Multi‐level patterns in population genetics: Variogram series detects a hidden isolation‐by‐distance‐dominated structure of Scandinavian brown bears Ursus arctos

• Published in: Methods in Ecology and Evolution
• Main Authors: Schregel, Julia, Remm, Jaanus, Eiken, Hans Geir, Swenson, Jon E., Saarma, Urmas, Hagen, Snorre B.
• Other Authors: Gaggiotti, Oscar
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2018
• Subjects: Norway; Ursus arctos
• Online Access: http://dx.doi.org/10.1111/2041-210x.12980; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2F2041-210X.12980; https://onlinelibrary.wiley.com/doi/pdf/10.1111/2041-210X.12980; https://onlinelibrary.wiley.com/doi/full-xml/10.1111/2041-210X.12980; https://besjournals.onlinelibrary.wiley.com/doi/pdf/10.1111/2041-210X.12980

Abstract Large‐scale pattern‐oriented approaches are useful to understand the multi‐level processes that shape the genetic structure of a population. Matching the scales of patterns and putative processes is both a key to success and a challenge. We have developed a simple statistical approach, based on variogram analysis, that identifies multiple spatial scales where the population pattern, in this case genetic structure, have highest expression (i.e. the spatial scales at which the strength of patterning of isolation‐by‐distance ( IBD ) residual variance reached maximum) from empirical data and, thus, at which scales it should be studied relative to the underlying processes. The approach is applicable to any spatially explicit pairwise data, including genetic, morphological or ecological distance or similarity of individuals, populations and ecosystems. To exemplify possible applications of this approach, we analysed microsatellite genotypes of 1,530 brown bears from Sweden and Norway. The variogram approach identified two scales at which population structure was strongest, thus indicating two different scale‐dependent processes: home‐range‐related processes at scales <35 km, and subpopulation division at scales >98 km. On the basis of this, we performed a scale‐explicit analysis of genetic structure using DR esD analysis and compared the results with those obtained by the Bayesian clustering implemented in structure . We found that the genetic cluster identified in central Scandinavia by Structure is caused by IBD , with distinct gene flow barriers to the south and north. We discuss possible applications and research perspectives to further develop the approach.