Erratum: Detecting reconnection events in kinetic Vlasov hybrid simulations using clustering techniques (Astrophysical Journal (2021) 908 (107) DOI: 10.3847/1538-4357/abd24b)

While the physical picture that emerges is correct, we regret the presence of a bug in the code for the estimation of the aspect ratio of the candidate structures found. In particular, a factor √2 was inadvertently set in the estimation of the thickness, leading to an underestimation of the AR by a factor 1/√2. The revised Tables 3, 4, and 5 with the corrected values of AR are reported in the following, together with the corrected Figure 6. As anticipated the physical picture is the same, with only a slightly different value of the "optimal"AR threshold. The new "optimal"AR threshold is ≃26. In the published article we shared the link to our codes, which will become a tool, namely Unsupervised ML Reconnection, in Python language, in the context of the EU AIDA project. We share here the links to the new version of the scripts where the small bug has been corrected: https://gitlab.com/aidaspace/aidapy/-/tree/unsuperv_rec_2/aidapy/unsupmr and, for Zenodo, https://doi.org/10.5281/zenodo.5118596 (version 2.0.0). We also take advantage of this opportunity to signal a misspelling: "Davis-Bouldin index"should be "Davies-Bouldin index." (Table Presented) (Figure Presented).