I am having problems tracing BH mergers from blackhole_mergers.hdf5 files from TNG100-1. In all there /tree/* datasets I am encountering references to indices that don't exist. For example there are 18374 entries in each /tree/* dataset but the highest value they hold is 24183, 24091 and 24747 for prev_in, prev_out and next dataset respectively. The Data specifications page warns that there are inconsistencies in TNG100-1. I also checked if the previously described /tree/* issue appears in TNG300-1 blackhole_mergers.hdf5 file and it does.
Is this issue a general property of these files? The Data specifications page about these files is quite clear so I hope that the issue is not in my understanding of how to read the dataset.
Hi Jon, I'm sorry for the trouble you're having. I'll look into this issue in the next few days and get back to you as soon as possible.