The genetic consequences of demography and disturbance in small mammal populations

Abstract

Social, demographic and ecological processes shape patterns of genetic diversity. These patterns can therefore reveal insights into the biology of species and the response of populations to disturbance. During my PhD, I used a combination of computer simulations, molecular techniques and field-based experiments to explore how biological and ecological processes shape populations and their underlying genetic diversity.

Dispersal and mating systems have long been known to shape population-level patterns of genetic structure. However, few studies focus on how these processes shape spatial genetic patterns within populations. Using the agile antechinus (Antechinus agilis) as a model, I carried out computer simulations to investigate how dispersal and mating behaviour shape fine-scale genetic structure (over the scale of metres) across autosomal, mitochondrial and Y chromosome markers. While dispersal was the major driver of fine-scale genetic structure, variation in mating behaviour also created differences in the level of structure detected at uniparentally inherited markers. Thus, comparing sex-specific patterns across markers with differing modes of inheritance can help elucidate demographic processes occurring within populations.

In addition to microsatellite, mitochondrial and Y chromosome markers, high throughput sequencing data is becoming increasingly accessible for ecological research. However, decisions about marker choice, bioinformatic pipelines and filtering can be overwhelming for experts and non-experts alike. Through my empirical research focusing on a native Australian rodent, the pale field-rat (Rattus tunneyi), in the Kimberley region of Western Australia, I explored how marker choice and bioinformatic methods influence biological conclusions. Genetic analyses revealed low levels of genetic structure across this disturbance-prone landscape. While population-level estimates of genetic structure were fairly robust, measures of heterozygosity and diversity differed among marker types and filtering criteria. This demonstrates the importance of understanding how methodological decisions can impact biological inference from genetic data.

The pale field-rat is one of many small mammals declining across northern Australia. This is due, in part, to the interaction between altered fire regimes and other key threats. To better understand this decline, I investigated habitat preferences, fire response and post-fire population recovery using a replicated fire experiment and capture-mark-recapture study. Mixed modelling showed that capture rate was negatively correlated with the extent of experimental fire, and that pale field-rat habitat preferences did not change in the post-fire landscape. However, all populations completely recovered one year after fire.

The fire experiment suggested that spatial recovery processes differed according to the size and spatial pattern of fires. To test these different recovery hypotheses, I used parentage and genetic spatial autocorrelation analysis to explore patterns of relatedness before and after fire. This indicated that post-fire recovery after patchy fires was driven by in situ survivors from within unburnt refuges, compared to recolonisation after thorough fires. Furthermore, changes in female dispersal strategies appeared to be driving these different recovery patterns. These results suggest that fire management should aim to maximise the patchiness of burns and limit their extent in order to facilitate recovery of small mammals in this system.

My thesis demonstrates that the combined use of computer simulations, direct field research and genetic analyses can reveal novel insights into the demographic processes occurring within populations and the response of populations to disturbance. I discuss how these insights add to our understanding of mammal declines in northern Australia and can be used to inform fire management. Show more