Weighting unit in 2D

[pordyna]

It looks to me that the particle weight in an output from a 2D simulation (as well as the accumulated value in a particle histogram) is actually not the particle number, but rather linear particle density along the 3rd not simulated direction. To be honest, this makes a lot of sense, but in the openPMD output the weighting is marked as dimensionless. Could this be a bug or is it intentional?

[pordyna]

What also could happen is that I was supposed to set somewhere the cell size in the not simulated dimension but didn't, and it was set to 1m by default?

Btw. I'm actually not sure of the openPMD unitDiemension for the weighting attribute (can check it once perlmutter is back online), but I had a particle field defined as:
fields.particle_fields.particle_number(x,y,z,ux,uy,uz) = 1 and it has a unitless unitDimension in the output. While it seems to be either a linear density (or a number of particles in a 1m long cell :D in that case dimensionless is correct)
Here is more context for the field output:

fields.diag_type = Full
fields.fields_to_plot = By Ex Ez
fields.file_min_digits = 8
fields.format = openpmd
fields.intervals = ::26631,57587:73566:533,36282:57587:5326,62913:62913:,73566::5326,5656:17640:533,9650:9650:,17640:36282:5326
fields.openpmd_backend = bp
fields.particle_fields.average_gamma(x,y,z,ux,uy,uz) = sqrt(1 + ux**2+uy**2+uz**2)
fields.particle_fields.average_gamma.do_average = 1
fields.particle_fields.average_momentum_x(x,y,z,ux,uy,uz) = ux
fields.particle_fields.average_momentum_x.do_average = 1
fields.particle_fields.average_momentum_z(x,y,z,ux,uy,uz) = uz
fields.particle_fields.average_momentum_z.do_average = 1
fields.particle_fields.particle_number.do_average = 0
fields.particle_fields_to_plot = average_gamma average_momentum_x average_momentum_z particle_number
fields.write_species = 0

[ax3l]

Hi @pordyna,

Thanks for the report. Yes, you got this right. Weighting is actually a number of particles (count, unitless) only in 3D, and in 2D a count per length and in 1D a count per surface area.

We need to change this in both WarpX:

• openPMD: Fix Particle Weight Dims in 1D/2D #4717
  and openPMD-standard:
• Particle Weight: Clarify Low-D Dimension openPMD/openPMD-standard#285

Thanks for the report!!