The diffuse Galactic emission is the fog of high-energy astrophysics. It arises when cosmic rays, protons and electrons accelerated by sources such as supernova remnants, collide with interstellar gas, dust grains, magnetic fields, and the sea of starlight that permeates the Galaxy. Protons smashing into gas produce neutral pions that decay into gamma rays, while electrons scatter low-energy photons up to gamma-ray energies through inverse Compton processes and radiate synchrotron light in magnetic fields. Disentangling these components is notoriously difficult because their contributions overlap and depend on poorly constrained distributions of gas, radiation, and cosmic rays. Traditional models, such as the widely used GALPROP code, solve the cosmic-ray transport equation with detailed parameterizations, but they require extensive tuning and still struggle in the structurally complex inner Galaxy.

The research team, led by Xi Liu and colleagues at Sun Yat-sen University, took a deliberately different route. Rather than assuming a physical model from first principles, they trained two standard machine learning algorithms, Random Forest and K-Nearest Neighbors regression, to learn a direct mapping between Planck’s nine all-sky foreground maps, spanning 30 to 857 gigahertz, and the gamma-ray intensity predicted by the Fermi-LAT interstellar emission model across 28 energy bins from 50 megaelectronvolts to 814 gigaelectronvolts. Each Planck band traces a different physical regime: the lowest frequencies are dominated by synchrotron radiation from cosmic-ray electrons, intermediate bands mix free-free and spinning-dust emission, and the highest frequencies are dominated by thermal dust, a reliable tracer of interstellar gas.

The results are striking. In the 0.1 to 10 gigaelectronvolt range, where the diffuse gamma-ray sky is brightest, the models achieve coefficients of determination above 0.90, peaking at 0.96 at 687 megaelectronvolts. The predicted full-sky maps reproduce the major morphological features of the gamma-ray Galaxy, and the extracted spectral energy distribution of the inner Galactic plane matches the reference model closely. The team verified that the learned relationship is not a local accident: when a model trained on one Galactic hemisphere was tested on the opposite side, performance varied by no more than 6 percent regardless of how large a training region was used, indicating that the microwave-to-gamma-ray connection reflects a robust, large-scale property of the Milky Way rather than a statistical fluke of nearby structures.

Perhaps the most physically revealing result comes from asking which Planck frequencies matter most. When the high-frequency bands, dominated by thermal dust emission, were used alone, they achieved predictive power nearly identical to that of all nine bands in the 0.1 to 10 gigaelectronvolt range, while the low-frequency synchrotron channels performed markedly worse. This is exactly what one would expect if the diffuse gamma-ray emission at these energies is hadronically dominated: gamma-ray intensity is proportional to the product of gas column density and cosmic-ray density, and dust is an excellent proxy for the gas that serves as the target material. Above 10 gigaelectronvolts, however, the predictive power of the low-frequency bands rises, consistent with a growing leptonic contribution in which the synchrotron-emitting electron population also produces gamma rays through inverse Compton scattering. The analysis thus provides an independent, data-driven confirmation of the standard picture of Galactic gamma-ray production.

The residual maps, showing where the learned mapping breaks down, read like a tour of the Galaxy’s most enigmatic structures. Positive residuals appear toward the Magellanic Clouds and Centaurus A, where abundant gas and dust in nearby systems tempt the model to predict hadronic gamma-ray emission that the reference maps treat separately. Warm ionized gas complexes such as Barnard’s Loop in Orion and the Gum Nebula also show over-predictions, because free-free emission is subdominant across the Planck bands. On the other side, negative residuals trace Loop I and the North Polar Spur, hinting at inverse-Compton-dominated radiation fields or enhanced electron populations that submillimeter dust tracers cannot capture. Around the Galactic Center, the residuals mirror the shape of the Fermi Bubbles, the giant gamma-ray lobes whose faint, hard-spectrum microwave counterpart is too subtle for the model to disentangle without explicit priors.

The team also uncovered a curious hemispheric asymmetry: at high Galactic latitudes, the model systematically under-predicts the north while over-predicting the south by 10 to 30 percent on average. Because the discrepancy is spatially coherent rather than random, it may point to genuine physical differences between the two Galactic hemispheres, such as variations in cosmic-ray density or radiation fields. A segmented analysis of the inner Galaxy further showed that the asymmetry is concentrated within 30 degrees of the Galactic Center, likely linked to the Galactic bar and spiral-arm tangents, while the outer disk behaves with remarkable uniformity.

In a direct head-to-head comparison at roughly 4.3 gigaelectronvolts, the machine learning approach outperformed a GALPROP simulation tuned to the latest AMS-02 cosmic-ray data in the inner disk and Galactic Center region, achieving a coefficient of determination of 0.9465 and a mean absolute relative error of 14.7 percent, compared with 0.8624 and 22.3 percent for the physical model. GALPROP, by contrast, held a slight edge in the more homogeneous outer disk and in parts of the halo, confirming its reliability under simpler conditions. The authors emphasize that the two approaches are complementary: the data-driven model absorbs non-linear multi-frequency correlations that static gas distributions miss, while the physical simulation remains indispensable for interpretation and extrapolation.

An appendix to the study adds a compelling validation using raw Fermi-LAT photon counts rather than the smoothed emission model. As exposure accumulated over 40 weeks of mission data, the predictive performance climbed steadily and stabilized, and the negative residuals in the predicted maps turned out to coincide with 86 known gamma-ray point sources cataloged by Fermi, including pulsars, blazars, and a radio galaxy. Because the machine learning model learned the diffuse background purely from gas and dust maps without ever seeing point-source information, any compact gamma-ray emitter naturally surfaces as a localized deficit, offering an independent check on standard source-detection pipelines.

Beyond its immediate results, the work positions machine learning as a physically interpretable instrument rather than a black box. The learned mapping encodes real relationships between interstellar matter, radiation fields, and cosmic-ray processes, and its failures mark precisely the regions where conventional templates are incomplete or biased. The authors propose extending the framework with low-frequency radio surveys such as the Haslam 408-megahertz map, polarized microwave channels, and hydrogen-alpha data on ionized gas, and eventually integrating X-ray, ultra-high-energy gamma-ray, and neutrino observations. Such a multi-messenger, data-driven baseline could help separate standard emission from exotic components, sharpen constraints on cosmic-ray propagation, and, in an era when IceCube has detected Galactic neutrinos and LHAASO has measured PeV-scale diffuse emission, provide the empirical foundation that next-generation high-energy astrophysics will demand.

Subject of Research: Data-driven machine learning modeling of Galactic diffuse gamma-ray emission using multi-wavelength Planck observations

Article Title: Data-driven modeling of Galactic diffuse emission with multi-wavelength observations

Article References: Data-driven modeling of Galactic diffuse emission with multi-wavelength observations. (n.d.). https://doi.org/10.1140/epjc/s10052-026-16408-2

Image Credits: AI Generated

DOI: 10.1140/epjc/s10052-026-16408-2

Keywords: Galactic diffuse emission, machine learning, Planck satellite, Fermi-LAT, cosmic rays, gamma-ray astronomy, interstellar medium, inverse Compton, hadronic emission, GALPROP, Fermi Bubbles, multi-messenger astrophysics

The observational foundation of this ecosystem view rests on decades of precision astrometry. Near-infrared monitoring of the innermost parsec, most notably by the GRAVITY instrument on the European Southern Observatory’s Very Large Telescope Interferometer, has tracked stars on bound orbits around the compact radio source Sagittarius A star, delivering the most convincing dynamical evidence that this object is indeed a supermassive black hole as described by general relativity. The S-cluster of stars, dominated by the luminous star S2 on its sixteen-year elliptical orbit, has been used to test relativistic effects including gravitational redshift and the Schwarzschild precession, while also pinning down the distance to the Galactic Centre to within a few percent. These measurements transform the central parsec into a precision laboratory where the interplay between stellar dynamics, black hole mass and relativistic gravity can be examined against exact predictions rather than qualitative expectations.

Yet the S-cluster presents one of the field’s most persistent puzzles: the so-called paradox of youth. The central parsec contains massive, luminous B-type and Wolf-Rayet stars whose lifetimes are measured in millions of years, far too short for them to have migrated from their presumed birth sites in the surrounding disk through standard dynamical friction. At the Brno symposium, participants revisited candidate solutions, including in-situ star formation within the dense accretion disk fragments of the past, tidal disruption of binary stars passing close to the black hole which leaves one captured star in a tight orbit, and exchange interactions in which the black hole swaps into a hard massive binary. Each mechanism leaves a different fingerprint in the distribution of orbital eccentricities and inclinations, and the latest modeling of the S-cluster’s phase-space structure suggests that no single channel explains every observed orbit, hinting at a layered formation history stretching back many millions of years.

Beyond the S-cluster lies the nuclear star cluster, a dense concentration of roughly ten million solar masses packed within about ten parsecs of the black hole. New analyses presented at the meeting explored how this cluster’s complex structure, with young stars preferentially rotating in a disk-like configuration at larger radii and older populations dominating closer in, encodes the history of gas inflow episodes triggered by the galactic bar. The nuclear cluster also serves as a gravitational anchor for less massive objects, and speakers examined the fate of stellar-mass black holes, neutron stars and white dwarfs that segregate toward the centre through mass segregation. Their predicted number densities, potentially tens of thousands of stellar remnants within the central parsec, carry consequences for gravitational-wave predictions and for the rates of tidal disruption events, in which stars venturing too close to Sagittarius A star are torn apart and briefly outshine much of the surrounding galaxy.

Gas dynamics formed the second pillar of the symposium. The circumnuclear disk, a ring of dense molecular material at radii of a few parsecs, regulates the flow of gas toward the central engine and episodically feeds, or starves, both accretion onto the black hole and star formation in the central molecular zone. Recent work on the ionized and neutral gas kinematics within the central parsec, including the mini-spiral structures that thread hot plasma through the cavity inside the circumnuclear disk, illustrated how inflow is inefficient and turbulent, with only a small fraction of the supplied material ever reaching the accretion flow. Sagittarius A star today accretes at a rate roughly a billion times below the Eddington limit, producing the faint, radiatively inefficient glow observed by near-infrared polarimetry and by the Event Horizon Telescope, which resolved the ring-like shadow of the black hole’s photon emission region in 2022.

That quiescent present contrasts dramatically with a far more active past, and the evidence is written in the largest structures of the Galactic Centre region. The Fermi bubbles, two gamma-ray-emitting lobes extending tens of thousands of light-years above and below the Galactic plane, together with the softer X-ray counterpart bubbles detected by the eROSITA instrument, testify to an energetic event several million years ago, plausibly a phase of rapid accretion onto Sagittarius A star or an intense nuclear starburst. Meeting discussions emphasized how the spectral gradients across these bubbles constrain the timing, energy budget and particle acceleration mechanisms of the outburst, with estimates of total injected energy in the range of tens of millions of supernova equivalents, sufficient to reshape the thermal history of the entire Galactic halo.

On smaller scales, X-ray observations of light echoes provide a time-lapse record of more recent activity. Reflections of X-ray photons from past flares, scattered by dense molecular clouds such as the famous MC2 complex near the Sagittarius B region, have been mapped as moving fronts of fluorescing iron, indicating that the central engine brightened substantially within the last few centuries. Modeling of these reverberation signals, combined with polarization measurements that can distinguish forward-scattered from back-scattered radiation, suggests flare luminosities that transiently approached a significant fraction of the Eddington limit, a striking reversal of the present-day faintness. Speakers noted that continuous monitoring of these echoes offers an essentially forensic technique for reconstructing the accretion history of a low-luminosity galactic nucleus, a technique now being extended to nearby external galaxies where analogous echoes betray dormant black holes in action.

The Galactic Centre also functions as the Rosetta stone for galactic nuclei everywhere. Because the Milky Way’s central black hole is roughly ten thousand times less massive than those powering the most luminous quasars, phenomena observed at different scales and timescales in active galaxies can be scaled and tested locally. Symposium sessions drew explicit connections between the scaling of accretion flows, the physics of relativistic jets, the coupling between black hole feedback and star formation, and the secular evolution of gas driven inward by galactic bars. Comparative studies of nearby low-luminosity nuclei, combined with the Milky Way’s uniquely resolvable environment, allow astronomers to trace how a galactic nucleus transitions between quiescence and activity, and how feedback from the central engine regulates the fuel supply in a self-limiting loop that shapes the growth of galaxies over cosmic time.

The meeting also spotlighted the computational and observational infrastructure driving the field forward. General relativistic magnetohydrodynamic simulations of the accretion flow around Sagittarius A star, radiative transfer modeling of its polarization variability, and N-body models of the nuclear star cluster’s formation now operate at resolutions and statistical sophistication that can be directly confronted with GRAVITY, the Event Horizon Telescope, ALMA, Chandra, XRISM and the upcoming ELT datasets. High-cadence monitoring campaigns of the black hole’s infrared and submillimeter flares are converging on a picture in which magnetic reconnection and orbiting plasma instabilities generate the observed variability, linking microphysical processes near the horizon to macroscopic structures visible across the electromagnetic spectrum. As these instruments mature over the coming decade, the Galactic Centre ecosystem that the Brno symposium so comprehensively surveyed will be tested not as a collection of separate puzzles, but as a single evolutionary system whose past activity, present faintness and future episodes of renewed feeding can be read directly from the stars, gas and echoes it leaves behind.

Subject of Research: The Milky Way Galactic Centre as a laboratory for studying supermassive black holes, stellar dynamics and multiphase gas in galactic nuclei.

Article Title: Traversing the Galactic Centre in space and time

Article References: Zajaček, M., Czerny, B., Mondek, M., Mitra, S., Labaj, M., Ondro, T., Janík, J., & Dušek, J. (2026). Traversing the Galactic Centre in space and time. Nature Astronomy. https://doi.org/10.1038/s41550-026-02958-8

Image Credits: AI Generated

DOI: 10.1038/s41550-026-02958-8

Keywords: Galactic Centre, Sagittarius A*, supermassive black hole, nuclear star cluster, GRAVITY, stellar dynamics, circumnuclear disk, Fermi bubbles, X-ray light echoes, IAU Symposium 405, black hole accretion, paradox of youth