Number density discrepancy when loading particles from file with periodic boundary conditions

[n-loo]

Hello,

I am loading particles from a binary data file into my simulation. It varies spatially in the x direction, but is uniform in the y direction. When I select open boundary conditions for the top and bottom (bc_y_max, bc_y_min) I get the expected density distribution. If instead I select periodic boundary conditions I get an anomalously high number density on the two rows of y edge cells. even with identical particle data files. See below for pictures of t=0 of the 2D simulation. Red is higher density, blue is lower density.
in my output block: number_density = always + species

The number of particles per cell is identical between the two cases. Red is 4 particles per cell (as specified in the data file) and blue is 0 particles per cell.
in my output block ppc = always + species

data is visualized in VisIT.

What is the reason for the number density discrepancy in edge cells with periodic boundary conditions?

Best,
Niels

[TomGoffrey]

Could you attach the input deck and binary file you are using please?

[n-loo]

zip.zip
Hi! Please see attached.