores, but do not have enough mass to sustain stable hydrogen fusion like true stars. Because of this, they are very dim and cold in the optical range, making them extremely difficult to detect at great distances – especially if they are hidden behind dust clouds.

This is exactly where Webb’s sensitivity in the infrared spectrum comes to the fore. In Westerlund 2, Webb’s data reveal for the first time a whole series of brown dwarfs in an extremely massive young cluster, including objects whose mass is only about ten Jupiter masses. Such objects are beyond the reach of most previous instruments, so they have been practically invisible at these distances until now.

The census of these substellar cluster members is particularly important for understanding the so-called initial mass function – the distribution of masses with which nature “births” stars and substellar objects in different environments. In quieter, less dense stellar nurseries, astronomers have already been able to study how the number of small stars and brown dwarfs compares to the number of massive stars. But in extreme, supermassive clusters like Westerlund 2, where very massive stars and strong radiation dominate, such detailed studies have been impossible until now.

Webb’s data now allow determining for Westerlund 2 to what extent this “extreme neighborhood” encourages or suppresses the formation of brown dwarfs. If it turns out that their number is similar to that in quieter areas, this would suggest that nature creates substellar objects at a similar pace regardless of the environment. But if there are significantly fewer or more of them, it would mean that the environment – gas density, turbulence, radiation – directly changes the way gas fragments and collapses into new objects.

Disks in a dangerous neighborhood: where planets are born and disappear

Along with brown dwarfs, Webb’s data also build upon years of Hubble monitoring of protoplanetary disks in Westerlund 2. Disks are flat, rotating structures of gas and dust surrounding young stars; it is precisely in these disks that planetary systems form. However, in the center of Westerlund 2, where massive stars constantly bombard the surroundings with ultraviolet radiation and powerful winds, disks are exposed to an environment that threatens them with dispersal and evaporation.

Hubble showed in a three-year program that disks of stars closer to the cluster center lose material much faster than disks in peripheral zones. Webb now continues this story in the infrared range: thanks to sensitivity to warm dust and molecular gas, astronomers can identify several hundred stars with disks in various stages of development throughout the cluster, from dense, still massive disks to already quite stripped structures on the brink of survival.

Such a comparison provides a rare opportunity to directly observe how the environment affects the future of potential planetary systems. In the center of Westerlund 2, planets – if they form at all – are likely exposed to strong radiation and frequent close encounters with neighboring stars. The result can be destabilized orbits, ejected planets, or very exotic systems that hardly resemble our Solar System. At the edges of the cluster, where the environment is calmer, disks have more time to cool down, collect dust, and build planets before surrounding giants destroy them.

Westerlund 2 thus serves as an extreme testing ground: it shows how planet formation takes place in a “bad neighborhood” of the galaxy, under conditions that likely resemble environments where the first generations of stars and planets were born in the early history of the universe. By comparing with calmer, smaller clusters, astronomers can reconstruct how “protected” our Solar System was relative to such violent regions.

EWOCS: a major project mapping supermassive star clusters

The new Webb image of Westerlund 2 is part of broader observations conducted within the framework of the Extended Westerlund 1 and 2 Open Clusters Survey (EWOCS) project. It is a large international program that combines data from the James Webb Space Telescope, the Chandra X-ray Observatory, and other instruments to study in detail the most massive young clusters in our galaxy – Westerlund 1 and Westerlund 2.

EWOCS was designed with several key goals. First, it aims to determine the complete mass distribution of stars and substellar objects in these clusters, from the most massive stars to brown dwarfs. Second, the intention is to identify and characterize protoplanetary disks in such an extreme environment, where strong radiation and winds from massive stars constantly change the local “climate.” Third, the project strives to track how interactions between stars and the environment – radiation, shocks, gravitational close passes – reflect on the evolution of the stars themselves and their systems.

For Westerlund 1, EWOCS has already delivered a detailed catalog of X-ray sources and a spectacular Webb image showing the core of the supermassive cluster with thousands of stars of different masses. The same approach is now being applied to Westerlund 2, where Webb’s infrared observations (including a dedicated JWST program on Westerlund 2) are designed to reach even the dimmest, low-mass members of the cluster, but also to highlight disks and gas structures.

Besides producing impressive images for the public, EWOCS also generates extensive catalog data – positions, brightnesses, spectra, and other parameters of thousands of sources. These data enable statistical studies that will answer a series of questions in the coming years: how many brown dwarfs are born in such supermassive clusters, how long disks survive near massive stars, what the internal dynamics of the cluster are, and how all this affects the prospects for planet formation.

Link with Hubble, Chandra, and future missions

Westerlund 2 is not a new object for astronomers – but Webb gives it a completely new dimension. Hubble published a famous image of Westerlund 2 and the Gum 29 nebula back in 2015 on the occasion of its own 25th anniversary, showing a “stellar fireworks display” in the visible and near-infrared range. The Chandra X-ray Observatory completed the picture in the high-energy range, revealing how many young and massive stars shine in X-rays as well, along with diffuse emission of hot gas between them.

Now Webb combines and expands all this information in the infrared, the range where stars with disks, brown dwarfs, and warm dusty gas are most easily observable. In combination with Hubble and Chandra data, astronomers can build an almost “multilayered” map of Westerlund 2: visible light shows illuminated cloud surfaces, infrared radiation reveals hidden young stars and disks, and X-ray radiation speaks of the most energetic processes and shocks.

Such multiple comparisons will have far-reaching consequences for future missions as well. Westerlund 2 serves as a reference sample for understanding star clusters in other galaxies, which Webb is already observing in the infrared, but as tiny, barely resolvable dots. Knowing in detail what one local, but extreme cluster looks like “from the inside,” astronomers will be better able to interpret signals from much more distant objects in the early universe.

Westerlund 2 as a window into the past of the Milky Way

Westerlund 2 and its “twins” like Westerlund 1 represent a type of environment thought to have been common in the early history of the Milky Way. In that era, the galaxy experienced periods of intense star formation, creating supermassive clusters where star density and radiation levels were much higher than in most present-day open clusters.

Understanding how stars and planets form and survive in such conditions helps answer questions like: how common is it at all for planetary systems like our Solar System to form and survive in the galaxy? If most stars form in clusters similar to Westerlund 2, then a significant portion of potential planets are exposed from the start to violent conditions that can limit their longevity and stability.

Westerlund 2 also allows testing theories about how massive stars are born and die. In such dense environments, close interactions and mergers of stars are possible, which can lead to the creation of exceptionally massive objects and, ultimately, very powerful supernovae or even black holes of greater masses than in average cases. Webb’s detailed infrared data will help identify the most massive and evolved stars in the cluster, as well as their interactions with the surrounding gas.

For astronomers dealing with galaxy evolution on cosmic scales, Westerlund 2 represents a local “model” of a starburst environment – an area of extremely intense star formation. By comparison with distant starburst galaxies, Webb allows the macroscopic picture (global star formation rates and infrared shimmering of the entire galaxy) to be compared with the microscopic picture (individual clusters, disks, and brown dwarfs).

Cosmic scene for the public, data treasure for science

Although this “dazzling fireworks display of dwarf stars” will remain primarily a visual delight for many, behind it stands a vast amount of data and pre-planned scientific goals. Each pixel of the image was created from a series of observations in different filters, complex calibrations, and careful processing. The data are then converted into catalogs with positions, brightnesses, and colors of thousands of stars, and into spectra and additional products that enable detailed analyses.

For scientists, this image is just the beginning – a sort of cover page for extensive catalogs and papers that follow. In the coming years, detailed studies of the mass function in Westerlund 2, papers on disk development in extreme conditions, analyses of the structure of the Gum 29 nebula, but also comparisons with other clusters in the Milky Way and beyond are expected. Westerlund 2, which already marked Hubble’s 25th anniversary, now also becomes one of the most visually impressive symbols of Webb’s ability to unveil the hidden layers of our cosmos.

For the wider public, Webb’s final image of the month in 2025 provides something of a deeper reminder: behind every shining star in the night hides a whole history of formation and evolution, largely shaped by the environment in which that star was born. Westerlund 2 shows how dramatic that environment can be – and how much knowledge is needed for us to truly understand it.