Local Bubble: Structure, Evolution & Diagnostics

• Local Bubble is a low-density cavity in the interstellar medium, characterized by hot (≈10^6 K) plasma and irregular, tracer-dependent boundaries.
• 3D reconstructions and multi-wavelength diagnostics map its complex morphology, with effective scales from ~100 to over 600 pc influenced by sequential supernova events.
• Numerical models integrating dust mapping, Faraday rotation, and X-ray observations provide insights into its magnetic structure, cosmic ray contribution, and star formation impact.

Searching arXiv for recent and foundational Local Bubble papers to ground the article. arXiv.search query="Local Bubble interstellar medium synchrotron dust X-ray cosmic rays" max_results=10

Searching arXiv more specifically for 3D structure, magnetism, and high-energy diagnostics of the Local Bubble. arXiv.search query="ti:\"Local Bubble\" OR abs:\"Local Bubble\" 3D dust magnetic field X-ray Faraday" max_results=10

The Local Bubble (LB) is a low-density cavity in the local interstellar medium surrounding the Solar System, generally described as a supernova-driven region of hot plasma with temperature $T \sim 10^6$ K and density $n \sim 10^{-2}\,\mathrm{cm}^{-3}$, bounded by colder dusty gas and neutral material (Erlykin et al., 2016, Farhang et al., 2019). Reported characteristic scales depend on tracer and shell definition: the LB has been described as extending about $100$ pc in the Galactic plane and several hundred parsecs vertically, as having an effective radius of $200$ pc, and, in recent dust-traced reconstructions, as having a peak extinction surface at an average distance of $170$ pc from the Sun, spanning $70$–$600+$ pc (Loon et al., 2015, Erlykin et al., 2016, O'Neill et al., 2024). Across extinction, diffuse interstellar band (DIB), dust-polarization, X-ray, radio, Faraday-rotation, cosmic-ray, and radioisotopic studies, the LB emerges not as an empty, uniform void, but as an irregular, evolving superbubble whose shell structure, ionization state, magnetic geometry, and supernova history remain active subjects of quantitative reconstruction and debate.

1. Morphology and defining scales

The LB is commonly characterized as a cavity of hot, low-density gas in which the Solar System resides, surrounded by a shell of colder, dusty material (Pelgrims et al., 2019). One synthesis describes it as an approximately spherical region with an effective radius of $200$ pc, temperature $T \sim 10^6$ K, and density $n \sim 10^{-2}\,\mathrm{cm}^{-3}$ (Erlykin et al., 2016). Optical DIB work similarly describes a cavity extending about $n \sim 10^{-2}\,\mathrm{cm}^{-3}$0 pc in the plane of the Galaxy and hundreds of parsecs vertically, while also emphasizing a highly irregular structure and the presence of neutral clouds within the otherwise tenuous hot gas (Loon et al., 2015).

Recent parsec-resolution dust mapping gives a more asymmetric picture. In that reconstruction, the peak extinction surface falls at an average distance of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$1 pc from the Sun, with a span of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$2–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$3 pc, a typical shell thickness of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$4 pc, a total dust-traced mass of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$5, and a cavity volume of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$6 pc$n \sim 10^{-2}\,\mathrm{cm}^{-3}$7, equivalent to a sphere of radius $n \sim 10^{-2}\,\mathrm{cm}^{-3}$8 pc (O'Neill et al., 2024). The shell is highly irregular and asymmetric, extends from $n \sim 10^{-2}\,\mathrm{cm}^{-3}$9 to $100$0 pc, $100$1 to $100$2 pc, and $100$3 to $100$4 pc, and contains a prominent northern extension morphologically consistent with a “Local Chimney” that reaches $100$5 pc into the lower Galactic halo (O'Neill et al., 2024).

Closer-range extinction mapping also shows that the LB boundary is not at a unique distance in all directions. Within $100$6 pc, median extinction $100$7 reaches up to $100$8 mag in the northern part of the first and second Galactic quadrants and is as low as $100$9 mag in the southern part of the third and fourth quadrants; a practical threshold $200$0 mag was used as an indicator of the onset of the LB wall (Knude, 2013). That result already implied an irregular cavity whose wall is closer than $200$1 pc in some directions and farther than $200$2 pc in others (Knude, 2013).

DIB tomography further undermines the notion of a cleanly evacuated cavity. The $200$3 DIB carrier is present both inside and outside the LB, while the $200$4 carrier is much less abundant inside and is largely confined to the walls or denser cloudlets (Farhang et al., 2019). The “Local Chimney” and narrow tunnels to Loop I are not devoid of DIBs; instead, filaments of the $200$5 DIB fill these structures (Farhang et al., 2019). This establishes the LB as a structured cavity with tracer-dependent boundaries rather than a single, sharply defined surface.

2. Reconstruction methods and geometric modeling

The LB has been mapped with several inverse and forward-modeling strategies. A classical extinction approach used Hipparcos parallaxes, $200$6 photometry, and spectral classification, together with the relation

$200$7

to estimate local extinction and identify the first indications of the LB boundary (Knude, 2013). In that framework, the relatively shallow extinction in the Hipparcos sample allowed dwarf/giant separation directly in a color–magnitude diagram, and stars with $200$8 were excluded before constructing maps of median $200$9 (Knude, 2013).

Three-dimensional DIB mapping casts the reconstruction as an inverse problem. Using 637 early-type stars, with 359 retained for final mapping after quality selection, DIB equivalent widths were inverted with a Bayesian, regularized formalism and a smoothing length $170$0 pc (Farhang et al., 2019). The sightline relation was written as

$170$1

and the ratio $170$2 was used as an environmental diagnostic: low ratios $170$3 identify “$170$4” type clouds in strongly irradiated regions, whereas high ratios $170$5 identify more shielded “$170$6” type clouds (Farhang et al., 2019).

Dust-based shell extraction methods use differential extinction directly. One model sampled the radial profile $170$7 in each HEALPix direction, identified the inner shell surface $170$8 by the first inflection point where $170$9, smoothed the profiles with a Gaussian kernel of $70$0 pc, and then represented the shell surface with a spherical-harmonic expansion

$70$1

with $70$2–$70$3 to control complexity (Pelgrims et al., 2019). The corresponding ordered magnetic field model was then fit to \textit{Planck} 353 GHz polarized dust emission using MCMC (Pelgrims et al., 2019).

A distinct class of models treats the LB as an expanding thin shell in a stratified medium. In the thin layer approximation, momentum conservation is written

$70$4

and asymmetric shapes are obtained by embedding the expansion in exponential, Gaussian, inverse-square, or Navarro–Frenk–White density profiles (Zaninetti, 2020). Comparison to observed asymmetric cuts of the LB boundary was quantified with an “observational percentage of reliability,” with reported values of $70$5 for the exponential profile, $70$6 for the Gaussian profile, $70$7 for the inverse-square profile, and $70$8 for the NFW profile (Zaninetti, 2020).

3. Supernova origin and age estimates

The standard physical picture is that the LB was created by multiple supernovae and stellar winds. One interpretation describes it as the result of a succession of about $70$9 supernovae, starting approximately $600+$0 years ago, plus the combined effects of winds from massive stars (Erlykin et al., 2016). A radioisotopic reconstruction instead modeled the LB as the product of $600+$1 supernovae over the last $600+$2 Myr, with progenitor masses $600+$3–$600+$4, and inferred a present-day radius of $600+$5–$600+$6 pc in an inhomogeneous medium (Schulreich et al., 2018). A later Gaia EDR3-based study identified $600+$7 supernova explosions, with $600+$8 in UCL/LCC and one in V1062 Sco, and found that the Solar System entered the LB about $600+$9 Myr ago (Schulreich et al., 2023).

These reconstructions do not use the same temporal marker. Some estimate the interval since the first supernova in the sequence, whereas others constrain the much shorter time since the last event reheated the cavity. High-resolution non-equilibrium ionization simulations of the joint evolution of the LB and Loop I used $200$0 massive stars for the LB and $200$1 for Loop I, and bracketed the current LB evolution time between $200$2 and $200$3 Myr since the last supernova reheated the cavity (Avillez et al., 2012). A related simulation found that observed OVI columns are reproduced for $200$4 Myr since the last SN (Avillez et al., 2011).

A major current controversy concerns the total age and supernova budget. A 2025 SISSI analysis, combining 3D dust maps with simulations of supernova remnants in a stratified, shearing ISM, argued that $200$5 SNe over $200$6 Myr are required and derived an LB age of $200$7 to $200$8 Myr, in tension with previous older estimates and with the assumption that the LB was powered solely by the nearby Scorpius–Centaurus OB association (Romano, 4 Sep 2025). The same work states that these results cast serious doubts on the claim that star formation in the solar neighborhood was driven by LB expansion and suggests it might instead have been quenched (Romano, 4 Sep 2025).

4. Plasma state, ionization, and X-ray diagnostics

The LB’s hot phase is central to its observational definition, but its ionization structure is not well described by collisional ionization equilibrium alone. Non-equilibrium ionization (NEI) simulations using E(A+M)PEC and adaptive mesh refinement showed that delayed recombination is essential for understanding the spatial distributions of C IV, N V, and O VI in a cooling post-supernova cavity (Avillez et al., 2012). The observationally required constraints

$200$9

are jointly satisfied for a narrow interval $T \sim 10^6$0–$T \sim 10^6$1 Myr after the last local supernova (Avillez et al., 2012). In a related study, simulated OVI columns increase with time after the last SN owing to continued recombination, and the observed range $T \sim 10^6$2 is matched for $T \sim 10^6$3 Myr (Avillez et al., 2011).

The X-ray sky expected from an LB analogue is highly intermittent. In a magnetohydrodynamical simulation, shortly after a supernova approximately $T \sim 10^6$4 of the X-ray luminosity originates from less than $T \sim 10^6$5 of the bubble volume, concentrated in hot regions around recent SN sites (Girichidis et al., 23 Mar 2026). During quiescent phases without recent SNe, X-ray-bright regions become more volume-filling; the total flux varies by several orders of magnitude on Myr timescales, and SN-driven peaks fade within $T \sim 10^6$6 years (Girichidis et al., 23 Mar 2026). Observable soft X-rays are also strongly modulated by line-of-sight absorption: gas with $T \sim 10^6$7 efficiently absorbs soft X-ray photons and limits the visible depth (Girichidis et al., 23 Mar 2026).

A further complication is that the soft X-ray foreground need not be dominated by the LB at all. Uniform XMM-Newton and Suzaku shadowing analyses found that solar wind charge exchange (SWCX) can mimic the thermal LB foreground at CCD resolution and placed an upper limit on the LB O VII intensity of $T \sim 10^6$8 photons/cm$T \sim 10^6$9/s/sr at $n \sim 10^{-2}\,\mathrm{cm}^{-3}$0 confidence (Henley et al., 2015). The same analysis showed that, if the foreground is in fact SWCX-dominated and brighter than $n \sim 10^{-2}\,\mathrm{cm}^{-3}$1 erg/cm$n \sim 10^{-2}\,\mathrm{cm}^{-3}$2/s/deg$n \sim 10^{-2}\,\mathrm{cm}^{-3}$3 in $n \sim 10^{-2}\,\mathrm{cm}^{-3}$4–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$5 keV, using an LB foreground model can bias the inferred halo temperature upward by $n \sim 10^{-2}\,\mathrm{cm}^{-3}$6–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$7 K and the $n \sim 10^{-2}\,\mathrm{cm}^{-3}$8–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$9 keV halo surface brightness downward by $n \sim 10^{-2}\,\mathrm{cm}^{-3}$00–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$01 erg/cm$n \sim 10^{-2}\,\mathrm{cm}^{-3}$02/s/deg$n \sim 10^{-2}\,\mathrm{cm}^{-3}$03 (Henley et al., 2015).

At much smaller scales, ENA observations from IBEX have also been connected to the LB. One proposed mechanism attributes the ribbon to charge exchange between neutral H at the nearby edge of the Local Interstellar Cloud and hot LB protons, requiring an interface $n \sim 10^{-2}\,\mathrm{cm}^{-3}$04–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$05 AU away and an LB proton density $n \sim 10^{-2}\,\mathrm{cm}^{-3}$06 with $n \sim 10^{-2}\,\mathrm{cm}^{-3}$07 non-thermal protons over the IBEX energy range (Grzedzielski et al., 2010).

5. Magnetic structure, polarization, and radio/Faraday signatures

The LB is not only a thermal cavity but also a magnetized structure that shapes the polarized sky. Dust-based modeling of the LB shell extracted its inner surface from 3D extinction maps and fit an analytical shell-field model to \textit{Planck} 353 GHz polarization over the Galactic polar caps (Pelgrims et al., 2019). In that framework the shell is continuous, thick ($n \sim 10^{-2}\,\mathrm{cm}^{-3}$08–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$09 pc), and present even at high Galactic latitudes, and the preferred initial large-scale field orientation is $n \sim 10^{-2}\,\mathrm{cm}^{-3}$10, $n \sim 10^{-2}\,\mathrm{cm}^{-3}$11 (Pelgrims et al., 2019). The resulting shell produces a dominant large-angular-scale polarized foreground at high latitudes, described as a “magnetic veil,” which is important for CMB foreground separation (Pelgrims et al., 2019).

Radio synchrotron studies show that the LB contributes to the radio sky but is very unlikely to explain the full low-frequency excess reported by ARCADE 2. The observed excess antenna temperature is $n \sim 10^{-2}\,\mathrm{cm}^{-3}$12 mK at 3 GHz, and the observed antenna-temperature spectrum is described by $n \sim 10^{-2}\,\mathrm{cm}^{-3}$13 with $n \sim 10^{-2}\,\mathrm{cm}^{-3}$14, corresponding to $n \sim 10^{-2}\,\mathrm{cm}^{-3}$15 (Krause et al., 2021). Although an IC 10-based scaling argument gives a predicted LB emission of about $n \sim 10^{-2}\,\mathrm{cm}^{-3}$16 K at $n \sim 10^{-2}\,\mathrm{cm}^{-3}$17 MHz, overpredicting the measured background by a factor of two, more precise modeling constrained by Voyager 1, AMS, pulsar Faraday rotation, and turbulence indicates that the LB contribution is at most at the percent level unless unrealistically large magnetic fields or cosmic-ray densities are adopted (Krause et al., 2021). The same work argues that matching the ARCADE 2 background would require a mean magnetic field strength $n \sim 10^{-2}\,\mathrm{cm}^{-3}$18 between $n \sim 10^{-2}\,\mathrm{cm}^{-3}$19–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$20 nT, but that such a state would imply strong directional variations and high polarization inconsistent with several observations; equipartition arguments instead suggest $n \sim 10^{-2}\,\mathrm{cm}^{-3}$21–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$22 nT (Krause et al., 2021).

Faraday-rotation modeling reinforces the local importance of the LB shell. Synthetic RM maps for an observer placed inside an LB analogue show that the cavity walls and nearby superbubble edges are of fundamental importance for interpreting the global Faraday sky, that the LB has a non-negligible contribution to the sinusoidal RM patterns as a function of Galactic longitude, and that the RM signal from diffuse synchrotron emission strongly corresponds to the RM generated by the LB candidate walls (Maconi et al., 13 Apr 2025). Prospective SKA-Low surveys over $n \sim 10^{-2}\,\mathrm{cm}^{-3}$23–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$24 MHz are designed to exploit this regime: high-sensitivity polarized images, RM synthesis, and Faraday depth cubes are expected to reveal the 3D structure of the magnetized medium in the LB (Sun et al., 24 Jun 2026).

6. Cosmic rays, star formation, and terrestrial records

The LB is a prominent ingredient in local cosmic-ray interpretations. One model proposes that the bulk of cosmic rays below about $n \sim 10^{-2}\,\mathrm{cm}^{-3}$25 GV of rigidity comes from a modest number of supernova remnants in the LB, while a separate “Local Source,” a single supernova remnant generated about $n \sim 10^{-2}\,\mathrm{cm}^{-3}$26 years ago, dominates from $n \sim 10^{-2}\,\mathrm{cm}^{-3}$27 GV up to at least $n \sim 10^{-2}\,\mathrm{cm}^{-3}$28 GV (Erlykin et al., 2016). In that scenario, the LB’s hot $n \sim 10^{-2}\,\mathrm{cm}^{-3}$29 K environment increases injection efficiency, potentially toward $n \sim 10^{-2}\,\mathrm{cm}^{-3}$30, and anomalous diffusion lowers the effective diffusion coefficient $n \sim 10^{-2}\,\mathrm{cm}^{-3}$31, enhancing low-energy particle residence times (Erlykin et al., 2016). Radio synchrotron calculations, however, indicate that this does not translate into dominance of the unexplained radio background; the LB remains a subdominant synchrotron contributor (Krause et al., 2021).

The relation between LB expansion and recent star formation is disputed. In the Taurus region, the age–distance-to-the-Local-Bubble relation for Distributed Young Stellar Objects shows a clear trend: the farther they are from the Local Bubble, the younger they are, which is consistent with the supernovae-driven formation scenario of the Local Bubble (Liu et al., 2024). Grouped Young Stellar Objects are significantly younger, mostly $n \sim 10^{-2}\,\mathrm{cm}^{-3}$32–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$33 Myr old and confined to $n \sim 10^{-2}\,\mathrm{cm}^{-3}$34–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$35 pc, and may also be products of the Local Bubble but formed in more recent and localized events (Liu et al., 2024). By contrast, the SISSI reconstruction argues that LB expansion likely quenched rather than triggered star formation in the solar neighborhood (Romano, 4 Sep 2025). The observational and numerical literature therefore supports both triggered and quenched local-star-formation interpretations, depending on the adopted LB age and dynamical history.

Terrestrial and lunar radioisotopes supply an independent clock for nearby supernova activity. Deep-sea archives show an enhanced concentration of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$36Fe in layers dating from about $n \sim 10^{-2}\,\mathrm{cm}^{-3}$37 Myr ago, with a main signal spanning $n \sim 10^{-2}\,\mathrm{cm}^{-3}$38–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$39 Myr ago and a smaller signal around $n \sim 10^{-2}\,\mathrm{cm}^{-3}$40–$n \sim 10^{-2}\,\mathrm{cm}^{-3}$41 Myr ago (Schulreich et al., 2018). Hydrodynamical simulations that inject sequential supernovae at kinematically reconstructed positions reproduce both the timing and intensity of these signals and link them to LB formation and to passage through neighboring superbubbles (Schulreich et al., 2018, Schulreich et al., 2023). In the Gaia EDR3-based model with $n \sim 10^{-2}\,\mathrm{cm}^{-3}$42 explosions, the Solar System entered the LB about $n \sim 10^{-2}\,\mathrm{cm}^{-3}$43 Myr ago, the present-day LB has a radius of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$44 pc, and the measured recent influx of $n \sim 10^{-2}\,\mathrm{cm}^{-3}$45Fe is naturally explained by turbulent radioisotopic transport (Schulreich et al., 2023). These results make the LB an unusually well-constrained superbubble: its history is recorded not only in interstellar gas and dust, but also in the geological archive of the Earth.