SISSI: Supernovae in a stratified, shearing interstellar medium. II. Star formation near the Sun is quenched by expansion of the Local Bubble

TL;DR

This work reevaluates the Local Bubble's history and its impact on nearby star formation by linking LB momentum and size, derived from 3D dust maps, to a suite of high-resolution SNR simulations within the SISSI framework. It supplements these constraints with a Gaia DR3-based census of nearby star clusters to recover the star-formation history and SN-rate in the solar neighborhood, including the $\alpha$Persei family and Sco-Cen. The authors find a substantially younger LB age, $t_{ ext{age}} \sim 2.8-5.8\,\text{Myr}$, requiring $N_{ ext{SN,LB}} \sim 7-59$ SNe, which is difficult to reconcile with Sco-Cen-alone powering and suggests pre-LB SF dominance with subsequent quenching from LB expansion. They conclude that, on large scales, SN-driven SBs likely suppress star formation, though localized triggering near the current LB edge may still occur, motivating further galaxy-scale simulations of old superbubbles and their feedback on the ISM. Overall, the study reframes the LB as a younger, more SN-rich structure whose expansion predominantly quenches local SF rather than triggering it, while leaving room for edge-cloud interactions.

Abstract

The age of the Local Bubble (LB) constrains the timescale on which the interstellar medium in the solar neighborhood evolves. Previous estimates placed the age of the LB at \sim 14 Myr, and attributed its expansion to \sim 15-20 supernovae (SNe), yet a companion paper suggests this age may be overestimated. We place new constraints on the age of the LB and re-evaluate the question whether its expansion triggered or suppressed local star formation. We reconstruct the LB's geometry and momentum using publicly available 3D dust maps and compare them to the high-quality sample of simulated supernova remnants in the SISSI project. Independent constraints on the star-formation history and supernova rate are obtained from a Gaia DR3-based census of nearby star clusters. We find that \sim 7-59 SNe over \sim 5.8 Myr to \sim 2.8 Myr, respectively, are required to explain both the LB's momentum and size and confirm that such a high supernova rate can be sustained by local star clusters. Our analysis yields a substantially smaller LB age than previous estimates, requiring a correspondingly larger number of SNe, driving its expansion. We show that this result is in tension with the conclusion that the LB is powered solely by SNe from the Scorpius-Centaurus OB association, which ceased star formation around the time the LB formed. If our estimates are correct, it follows that the majority of star formation in the solar neighborhood happened before the formation of the LB and was not triggered by its expansion. Instead, the SNe that powered the LB appear to overall have quenched the ongoing star formation process. This does not rule out that star formation in the clouds, located near its current edge, could have been affected by the LB expansion.

Figures (9)

• Figure 1: Time evolution of effective size of the simulated sample of SNRs of the SISSI sample between 2 and 10 Myr, for SNRs with ambient densities within 0.3 dex of the LB (Tab. \ref{['tab:local_bubble']}). Gray, red and blue lines correspond to different explosion models. Solid lines correspond to the median size of the simulated bubbles, with shaded areas corresponding to the range between the 30th and 70th percentiles. Dotted lines and the hatched blue contour correspond to theoretical models based on radiative blastwaves in uniform media (Appendix \ref{['app:SB_models']}). For model N10, the size of the LB corresponds to an age of $\sim 4.5\,\text{Myr}$, while for N1x10 it corresponds to $\sim 6-7\,\text{Myr}$.
• Figure 2: Effective size as a function of age / number of SNe extrapolated from the model N1x10 and the mean of the momentum constraint Eq. \ref{['eq:N_SN']} for various values of $\alpha$, namely, $R_{\alpha} = R_{\text{N1x10}}\left(t = t_{\text{age}}\right) \times \left[\dot{N}_{\text{SN, LB}}\left(t_{\text{age}}\right) \, / \, 1 \, \text{Myr}^{-1}\right]^\alpha$, where $\dot{N}_{\text{SN, LB}} = N_{\text{SN, LB}} / t_{\text{age}}$. For $\alpha \lesssim 0.225$ the extrapolated sizes match the LB for ages in the range $3.5 - 5.5 \, \text{Myr}$ while for $\alpha \gtrsim 0.225$ the two constraints on the size and momentum cannot be satisfied simultaneously.
• Figure 3: Smoothed density profiles using different boundary treatments for two characteristic lines-of-sight. The boundary is indicated by the dashed vertical line. The original data is shown as a gray line, and the extrapolations beyond the boundary are shown at decreased opacity. The top panel shows a line-of-sight with a peak near the inner boundary, while the bottom panel shows a line of sight with a steep decline near the boundary. With reflective boundary conditions, the peak is missed entirely, while simple normalization adds a spurious peak even in the case of the steep decline.
• Figure 4: Comparison of the geometric momentum estimate Eq. \ref{['eq:momentum']} to the true momentum for the simulated sample of SBs from the SISSI simulation at different points in time, as indicated by the color scheme. A representative error bar corresponding to uncertainties due to different ways of defining the center (geometric center vs. center of mass) of the SBs and their volume (threshold value for passive-scalar tracer-variable) is shown in the upper-left corner. The geometrical momentum estimate is slightly biased towards larger values, by a factor of $\sim 2$.
• Figure 5: Time-evolution of the SN count estimate Eq. \ref{['eq:N_SN']} for the simulated sample of SBs in different density ranges. The true SN count is shown as a black dashed line. At extremely high and low ambient densities the estimate appears to overestimate the number of SNe, while for moderate densities, around the density of the LB, the estimate is quite accurate, within about $\lesssim 50\%$.