This was first published in the Winter 2021 edition of BTO News, the excellent membership magazine of the British Trust for Ornithology. The article describes work investigating disturbance of wintering birds by drones.
We carefully watch the distant godwits. Through the autumnal greyness of the Scottish lowlands, their heads intermittently appear above the stubbled cereal crop as they stab the earth for worms. Beyond, smoke drifts across the Firth of Forth from distant oil refineries. Perhaps the godwits are spooked by the whirr of the drone or the unfamiliar shape approaching, but with the drone still more than a hundred metres away their heads are jerking this way and that. One birds calls, and soon the flock is wheeling high over the fields. We fly the drone back to our vantage point and after a minute or so the flock settles again two fields away.
With drones getting cheaper and more popular, BTO often fields enquiries from researchers and surveyors interested in using drones to count flocks of wintering birds. Members of the public also sometimes report that they have observed wild birds being disturbed by drones. With that in mind, we set out to investigate how flocks of wintering waterbirds respond to drones. High- quality roosting and feeding areas for the large numbers of waders, wildfowl and gulls that winter in the UK are limited and repeated disturbance could have significant impact on the condition of birds, or cause feeding or roosting sites to be abandoned.
We approached flocks with the drone in intertidal coastal areas, at small inland lochs, and in cereal stubble fields. The drone was flown at a standardised speed and height, and flock responses like alarm calling or flushing were recorded. We soon discovered that flocks feeding in arable stubble fields, like those godwits, were extremely sensitive – on almost every approach they flushed at significant distance. Birds need to be more vigilant when feeding in stubble fields because vegetation can obscure their field of view, making them more susceptible to predation, while mammalian predators are an added danger. At coastal sites results were mixed, with larger flocks likely to flush and smaller flocks less likely to do so. At inland lochs where there was already lots of recreational disturbance,
waterbirds rarely responded to the approaching drone.
Interpreting how wild animals respond to disturbance stimuli is challenging. Had our godwits continued feeding with the drone hovering above them, it would be tempting to simply conclude that the drone caused them little distress. But what if the birds were in poor physical condition and did not have the energy to take off? This question was neatly investigated by scientists at the University of Glasgow in 2004. They gave supplementary food to some Turnstone flocks but not others, and found that the flocks which had received supplementary food were more likely to flush in response to experimental disturbance. This demonstrated that responses to disturbance could be driven by the condition of birds. And what about lochs in busy recreational areas where birds rarely responded to the drone? Individuals that use these sites are subject to a high level of disturbance but can exploit resources without competition from more sensitive individuals. Because these sites are subject to more human activity, predators may be less likely to hunt here, so a lower level of vigilance is required. The threat of aerial predation is also lower for birds sitting on water. Additionally, while disturbance may initially cause a strong response, if repeated stimuli are benign then some degree of habituation may occur, and behavioural responses weaken. All these factors probably contribute to the low response level we saw in our tests at the inland loch sites. Another important factor driving responses may be the availability of alternative habitat; in cases where birds don’t respond, individuals may have very limited choices of feeding or roosting habitat, meaning that they are very reluctant to leave a site irrespective of any stress caused by the disturbance event.
Trying to unravel complex interactions between habitat availability, habituation, bird condition and predation risk is difficult, and identifying variations in sensitivity between species is also challenging when a mix of different species will be present at most sites (as in this study). As such, we must be cautious when drawing conclusions. However, we also need to be clear about what we do know: the coastline of Britain and Ireland provides wintering habitat for internationally important populations of many species, and there are various examples of other types of disturbance that already negatively affect local populations. Additional disturbance from drones could act in synergy with other types, and it is very difficult to identify ‘safe’ disturbance thresholds until they are surpassed. So, when considering drone use at coastal sites where large flocks of wintering waterbirds congregate, at least for now it would be prudent to err on the side of caution.
This work was generously funded by The Sound Approach.
コメント