279
Both algorithms have proven to be very efficient in delineating and targeting features in digital image analysis (Foody and Mathur 2004; Pal 2005). These approaches
can be used for any other colonial species to provide quick status assessment and
colony size estimates. Automatic techniques based on pattern recognition and image
classification have been previously applied for bird detection in flocks (AbdElrahman 2005; Grenzdörffer 2013) showing acceptable results.
Additionally, visual assessment allowed us to identify standing birds from lying
birds and, for a few cases, count eggs lying exposed on the ground. We were also
able to identify other species breeding in the colony, such as the Black-headed Gull
(Chroicocephalus ridibundus) and Yellow-legged Gull (Larus michahellis), the
latter being an active chick predator of the other two species.
Background
Two of the most critical factors in assessing the conservation status of a bird population are the number of breeding pairs and annual breeding success. Usually, managers and scientists estimate changing population sizes by counting nesting pairs and
fledging chicks throughout the breeding season and from year to year. Additional
research or monitoring activities are however needed to obtain ancillary information
over the course of the breeding season, such as clutch size and the number of born
and fledging chicks per nesting colony. This information requires very intensive
sampling but generally involves entering the breeding colonies. This usually leads
to significant disturbances (Anderson and Gaston 2013), to which Slender-billed
Gulls are particularly sensitive. (Oro and Tavecchia 2008).
As an alternative to ground sampling, manned aerial surveys have been used
widely to reduce disturbances in deriving counts while also providing synoptic
information (Frederick et al. 1996; McEvoy et al. 2016). This is the case in Doñana
Natural Space, where aquatic bird populations are estimated on a monthly basis by
manned flights over wetlands (Fig. 1). In addition to these observations, photos are
usually taken during the flight for after-flight assessment. However, these images
are not usually acquired with adequate planimetric planning (zenithal view, appropriate overlapping and pixel size) or illumination conditions, making them an
unsuitable source for mapping – shadowed and overlapping individuals are usually
the main constraints.
In recent years, many studies have shown the usefulness of UAVs for deriving
counts of colony-nesting birds (Chabot et al. 2015; Hodgson et al. 2016; SardàPalomera et al. 2012). The recent increase in the availability and ease of use of
UAVs has also motivated scientists to use them for ecological studies (Anderson
and Gaston 2013). Professional UAV campaigns may resolve these kinds of issues
since proper mission planning can provide accurate and spatially explicit information. UAVs have also proven efficient in overflying bird colonies and providing
valuable scientific information (Jones et al. 2006; Sardà-Palomera et al. 2012).
However, in the context of bird surveys, the use of UAVs has not been sufficiently
Using UAVs to Map Aquatic Bird Colonies
Both algorithms have proven to be very efficient in delineating and targeting features in digital image analysis (Foody and Mathur 2004; Pal 2005). These approaches
can be used for any other colonial species to provide quick status assessment and
colony size estimates. Automatic techniques based on pattern recognition and image
classification have been previously applied for bird detection in flocks (AbdElrahman 2005; Grenzdörffer 2013) showing acceptable results.
Additionally, visual assessment allowed us to identify standing birds from lying
birds and, for a few cases, count eggs lying exposed on the ground. We were also
able to identify other species breeding in the colony, such as the Black-headed Gull
(Chroicocephalus ridibundus) and Yellow-legged Gull (Larus michahellis), the
latter being an active chick predator of the other two species.
Background
Two of the most critical factors in assessing the conservation status of a bird population are the number of breeding pairs and annual breeding success. Usually, managers and scientists estimate changing population sizes by counting nesting pairs and
fledging chicks throughout the breeding season and from year to year. Additional
research or monitoring activities are however needed to obtain ancillary information
over the course of the breeding season, such as clutch size and the number of born
and fledging chicks per nesting colony. This information requires very intensive
sampling but generally involves entering the breeding colonies. This usually leads
to significant disturbances (Anderson and Gaston 2013), to which Slender-billed
Gulls are particularly sensitive. (Oro and Tavecchia 2008).
As an alternative to ground sampling, manned aerial surveys have been used
widely to reduce disturbances in deriving counts while also providing synoptic
information (Frederick et al. 1996; McEvoy et al. 2016). This is the case in Doñana
Natural Space, where aquatic bird populations are estimated on a monthly basis by
manned flights over wetlands (Fig. 1). In addition to these observations, photos are
usually taken during the flight for after-flight assessment. However, these images
are not usually acquired with adequate planimetric planning (zenithal view, appropriate overlapping and pixel size) or illumination conditions, making them an
unsuitable source for mapping – shadowed and overlapping individuals are usually
the main constraints.
In recent years, many studies have shown the usefulness of UAVs for deriving
counts of colony-nesting birds (Chabot et al. 2015; Hodgson et al. 2016; SardàPalomera et al. 2012). The recent increase in the availability and ease of use of
UAVs has also motivated scientists to use them for ecological studies (Anderson
and Gaston 2013). Professional UAV campaigns may resolve these kinds of issues
since proper mission planning can provide accurate and spatially explicit information. UAVs have also proven efficient in overflying bird colonies and providing
valuable scientific information (Jones et al. 2006; Sardà-Palomera et al. 2012).
However, in the context of bird surveys, the use of UAVs has not been sufficiently
Using UAVs to Map Aquatic Bird Colonies