Cross-match files

THESAN provides a number of simulation runs sharing the same initial conditions but with differences in their physical model. We provide files that allow to easily match haloes in different runs.
These files are produced in the same way as the merger trees, with the difference that haloes in coeval snapshots were linked together, rather than haloes in successive snapshots within the same simulation. The results were then re-arranged into a HDF5 file for each simulation run.

The cross-link file for a run SIM1 contains a group for each other THESAN runs (SIM2) with the same initial conditions. These are named to_SIM2. Within each of these groups, there is one dataset for each snapshot of the simulation. Each dataset contains a number of integer values equal to the number of subhaloes in SIM1 for that snapshot. Each number in the dataset corresponds to the global index (see offsets) of the subhalo in SIM2 matched to the currespondent subhalo in SIM1 in the same snapshot.
This means that the subhalo with global index IDX1 in the snapshot SNAP in SIM1 is associated to the subhalo in SIM2 with global index cross_link_file/to_SIM2/snap_SNAP[IDX1].