Contrasting effects of mosaic structure on alpha and beta diversity of bird assemblages in a human‐modified landscape

Abstract

Habitat loss and fragmentation are key processes causing biodiversity loss in human‐modified landscapes. Knowledge of these processes has largely been derived from measuring biodiversity at the scale of ‘within‐habitat’ fragments with the surrounding landscape considered as matrix. Yet, the loss of variation in species assemblages ‘among’ habitat fragments (landscape‐scale) may be as important a driver of biodiversity loss as the loss of diversity ‘within’ habitat fragments (local‐scale). We tested the hypothesis that heterogeneity in vegetation cover is important for maintaining alpha and beta diversity in human‐modified landscapes. We surveyed bird assemblages in eighty 300‐m‐long transects nested within twenty 1‐km2 vegetation ‘mosaics’, with mosaics assigned to four categories defined by the cover extent and configuration of native eucalypt forest and exotic pine plantation. We examined bird assemblages at two spatial scales: 1) within and among transects, and 2) within and among mosaics. Alpha diversity was the mean species diversity within‐transects or within‐mosaics and beta diversity quantified the effective number of compositionally distinct transects or mosaics. We found that within‐transect alpha diversity was highest in vegetation mosaics defined by continuous eucalypt forest, lowest in mosaics of continuous pine plantation, and at intermediate levels in mosaics containing eucalypt patches in a pine matrix. We found that eucalypt mosaics had lower beta diversity than other mosaic types when ignoring relative abundances, but had similar or higher beta diversity when weighting with species abundances. Mosaics containing both pine and eucalypt forest differed in their bird compositional variation among transects, despite sharing a similar suite of species. This configuration effect at the mosaic scale reflected differences in vegetation composition among transects. Maintaining heterogeneity in vegetation cover could help to maintain variation among bird assemblages across landscapes, thus partially offsetting local‐scale diversity losses due to fragmentation. Critical to this is the retention of remnant native vegetation.

Monitoring biodiversity in the Anthropocene using remote sensing in species distribution models

“…Besides land-use composition, land-use configuration can in some cases represent a good proxy for those species requiring corridors and landscape borders to survive (e.g., Neilan et al, 2019;Vinter et al, 2016). Accordingly, the heterogeneity of land-use or of satellite reflectance data has been widely assessed in the past, using various algorithms and metrics such as multivariate statistical analysis (Feilhauer and Schmidtlein, 2009), the spectral species concept (Féret and Asner, 2014), self-organizing feature maps (Foody, 1999), multidimensional distance metrics (Rocchini et al, 2016), and Rao's Q diversity (Rocchini et al, 2017).…”

Matrix composition mediates effects of habitat fragmentation: a modelling study

“…It has, as a consequence of these qualities, been suggested that FPS may have a lesser effect if the matrix includes land covers that are of a similar type to the focal-habitat (Miller-Rushing et al 2019). Matrix heterogeneity may also help maintain variation in species across landscapes (beta-diversity) and offset negative effects of FPS on gamma-diversity (Neilan et al 2018). As mentioned, land cover diversity and positive edge effects have been identified as potential positive mechanisms of FPS (Fahrig et al 2019).…”

Why We Need to Invest in Large-Scale, Long-Term Monitoring Programs in Landscape Ecology and Conservation Biology

“…For example, forest and other land clearing often occurs at large spatial scales [9,10] and can have profound impacts on biodiversity and ecological processes over extensive areas [11,12]. Quantifying spatial patterns of biodiversity occurrence like beta-diversity and its relationships to spatial patterns of vegetation cover also can demand large-scale monitoring [13]. Edge effects from changes in land cover can propagate over extensive areas [14] and their impacts can demand monitoring at correspondingly large scales.…”