AUTHOR=Borthwick Richard , de Flamingh Alida , Hesselbarth Maximilian H. K. , Parandhaman Anjana , Wagner Helene H. , Abdel Moniem Hossam E. M. 

TITLE=Alternative Quantifications of Landscape Complementation to Model Gene Flow in Banded Longhorn Beetles [Typocerus v. velutinus (Olivier)]

JOURNAL=Frontiers in Genetics

VOLUME=Volume 11 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2020.00307

DOI=10.3389/fgene.2020.00307

ISSN=1664-8021

ABSTRACT=Rapid progression of human socio-economic activities has altered structure and function of natural landscapes. Species that rely on multiple, complementary habitat types to complete their life cycle may be especially at risk. However, such landscape complementation has received little attention in the context of landscape connectivity modelling. A previous study on flower longhorn beetles (Cerambycidae: Lepturinae) integrated landscape complementation into a continuous habitat suitability ‘surface’, which was used to quantify landscape connectivity between pairs of sampling sites using gradient-surface metrics. This connectivity model was validated with molecular genetic data collected for the banded longhorn beetle (Typocerus v. velutinus) in Indiana, USA. However, this approach has not been compared to alternative models in a landscape genetics context. Here, we used a discrete land cover map to calculate landscape metrics related to landscape complementation based on a patch mosaic model as an alternative to the previously published, continuous habitat suitability model. We evaluated the habitat suitability model surface with gradient surface metrics and with two resistance-based models based on least cost path and commute distance, in addition to an isolation-by-distance model based on Euclidean distance. We compared the ability of these competing models of connectivity to explain pairwise genetic distances (RST) previously calculated from ten microsatellite genotypes of 454 beetles collected from 17 sites across Indiana. Model selection with maximum likelihood population effects models found that gradient surface metrics were most effective at explaining pairwise genetic distances as a proxy for gene flow across the landscape, followed by landscape metrics calculated from the patch mosaic model, whereas the least cost path model performed worse than the commute distance and the isolation by distance model. We argue that the analysis of a continuous habitat suitability model with gradient surface metrics might perform better because of their combined ability to effectively represent and quantify the continuous degree of landscape complementation (i.e. availability of complementary habitats in vicinity) found at and in-between sites, on which these beetles depend. Our findings may inform future studies that seek to model habitat connectivity in complex heterogeneous landscapes as natural habitats continue to become more fragmented in the Anthropocene.