Voyager 1 Data Reveals Signatures of the Local Gas and Cosmic-Ray Source Distributions

TL;DR

This work addresses how the local interstellar medium (VLISM), particularly the Local Bubble’s gas structure and a deficit of nearby cosmic-ray sources, biases the interpretation of low-energy CR spectra. It employs the GALPROP framework with data-driven 3D gas distributions (GP2D and S25) and a local source deficit to fit Voyager 1 measurements, assessing how local structure affects inferred propagation. The study finds that reproducing the Voyager low-energy spectra requires both an underdense interior and a nearby source deficit, with the nearest contributing sources around 150–200 pc away, and it supports the presence of a primary Boron component alongside Be/O as a secondary constraint. The results underscore the necessity of incorporating realistic VLISM structure when deriving Galactic propagation parameters from local CR data, with implications for interpreting low-energy CRs and global propagation models.

Abstract

We investigate the effects of the nearby interstellar medium (ISM) on the locally measured cosmic-ray (CR) spectra. Using the GALPROP code we explore how variations in the local gas and source distributions affect spectral features at low energies. Comparing with recent Voyager 1 data taken in the local ISM, we show that for a realistic interstellar gas distribution, the nearest source of the low energy CR particles observed nearby the Solar system is constrained to be within the range ~150-200 pc distant. We find that the modelling supports the conclusion of Cummings et al. (2025) that there is a significant fraction of primary Boron in its observed spectrum at low energies. Our study shows that detailed modelling of the immediate Galactic environment is required to robustly infer Galactic CR propagation parameters from local measurements, and that accounting for nearby ISM structure can alleviate tensions between direct CR data and global propagation models.

Figures (6)

• Figure 1: Model configuration components. Left panel: profiles for the CR source and GP2D gas surface densities along the $Y=0$ kpc direction. Line styles: source density, solid; H i surface density, dash-dotted; H$_2$ surface density, long dashed. Right panel: S25 model gas surface density for the nearby Galactic neighbourhood. The Solar system is marked by the Sun symbol in the centre, and the dashed line is a circle of radius 200 pc about it.
• Figure 2: Model combination results together with V1 data for 2013/1--2021/365 in the VLISM from Cummings_2025 for elemental spectra Z=1--10. Line styles: 2D, 3D/0, solid; 3D/100, short dashed; 3D/200, long dashed; S25/0, dotted; S25/100, short dash dotted; S25/200, long dash dotted.
• Figure 3: Model combination results together with V1 data for 2013/1--2021/365 in the VLISM from Cummings_2025 for B/C. Line styles as for Fig. \ref{['fig1a']}.
• Figure 4: Model combination results together with V1 data for 2013/1--2021/365 in the VLISM from Cummings_2025 for elemental spectra Z=11--19 (left) and Z=20--28 (right). Line styles as for Fig. \ref{['fig1a']}.
• Figure 5: Elemental ratios for He, O, and Fe compared with V1 data. For each ratio, the corresponding V1 ratio has been derived from the spectra using power-law interpolation at the geometric mean energy for the elemental denominator species. Uncertainties are obtained by combining in quadrature the 1-sigma uncertainties for each elemental species interpolated using the same method. Left panel: ratios for combinations with 3D/GP2D and S25 gas models with source density hole sizes 0, 100, and 200 pc. Right panel: ratios for S25 gas model with source density hole sizes 0, 100, 150, 200, and 400 pc. Line styles are same as for Fig. \ref{['fig1a']}, with the addition of double-dot dashed and triple-dotted dashed lines for 150 pc and 400 pc source density holes, respectively.