Testing concordance and conflict in spatial replication of landscape genetics inferences.

The degree to which landscape genetics findings can be extrapolated to different areas of a species range is poorly understood. Here, we used a broadly distributed ectothermic lizard (Sceloporus occidentalis, Western Fence lizard) as a model species to evaluate the full role of topography, climate, vegetation, and roads on dispersal and genetic differentiation. We conducted landscape genetics analyses with a total of 119 individuals in five areas within the Sierra Nevada mountain range. Genetic distances calculated from thousands of ddRAD markers were used to optimize landscape resistance surfaces and infer the effects of landscape and topographic features on genetic connectivity. Across study areas, we found a great deal of consistency in the primary environmental gradients impacting genetic connectivity, along with some site-specific differences, and a range in the proportion of genetic variance explained by environmental factors across study sites. High-elevation colder areas were consistently found to be barriers to gene flow, as were areas of high ruggedness and slope. High temperature seasonality and high precipitation during the winter wet season also presented a substantial barrier to gene flow in a majority of study areas. The effect of other landscape variables on genetic differentiation was more idiosyncratic and depended on specific attributes at each site. Across study areas, canyon valleys were always implicated as facilitators to dispersal and key features linking populations and maintaining genetic connectivity, though the relative importance varied in different areas. We emphasize that spatial data layers are complex and multidimensional, and careful consideration of spatial data correlation structure and robust analytic frameworks will be critical to our continued understanding of spatial genetics processes.

Temperate Zone Isolation by Climate: An Extension of Janzen's 1967 Hypothesis.

AbstractOne of the most stunning patterns of the distribution of life on Earth is the latitudinal biodiversity gradient. In an influential article, Janzen (1967) predicted that tropical mountains are more effective migration barriers than temperate mountains of the same elevation, because annual temperature variation in the tropics is lower. A great deal of research has demonstrated that the mechanism envisioned by Janzen operates at broad latitudinal scales. However, the extent that the mechanism mediates biodiversity generally, and at smaller scales, is far less understood. We investigated whether climate overlap is associated with genetic similarity between populations within temperate regions using lizards in the Sierra Nevada mountain range of California as a study system. By comparing genetic differentiation between high- and low-elevation populations, we found that in addition to the expected strong pattern of isolation by distance, high climate overlap was negatively associated with genetic differentiation, indicating that population pairs that inhabit climatically similar environments are less genetically differentiated. Moreover, while climate overlap between high- and low-elevation sites is predicted to increase from the equator to temperate regions, we find that in adjacent mountain ranges at the same latitude in temperate regions, climate overlap values can vary widely. This study suggests that in addition to the well-studied main effect of latitude on climate overlap and population differentiation, local climate factors within bioclimatic regions can also influence genetic differentiation between populations and do so by the same general mechanism that operates at larger geographic scales.

Biogeographic inferences across spatial and evolutionary scales.

The field of biogeography unites landscape genetics and phylogeography under a common conceptual framework. Landscape genetics traditionally focuses on recent-time, population-based, spatial genetics processes at small geographical scales, while phylogeography typically investigates deep past, lineage- and species-based processes at large geographical scales. Here, we evaluate the link between landscape genetics and phylogeographical methods using the western fence lizard (Sceloporus occidentalis) as a model species. First, we conducted replicated landscape genetics studies across several geographical scales to investigate how population genetics inferences change depending on the spatial extent of the study area. Then, we carried out a phylogeographical study of population structure at two evolutionary scales informed by inferences derived from landscape genetics results to identify concordance and conflict between these sets of methods. We found significant concordance in landscape genetics processes at all but the largest geographical scale. Phylogeographical results indicate major clades are restricted to distinct river drainages or distinct hydrological regions. At a more recent timescale, we find minor clades are restricted to single river canyons in the majority of cases, while the remainder of river canyons include samples from at most two clades. Overall, the broad-scale pattern implicating stream and river valleys as key features linking populations in the landscape genetics results, and high degree of clade specificity within major topographic subdivisions in the phylogeographical results, is consistent. As landscape genetics and phylogeography share many of the same objectives, synthesizing theory, models and methods between these fields will help bring about a better understanding of ecological and evolutionary processes structuring genetic variation across space and time.

Properties of Markov Chain Monte Carlo Performance across Many Empirical Alignments.

Nearly all current Bayesian phylogenetic applications rely on Markov chain Monte Carlo (MCMC) methods to approximate the posterior distribution for trees and other parameters of the model. These approximations are only reliable if Markov chains adequately converge and sample from the joint posterior distribution. Although several studies of phylogenetic MCMC convergence exist, these have focused on simulated data sets or select empirical examples. Therefore, much that is considered common knowledge about MCMC in empirical systems derives from a relatively small family of analyses under ideal conditions. To address this, we present an overview of commonly applied phylogenetic MCMC diagnostics and an assessment of patterns of these diagnostics across more than 18,000 empirical analyses. Many analyses appeared to perform well and failures in convergence were most likely to be detected using the average standard deviation of split frequencies, a diagnostic that compares topologies among independent chains. Different diagnostics yielded different information about failed convergence, demonstrating that multiple diagnostics must be employed to reliably detect problems. The number of taxa and average branch lengths in analyses have clear impacts on MCMC performance, with more taxa and shorter branches leading to more difficult convergence. We show that the usage of models that include both Γ-distributed among-site rate variation and a proportion of invariable sites is not broadly problematic for MCMC convergence but is also unnecessary. Changes to heating and the usage of model-averaged substitution models can both offer improved convergence in some cases, but neither are a panacea.

Turnover of hydrogen isotopes in lake sturgeon blood: implications for tracking movements of wild populations.

Naturally occurring deuterium ((2)H) in biota can be used to trace movement, migration and geographic origin of a range of organisms. However, to evaluate movements of animals using δ(2)H measurements of tissues, it is necessary to establish the turnover time of (2)H in the tissues and the extent of isotopic discrimination from different environmental (2)H sources to those tissues. We investigated the turnover of (2)H in lake sturgeon (Acipenser fulvescens) blood by manipulating both environmental water δ(2)H and diet δ(2)H over a four-month period. The half-life of deuterium in lake sturgeon blood was 37.9 days after an increase in the environmental water δ(2)H of +714 ‰. However, no clear turnover in blood (2)H occurred over the same period in a separate trial following a change of -63.8 ‰ or +94.2 ‰ in diet. These findings suggest that environmental water (2)H exchanges much faster with blood than diets and that blood δ(2)H values can be used to trace movements of sturgeon and other fish moving among isotopically distinct waters.