Provenance (reproducibility)

Six months after you make a figure, the question is always the same: which snapshot, which Mera version, what units produced this? provenance answers it. It reads the metadata every Mera result already carries (its InfoType) and returns a compact, deterministic record you can print, compare, or stamp onto a figure or a FITS header.

using Mera
gas = gethydro(getinfo(100, "/data/sim"))

provenance(gas)
# Provenance:
#   Mera version : 1.8.0
#   simulation   : /data/sim
#   output       : 100  (RAMSES, written 2025-06-21T18:31:55.533)
#   time         : 148.08 Myr
#   box / levels : L=100.0  ndim=3  levels 3–7
#   scale type   : ScalesType003

The time is human-readable: physical time in Myr/Gyr for a normal run, and redshift (plus expansion factor and age) for a cosmological one:

#   time         : z=0.1426  (aexp=0.8752, age 11.925 Gyr)

Where it applies

provenance works on any object that carries an InfoType — not just projections. That is every data object, projection map, and LOS/velocity cube, plus an InfoType itself:

make it with	type	provenance?
`getinfo`	`InfoType`	✓
`gethydro`	`HydroDataType`	✓
`getparticles`	`PartDataType`	✓
`getgravity`	`GravDataType`	✓
`getclumps`	`ClumpDataType`	✓
`getrt`	`RtDataType`	✓
`projection`	`AMRMapsType` (the map)	✓
`velocity_cube` / `los_cube`	`LosCubeType`	✓
`pdf` (cell/particle or 2D-map form)	`NamedTuple` with `.info`	✓
`position_velocity`	`NamedTuple` with `.info`	✓
`face_on` / `edge_on`	`GalaxyFrame`	✓

provenance(getparticles(info))         # particles
provenance(projection(gas, :sd))       # a projection map
provenance(velocity_cube(gas))         # a LOS / velocity cube
provenance(pdf(gas, :rho))             # a PDF result
provenance(face_on(gas))               # a galaxy frame
provenance(gas.info)                   # the InfoType directly

These derived results carry an .info field, so the same provenance call works on them. The exceptions, which keep no source snapshot, raise a clear error (pointing you to provenance of the source data):

the raw-matrix form pdf(map2d) (a bare 2D array has no snapshot);
a timeseries table — it spans many outputs, so a single provenance does not describe it; inside the reducer you have the full data object, so e.g. timeseries(path, d -> (prov = provenance_string(d), …)) records per-snapshot provenance.

Stamping a figure or FITS header

provenance_string renders a one-liner — drop it into a figure caption, a log, or a COMMENT card when you savefits:

provenance_string(gas)
# "Mera v1.8.0 | sim/output_00100 | 148.08 Myr | L=100.0 ndim=3 lmin=3 lmax=7 | ScalesType003"
# cosmological run → "… | z=0.14256 | …"

# e.g. as a caption
text(0, 0, provenance_string(gas); fontsize=8)

What it records

field	meaning
`mera_version`	the Mera version that read the data
`path`, `output`, `simcode`	which simulation, output number, and code (RAMSES)
`cosmological`	whether the run is cosmological (`iscosmological`)
`time_myr`	physical snapshot time in Myr (the age of the universe for a cosmological run)
`redshift`, `aexp`	cosmological redshift `z = 1/aexp − 1` and expansion factor (`0` / `1` for a normal run)
`boxlen`, `ndim`, `levelmin`, `levelmax`	box size, dimensionality, AMR level range
`scale_type`	the serialized scale-type version (e.g. `:ScalesType003`) — relevant for reading older mera files
`file_ctime`	when the snapshot was written on disk

The record is deterministic — it depends only on the snapshot's own metadata, never on the wall clock — so two runs over the same output produce identical provenance, and it is safe to use in tests and comparisons.

Provenance (reproducibility)

Where it applies

Stamping a figure or FITS header

What it records

See also