Landscape Genetics Andrew Storfer, Washington State University, Pullman, WA, USA r 2013 Elsevier Inc All rights reserved Glossary Analysis of molecular variance Similar to the statistical analysis of variance (ANOVA), but involves partitioning of molecular genetic data in different groups (e.g., within and among populations) based on hierarchical sampling procedures Extent Generally used in landscape ecology, referring to the overall size of a study area Functional connectivity Describes the processes of dispersal and gene flow of organisms among habitat patches Gene flow Refers to the movement of organisms among populations or locales followed by successful breeding event(s), resulting in the transfer of alleles from one population to another Grain Generally used in landscape ecology, referring to the spatial scale of individual observations Least-cost path A line (vector) of ‘least resistance’ is generated between two observation points The resistance refers to a resistance surface, or cost grid in a landscape matrix, which is parameterized according to some Introduction An increasing goal of ecology and evolutionary biology is to understand spatial patterns of genetic variation However, until recently there have been both technological and analytical gaps between the fields of landscape ecology and population genetics to test the influence of spatial processes on population genetic structure (Manel et al., 2003) That is, whereas landscape ecology improved greatly in its ability to understand structural connectivity of landscapes with advances in geographic information systems (GIS) and remote sensing, understanding functional connectivity among biological populations remained a challenge because of difficulties associated with accurately estimating migration rates (e.g., low recapture probabilities or estimating breeding without dispersal; Holderegger and Wagner, 2008) However, our ability to obtain high resolution genetic data has dramatically increased, but population genetics had been focused on estimating functional connectivity (i.e., via estimating gene flow, which incorporates dispersal and breeding among populations), with little regard to structural connectivity of the intervening habitat matrix between populations Landscape genetics emerged as an interdisciplinary analytical framework to bridge this gap (Holderegger and Wagner, 2008; Manel et al., 2003) by drawing on analysis methods from landscape ecology, spatial statistics, geography, and population genetics to ‘‘yexplicitly quantify the effects of landscape composition, configuration and matrix quality 508 understanding of species’ dispersal habits The least-cost distance can be expressed in terms of geographic distance or total resistance along this path Mantel test A statistical test of the correlation between two matrices; in landscape genetics, these are a geographic and genetic distance matrix Significance of the correlation is determined by randomly permuting the values in the rows and columns of the matrices to create a distribution of values to which the observed value can be compared Spatial autocorrelation Observations are not independent identically distributed, but are correlated over some distance in space Positive autocorrelation indicates that nearby values are similar; negative autocorrelation indicates that nearby values are dissimilar Structural connectivity Generally refers to the distance between habitat patches and the nature of the intervening landscape matrix Tessellation A pattern of nonoverlapping polygons is used to partition the area leaving no gaps In landscape genetics, tessellation is generally used to create polygons around each point where genotypes are collected on gene flow and spatial genetic variation’’ (Storfer et al., 2007) Whereas population genetics studies had previously been limited in spatial inference to tests of isolation-by-distance (IBD), landscape genetics studies test explicitly the relative influence of landscape and environmental features on gene flow, genetic discontinuities (Guillot et al., 2008) and genetic population structure (Manel et al., 2003) This area of inquiry has met with early success, as landscape genetic models generally provide a more complete picture of population genetic structure by typically explaining a significantly higher proportion of variation in gene flow than classic, straight-line Euclidean IBD models (Figure 1; Storfer et al., 2010) Further, recent simulation studies have shown that a variety of landscape genetics techniques were able to detect the simulated scenario with relatively high power and accuracy (Balkenhol et al., 2009a; Cushman and Landguth, 2010; Murphy et al., 2008) As a result, understanding landscape effects on functional (genetic) connectivity can provide insight into fundamental biological processes such as: metapopulation dynamics, (ecological) speciation, and ultimately limits to species’ geographic distributions The integrative approach of landscape genetics has tested both basic and applied research questions, including: identifying barriers to dispersal (Guillot et al., 2008; Latch et al., 2008; Manni et al., 2004), modeling functional connectivity across different spatial scales (Murphy et al., 2010; Rasic and Keyghobadi, 2012), inferring the effect of landscape change Encyclopedia of Biodiversity, Volume http://dx.doi.org/10.1016/B978-0-12-384719-5.00386-5