Amy G. Vandergast

Understanding the genetic effects of recent habitat fragmentation in the context of evolutionary history: Phylogeography and landscape genetics of a southern California endemic Jerusalem cricket (Orthoptera: Stenopelmatidae: Stenopelmatus)

Habitat loss and fragmentation due to urbanization are the most pervasive threats to biodiversity in southern California. Loss of habitat and fragmentation can lower migration rates and genetic connectivity among remaining populations of native species, reducing genetic variability and increasing extinction risk. However, it may be difficult to separate the effects of recent anthropogenic fragmentation from the genetic signature of prehistoric fragmentation due to previous natural geological and climatic changes. To address these challenges, we examined the phylogenetic and population genetic structure of a flightless insect endemic to cismontane southern California, Stenopelmatus‘mahogani’ (Orthoptera: Stenopelmatidae). Analyses of mitochondrial DNA sequence data suggest that diversification across southern California began during the Pleistocene, with most haplotypes currently restricted to a single population. Patterns of genetic divergence correlate with contemporary urbanization, even after correcting for (geographical information system) GIS‐based reconstructions of fragmentation during the Pleistocene. Theoretical simulations confirm that contemporary patterns of genetic structure could be produced by recent urban fragmentation using biologically reasonable assumptions about model parameters. Diversity within populations was positively correlated with current fragment size, but not prehistoric fragment size, suggesting that the effects of increased drift following anthropogenic fragmentation are already being seen. Loss of genetic connectivity and diversity can hinder a populations ability to adapt to ecological perturbations commonly associated with urbanization, such as habitat degradation, climatic changes and introduced species. Consequently, our results underscore the importance of preserving and restoring landscape connectivity for long‐term persistence of low vagility native species.

Influence of volcanic activity on the population genetic structure of Hawaiian Tetragnatha spiders: fragmentation, rapid population growth and the potential for accelerated evolution

Volcanic activity on the island of Hawaii results in a cyclical pattern of habitat destruction and fragmentation by lava, followed by habitat regeneration on newly formed substrates. While this pattern has been hypothesized to promote the diversification of Hawaiian lineages, there have been few attempts to link geological processes to measurable changes in population structure. We investigated the genetic structure of three species of Hawaiian spiders in forests fragmented by a 150‐year‐old lava flow on Mauna Loa Volcano, island of Hawaii: Tetragnatha quasimodo (forest and lava flow generalist), T. anuenue and T. brevignatha (forest specialists). To estimate fragmentation effects on population subdivision in each species, we examined variation in mitochondrial and nuclear genomes (DNA sequences and allozymes, respectively). Population subdivision was higher for forest specialists than for the generalist in fragments separated by lava. Patterns of mtDNA sequence evolution also revealed that forest specialists have undergone rapid expansion, while the generalist has experienced more gradual population growth. Results confirm that patterns of neutral genetic variation reflect patterns of volcanic activity in some Tetragnatha species. Our study further suggests that population subdivision and expansion can occur across small spatial and temporal scales, which may facilitate the rapid spread of new character states, leading to speciation as hypothesized by H. L. Carson 30 years ago.

Genetic Landscapes GIS Toolbox: tools to map patterns of genetic divergence and diversity.

The Landscape Genetics GIS Toolbox contains tools that run in the Geographic Information System software, ArcGIS®, to map genetic landscapes and to summarize multiple genetic landscapes as average and variance surfaces. These tools can be used to visualize the distribution of genetic diversity across geographic space and to study associations between patterns of genetic diversity and geographic features or other geo‐referenced environmental data sets. Together, these tools create genetic landscape surfaces directly from tables containing genetic distance or diversity data and sample location coordinates, greatly reducing the complexity of building and analyzing these raster surfaces in a Geographic Information System.

Loss of genetic connectivity and diversity in urban microreserves in a southern California endemic Jerusalem cricket (Orthoptera: Stenopelmatidae: Stenopelmatus n. sp. ''santa monica'')

Microreserves may be useful in protecting native arthropod diversity in urbanized landscapes. However, species that do not disperse through the urban matrix may eventually be lost from these fragments. Population extinctions may be precipitated by an increase in genetic differentiation among fragments and loss of genetic diversity within fragments, and these effects should become stronger with time. We analyzed population genetic structure in the dispersal limited Jerusalem cricket Stenopelmatus n. sp. “santa monica” in the Santa Monica Mountains and Simi Hills north of Los Angeles, California (CA), to determine the impacts of fragmentation over the past 70 years. MtDNA divergence was greater among urban fragments than within contiguous habitat and was positively correlated with fragment age. MtDNA genetic diversity within fragments increased with fragment size and decreased with fragment age. Genetic divergence across 38 anonymous nuclear Inter-Simple Sequence Repeat (ISSR) loci was influenced by the presence of major highways and highway age, but there was no effect of additional urban fragmentation. ISSR diversity was not correlated with fragment size or age. Differing results between markers may be due to male-biased dispersal, or different effective population sizes, sorting rates, or mutation rates among sampled genes. Results suggest that genetic connectivity among populations has been disrupted by highways and urban development, prior to declines in local population sizes. We emphasize that genetic connectivity can rapidly erode in fragmented landscapes and that flightless arthropods can serve as sensitive indicators for these effects.

Effects of Natural Forest Fragmentation on a Hawaiian Spider Community

Abstract The kipuka system, a network of forest fragments surrounded by lava flows on the island of Hawaii, offers an opportunity to study the natural, long-term fragmentation of a native ecosystem. We examined the impacts of habitat edges upon the community structure of nocturnally active native spiders, primarily in the genus Tetragnatha. We measured plant and spider species distributions across the edges of four small fragments and one large continuously forested area that were surrounded by a lava flow in 1855. Results indicated that an ≈20 m edge ecotone surrounds core forest habitat. Spider community structure changed across the edge, with a decrease in total species richness and diversity at the forest/lava boundary, and a change in the dominant taxon from native Tetragnatha (Tetragnathidae) to native Cyclosa (Araneidae). Severe habitat restrictions were found for some spider species. In addition, nearly all of the spiders captured were endemic species, and the few introduced species were limited to the younger and more open lava flows. Our results suggest that species responses to edges can vary, and that core habitat specialists may decline in fragmented conditions.

Refugial isolation and divergence in the Narrowheaded Gartersnake species complex (Thamnophis rufipunctatus) as revealed by multilocus DNA sequence data

Glacial–interglacial cycles of the Pleistocene are hypothesized as one of the foremost contributors to biological diversification. This is especially true for cold‐adapted montane species, where range shifts have had a pronounced effect on population‐level divergence. Gartersnakes of the Thamnophis rufipunctatus species complex are restricted to cold headwater streams in the highlands of the Sierra Madre Occidental and southwestern USA. We used coalescent and multilocus phylogenetic approaches to test whether genetic diversification of this montane‐restricted species complex is consistent with two prevailing models of range fluctuation for species affected by Pleistocene climate changes. Our concatenated nuDNA and multilocus species analyses recovered evidence for the persistence of multiple lineages that are restricted geographically, despite a mtDNA signature consistent with either more recent connectivity (and introgression) or recent expansion (and incomplete lineage sorting). Divergence times estimated using a relaxed molecular clock and fossil calibrations fall within the Late Pleistocene, and zero gene flow scenarios among current geographically isolated lineages could not be rejected. These results suggest that increased climate shifts in the Late Pleistocene have driven diversification and current range retraction patterns and that the differences between markers reflect the stochasticity of gene lineages (i.e. ancestral polymorphism) rather than gene flow and introgression. These results have important implications for the conservation of T. rufipunctatus (sensu novo), which is restricted to two drainage systems in the southwestern US and has undergone a recent and dramatic decline.

The value of DNA sequence data for studying landscape genetics.

In a recent Opinion article in Molecular Ecology, Wang (2010) emphasizes the fact that current patterns of genetic differentiation among populations reflect processes that have acted over temporal scales ranging from contemporary to ancient. He draws a sharp distinction between the fields of phylogeography (as the study of historical processes) and landscape genetics (which he restricts to very recent processes). Wang characterizes DNA sequence data as being inappropriate for the study of contemporary population processes and further states that studies which only include mitochondrial DNA or chloroplast DNA data cannot be considered part of landscape genetics. In this response, we clarify the generally accepted view that DNA sequence data can be analysed with methods that separate contemporary and historical processes. To illustrate this point, we summarize the study of Vandergast et al. (2007), which Wang mischaracterizes as being confused in terms of temporal scale. Although additional focus should be placed on the important issue of correct data interpretation, we disagree strongly with the implication that contemporary and historic processes cannot be separated in the analyses of DNA sequence data.

Habitat fragmentation in coastal southern California disrupts genetic connectivity in the cactus wren (Campylorhynchus brunneicapillus).

Achieving long‐term persistence of species in urbanized landscapes requires characterizing population genetic structure to understand and manage the effects of anthropogenic disturbance on connectivity. Urbanization over the past century in coastal southern California has caused both precipitous loss of coastal sage scrub habitat and declines in populations of the cactus wren (Campylorhynchus brunneicapillus). Using 22 microsatellite loci, we found that remnant cactus wren aggregations in coastal southern California comprised 20 populations based on strict exact tests for population differentiation, and 12 genetic clusters with hierarchical Bayesian clustering analyses. Genetic structure patterns largely mirrored underlying habitat availability, with cluster and population boundaries coinciding with fragmentation caused primarily by urbanization. Using a habitat model we developed, we detected stronger associations between habitat‐based distances and genetic distances than Euclidean geographic distance. Within populations, we detected a positive association between available local habitat and allelic richness and a negative association with relatedness. Isolation‐by‐distance patterns varied over the study area, which we attribute to temporal differences in anthropogenic landscape development. We also found that genetic bottleneck signals were associated with wildfire frequency. These results indicate that habitat fragmentation and alterations have reduced genetic connectivity and diversity of cactus wren populations in coastal southern California. Management efforts focused on improving connectivity among remaining populations may help to ensure population persistence.

Species Differentiation on a Dynamic Landscape: Shifts in Metapopulation Genetic Structure Using the Chronology of the Hawaiian Archipelago

Species formation during adaptive radiation often occurs in the context of a changing environment. The establishment and arrangement of populations, in space and time, sets up ecological and genetic processes that dictate the rate and pattern of differentiation. Here, we focus on how a dynamic habitat can affect genetic structure, and ultimately, differentiation among populations. We make use of the chronology and geographical history provided by the Hawaiian archipelago to examine the initial stages of population establishment and genetic divergence. We use data from a set of 6 spider lineages that differ in habitat affinities, some preferring low elevation habitats with a longer history of connection, others being more specialized for high elevation and/or wet forest, some with more general habitat affinities. We show that habitat preferences associated with lineages are important in ecological and genetic structuring. Lineages that have more restricted habitat preferences are subject to repeated episodes of isolation and fragmentation as a result of lava flows and vegetation succession. The initial dynamic set up by the landscape translates over time into discrete lineages. Further work is needed to understand how genetic changes interact with a changing set of ecological interactions amongst a shifting mosaic of landscapes to achieve species formation.

Identification of co‐occurring Branchinecta fairy shrimp species from encysted embryos using multiplex polymerase chain reaction

Morphological identification of many fairy shrimp species is difficult because distinguishing characters are restricted to adults. We developed two multiplex polymerase chain reaction assays that differentiate among three Branchinecta fairy shrimp with distributional overlap in southern California vernal pools. Two of the species are federally listed as threatened. Molecular identification of Branchinecta from cysts allows for species surveys to be conducted during the dry season, expanding the timeframe for population assessment and providing a less intrusive method of sampling sensitive vernal pool habitats.