The standard methodology for recording butterfly numbers is the "Pollard walk",  simply noting all butterflies seen on a regular walking route, with the walk repeated every week.  This is known as a butterfly transect, although the route chosen is probably too long (at over 4.6 kilometres) to be acceptable as a formal transect.  In order that the data collected is consistent over time, the methodology has strict requirements on the conditions, and how far away from the observer butterflies can be counted.  However, I have taken a more flexible view, and count everything I can see and identify, although I try to stay within the requirements for time of day, percentage direct sunshine, and wind-speed.

 

The walk is repeated once a week, for each of the 26 weeks between 1st April and the end of September.  Over the seven year period there have only been three weeks (back in early 2015) when the route has not been walked.  The counts only reflect actual walks, so the results for 2015 are slightly lower than they may have been had I succeeded in walking all 26 weeks.

 

The count is a simple log of the total number of each species seen in each section of the route, but this clearly requires me to correctly identify each butterfly.  Where I have been unable to be certain of the species (such as distinguishing in flight between Small White and Green-veined White) a best guess is recorded, based on previous sightings, knowledge of the immediate vegetation conditions and micro-climate, and an assessment of the probability of seeing a particular species at that time of year.