The role of urban scavengers in estimating the number of birds killed by window collisions

Most people can remember an instance where a bird collided with one of the windows of their home. Most people don’t know is that this has been identified as one of the largest human-related causes of bird deaths in Canada. Many studies have attempted to estimate the exact number of mortalities however, instead of coming up with an accurate number, multiple biases have been identified. The largest of which being the removal of a dead bird from below a window by a scavenger before it can be recorded.


-A house cat drags a bird carcass away from a window where it had collided

The number of birds being removed by scavengers has been accounted for in previous estimates, however these previous scavenger studies have all taken place at wind turbines. There have been few carcass removal studies done in an urban environment and none of these have been used to determine a correction factor that can be used in determining a more accurate window collision estimate.

To learn more about the role scavengers play in an urban environment we conducted a carcass removal study at houses within Edmonton throughout 2015. The premise was simple: a dead bird and a time-lapse camera were placed below a window in the front yard of each house. After 1 week we returned to see if the carcass had been removed.

In the end 67.5% of carcasses were removed within 1 week, with the average time to removal being 3.46 days. The most common scavengers were Black-billed Magpies (61.6% of removals) and domestic or feral cats (16.1% of removals). There were also removals by American Crows, Blue Jays and Red squirrels.


-Black-billed magpies and squirrels were some of the scavengers responsible for removing bird carcasses from collision sites

Carcasses were less likely to be removed in the winter and the relative probability of a carcass removal was 7.6 times higher during mid-summer compared to mid-winter. Newer houses experienced a lower probability of removal compared to houses built before 1970. As well, developed neighbourhoods saw a lower probability of removal than undeveloped ones. These factors are similar to those factors we had previously identified as having a large effect on the likelihood of a bird-window collision suggesting those homes experiencing a larger number of collisions are also experiencing a higher number of scavenging events.

From these results, we came up with a correction factor for carcass removal by scavengers. 31.8% of carcasses were removed in the first 24 hours, which results in a 1.47 carcass removal rate. This means the number of carcasses detected in the first 24 hours needs to be adjusted by 1.47 to account for removal by scavengers. This rate is lower than the one developed from wind turbine studies that was used in creating the current estimate of bird-window collision mortality in Canada.

Using this removal rate and the citizen science data previously collected by the Birds and Windows project we estimated 957,440 (± 59,280 SD) birds are killed from window collisions at houses in Alberta each year. This is the most detailed estimate of bird-window collision fatalities in Canada as it’s based on the most detailed window collision study at houses to date and a carcass removal study located in the same area. Unlike previous studies, we did not extrapolate our results across the country. Our estimate is for Alberta, the area from which the data was collected. If we are to improve the current bird-window collision mortality estimate for Canada, more localized studies like ours are needed. Completing studies in each of the provinces will help reduce several of the existing biases in the fatality estimate at houses.

-Post and photos by Justine Kummer

Link to full text: http://www.ace-eco.org/vol11/iss2/art12/

Kummer, J. A., C. J. Nordell, T. M. Berry, C. V. Collins, C. R. L. Tse, and E. M. Bayne. 2016. Use of bird carcass removals by urban scavengers to adjust bird-window collision estimates. Avian Conservation and Ecology 11(2):12.

Leave a Reply

Your email address will not be published. Required fields are marked *