In that thread from august I find two objections (Maybe this isn’t the place to continue that discussion, anyway here I go): 1. You seem to think that Svensmark used an invalid method to select Forbush events. 2. You seem to think that there a too few events in the analysis.

Regarding number 1. The events were ranked according to the level of ionization of the lower atmosphere. I don’t see what’s wrong with that? Would GCRs count be a better method of selection or how are you supposed to do this?

Regarding number 2. There are not that many events to analyze to start with and the Aeronet network has taken measurements only from 1998 onwards. In other scientific areas it is a valid method to look at strongest events while studying a phenomenon, e.g. you can study the strongest earthquakes, strongest hurricanes, strongest meteorite impacts etc. Why isn’t it OK to analyze the five strongest Forbush decreases?]]>