It is a good read representing the results from 3,430 randomly allocated 1km squares. Each square is surveyed twice, the early surveys, in April to mid May, are likely to record fewer Swifts than the later ones in mid May to late June when the Swift population is at its peak.
Unlike song birds, Swifts recorded in a 1km square, out in the country, are unlikely to be breeding in that square. Indeed, they are almost as likely to be non-breeders as breeders, but one hopes that the frequency of occurrence of Swifts during BBS surveys is a good proxy for the general population level. For Swifts it is more a total population level index than a breeding bird index.
|BBS Index for Swift, UK, 1994 - 2012|
Index values are scaled so that the smoothed index in 1995 is 100
So the 2011 UK population is estimated at 61% of the 1995 population. Note that this is calculated from the values behind the smoothed red trend line. The years 1994 and 2012 are not used as the smoothed values at the extremes are not considered reliable. Thus we have scaled the graphs so the smoothed index value in 1995 is set to 100.
Although the 2012 annual index is 120% of the 2011 one, it would not be appropriate to multiply the 2011 smoothed value by 120%, as the 120% figure comes from the annual index values
|See BBS 2012 report for confidence limits|
However, all of the numbers in this table have wide confidence limits, so we will have to wait for 2013 and beyond for smoothed index values in 2012.
It is tempting to speculate that the long wet summer of 2012 forced Swifts out into the countryside to feed in areas where there were BBS survey squares. However, this does not explain the situation in Scotland.
Thank you Kate Risely, BBS National Organiser, for providing the data for the graphs.