Consequences of within population variation have recently attracted an increased interest in evolutionary ecology research. Theoretical models suggest important population-level consequences, but many of these predictions still remain to be tested. These issues are important for a deepened understanding of population performances and persistence, especially in a world characterized by rapid fragmentation of natural habitats and other environmental changes. I review theoretical models of consequences from intra population genetic and phenotypic variation. I find that more variable populations are predicted to be characterized by broader resource use, reduced intraspecific competition, reduced vulnerability to environmental changes, more stable population dynamics, higher invasive potential, enhanced colonization and establishment success, larger distribution ranges, higher evolvability, higher productivity, faster population growth rate, decreased extinction risk, and higher speciation rate, compared with less variable populations. To test some of these predictions I performed experiments and compared how different degree of colour polymorphism influences predation risk and establishment success in small groups. My comparisons of predation risk in mono- and polymorphic artificial prey populations showed that the risk of being eaten by birds does not only depend on the coloration of the individual prey item itself, but also on the coloration of the other members of the group. Two experiments on establishment success in small founder groups of Tetrix subulata pygmy grasshoppers with different degree of colour morph diversity show that establishment success increases with higher degree of diversity, both under controlled conditions in outdoor enclosures and in the wild. These findings may be important for re-stocking of declining populations or re-introductions of locally extinct populations in conservation biology projects. I report on remarkably rapid evolutionary shifts in colour morph frequencies in response to the changed environmental conditions in replicated natural populations of pygmy grasshoppers in fire ravaged areas. This finding 1 illustrates the high adaptive potential in a polymorphic species, and indicates the importance of preserved within-species diversity for evolutionary rescue. Finally, I review if theoretical predictions are supported by other published empirical tests and find strong support for the predictions that more variable groups benefit from reduced vulnerability to environmental changes, reduced population fluctuations and extinction risk, larger distribution ranges, and higher colonization or establishment success. In conclusion, my thesis illustrates how within-population variation influences ecological and evolutionary performances of populations both in the short and long term. As such, it emphasizes the need for conservation of biodiversity also within populations.

Biomarker testing, wherein the concentration of a biochemical marker is measured to predict the presence or absence of a certain binary characteristic (e.g., a disease) in a subject, is an essential component of public health screening. For many diseases, the concentration of disease-related biomarkers may exhibit a wide range, particularly among the disease positive subjects, in part due to variations caused by external and/or subject-specific factors. Further, a subject's actual biomarker concentration is not directly observable by the decision maker (e.g., the tester), who has access only to the test's measurement of the biomarker concentration, which can be noisy. In this setting, the decision maker needs to determine a classification scheme in order to classify each subject as test negative or test positive. However, the inherent variability in biomarker concentrations and the noisy test measurements can increase the likelihood of subject misclassification. We develop an optimal data-driven framework, which integrates optimization and data analytics methodologies, for subject classification in disease screening, with the aim of minimizing classification errors. In particular, our framework utilizes data analytics methodologies to estimate the posterior disease risk of each subject, based on both subject-specific and external factors, coupled with robust optimization methodologies to derive an optimal robust subject classification scheme, under uncertainty on actual biomarker concentrations. We establish various key structural properties of optimal classification schemes, show that they are easily implementable, and develop key insights and principles for classification schemes in disease screening. As one application of our framework, we study newborn screening for cystic fibrosis in the United States. Cystic fibrosis is one of the most common genetic diseases in the United States. Early diagnosis of cystic fibrosis can substantially improve health outcomes, while a delayed diagnosis can result in severe symptoms of the disease, including fatality. We demonstrate our framework on a five-year newborn screening data set from the North Carolina State Laboratory of Public Health. Our study underscores the value of optimization-based approaches to subject classification, and show that substantial reductions in classification error can be achieved through the use of the proposed framework over current practices. / Doctor of Philosophy / A biomarker is a measurable characteristic that is used as an indicator of a biological state or condition, such as a disease or disorder. Biomarker testing, where a biochemical marker is used to predict the presence or absence of a disease in a subject, is an essential tool in public health screening. For many diseases, related biomarkers may have a wide range of concentration among subjects, particularly among the disease positive subjects. Furthermore, biomarker levels may fluctuate based on external factors (e.g., temperature, humidity) or subject-specific characteristics (e.g., weight, race, gender). These sources of variability can increase the likelihood of subject misclassification based on a biomarker test. We develop an optimal data-driven framework, which integrates optimization and data analytics methodologies, for subject classification in disease screening, with the aim of minimizing classification errors. We establish various key structural properties of optimal classification schemes, show that they are easily implementable, and develop key insights and principles for classification schemes in disease screening. As one application of our framework, we study newborn screening for cystic fibrosis in the United States. Cystic fibrosis is one of the most common genetic diseases in the United States. Early diagnosis of cystic fibrosis can substantially improve health outcomes, while a delayed diagnosis can result in severe symptoms of the disease, including fatality. As a result, newborn screening for cystic fibrosis is conducted throughout the United States. We demonstrate our framework on a five-year newborn screening data set from the North Carolina State Laboratory of Public Health. Our study underscores the value of optimization-based approaches to subject classification, and show that substantial reductions in classification error can be achieved through the use of the proposed framework over current practices.

Lateralized behaviour is the most conspicuous manifestation of hemispheric specialization of the brain and has been reported in a variety of taxa. Only a few studies have so far assessed lateralized behaviours in horses. Therefore, I observed ten domestic horses for 16 weeks for an array of spontaneously occurring motor behaviours as well as stimulus-induced behavioural responses to determine if they display side preferences at the individual or population level and to assess possible correlations between lateralized behaviours. Significant side preferences were found for certain behaviours at the individual level, ranging from standing and flexing, to auditory stimuli, and olfactory stimuli. All horses showed task-dependent changes in their side preferences and no significant side preferences were found at the population level for any behaviours. Similarly, no significant correlations were found between behaviours. Taken together, the results of the present study suggest that horses, like all other species studied so far except humans and some great apes, only display lateralized behaviour at the individual, but not at the population level.

Rerouting highways to accommodate construction of the Jordanelle Reservoir in northeastern Utah caused a dramatic increase in vehicle collisions with mule deer (Odocoileus hemionus). I evaluated the effectiveness of a new system of highway crosswalk structures installed to reduce deer losses and preserve seasonal migrations. In addition, I constructed computer simulation models to investigate how highway mortality has impacted the Jordanelle deer population. The crosswalk system restricted deer crossings to specific, well-marked areas along highways where motorists could anticipate them. Subsequent to installation, mortality declined 42.3% and 36.8% along a four-lane and two-lane highway, respectively. I was unable to statistically demonstrate that observed mortality reductions were a direct result of the crosswalk system. The potential applicability of the structures, however, should not be dismissed. Reduced deer use of the highway right-of-way (ROW), the apparent maintenance of migratory behavior, and observations of animals crossing within crosswalk boundaries indicate the system warrants further testing. Lack of motorist response to crosswalk warning signs, the tendency for foraging deer to wander outside crosswalk boundaries, and the ineffectiveness of ROW escape gates contributed to most treatment area mortalities. I offer design modifications that address these shortcomings. Four years of field data revealed that highway mortality at Jordanelle was inversely density-dependent, removed between 5.6% and 17.4% of the population each year, and disproportionately impacted bucks. I incorporated this information into 3 competing simulation models in which highway losses operated in a strictly additive, partially compensatory, or strictly compensatory manner. The partial compensation model most closely tracked observed population dynamics, suggesting that highway losses were not completely offset by reductions in other mortality sources. Highway mortality apparently worsened a population crash initiated by severe winter conditions, and may be slowing the recovery. The disproportionate loss of bucks along roads altered sex ratios of simulated populations. Mitigative efforts should target road-kill reductions >60% to avoid population declines predicted by the partial compensation model. Annual variation in demographic parameters offset the impacts of highway mortality at high population levels. At low population levels, however, highway mortality was severe enough to drive declining population trends.