Landscape genetics, phylogeography, and demographic history of a pollinator longhorn beetle (Typocerus v. velutinus)

Hossam Eldien Mohammed Abdel Moniem, Purdue University

Abstract

One of the central problems in contemporary ecology and conservation biology is the drastic change of landscapes induced by anthropogenic activities, resulting in habitat loss and fragmentation. For many wild living species, local extinctions of fragmented populations are common and re-colonization is critical for regional survival. Thus, habitat fragmentation in the landscape is a major threat to biodiversity, of which insects are a major proportion. Understanding the link between patterns, processes and population genetic continuity in the landscape is crucial for conserving genetic diversity within species. This is important for species persistence, for ecosystem functioning, and for future evolution. Herein, I use a newly introduced landscape gradient paradigm with surface metrology metrics, phylogeography, and landscape genetics to evaluate the influence of contemporary events (e.g. habitat fragmentation in the landscape) and pre-historic events (e.g. Quaternary glaciation) on the demography and population genetic structure of a pollinator longhorn beetle [Typocerus v. velutinus (Olivier)] in Indiana, USA and Canada. Landsat 7 ETM+ imagery products provide researchers in many fields with a large amount of remotely sensed data that serves many applications. However, a malfunction of the scan line corrector (SLC) onboard Landsat 7 causes substantial data gaps and data are available only as is, in the SLC-off mode. These data gaps may form an obstacle in using Landsat 7 ETM+ in many research disciplines. Several methods have been proposed to fix data gaps in Landsat 7 ETM+ imagery. These methods yield reliable results, but require sophisticated analyses and intensive computations and are still accompanied by some caveats. In the second chapter of this dissertation I demonstrate a spatial replacement method that is based on a simple neighborhood interpolation (SNI) approach. The results suggest that SNI provides an easily applicable, relatively quick and potentially reliable correction for the missing data patterns in Landsat 7 ETM+ data. I demonstrate the efficiency of the technique for two color bands across Indiana, USA. I tested the corrected imagery in calculating the normalized difference vegetation index (NDVI). Measuring habitat connectivity in complex landscapes is a major focus of landscape ecology and conservation research. Most studies use a binary landscape or patch mosaic model for describing spatial heterogeneity and understanding pattern-process relationships. While the value of a landscape gradient approach is recognized, applications of the newly proposed three-dimensional surface metrics remain extremely under-used. In the third chapter, I created a surface habitat quality from several GIS layers and applied surface metrics to measure connectivity between 67 locations in Indiana, USA that were surveyed for one group of ecosystem service providers, flower longhorn beetles (Cerambycidae: Lepturinae). The results demonstrated great potential of surface metrics of connectivity to explain the differences of lepturine assemblages among the 2211 studied landscapes. Surface kurtosis and its interaction with geographic distance were among the most important metrics. This approach provided unique information about the landscape through four configuration metrics. There were some uniform trends of the responses of many species to some of surface metrics, however some species responded differently to other metrics. I suggest that surface metrics of connectivity applied to a habitat surface map created with insight into species requirements is a valuable approach for understanding the spatial dynamics of species, guilds, and ecosystem services. Historical geological processes have shaped the contemporary distribution of genetic variation in many species. However, there have been few empirical appraisals of cerambycid phylogeography despite of their economic importance and the fact that many geological processes (e.g., glaciations) should have had pronounced impacts on these insects as well as other taxa. In chapter four, I aimed to quantify phylogeographic effects on the contemporary gene pool of Typocerus v. velutinus. The beetle was collected from sites that were glaciated and unglaciated during the Pleistocene to determine genetic structure within and among populations from the US and Canada, to elucidate phylogenetic relationships among demes, and to determine divergence times between populations. A total of 451 beetles were sampled from 14 sites and sequenced at a mitochondrial DNA (mtDNA) gene. Maximum likelihood and Bayesian approaches were applied to analyze the mtDNA genealogy and to reconstruct phylogenetic trees whereas Bayesian skyline analyses were used to estimate divergence time. A total of sixteen haplotypes revealed weak geographical population structuring among most populations, but statistical tests identified significant differences between the Canadian and US populations. As a result of post-glacial recolonization, the US populations appear to have experienced demographic expansion while the Canadian population was influenced by a bottleneck. The results suggest that Canadian population diverged from more southern populations around the time of last glacial maximum (~17,500 ybp). Understanding the underling patterns and processes in the landscape that are affecting the population genetic structure and population connectivity is a major discipline in landscape genetics research. A vast number of these studies have implemented categorical approaches in analyzing both landscape and genetic data. In chapter five, I adopted a landscape gradient model and used the surface metrics of connectivity to model the genetic continuity between populations of the beetle (Typocerus v. velutinus) that was collected at 17 sites across a fragmentation gradient from Indiana, USA. I tested the hypothesis that landscape structure and habitat connectivity facilitate beetle movement and thus gene flow between the beetle populations against a null model of isolation by distance (IBD). I used next-generation sequencing and developed 10 polymorphic microsatellite loci and genotyped the population. Genetic dissimilarities between sites were calculated using RST and the population genetic structure was assessed using both non-spatial and spatial explicit Bayesian techniques. The connectivity in 137 landscapes was measured using surface topology metrics. The results indicated that panmixia was not evident with the beetle population. The source of genetic variation was mainly within rather than among populations. The surface metrics were found to significantly explain the variance in genetic dissimilarities between beetle populations 30 times better than IBD. I concluded that surface metrics of connectivity is a powerful extension in landscape genetics tools and need more attention especially to understand the configuration metrics. This approach might yield insightful applications in conservation management.

Degree

Ph.D.

Advisors

Holland, Purdue University.

Subject Area

Ecology|Entomology|Genetics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS