Aerial photogrammetry of seabirds from digital aerial video images using relative change in size to estimate flight height

Calculating the height at which birds fly over the sea is a challenging task, but remains important to assessing collision risk in proposed offshore wind farm areas for consenting purposes. This could be done by several methods (e.g. GPS or laser rangefinders), but each have biases that make it difficult to generate site-wide assessments. Digital video aerial surveys, which quickly cover large areas, were used to assess flight heights of northern gannets and black-legged kittiwakes using a photogrammetric technique combined with a semi-automated measurement tool. The lengths of birds known to be at sea surface, as identified by reflection on the water, were compared to lengths of birds at unknown height to generate individual flight height profiles. Validation of the method on kittiwake found that LiDAR produced mean flights that were not significantly different to the photogrammetry approach. Validation of the flight height method using man-made objects of known dimensions and heights suggested a 9–18% error (3–6 m at ~ 30 m height). However, the profiles of mean flight height distribution matched patterns in previous work. This method was able to estimate the flying heights of 65% and 75% of flying gannets and kittiwakes, respectively, in this case study. The annual percentage of gannets at collision risk height for a set of turbines with a 30 m air gap was estimated at 29.8%, and 16.1% for kittiwake. This technique can greatly improve our knowledge of the spatial distribution of flight height patterns in marine ecosystems, but also allows stakeholders to assess collision risk more easily within the sphere of offshore wind for the consenting process.