All gone

Managed to find the structure of the poles in my marginal “cirkus tent” distribution. Before and after graphs: