Coordinate transformation and binning procedure of the model: from the simple conical geometry in the source reference frame, uniformly populated with clouds, to the 3D model cube in the observer’s reference frame. The model is then binned to assign to each cloud a position in the (x, y, v_z) space, as highlighted for the purple cloud. The model cube has the same spatial and spectral resolution as the observed data cube in order to weight the clouds according to the observed flux in the corresponding spaxel. Dashed grey lines show two adjacent voxels, corresponding to different spatial and/or spectral channels.