This may be source of your confusion. No - its not correct that if you tag all emissions, O3_OTHR should be zero.
As I wrote above: “O3_OTHER” is concentrations attributed to all other emissions streams and other processes in the model.
First, there are internally calculated emissions that you can tag, but didn’t in your control file. These are things like inline biogenics, inline lightning, dust, etc.
Second, there are processes in the model that create O3 not just from user specified emissions streams like your “GRIDDED_EMIS” and also not just from inline emissions I listed above. The most obvious examples are actually tracked - these are ICON and BCON. Additionally, there ozone that comes from background methane that is specified in the model as a constant.
Finally, ISAM is an approximation for attribution. In the formulation, assumptions are made about which species are most important in ozone formation. However, pretty much ALL species in the gas-phase chemical mechanism of your choice contribute to some degree to ozone formation. Their contribution will go into “OTHER” even though you specified their source emissions stream. Now, this gets into the Known Issue thing I posted. There is currently a problem where “OTHER” from such sources is actually over-estimated and we are working on this issue.
There is a test you can do to illustrate all this stuff if you really want. You can zero out all of the emissions in your input files and lower the values in your ICON/BCON files to close to zero. When you run the model, you will still get ozone. This is a very rough approximation of the “OTHER” ozone.