Individual study: Effect of pane angle, tinted glass and feeder placement on rate of bird collisions with windows in Lehigh County, Pennsylvania, USA
Klem D. Jr., Keck D.C., Marty K.L., Miller Ball A.J., Niciu E.E. & Platt C.T. (2004) Effects of window angling, feeder placement, and scavengers on avian mortality at plate glass. Wilson Bulletin, 116, 69-73
This study is summarised as evidence for the intervention(s) shown on the right. The icon shows which synopsis it is relevant to.
Angle windows to reduce collisions by birds
A randomised, replicated and controlled experiment in 1991 in Pennsylvania, USA (Klem et al. 2004), found that a fewer birds collided with windows angled at 20o or 40o from the vertical (28% and 15% of 53 recorded collisions respectively) than with vertical windows (57% of collisions). Six plate glass, wooden framed windows (1.4 x 1.2 m, 1.2 m off the ground, 15-43 m apart) were used, between January and May, on the edge of deciduous woodland and farmland.
Place feeders close to windows to reduce collisions
A randomised, replicated and controlled experiment between October and December 1991 in Pennsylvania, USA (Klem et al. 2004), found that there were fewest collisions and fatal collisions with windows when platform feeders were placed 1 m away from the window (24% of the 105 collisions, none fatal), compared with when feeders were 5 m (28% of collisions, 33% of fatalities) or 10 m from the window (48% and 67%). Similarly, in a repeat experiment in February 1992, there were fewest collisions and fatalities when feeders were 2m from the window (23% of 197 collisions, 5% of 21 fatalities) than 3 m (46% and 43%) or 4 m (31% and 52%) away. The proportion of collisions that were fatal increased with distance that feeders were from windows from 0% at 1 m away and 2% at 2 m to 59% at 5 m and 69% at 10 m. Six plate glass, wooden framed windows (1.4 x 1.2 m, 1.2 m off the ground, 55 m apart) and six platform feeders (with the platform level with the base of each window) were used on the edge of deciduous woodland and farmland.
Mark or tint windows to reduce collision mortality
A randomised, replicated and controlled experiment between January and May 1991 in Pennsylvania, USA (Klem et al. 2004), found that a smaller proportion of collisions were with tinted windows (32% of 53 recorded collisions) than with clear windows (68% of collisions). The same study found that, when platform feeders were placed at varying distances in front of the windows (see ‘Provide supplementary food - Place feeders close to windows to reduce collisions’), only four of 52 fatal collisions (8%) occurred with tinted windows, the rest with clear glass windows. However, the authors note that the tinted glass was of a poor reflective quality and they believe this may have resulted in fewer fatalities than a highly reflective tinted glass. Experiments used six plate glass, wooden framed windows (1.4 x 1.2 m, 1.2 m off the ground, 15-43 m apart) on the edge of deciduous woodland and farmland.