2 Experimental design
The 45 lofts included in the trial were those which had previously reported a "high level" of Sparrowhawk attacks to the Scottish Homing Union (based on subjective judgements made by the loft owners themselves). For logistical reasons, lofts were restricted to particular geographic regions in order to allow the single contract falconer to travel between lofts (See Fig 1a and 1b). Lofts were then allocated randomly to one of the three treatment groups, although in several instances it was necessary to swap lofts between treatments (eg. from the Larsen trap treatment when owners were unable to commit long periods at the loft overseeing the trap; SHUpers. comm). The type of racing system used by individual lofts (eg. 'natural' where lofts are kept open for long periods and birds come and go freely versus 'roundabout' where lofts are open for short intervals within a day) was not formally factored into the design, nor was the type of landscape surrounding the loft (eg. amount of forestry within close proximity) or loft size (number of pigeons released for exercising). All these latter factors could in theory affect attack rates of pigeons by Sparrowhawks. The three treatments were:
1) Visual Deterrents
The use of Mylar tape as a deterrent at the loft.
2) Translocation
The trapping, transfer and release of Sparrowhawks away from the loft.
3) Control
No deterrents used and no Sparrowhawk translocations were carried out.
It was proposed from the onset of the trial to have 15 lofts allocated to each of these three treatment groups.