Identifying Merger-Driven Long Gamma-Ray Bursts based on Machine Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909645178667008 |
|---|---|
| author | Zhu, Si-Yuan Deng, Hui-Ying Zhang, Fu-Wen Mo, Qian-Zi Tam, Pak-Hin Thomas |
| author_facet | Zhu, Si-Yuan Deng, Hui-Ying Zhang, Fu-Wen Mo, Qian-Zi Tam, Pak-Hin Thomas |
| contents | Gamma-ray bursts (GRBs) are classified as Type I GRBs originated from compact binary mergers and Type II GRBs originated from massive collapsars. While Type I GRBs are typically shorter than 2 seconds, recent observations suggest that some extend to tens of seconds, forming a potential subclass, Type IL GRBs. However, apart from their association with kilonovae, so far no rapid identification is possible. Given the uncertainties and limitations of optical and infrared afterglow observations, an identification method based solely on prompt emission can make such identification possible for many more GRBs. Interestingly, two established Type IL GRBs: GRB 211211A and GRB 230307A, exhibit a three-episode structure: precursor emission (PE), main emission (ME), and extended emission. Therefore, we comprehensively search for GRBs in the Fermi/GBM catalog and identify 29 three-episode GRBs. Based on 12 parameters, we utilize machine learning to distinguish Type IL GRBs from Type II GRBs. Apart from GRB 211211A and GRB 230307A, we are able to identify six more previously unknown Type IL GRBs: GRB 090831, GRB 170228A, GRB 180605A, GRB 200311A, GRB 200914A, and GRB 211019A. We find that Type IL GRBs are characterized by short duration and minimum variability timescale of PE, a short waiting time between PE and ME, and that ME follows the $E_{\rm p,z}$--$E_{\rm iso}$ correlation of Type I GRBs. For the first time, we identify a high-significant PE in the confirmed Type IL GRB 060614. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_08675 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Identifying Merger-Driven Long Gamma-Ray Bursts based on Machine Learning Zhu, Si-Yuan Deng, Hui-Ying Zhang, Fu-Wen Mo, Qian-Zi Tam, Pak-Hin Thomas High Energy Astrophysical Phenomena Gamma-ray bursts (GRBs) are classified as Type I GRBs originated from compact binary mergers and Type II GRBs originated from massive collapsars. While Type I GRBs are typically shorter than 2 seconds, recent observations suggest that some extend to tens of seconds, forming a potential subclass, Type IL GRBs. However, apart from their association with kilonovae, so far no rapid identification is possible. Given the uncertainties and limitations of optical and infrared afterglow observations, an identification method based solely on prompt emission can make such identification possible for many more GRBs. Interestingly, two established Type IL GRBs: GRB 211211A and GRB 230307A, exhibit a three-episode structure: precursor emission (PE), main emission (ME), and extended emission. Therefore, we comprehensively search for GRBs in the Fermi/GBM catalog and identify 29 three-episode GRBs. Based on 12 parameters, we utilize machine learning to distinguish Type IL GRBs from Type II GRBs. Apart from GRB 211211A and GRB 230307A, we are able to identify six more previously unknown Type IL GRBs: GRB 090831, GRB 170228A, GRB 180605A, GRB 200311A, GRB 200914A, and GRB 211019A. We find that Type IL GRBs are characterized by short duration and minimum variability timescale of PE, a short waiting time between PE and ME, and that ME follows the $E_{\rm p,z}$--$E_{\rm iso}$ correlation of Type I GRBs. For the first time, we identify a high-significant PE in the confirmed Type IL GRB 060614. |
| title | Identifying Merger-Driven Long Gamma-Ray Bursts based on Machine Learning |
| topic | High Energy Astrophysical Phenomena |
| url | https://arxiv.org/abs/2506.08675 |