Fig 1.
Investigated non-crop habitat islands are highlighted in black. The red arrow indicates the smallest investigated island (with an area of 50 square metres). The hatched island was ultimately omitted from data analyses due to substantial trap destruction by large mammals. Crosses indicate the positions of arable field sites investigated in this study.
Fig 2.
The effects of habitat type and sampling period on the activity density and recorded species richness of ground-dwelling arthropods.
Mean number of A) carabid beetles and B) spiders collected per trap (i.e., activity density) and mean recorded species richness of C) carabid beetles and D) spiders per trap ± SE are shown for both sampling periods investigated (spring to early summer and peak summer).
Fig 3.
The effects of non-crop habitat island characteristics and sampling period on the activity density of ground-dwelling arthropods.
The relationships between A) number of collected carabids and non-crop habitat island area, B) number of collected carabids and non-crop habitat island shrub cover and C) number of collected spiders and non-crop habitat island area are shown. Full circles (solid line) represent the first sampling period (spring to early summer) and open circles represent the second sampling period (peak summer).
Fig 4.
The effects of non-crop habitat island characteristics and sampling period on the recorded species richness of ground-dwelling arthropods.
The relationships between A) number of carabid species recorded and non-crop habitat island area, B) number of carabid species recorded and non-crop habitat island shrub cover and C) number of spider species recorded and non-crop habitat island grass cover are shown. Full circles (solid line) represent the first sampling period (spring to early summer) and open circles represent the second sampling period (peak summer).
Fig 5.
The effects of habitat type and non-crop habitat island characteristics on the species composition of ground-dwelling arthropod assemblages.
The ordination diagrams based on the correspondence canonical analysis visualising A) the effects of habitat type and sampling period on carabid beetle assemblages, B) the effects of habitat type and sampling period on spider assemblages, C) the effects of non-crop habitat island characteristics and sampling period on carabid beetle assemblages and D) the effects of non-crop habitat island characteristics and sampling period on spider assemblages. Fifteen of the most influential species for each analysis are displayed and named using abbreviations made from the first three letters of the genus name followed by the first three letters of the species name (for full names see S1 and S2 Tables).
Fig 6.
The temporal shift in the species composition of ground-dwelling arthropod assemblages inhabiting agricultural landscapes.
The ordination diagrams based on the detrended canonical analysis visualising similarity in the species composition of A) carabid beetle and B) spider assemblages originating from particular sampling sites are shown. The position of a particular site sampled in the first sampling period (spring to early summer) is in the ordination space indicated by circle (non-crop habitat island) or square (arable field site) symbols. The temporal shift in species composition of a particular sampling site between the first and the second sampling period (peak summer) is indicated by the line originating in the centre of particular symbol. The size of circle symbols indicates the area of particular non-crop habitat islands. The red arrow indicates the smallest non-crop habitat island. The first two ordination axes explain 27.2% of the total variation in species data in the case of carabid beetles and 27.7% in the case of spiders. The ten most influential species are displayed and named using abbreviations made from the first three letters of the genus name followed by the first three letters of the species name (for full names see S1 and S2 Tables).
Fig 7.
The effects of non-crop habitat island characteristics on the activity density and recorded species richness of non-crop habitat arthropod specialists.
The relationships between A) activity density of carabid non-crop specialists and non-crop habitat island area, B) activity density of spider non-crop specialists and habitat island area, C) activity density of spider non-crop specialists and habitat island tree cover, D) number of carabid non-crop specialist species recorded and non-crop habitat island area, E) number of spider non-crop specialist species recorded and non-crop habitat island area, F) number of spider non-crop specialist species recorded and habitat island tree cover are shown. Full circles (solid line) represent the first sampling period (spring to early summer) and open circles represent the second sampling period (peak summer).