Using risk prediction models and species sensitivity maps for large-scale identification of infrastructure-related wildlife protection areas: The case of bird electrocution

Abstract

A B S T R A C TThe use of systematic area-selection procedures to design protected areas can help optimize conservation actions. However, this process has rarely been used to identify high-risk mortality areas to protect wildlife from human impacts. Electrocution on power lines is one the most important human-related causes of bird mortality worldwide, especially for raptors. Identifying and correcting dangerous individual pylons can significantly reduce the number of electrocution victims, but applying this procedure at a large spatial scale is impractical. In this paper we describe a new selection process that allows for identification of high-risk mortality areas at large scales, combining spatial electrocution risk models with maps of species sensitivity to such an impact. We used the Valencia Region (eastern Spain) as our study system. The risk prediction map was built using bird electrocution records on 1 km × 1 km grids from 2000 to 2009 and the species sensitivity map was built using data on presence and use of four raptor species. The combination of both maps was compared to the distribution of Special Protected Areas and validated by local experts to identify prediction errors or gaps. The final proposal of high priority areas to protect birds from electrocution covered 16.3% of the Valencia Region. Our work supports the use of predictive models and sensitivity maps in the decision-making process to locate high priority infrastructure-related wildlife protection areas at a large scale.

Where to start? Development of a spatial tool to prioritise retrofitting of power line poles that are dangerous to raptors

“…The results demonstrate that the power distribution lines cause a large number of deaths among many of the most threatened raptor species in Spain (Tables and ), especially considering that a large number of victims may have passed undetected (Ferrer, de la Riva, & Castroviejo, ; Kemper et al., ). We reported a high occurrence of electrocution in many raptor species, particularly the common buzzard, griffon vulture and the red kite, as has been found in other areas in Spain (Guil et al., ; Janss & Ferrer, ; Pérez‐García et al., ; Tintó et al., ). Possibly their biological traits (e.g., size, abundance) and other characteristics, such as hunting behaviour, make these species highly sensitive to electrocution (Lehman et al., ).…”

“…Our results indicate that poles placed in areas with a greater proportion of land occupied by crops, middle slopes, low forest cover and scrub–grass cover accumulate the highest number of electrocuted raptors. The role of crop lands in our model was largely a function of the high proportion of sit‐and‐wait raptor electrocutions contributing to the dataset (Pérez‐García et al., ). These raptors often choose poles as perching points in ecotone areas, between natural habitat and the more anthropogenically modified habitats (principally agricultural) (Sánchez‐Zapata & Calvo, ).…”

“…This result confirms that large‐scale spatial patterns of raptor electrocutions depend on the combination of how the pole‐specific features are arranged in the landscape and the surrounding habitat. Our model improves the previous approaches that use pole density and/or length of power lines as key risk factors for avian electrocutions on a regional scale (Dwyer et al., ; Pérez‐García et al., ).…”

“…One key strength of our model, relative to studies elsewhere (Dwyer et al., , ; Guil et al., ; Mañosa, ; Pérez‐García et al., ), is that the electrocution risk assessment was applied to an extensive geographical area (65,000 km 2 ) with 188,741 poles. The availability of detailed information about the pole‐specific features of large power distribution networks creates new possibilities to generate more accurate risk maps of avian electrocution at large geographical scales.…”

“…It could be assumed that this database is biased towards larger birds (electrical incident) or more frequented areas (greater detectability). As in other studies conducted elsewhere (e.g., Pérez‐García et al., ), the two sources of data are complementary: IC records cover incidents in less frequented or remote areas (they review all incidents based on security and maintenance protocols) and records made by wildlife rangers and citizens are independent of electrical incidents. Finally, in this study, we focus on electrocution risk modelling of raptor species such as eagles, vultures, hawks and owls.…”