Magnetic structure and asymmetric eruption of a 500 Mm filament rooted in weak-field regions

Context. The eruption of large-scale solar filaments that extend beyond the core of an active region (AR) into weak-field regions presents a valuable opportunity to investigate flare and eruption dynamics across various magnetic environments. Aims. We performed a detailed analysis of the magnetic structure and asymmetric eruption of a large (∼500 Mm in extent) inverse S-shaped filament partially located in AR 13229 on February 24, 2023. Our primary goal was to relate the filament’s pre-eruptive magnetic configuration to the observed dynamics of its eruption and the formation of a large-scale coronal dimming in a weak-field region. Methods. The evolution of the eruption and associated coronal dimming were analyzed using multiwavelength observations from the Atmospheric Imaging Assembly (AIA) and Hα observations from Kanzelhöhe Observatory. A detailed dimming analysis was performed based on an AIA 211 Å logarithmic base-ratio image sequence. To reconstruct the coronal magnetic field, we applied a physics-informed neural network (PINN)-based nonlinear force-free field (NLFFF) extrapolation method to a large computational volume (∼730 Mm × 550 Mm), using a pre-eruption photospheric vector magnetogram from the Helioseismic and Magnetic Imager (HMI) as the lower boundary. Results. Our results show a highly asymmetric eruption. The eastern part of the filament erupts freely, creating a coronal dimming associated with its footprint, which subsequently expands (total area of ∼9 × 10⁹ km²) together with an inverse J-shaped flare ribbon. The eastern dimming area covers a weak magnetic field region with a mean unsigned flux of only ∼5 G. The NLFFF extrapolation shows the presence of a large-scale magnetic flux rope (MFR) of ∼500 Mm in length, consistent with the observed filament. We identified an extended MFR footprint to the east in the NLFFF extrapolation that connects to an inverse J-shaped flare ribbon (hook) observed during the eruption, outlining the area from which the coronal dimming originated. Overlying strapping fields connect to the region into which the coronal dimming and flare ribbon subsequently expand. This configuration offers a plausible explanation for the formation of the dimming as a stationary flux rope and strapping flux dimming. The subsequent expansion of the stationary flux rope dimming is caused by the growth of the MFR footprint through strapping-strapping reconnection. In contrast, the western filament leg shows multiple anchor points along a narrow channel and potentially strong overlying magnetic fields, which could have resulted in the suppressed dimming and partial confinement by overlying loops observed on this side of the filament during the eruption. Conclusions. The reconstructed pre-eruptive NLFFF configuration provides a clear physical explanation for the observed asymmetries in the eruption, flare geometry, and coronal dimming. This successful application shows that PINN-based NLFFF extrapolation can be effective for modeling large-scale filaments extending into weak-field regions, and that combining this method with detailed observational analysis can greatly improve our understanding of complex solar eruptions.