The Impact of Predation on Songbirds

New Study Challenges 10-year-old Methodology which may have led to Inaccurate Conclusions

The quantification of birdsong decline is well documented, yet the impact of predation on songbirds has been previously contested in studies. For example, the RSPB claims that songbirds are not limited by avian predators [1].

In 2010, Newson et al [2]. attempted to diagnose the underlying cause of songbird decline and concluded that predation does not affect songbird populations.  Ten years on, a new study by Dr Christopher Paul Bell (published in Frontiers in Ecology and Evolution - here) revisited the study and applied a different methodology using Common Bird Census data (CBC) to measure the impact of predation on songbirds. The results highlight that there are significant limitations with the Newson 2010 study (which used CBC and Breeding Bird Survey Data) and that further research is required to re-evaluate the factors behind songbird population decline. A comparison of methodology between Newson et al. (2010) and Bell (2020) provides some key learnings:

Fig 1.jpg

1. Predator research required

Increased predator abundance has been dismissed by Newson et al. as a cause of bird population decline, leaving changes in farming practice as the only viable explanation. Yet, agri-environment schemes have failed to reverse the decline of bird populations. Further rigorous, and preferably experimental research on predation therefore needs to be implemented immediately.

2. Inaccurate estimations of predator numbers

Newson et al. (2010) concluded that bird census data, “Provides little underlying evidence for large-scale impacts of widespread avian predators on avian prey populations”. These results were dependent on the method used to estimate predator population change: Raptor population estimates were averaged across a wide scale, which does not accurately reflect density at individual sites. E.G. Predator population density estimates: the overall density is 0.2, but the density of quadrants is (clockwise from the top left) 0.3, 0.25, 0.06 & 0.125.

fig 2.jpg

Positive relationships between predator and prey populations (e.g. increases in both) do not automatically mean that prey populations haven’t been impacted by predators - an assumption made by Newson et al. In this example, the habitat improved over time, resulting in an increase in both sparrows and hawks, (a positive correlation)...


3.    The relationship between predators and prey is complex. Increasing predator numbers can negatively

impact prey numbers even where the latter are also increasing.

fig 3.png
fig 3a.png

… But if the hawks had not increased in number, there may have been an even greater increase in the number of sparrows.

4. Predator and prey populations can fluctuate independently

Even if we find correlations between changes in predator and prey populations, this does not necessarily mean that the two are interacting:  Here, change in sparrow numbers across two sites is negatively correlated with change in hawk numbers, but this is because a sparrow population which is dwindling for other reasons becomes extinct in the site with poorer habitat, which is simultaneously colonised by an expanding hawk population

fig 4.png
fig 4a.png


For two decades, inferences derived from monitoring studies have been used to deny a role for increased predator numbers as a cause of bird population decline. This leaves farming practices as the only plausible explanation, but agri-environment schemes built on this premise have failed to reverse the decline in bird numbers.


As with every scientific enterprise, conservation science should always be subject to criticism and revision. Nature is often more complex than anything that can be captured by analysis of monitoring data, and the messy complexities of natural systems may be intelligible only through carefully designed experimental studies.


Meanwhile, the move towards more devolved stewardship schemes for the countryside, in which stakeholders can choose priorities and objectives and be rewarded for results rather than box-ticking, represents a belated recognition that science should complement rather than replace local knowledge and hands-on expertise.





[1] Roos, S., Smart, J., Gibbons, D. W. and Wilson, J. D. (2018). A review of predation as a limiting factor for bird populations in mesopredator‐rich landscapes: a case study of the UK. Biological Reviews 93: 1915-1937. doi/abs/10.1111/brv.12426.


[2] Newson, S. E., Rexstad, E. A., Baillie, S. R., Buckland, S. T., and Aebischer, N. J. (2010). Population change of avian predators and grey squirrels in England: is there evidence for an impact on avian prey populations? J. Appl. Ecol. 47, 244–252. doi: 10.1111/j.1365-2664.2010.01771.x