Fig 1.
A. Location of the study area and the observation points (white dots) in Castilla y León Region, Central Spain. Thin black lines represent 2 lane roads and bold black lines represent motorways. Main cities in the area are displayed with a star. Base map downloaded from Instituto Geográfico Nacional de España (IGN, http://www.ign.es). B. Detail of a sampling point either in a road or motorway, and the grid used for local scale analyses (shaded area) of 100 x 100 m cells inside the 500 m radius.
Table 1.
Predictor variables for the landscape scale analysis.
Fig 2.
Total abundance of common raptor species in sampling plots of each type of transport infrastructure. Details for all the species detected in the study can be seen in S1 Table.
Fig 3.
Two dimensional ordination (MDS) of general trends in raptor community.
Species composition of sampling plots represented by transport infrastructure types and season (mean ± SE of position in dimensions 1 and 2). The influence of most abundant species is also represented with arrows of length and direction obtained from their correlation with the axes. Circles represent control plots, triangles road plots, and squares motorway plots. Solid symbols for breeding season and empty symbols for winter. Stress value of the ordination of 0.25.
Fig 4.
Relative importance of foraging habitat selection predictor variables at landscape scale.
Akaike weights of predictors for each response variable in the community level analyses (diversity, abundance, richness) and the species-level analyses. Controlling variables are identified by asterisks and colored in light grey. The sign of the effect in the final average model is shown as positive (+) or negative (−). Variables in quadratic form are identified with “^2” and the symbols correspond to the simple and quadratic form, respectively. Variables with an averaged coefficient close to zero (≤|0.01|) are marked as “no effect” (NE). When the variable ADT is selected, the curve showing its effect in the response is also included. In this case, the x axis is always the amount of traffic (ADT), from 0 to 20.850 vehicles a day, and the y axis represents the value of the response variable. For a definition of the variables see Table 1 and the values of the averaged coefficients are in S10 and S11 Tables.
Table 2.
Selected models for foraging habitat selection with traffic effects at the landscape scale.
Table 3.
Raptors selection of asphalt cells within the sampling plots.