About three quarters of human emerging infectious diseases are caused by zoonotic pathogens, and many of them are spread by vectors such as mosquitoes. Mathematical models nowadays represent very powerful tools to make investigations and predictions for biological dynamical systems, providing helpful insights that can be extremely valuable for several aims. In this thesis, we will focus on a particular mosquito-borne zoonosis, West Nile virus (WNV), a flavivirus of emerging public health relevance in Europe and North America, and its main European vector, Culex pipiens mosquitoes. As the transmission of mosquito-borne diseases is largely driven by the abundance of the vector, to design appropriate control strategies it is crucial to understand the population dynamics of existing vector populations and evaluate how it depends on biotic and environmental factors. This thesis presents some new mathematical models that provide insights on several aspects of mosquito population dynamics by using different statistical and computational approaches, including for instance Linear Models and Markov chain Monte Carlo technique. Specifically, they aim to study the effect of biotic and abiotic factors on Cx. pipiens dynamics by using adult mosquito trapping data, gathered over several years in Northern Italy, to feed theoretical models. Furthermore, the effects of host competition and vector feeding preferences on the dynamics of a vector-borne infection (such as WNV) are investigated through a more theoretical study.

This thesis treats mathematical models for host-parasitoid interactions. It is composed of three parts. In the first part, a class of such models is analyzed theoretically. It focuses on the phenomena of multiple coexistence equilibria of competing parasitoid species. The second part is about a model for determining how a parasitoid release should be timed to optimally control the invasive fruit fly Drosophila suzukii. The third part analyzes an experiment for releasing parasitoids in a greenhouse which is infested by D.suzukii. The models presented are used to discuss how to improve such biological control strategies.

Freshwater lakes are among the most important ecosystems for both human and other biological communities. They account for about 87% of surface freshwater in the planet, thus constituting a major source of drinking water. They also provide a wide range of ecosystem services that go from the sustenance of a rich biodiversity to the regulation of hydrological extremes; from the provision of a means for recreation to the support of local economies, e.g., through tourism and fisheries, just to cite a few. Lakes are now also widely recognised as natural early warning systems, their responses potentially being effective indicators of local, regional and global scale phenomena such as acidification and climate change, respectively. This is because of their high sensitivity to environmental factors of the most diverse nature that can rapidly alter the course of their evolution. Examples of this are the observed abrupt shifts between alternative stable states in shallow lakes, which led them to become the archetype, go-to example in alternative stable state theory. Therefore, attaining a good scientific understanding of the many processes that take place within these ecosystems is fundamental for their adequate management. Among the tools that serve this purpose, ecological models are particularly powerful ones. Since their introduction in the 1960s, the development of mechanistic ecological models has been driven by their wide spectrum of potential applications. Nevertheless, these models often fall into one of the two following categories: overly simplistic representations of isolated processes, with limited potential to explain real-world observations as they fail to see the bigger picture; or overly complex and over-parameterised models that can hardly improve scientific understanding, their results being too difficult to analyse in terms of fundamental processes and controls. Moreover, it is now well known that an increased complexity in the mechanistic description of ecological processes, does not necessarily improve model accuracy, predictive capability or overall simulation results. To the contrary, a simpler representation allows for the inclusion of more links between model components, feedbacks which are usually overlooked in highly-complex models that partially couple a hydro-thermodynamic module to a biogeochemical one. However, ecological processes are now known to have the potential to significantly alter the physical response of aquatic ecosystems to environmental forcing. For example, steadily increasing concentrations of coloured dissolved organic carbon, a process known as brownification (also browning), as well as the intense phytoplankton blooms that characterise lakes undergoing severe nutrient enrichment, a process known as eutrophication, have been shown to have the potential to alter the duration of the stratified period, thermal structure and mixing regime of some lakes. In this thesis, with the aim of addressing the limitation of partially-coupled models to account for such feedbacks, we further develop a process-based model previously reported in scientific literature. Subsequent studies have already built upon this model in the last few years. In Chapter 2, we do so too by integrating hydro-thermodynamics and biogeochemistry in a reduced complexity framework, i.e., customising the model so that each version only includes the fundamental processes that, brought together, sufficiently describe the studied phenomena. Two case studies served the purpose of testing the adaptability and applicability of the developed model under different configurations and requirements. Limnological data for these two studies were measured at high spatial and temporal resolutions by means of an automated profiling system and recorded as part of two large-scale mesocosm experiments conducted in 2015 and 2016 at the IGB LakeLab in Lake Stechlin, Brandenburg, Germany. Meteorological datasets were also made available to us for both periods by the German Federal Environment Agency. The scope of the first experiment, which we describe in Chapter 3, was that of detecting any changes attributable to eutrophication and browning, in the competition for nutrients and light between four different groups of lake primary producers. These four groups are phytoplankton, periphyton, epiphyton and macrophytes. The model version for this study, therefore, includes equations for all four groups. By tailoring the model to these very specific needs with relative ease, we demonstrate its versatility and hint at its potential. The second experiment, described in Chapter 4, sought to shed light on the largely unknown effects of an increase in the diffuse luminance of the night sky that is due to artificial light at night (artificial skyglow) on lake metabolic rates, i.e., gross primary productivity, ecosystem respiration and net ecosystem productivity (the difference between the first two). For this purpose, an empirical equation for dissolved oxygen concentration was included, the parameters of which were estimated by means of a Markov Chain Monte Carlo sampling method within a Bayesian statistical framework, showing the compatibility, with these statistical methods, of our otherwise fully deterministic model. In Chapter 5, we present a theoretical study on the ecological controls of light and thermal patterns in lake ecosystems. A series of simulations were performed to determine in which cases ecological processes such as eutrophication and brownification may have an observable effect on the physical response of lakes to environmental forcing, which we assessed along a latitudinal gradient. Results show that, in general, across all examined latitudes, and consistent with previous studies, accounting for phytoplankton biomass results in higher surface temperatures during the warm-up phase, slightly lower water temperatures during the cool-down phase, and a shallower thermocline throughout the entire stratified period. This effect is relatively more important in eutrophic lakes where intense blooms are likely. This importance, however, decreases as lakes get browner. Finally, in line with the overall scope of the SMART EMJD, in Chapter 6 we illustrate the case of Ypacaraí Lake, the most important lake in landlocked Paraguay, hoping to provide an example of how interdisciplinary research and international intersectoral collaboration can help bridge the gap between science and management of freshwater ecosystems. This lake presents very special hydro-ecological conditions, such as very high turbidity that can impair phytoplankton growth despite its nutrient-based trophic state indices having consistently fallen within the hyper-eutrophic range in recent years. A strong interest in its complex functioning, through modelling, was taken early on. This led to a collaborative research line being established among several public and private institutions in Italy, Germany and Paraguay. Results so far include: • three concluded UniTN Master theses in Environmental Engineering, partly developed in Paraguay, the first two in collaboration with the “Nuestra Señora de la Asunción” Catholic University (UCNSA) and the third one with the National University of Asunción (UNA); • a collaborative UCNSA-UniTN research proposal submitted for consideration to receive funding through the PROCIENCIA Programme of the National Council of Science and Technology of Paraguay (CONACYT); and • the first multidisciplinary review that has ever been published about the case of Ypacaraí Lake, which highlights the importance of such a collaborative and integrative approach to further advance scientific knowledge and effectively manage this ecosystem.

Large carnivores are among the most challenging species to conserve in our modern and crowded world. Having large spatial requirements and living in low density, they generally require wide and relatively undisturbed areas. In Europe, one of the most anthropized areas of the planet, these needs must be fulfilled in a complex human-dominated landscape. The reintroduced brown bear population living in the Central Alps represents one of the most emblematic examples of a constrained carnivore: despite a steady population increase in the first few years after reintroduction, the population did not substantially expand its range, nor has the Alpine-Dinaric metapopulation been reestablished as envisioned. Although humans have lived in the Alps for centuries, little is known about their impact on the bear population. In other environments humans are known to function as a “super-predator” by changing habitats, competing for space, consuming resources, and harvesting, which alters the ecological niche of animals, especially large carnivores. This dissertation aims to evaluate this phenomenon by assessing the effects of human disturbance on brown bears in the Alps. Anthropogenic disturbance is generally assessed by structural proxies, such as infrastructure and land use, which overlook the impact of human presence. In the first Chapter, we developed the Cumulative Outdoor activity Index (COI) to derive anthropogenic disturbance using crowdsourced data by Strava and validated it with ground truth observations derived from a local camera trapping survey. The intensity of COI provided an effective measure of functional anthropogenic disturbance, and it outperformed all commonly-used proxies of structural disturbance in predicting bear habitat use. When displacement is not an option because of habitat limitations and social mechanisms, bear mobility may clash with human activity. During the moments of lowest mobility, such as resting periods, animals have decreased ability to cope with risky situations, and therefore the selection of suitable resting areas is crucial for the long-term survival of individuals. In the second Chapter, we measured multi-scale response to risk perception (i.e., COI) and resource proximity using bedding sites by GPS radio-collared adult brown bears in the Alps. To map resources across the study area, we developed a GIS-database combining spatial and non-spatial ecological information to map fruit availability. We observed that bears apply a security-food trade-off strategy, avoiding functional anthropogenic disturbance while in proximity to resources. In the third Chapter, we explicitly tested the effect of an abrupt interruption of human mobility during COVID-19 lockdown on bears’ use of ecological corridors. Using bear occurrences reported to local authorities during the recent COVID-19 outbreak, we observed that bears used human-dominated areas more frequently, approached more intensively hot spots for road crossing network, and used areas further from the population core areas more often than previous years, suggesting that connectivity increased with reduced human mobility. In a comparatively human-free system, for the fourth Chapter we used longitudinal morphometric data to analyze drivers of changes in body mass as part of an international collaboration with biologists studying the grizzly bear in the Greater Yellowstone Ecosystem. Specifically, we analyzed changes in lean body mass and fat percentage during years of major ecosystem perturbations. We observed that individual lean body mass during the last two decades was primarily associated with population density, but not body fat percentage, showing density-dependent factors. Our combined findings (Chapters 1-3) showed that brown bears have to adapt their space use, movement, and resource proximity as a result of functional anthropogenic disturbance. In Chapter 4 we explored one effect of unconstrained bear space use on individuals, as manifested through density-dependent effects on body size. In the Alps, however, we found multiple instances of the human-super predator outcompeting bears so as to make density-dependent effects likely less significant as compared to human-caused mortality. These effects could occur in a variety of socio-ecological contexts across Europe, jeopardizing the long-term establishment of both newly reintroduced bear populations, as well as spatially limiting those naturally present in the environment. In response to disturbance, bears have had to reduce their ecological niche in human-dominated landscapes. Allowing humans and bears to coexist in the same landscape is a challenging task, but it is essential for the long-term survival of this newly reintroduced population that are otherwise at risk of extinction.