Fleshy-fruit production is becoming more intensive worldwide, but how this affects frugivorous birds is poorly known. In the Mediterranean region, intensive and super-intensive olive orchards are fast expanding, potentially affecting millions of wintering songbirds. Here we test the idea that intensification may benefit frugivorous birds, at least locally, due to increased fruit availability, while negatively affecting the wider wintering bird community due to intensive management, structural simplification and landscape homogenisation. We estimated olive abundance and surveyed birds in early, mid- and late winter, at traditional, intensive, and super-intensive orchards in southern Portugal. We used Hierarchical Modelling of Species Communities to relate species richness, prevalence and abundance to management intensity, winter period, olive availability and landscape context, and evaluated the role of frugivory on observed responses. Olive availability was much higher throughout the winter in more intensive than in traditional orchards, both in trees and on the ground. Frugivorous bird abundance was higher in more intensive orchards, and the most abundant frugivorous species (blackcap, song thrush, robin) were positively affected by olive availability and/or increasing landscape cover by olive orchards, while intensification level had relatively minor effects after accounting for other variables. Non-frugivorous richness and abundance were higher in traditional orchards, and many non-frugivorous species were less prevalent in more intensive orchards or negatively affected by landscapes dominated by olive cultivation. Synthesis and applications. While negatively affecting the wider bird community, our results suggest that olive farming intensification can contribute to sustain large numbers of frugivorous birds in the Mediterranean region. As frugivorous birds are not seen as damaging by olive farmers, there is an opportunity to promote their conservation in intensive and super-intensive orchards, which requires management to increase habitat heterogeneity, and to reduce risks such as mortality associated with mechanical harvest and contamination with pesticide residues. Overall, we recommend that efforts to manage farmland biodiversity should consider the impacts and conservation opportunities of fruit crop intensification.

This dataset contains data to reproduce the analyses mentioned in the paper using the Hmsc modelling framework. Full details of the methods (collection, data processing, etc.) are presented in the paper.

Usage notes for each file: 

bird_presence_data.csv - Presence/ absence (1/0) of all recorded bird species (42 species) in each of the 669 sampling sessions (223 transects x 3 periods). Column headers: ID - code of the sampling session (a given transect at a given period); Alaarv to Upuepo - codes of species names (see species_names.csv for details).

bird_abundance_data.csv - Total number of individuals of each bird species recorded in each of the 669 sampling sessions (223 transects x 3 periods). Due to model convergence problems abundance analyses were only conducted for this subset of 16 species (see paper methods for details). Column headers: ID - code of the sampling session (a given transect at a given period). Antpra to Turphi - codes of species names (see species_names.csv for details).

environmental_data.csv - Environmental covariates matrix. Column headers: ID - code of the sampling session (a given transect at a given period). Int - intensification level of the olive orchard where the transect is located (A - traditional; B - Intensive; C - Super-intensive); olives_tree - availability of olives for birds on trees (tonnes/ha; log-transformed); olives_ground - availability of olives for birds on ground (tonnes/ha; log-transformed); period - sampling period (A - early winter, B - middle winter, C - late winter); landscape - proportion of cover by olive orchards within a 1km-radius buffer centred on each orchard sampling area (arcsine-transformed).

trait_data.csv - Species-trait matrix. Column headers: sp_code - codes of species names (see species_names.csv for details); frugivore - column indicating if a given species feeds primarily on fruits during the winter period (1) or not (0). Information is given for the 42 bird species.

phylogenetic_data.csv - Species-species matrix showing phylogenetic relationships among the 42 bird species. Column headers: sp_code - codes of species names (see species_names.csv for details);

study_design.csv - Matrix describing the hierarchical nature of the study design (with transects nested into orchards). Column headers: ID - code of the sampling session (a given transect at a given period); transect - transect code (S1-S223); orchard - orchard code (P1-P78). This information was used to set random effects that accounted for repeated measures in each transect during the winter and for multiple transects sampled in each orchard.

orchard_coordinates.csv - Geographical location of orchards. Column headers: orchard - orchard code (P1-P78); x.coordinate / y.coordinate - geographical coordinates of sampled orchards (Coordinate Reference System - Lisboa_Hayford_Gauss_IGeoE; Authority ID - ESRI:102164). This information was used to set the spatial random effect (to account for spatial autocorrelation).

species_names.csv - Correspondence between the used species codes and species full scientific names. Column headers: sp_scientific_name - bird species scientific name; sp_code - bird species code.