Landscape Population And Evolutionary Genomics Of Several Closely Related Species Of Mallard-Like Ducks

Divergence and speciation proceed through three major evolutionary forces (i.e., selection, genetic drift, and gene flow) that are often spatially and temporally heterogeneous across the landscape. Moreover, these forces can have differing but subtle effects within the genomes of diverging taxa, and therefore, disentangling the effects of these evolutionary mechanisms throughout the speciation process can be challenging. Here, I use a recent species radiation, the mallard complex, to investigate how strong, yet varied, evolutionary pressures influence the speciation process. The mallard complex consists of 14 mallard-like waterfowl species around the world that have some of the highest rates of hybridization among avian groups. In general, their wide distribution and lack of pre-zygotic reproductive barriers makes this group an excellent study system for investigating the genomic and adaptive effects of selection and gene flow during divergence. Additionally, anthropogenically-induced secondary contact between the mallard (Anas platyrhynchos) – one of the most ubiquitous species of waterfowl in the world – and the other mallard-like ducks provides unique natural experiments for testing the effects of gene flow from these non-native and captive-bred individuals into native congeners. Finally, improved sequencing and modelling technologies have increased our ability to sequence thousands of genomic markers from range-wide samples of non-model organisms.In Chapter 1, I investigated the population genetic structure of and hybridization rate between Mexican ducks (A. diazi) and mallards in southwest North America using double-digest Restriction-site Associated DNA sequencing (ddRAD-seq). I sequenced 3,189 ddRAD-seq autosomal and Z-chromosome loci, as well as a portion of the mtDNA control region across 387 samples of Mexican ducks, mallards, and putative hybrids. First, I confirmed that Mexican ducks and mallards are in fact genetically distinguishable, and that Mexican ducks exhibit strong population structure that suggests a history of sequential founder events from north to south. Moreover, Mexican ducks were characterized by three unique genetic clusters that show a southward cline in both mallard-like phenotypic traits and mallard genetic ancestry. Importantly, these results contradict past notions of pervasive introgressive hybridization from mallards into Mexican ducks, as I recovered only a few late-stage hybrids that were limited to the northern part of the Mexican duck’s range. I provide additional evidence that previous estimates of hybridization were likely biased by phenotypic variation, that I conclude was due to shared ancestry and not hybridization. Finally, I discuss how both genomic and morphological comparisons are necessary for investigating the evolution of complex traits in recent species radiations and consider how such insights can help future conservation efforts.In Chapter 2, I use range-wide samples sequenced in Chapter 1 to model evolutionary history, demographic history, and genotype-environment and genotype-phenotype associations in Mexican ducks. First, evolutionary models and genotype-environment associations (GEA) showed that Mexican ducks diverged from mallards during a glacial period in a local climate refugia around southwestern North America. Further reconstructing the demographic histories revealed that Mexican ducks diverged from mallards ~300,000 years ago during a glacial maximum and have since had cyclical population growth reflecting changing environmental conditions related to glacial and inter-glacial periods. Additionally, evolutionary and demographic models support recurring bouts of gene flow during secondary contact events, which artificially inflated Mexican duck effective population size during periods of contact. Importantly, a Mexican duck x mallard combined model of GEA showed that environmental selective pressures have played a key role in driving divergence, as there is significant genotypic turnover between species across environmental gradients. Finally, I used genotype-phenotype association testing and contend that sexual selection has acted as a co-evolutionary process, facilitating the development of reproductive barriers that initially arose due to strong ecological partitioning. Broadly, this chapter reveals that genomic and phenotypic patterns observed during the earliest stages of divergence are complex in the way that they contribute to the evolutionary trajectory of a lineage.In Chapter 3, I investigate the consequences of anthropogenic hybridization from introduced mallards in native New Zealand grey ducks (A. superciliosa superciliosa). Domestic mallards were introduced to New Zealand during the mid-1800s and have since become the dominant species of waterfowl throughout New Zealand. Alternatively, native grey ducks have been steadily declining and are at risk of complete lineage fusion due to extensive hybridization with mallards. First, I show that pockets of pure grey ducks have persisted in areas of more undisturbed habitat along the western and northwestern coasts of the South Island. In contrast, introduced mallards have experienced widespread introgression from native grey ducks, and now constitute a hybrid swarm. Additionally, I used GEA modelling to demonstrate that such extensive introgression into mallards has likely facilitated their rapid establishment and expansion throughout New Zealand. In fact, these strong selective pressures have resulted in a genetically unique New Zealand mallard that no longer resembles its original domestic stock and does not overlap the adaptive space of wild North American mallards. I showed that the New Zealand mallard hybrid swarm encompasses a more variable genetic niche space as compared to either parental species on their own, suggesting that these hybrids may be better adapted. I then measured the vulnerability of current GEAs to future climate change scenarios and found that grey ducks are at risk of losing critical adaptive habitat along the western coasts of the South Island. Finally, I discussed these findings in the larger context of conservation biology, and what should be prioritized as climate change and land use changes continue to increase cases of anthropogenic hybridization. In general, I argue that conservation efforts should be focused on preserving core habitats of native species, as this strategy will likely have the added benefit of strengthening reproductive barriers.While Chapter 3 emphasized the genetic and adaptive consequences of interspecific hybridization, Chapter 4 focusses on the effects of hybridization between wild and domestic conspecifics. Specifically, mallards were largely absent of eastern North America until the early 1900’s when government and private organizations began supplemental stocking programs. I report that extensive hybridization continues today, and a 4-fold decrease in the prevalence of pure wild mallards as compared to estimates from a decade ago. Next, I used GEA modelling to show that wild and game-farm hybrid mallards have differing adaptive breeding ranges, with game-farm hybrids showing significant genotypic turnover across the southern border of Canada. This is likely acting as a barrier to dispersal, as game-farm hybrid genotypes were found to be most strongly influenced by winter weather conditions. Moreover, when projecting GEAs across future climate conditions, I found that while wild mallards are unlikely to be impacted in their core breeding range of the central Canadian prairies, game-farm hybrids are poised to expand as temperatures in the region increase. This suggests that unless strong efforts are made to limit wild x game-farm mallard interactions, hybridization and introgression of maladaptive domestic traits into wild populations may increase conservation risk in the future.Overall, I provide insight into broad spatiotemporal responses to changing selective pressures acro...

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.